Customers` responses to service failures. Empirical studies on

Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
•j
LTK'í ERSITAT D' ALAC 'iNT
f-
1
Universitat d'Alacant
Universidad de Alicante
•
c ED 5 P
™1
i
O -1
i
Z ! ~ . L. - - '
I"!
i'IM-
L
TESIS DOCTORAL CON MENCIÓN DE DOCTOR EUROPEO
Alicante, Julio de 2005
Customers' responses to service failures
Empirical studies on prívate, voice and third-party
responses
Ana Belén Casado Díaz
Departamento de Economía Financiera, Contabilidad y Marketing
Facultad de Ciencias Económicas y Empresariales
Directores:
Francisco J. Mas Ruiz
Catedrático de Comercialización e Investigación de Mercados de la
Universidad de Alicante
J. D.P. Kasper
Professor of Services and Retail Management ofMaastricht University
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
!
,¿
:
•
a
f
¡:
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
A mis padres
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Acknowledgments
This is one of the most exciting moments in my life. I have dreamt many
times the writing this preface (though dreaming of writing it in Spanish) but,
after many years of hard work, here we are. Now, it is time to give recognition
to and to thank all the people that have made this work possible.
With respect to the development of this research, fírst of all, I would like to
express my sincerest gratitude to my co-supervisor Francisco José Mas Ruiz. He
has been crucial in all the stages of this work, giving me not only wise advice
but also his own time and hard work during all these years and, most
importantly, personal support and encouragement in the moments I needed. I
am deeply grateful to him.
Second, I would like to thank Peter S.H. Leeflang for the interest he has
always shown in the completion of this thesis. He introduced me to Hans
Kasper who agreed to be my co-supervisor without knowing anything about me.
Since then, Hans has patiently reviewed my work, giving me wise advice that
has improved this dissertation. I would especially like to thank him for having
made possible my stay at the Department of Marketing and Marketing Research
of Maastricht University. During the three months that I spent there I met
amazing people who shared their valuable knowledge and experience with me
and made my stay easier, and fun. Very special thanks go to Piet Pauwels, Vera
Blazevic, Lisa Deutskens and Sonja Wendel.
Third, I have to mention also my colleague Ricardo Sellers Rubio, co-author
of the study presented in the sixth chapter of this thesis. It has been a pleasure to
work with him and his contribution has been especially valuable with respect to
the implementation of the event study technique. I would also like to thank
Carlos Forner Rodríguez for his help with the bootstrap estimation in this
chapter.
Finally, I would like to thank the Department of Financial Economics,
Accounting and Marketing of the University of Alicante where I have found a
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
great place to work in. I would like to mention especially Juan Carlos Gómez
Sala, Francisco José Mas Ruiz, Joaquín Marhuenda Fructuoso, and Ángel León
Valle, for establishing the appropriate research orientation in the department
that has brought the physical and economic resources necessary to my
formation as researcher. I would like to thank also Juan I. España Valor and
Cristina Girones Ansuátegui for their valuable work, kindness, friendship, and
help; you can always count on them.
Regarding the personal support for developing this thesis, not only have I
found a great place to work; I have also found amazing people to work with at
the Department. I am probably the luckiest worker of this world, loving going to
work just to meet my colleagues: Juan Luis Nicolau (who has always
encouraged me with kind words), Mónica Espinosa (who introduced me in the
select 'morning-coffee group'), Ricardo Sellers, and so on. But especially
important for me are Felipe Ruiz, Paco Poveda, Carlos Forner, and "my" María
Jesús Pastor. They are my particular "sanedrín", my wise and extremely patient
friends. Above all, I want to thank my roommate Felipe for being there every
time I needed him, with his willingness to help/hear me and to make my life
easier. Having met them is one of the best things that has happened to me; no
doubt, they have made me a better person. I wish everyone friends like them!
From now on, I will continué in Spanish.
Me gustaría agradecer a toda mi familia su incondicional apoyo todos estos
años, la paciencia que todos han tenido conmigo, lo fácil que me han hecho la
vida, en resumen, el amor que siento que me tienen y que yo les tengo.
Especialmente quiero mencionar a mis padres, dos personas increíbles con las
que he tenido la suerte de crecer y a las que les debo todas las cosas buenas
que pueda haber en mí. A ellos les dedico esta tesis. A mis hermanos, que
siempre están ahí, apoyándome y escuchándome, son los mejores. A mis
sobrinos, a los que no he podido dedicar todo el tiempo que me habría gustado
en su primer año de vida pero que espero compensar a partir de ahora. A mis
cuñados y a mis suegros, que siempre me han tratado como a una hija más. A
mis tíos/as y primos/as, por el cariño que me han mostrado siempre.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
También quiero mencionar aquí a todos mis amigos por los ánimos que me
han dado durante todo este tiempo, especialmente Dominique, Javi, Natalia,
Jaime, Cristina, Paloma y el resto de la troupe valenciana (supongo que
respirarán aliviados cuando lean esto).
Para terminar, quiero agradecer a Pepe su apoyo, su paciencia, sus
ánimos, las veces que me ha hecho reír, las veces que me ha recordado las
cosas que de verdad importan en esta vida, su aguante infinito durante todos
estos años (sobre todo con las veces que lo he dejado solo), su sonrisa, sus
abrazos, ... supongo que todo.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Julio de 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Table of contents
Chapter 1. Introduction
1.1. SERVICE FAILURES: THE STARTING POINT
1.2 RESEARCH OBJECTIVES OF THE DISSERTATION
1.2.1 Objectives Chapter 4
1.2.2 Objectives Chapter 5
1.2.3 Objectives Chapter 6
1.3 DISSERTATION OUTLINE
Chapter 2. Service failures: theoretical considerations
11
11
14
16
17
17
18
21
2.1 INTRODUCTION
2. /. 1 The basic characteristics ofservices
2.1.2 Thefocus on customer satisfaction
21
22
24
2.2 SERVICE FAILURE ENCOUNTERS: DEFINITION AND NATURE
27
2.3 CUSTOMER (DIS)SATISFACTION IN SERVICE FAILURE ENCOUNTERS: DEFINITION
AND NATURE
31
2.3.1 Customer (dis)satisfaction as a response
32
2.3.2 Thefocus ofthe customer (dis)satisfaction response
33
2.3.3 The timing ofthe customer (dis)satisfaction response
34
2.4 ANTECEDENTS/DETERMINANTS OF CUSTOMER (DIS)SATISFACTION IN SERVICE
FAILURE ENCOUNTERS
35
2.4.1 Service features
36
2.4.2 Causal attributions
39
2.4.3 Customer emotions
40
2.4.4 Perceptions ofjustice
44
2.5 OUTCOMES OF CUSTOMER (DIS)SATISFACTION IN SERVICE FAILURE ENCOUNTERS
45
2.5.1 Prívate responses (Chapter 4 context)
47
2.5.2 Voice responses (Chapter 5 context)
48
2.5.3 Third-party responses (Chapter 6 context)
51
Chapter 3. Summary and description ofthe empirical applications
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
53
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
vi • Table of contente
Chapter 4. The consumer's reaction to delays in service
4.1 INTRODUCTION
4.2 THE MODELING OF THE SERVICE DELAY EVALUATIONS AND THE HYPOTHESES
56
56
58
4.2.1 Attribution theory: attribution of control
60
4.2.2 Attribution theory: attribution ofstability
61
4.2.3 Perceivedwaiting time and importance qfsuccessful service performance 62
4.2.4 Anger
65
4.2.5 Satisfaction with service
67
4.3 METHODOLOGY
68
4.3.1 Sample and data collection
68
4.3.2 Development ofmeasures
69
4.4 RESULTS AND DISCUSSION
4.4.1 Sample characteristics
4.4.2 Testing the proposed model
4.5 CONCLUSIONS
APPENDIX 4.1 MEASURES EMPLOYED IN THE STUDY
APPENDIX 4.2 DELAY CAUSES AND ATTRIBUTIONS
70
71
73
79
80
81
Chapter 5. Anger and distributive justice in a double deviation scenario:
explaining (dis)satisfaction in service failure and failed recovery contexts
83
5.1 INTRODUCTION
5.2 CONCEPTUAL FRAMEWORK AND RESEARCH HYPOTHESES
83
86
5.2.1 Service failure and failed recovery: a double deviation scenario
88
5.2.2 Determinants of (dis)satisfaction in a double deviation context
89
5.2.2.1 Direct effects and indirect effects, through cognitive and emotional antecedente,
of service failure-related variables
90
5.2.2.2 Indirect effects of service recovery-related variables through cognitive and
emotional antecedente
95
5.2.2.3 Direct and indirect effects of cognitive and emotional antecedente
99
5.3 METHODOLOGY
104
5.3.1 Sample and data collection
104
5.3.2 Development ofmeasures
106
5.3.3 Data analysis
109
5.3.3.1. General data analysis procedure
109
5.3.3.2 Analysis of the measurement models
110
5.4 RESULTS
113
5.4.1 Direct effects and indirect effects, through cognitive and emotional
antecedents, of service failure-related variables
116
5.4.2 Indirect effects, through cognitive and emotional antecedents, of service
recovery-related variables
118
5.4.3 Direct and indirect effects of cognitive and emotional antecedents
119
5.5 DISCUSSION
5.6 CONCLUSIONS
APPENDIX 5.1 MEASURES EMPLOYED IN THE STUDY
APPENDIX 5.2 FORMULATION OF MEASUREMENT AND STRUCTURAL MODELS
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
120
125
127
128
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Table of contents • vii
Chapter 6. Third-party complaints and banking market valué: the moderating
effects of quality corporate image and market concentration
131
6.1 INTRODUCTION
6.2 CONCEPTUAL FRAMEWORK AND RESEARCH HYPOTHESES
131
133
6.2.1 Relationship between thefirm 's appearance in the Annual Report on
Complaints andthefirm 's performance
755
6.2.2 Moderating effect of quality corporate image on the relationship between
thefirm 's appearance in the Annual Report on Complaints and thefirm 's
performance
136
6.2.3 Relationship between the number of third-party complaints andthefirm 's
performance
139
6.2.4 Moderating effect of market concentration on the relationship between the
number of third-party complaints and thefirm 's performance
141
6.3 METHODOLOGY
143
6.5.7 Sample
143
6.3.2 Analysis procedures
144
6.3.3 Data collection and measurement
149
6.3.4 Consumer complaint procedure
152
6.4 RESULTS AND DISCUSSION
153
6.4.1 Estimation ofreturn variation resultingfrom thefirm 's appearance in the
Annual Report on Complaints
153
6.4.2 Determinants ofreturn variation
156
6.4.2.1 Moderating effect of quality corporate image and direct effect of the number of
third-party complaints
156
6.4.2.2 Moderating effect of target market concentration
158
6.5 CONCLUSIONS
159
APPENDIX 6.1 P-VALÚES OBTAINED WITH BOOTSTRAP ESTIMATION
161
Chapter 7. Conclusión: summary, implications, limitations and future research 163
7.1 SYNOPSIS
7.2 THE CONSUMER'S REACTION TO DELAYS IN SERVICE
7.2.1 Main results and conclusions
7.2.2 Managerial implications
7.2.3 Limitations and future research
7.3 ANGER AND DISTRIBUTIVE JUSTICE IN A DOUBLE DEVIATION SCENARIO
7.5.7 Main results and conclusions
7.3.2 Managerial implications
7.3.3 Limitations and future research
7.4 THIRD-PARTY COMPLAINTS AND BANKING MARKET VALUÉ
7.4.1 Main results and conclusions
7.4.2 Managerial implications
7.4.3 Limitations and future research
7.5 FINAL CONCLUSIÓN
163
169
769
770
777
171
777
772
174
176
7 76
777
178
179
References
183
Resumen de la tesis doctoral
211
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Chapter 1
Introduction
In this chapter, we first introduce and describe the field this dissertation is
about. Next, we discuss the general research objectives as well as the specific
objectives of the individual studies. Finally, we conclude with an outline of the
remainder of this dissertation.
1.1. Service failures: the starting point
The understanding of the consequences of service failures is a key factor in
the strategic management of a service firm. Even the most customer-oriented
culture and the strongest quality program will not entirely eliminate mistakes
during service delivery (Kelley and Davis, 1994). Despite all the procedures,
some things may go wrong, especially since delivering services requires human
interaction. There is a popular saying: people may make or break the service.
Therefore, during the last years, service firms have made numerous attempts to
develop different strategies to deal with service failures (e.g., training
employees, starting customer affairs departments) with a double objective: to
prevent the same failure to occur again and to recover customers who complain
from their initial dissatisfaction. Thus, understanding the different elements that
affect (dis)satisfaction after service failure and subsequent behaviors derived
from this (dis)satisfaction can be useful for service managers to reduce the
impact of failures on firm performance.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
12 « Chapter 1
While the issue of consumer dissatisfaction is of importance to marketers in
general, some underlying characteristics of services make the topic especially
critical for services marketers. First, services are, to a greater degree than goods,
intangible, heterogeneous, and simultaneously produced/distributed and
consumed (Zeithaml et al., 1993). Second, in the performance of services both
customers and service personnel play a role (Solomon et al., 1985). These
characteristics increase the likelihood of errors (service failures) in the service
área both from an operational perspective as well as from the customer's
viewpoint and, therefore, increase the need for recovery (Brown et al., 1996).
Thus, service failure is defined as a customer's problem with a service
(Spreng et al., 1995) and is said to occur when the service experience falls short
of customer's expectations (Bell and Zemke, 1987). Traditionally, the service
literature considers failures to be inevitable or as Hart et al. (1990) stated
"mistakes are a critical part of every service" (p. 148). These failures in service
quality lead to dissatisfaction.
Existing research on customer (dis)satisfaction after a service failure occurs
can be divided into three major groups: (1) studies on the main antecedents and
consequences of customer (dis)satisfaction after service failure (e.g., Oliver,
1997; Westbrook, 1987), (2) studies on the main antecedents and consequences
of customer (dis)satisfaction after service failure and recovery (e.g., Smith et al.,
1999; Tax et al., 1998), and (3) studies on the main antecedents and
consequences of customer dissatisfaction response styles (e.g., Singh, 1988).
Different theories are behind the development of these studies such as the
expectancy-disconfírmation paradigm (Oliver, 1981), equity theory (Clemmer
and Schneider, 1996), emotion/affect theory (see Bagozzi et al., 1999, for a
review), or attribution theory (Weiner, 1985).
When a service failure occurs customers may respond in a variety of ways
from doing nothing at all to suing the company for millions of euros. This
process begins when the customer has evaluated a consumption experience and
ends when the customer has completed all behavioral and/or non-behavioral
responses to the experience (Day, 1980).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Introduction
•
13
Different taxonomies have been proposed to analyze the ways used to
express dissatisfaction (e.g., Day, 1980; Day and Landon, 1977; Richins, 1983,
1987; Singh, 1988). In this dissertation, we center on the framework proposed
by Singh (1988) which brings together the three empirical applications we have
carried out in a very clear and comprehensible manner. Singh (1988)
empirically proposed and tested a taxonomy in four different service categories:
grocery shopping, auto repair, medical care, and fínancial services. Thus, when
dissatisfaction occurs, three types of responses are likely to occur (see Figure
1.1): prívate response (e.g., repurchase intentions and/or word of mouth
communication to friends and relatives), voice response (e.g., seeking redress
from the seller and/or not taking any action), and third-party response (e.g.,
taking legal action and/or fíling a complaint with a Better Business Bureau). In
these three categories, there is a progression of the amount of effort involved in
complaining. For example, the prívate party objects are neither external to the
consumer's social network ñor are they directly involved in the dissatisfying
experience; the voice response (including no action or boycott) is primarily
directed against the seller; and the third party responses are directed toward
seeking redress from organizations (or courts) not directly involved in the
dissatisfying experience. Each of these three responses will receive full
attention in the theoretical and empirical part of this dissertation.
In the next section, we specifically address the overall objectives of this
dissertation and the specific objectives of the three studies we have carried out.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
14 « Chapter 1
Figure 1.1
Dissertation contextualization adapted from Singh's (1988) taxonomy
<
Dissatisfaction occurs:V
SER FICE FAIL
URE/
PRÍVATE
RESPONSES
(e.g., repurchase
intentions)
^
VOICE
RESPONSES
(e.g., redress seeking)
^_
Y
Specific context of
Chapter 4
Y
Specific context of
Chapter 5
THIRD PARTY
RESPONSES
(e.g., complain to a
public agency)
k_
"V
J
Specific context of
Chapter 6
Amount ofejfort involvedin complaining
+
1.2 Research objectives of the dissertation
The overall motivation behind this research is driven by the importance of
service failures in daily life, for firms as well as for customers. Therefore, the
aim of this dissertation is to contribute to the theoretical and empirical evolution
of service failures' research toward a better understanding of their
characteristics and, consequently, their implications for management. This
overall objective is divided into the following three general research questions.
The first two questions are examíned in the first two empirical applications
(Chapters 4 and 5), whereas the last question is dealt with in the third empirical
application (Chapter 6).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Introduction
•
15
Research Question 1. Which are the main variables that affect specific
customers' responses after a service failure (Le., prívate and voice responses)
and subsequentjudgments and/or behaviors?
We use empirical data to examine the main antecedents and consequences of
a dissatisfying experience following a specific service failure. In the first study
(Chapter 4), our service context is the airline industry and the service failure is a
flight delay. The second study (Chapter 5) is conducted in the banking industry
and the service failure is a banking failure. Using literature from a variety of
disciplines, such as marketing and (social) psychology, we first formúlate a
conceptual framework for every study. Subsequently, the substantive
relationships in these frameworks are tested.
Research Question 2. Which is the role played by negative emotions (Le.,
anger) vs. cognitive evaluations in customers' judgments and/or behaviors
following a service failure?
In the first two empirical applications (Chapters 4 and 5), we outline the
importance of studying the role of specific emotions in the formation of the
(dis)satisfaction judgment in a service failure context and, specifically, the role
of the negative emotion of anger. This focus on one specific emotion (i.e.,
anger) is in line with recent literature that focuses on the idiosyncratic elements
of specific emotions (Bougie et al., 2003; Louro et al., 2005; Tsiros and Mittal,
2000; Zeelenberg and Pieters, 2004). Accordingly, more insight into the specific
antecedents, phenomenology and consequences of different emotions (such as
anger) is needed (Lings et al., 2004). However, little attention has been paid to
the study of anger as the most frequent emotional reaction that arises in the
service failure contexts and its influence on customer's prívate (e.g., repurchase
intention) and voice (e.g., seeking redress from seller) responses. Specifically,
we examine the role of anger vs. different cognitive elements in the proposed
models.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
I
16 '
Chapter 1
Research Question 3. Do customers' third-party responses affect firm
performance?
In the third empirical application (Chapter 6), we examine the effect of thirdparty responses to service failures on firm performance. ín this study, we focus
on the investor's perspective, thus incorporating the financia! side into the
traditionai marketing view. Specifically, we use literature from different
disciplines, such as marketing, financial economics and signaling theory, for the
formulation of a conceptual framework. Then, the substantive relationships in
this framework are empirically tested.
To address these overall research questions effectively, we next formúlate
specific objectives for the different chapters in which we address the previous
three general problem statements from different perspectives.
1.2.1 Objectives Chapter 4
In Chapter 4, we focus on the airline industry to examine the impact of a
flight delay on the initial (dis)satisfaction judgment and subsequent behavioral
and complaining intentions {prívate responses). The objectives of this chapter
are: 1) to develop and empirically test a comprehensive conceptual framework
grounded in several research flelds that identifies the antecedents and
consequences of the (dis)satisfaction with the service failure (i.e., the flight
delay), 2) to examine the impact ofthe specific negative emotion of anger on the
previous framework, and 3) to explore the effects of different service-failure
related variables on (dis)satisfaction with service and on behavioral and
complaining intentions directly and indirectly through anger and
(dis)satisfaction with service.
As a new element, we jointly examine anger (emotional reaction) and
(dis)satisfaction with service failure (cognitive and emotional evaluation). Thus,
we analyze the impact of the initial negative emotion of anger on the initial
(dis)satisfaction judgment and subsequent behavioral and complaining
intentions.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
lntroduction • 17
1.2.2 Objectives Chapter 5
In Chapter 5, we go a step further and analyze the impact of a secondary
negative emotion (anger with service recovery) on secondary satisfaction
judgments (satisfaction with service recovery), in the specific context of double
deviation scenarios. In contrast to Chapter 4 which was centered on private
responses of customers, we base this research on voice responses of customers.
This means that we analyze data from customers who have complained to the
fírm after the service failure (i.e., they have voiced their dissatisfaction). From
these customers, we center on those who have experienced a failed recovery
after the initial service failure (i.e., double deviation). The objectives of this
chapter, which is focused on the banking industry, are: 1) to develop and
empirically test a comprehensive conceptual framework grounded in several
research flelds that identifles the antecedents of the (dis)satisfaction with
service recovery in the specific context of double deviation scenarios (i.e., failed
recoveries after service failures), 2) to examine the role of the secondary
emotion of anger (i.e., anger with service recovery) and the distributive
component ofjustice on the previous framework, and 3) to explore the direct
and indirect effects of service failure and service recovery-related variables on
(dis)satisfaction with service recovery judgments through the secondary
emotion of anger with service recovery and through the distributive justice
component.
The few studies in the service failure and recovery context that include
emotions in their proposals are centered on the emotions triggered by the initial
service failure (e.g., Andreassen, 2000; Bougie et al., 2003; Dubé and Maute,
1996; Smith and Bolton, 2002). Thus, this is the first attempt to model the effect
of specific secondary emotions on secondary (dis)satisfaction. It is also the first
attempt to empirically test a model of (dis)satisfaction with service recovery in
double deviation scenarios.
1.2.3 Objectives Chapter 6
In Chapter 6, we examine the impact of third-party complaints on company
performance. As in Chapter 5, we focus on the banking industry. Specifically,
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
18 • Chapter 1
we examine the complaints from the Bank of Spain's Complaints Service (the
third party), which publishes an Annual Report on Complaints to Spanish
banks. We propose that the reléase of this information about third-party
complaints is economically relevant to the stock market. The objectives of this
chapter are: 1) to determine the economic impact for the banks involved, in
terms of variation in stock prices, of appearing on the Annual Report on
Complaints ofthe Bank ofSpain 's Complaint Service, and 2) to examine to what
extent the variations in stock prices can be explained through the number of
complaints received by the bank in the Annual Report, the quality corporate
image, and the target market concentraron.
Until now, the influence of customer's third-party responses has been
analyzed from a customer perspective but not on the basis of its impact on firm
performance.
1.3 Dissertation outline
Chapter 2, Service failures: theoretical considerations, examines different
theoretical issues concerning the service failures. After defíning the basic
concepts, the antecedents of customer (dis)satisfaction in service failure
encounters are reviewed, focusing on the ones employed in this dissertation.
Finally, we present the outcomes of customer (dis)satisfaction in service failure
encounters following the taxonomy proposed by Singh (1988).
Chapter 3, Summary and description of the empirical applications, briefly
outlines the variables employed in the three empirical applications and how they
relate to the theoretical dimensions analyzed in Chapter 2. The main objective
of this short chapter is to give the reader a quick but complete view of what is
being studied in each ofthe empirical studies.
Chapter 4, The consumer's reaction to delays in service, centers on the
relationships that exist among attributions of control and stability, service features'
perceptions (perceived waiting time and punctuality importance), anger emotion,
(dis)satisfaction, and repurchase and complaining intentions of customers who
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Volver al Indice/Tornar a l'índex
Iníroduction
•
19
suffer delays in flights (airline industry). Thus, the service failure in this study is a
flight delay and the response examined is prívate.
Chapter 5, Anger and distributive justice in a double deviation scenario:
explaining (dis)satisfaction in service failure and failed recovery contexts,
analyzes the underlying mechanisms which contribute to (dis)satisfaction
formation in double deviation scenarios (i.e., failed recovery after service
failure). Accordingly, we propose and empirically test a framework that outlines
the roles of distributive justice (cognitive antecedent) and anger (emotional
antecedent) in determining (dis)satisfaction with service recovery (postrecovery attitude). Additionally, we examine how specifíc service failure and
service recovery-related variables influence customer (dis)satisfaction with
service recovery directly and/or indirectly through the cognitive and emotional
antecedents. This framework is applied to a cross-sectional sample of
dissatisfied banking customers (banking industry). Therefore, the service failure
in this study is a failed recovery and the response examined is the voice response.
Chapter 6, Third-party complaints and banking market valué: the
moderating effects of quality corporate image and market concentration,
examines the impact of third-party complaints on fírm performance.
Specifically, we analyze how the stock market (investors) reacts to the Annual
Report on Complaints to Spanish banks published by the Bank of Spain's
Complaints Service (the third-party). Additionally, we investígate the
explanatory power of the number of complaints received by the bank in the
Annual Report and the moderating roles played by quality corporate image and
market concentration. In sum, the service failure in this study is a failed recovery
and the response examined is the third-party one.
Finally, in Chapter 7, Conclusions: summary, implications, limitations and
future research, we provide a summary of the main theoretical and managerial
contributions, limitations, and directions for future research, of the three
empirical applications presented in this dissertation.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Chapter 2
Service failures: theoretical considerations
In this chapter, we examine the specific concepts, variables, and empirical
contexts that will be applied in the three applications discussed later in this
dissertation. Although each of these studies thoroughly reviews the existing
literature to define the study context and the variables employed, we feel that a
more general chapter will contribute to getting a broad perspective of the whole
present research.
2.1 Introduction
Services domínate most developed countries and, in the particular case of
most countries in the OECD (Organization for Economic Co-operation and
Development), the service sector accounts for 70% or more of aggregate
production and employment and continúes to grow (Wolfl, 2005). This growing
importance of the service sector has contributed to the development, over the
last three decades, of services marketing management (also known as services
marketing and/or services management) which has embraced other disciplines
such as human resources and operations (Swartz and Iacobucci, 2000).
Building on previous works in services marketing literature, Kasper et al.
(1999) propose the following defínitions of services:
'''Services include all economic activities whose output is not a physical
product or construction, is generally consumed at the time it is produced, and
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
22 » Chapter 2
provides added valué in forms (such as convenience, amusement, timeliness,
comfort or health) that are essentially intangible concerns ofitsfirst parchase"
(P-9)
"Services are originally intangible and relatively quickly perishable
activities whose buying takes place in an interaction process aimed at creating
customer satisfaction but during this Interactive consumption this does not
always leadto materialpossession" (p. 13)
Two main issues arise from the above definition: fírst, there are certain
characteristics that seem to differentiate services from goods and, second, in
line with authors such as Zeithaml and Bitner (2000), it seems that customer
satisfaction is the ultimate goal of service fírms. In the next subsections, we go
deeper into both aspects: the basic characteristics of services and the focus on
customer satisfaction.
2.1.1 The basic characteristics of services
Regarding the distinction between services and goods, from the early works
centered on the questions of 'if and 'how' services differed from goods, we
carne to the classic distinction between goods and services, based on the 4 I's:
intangibility, inseparability (as a degree of simultaneous production and
consumption), inconsistency (as a degree of heterogeneity), and inventory (as a
degree of perishability) of services compared to goods (Shostack, 1977).
The intangibility feature is the most dominant one in defíning services and
determines the other three characteristics (Kasper et al., 1999). However, the
differentiation between goods and services in terms of this feature is not easy.
Service organizations are trying to make tangible their intangible offer (and
even, many services can not be provided without tangibles), while many
manufacturers try to créate an (intangible) image around their goods.
Due to the intangibility, in many instances, customers find it hard to evalúate
services in advance. Furthermore, customers often cannot predict the outcome
of a service experience. These two aspects are strongly related with the risk
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 23
customers perceive in buying and using particular services. The literature
differentiates eight types of perceived risk: financial uncertainty (whether
customers pay more than they should); functional uncertainty (whether the
service really offers what it should); physical uncertainty (safety of the service
delivery); social uncertainty (the way in which the environment of the customer
will react to the choice of a certain service or particular service provider);
psychological uncertainty (the way in which a bad choice will damage the
image of the customer); life style uncertainty (similar to the social and
psychological uncertainty but especially focused on the expected or actual
consequences for one's own life style); time uncertainty (whether the time spent
searching for a service is wasted when the chosen service or service provider
does not perform according to expectations); and, environmental uncertainty
(the possible damage that the service or service delivery process may cause to
the environment). Generally, the (customer's total) perceived risk in a purchase
situation implies a mixture of these eight different kinds of uncertainty and it
can vary among services or service delivery processes. In any case, the
customer's perceived risk affects the way customers evalúate service
performance, also in service failure situations. The perceived risk is higher for
'fírst time buys' compared to 'repeated buys' due to customers' higher
uncertainty when consuming the service for the fírst time.
The second T , inseparability, or the degree of simultaneous production and
consumption, means that transferring the service usually requires the presence
and participation (i.e., the interaction) of the customer. This interaction may be
referred to as 'the service encounter', and can be mainly affected by the
environment in which the process of producing and consuming the service takes
place, the personnel involved, and the customer.
The third T , inconsistency, or the degree of heterogeneity, arises from the
active participation of the customer in the process of producing and consuming
the service, which makes the standardization of services quite diffícult.
Automation may reduce the impact of people and the environment on service
quality. However, there is still the problem that not only objective and/or
technical issues are evaluated by the customer but also subjective elements as
well as the amount of time used.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
24 » Chapter 2
Finally, inventory (the fourth T ) , or the degree of perishability, means that
usually services cannot be kept in stock. This reduces costs of warehousing, but
it also makes it difficult to face fluctuations in demand or in capacity.
Therefore, waiting is typical for service delivery processes. Probably,
information technology will contribute to better serve the customer in this sense.
2.1.2 The focus on customer satisfaction
Customer satisfaction, as the ultímate goal and the primary obligation of
service firms, is a defensible and appropriate company objective which allows
holding various corporate functions together and directs corporate resource
allocation (Zeithaml and Bitner, 2000).
Conceptually, virtually all company activities, programs, and policies should
be evaluated in terms of their contribution to satisfying customers (Peterson and
Wilson, 1992)1. The reason is that individual firms have discovered that
increasing levéis of customer satisfaction ('delight' customers) can be linked to
customer loyalty and profits (Anderson et al., 1997; Heskett et al., 1997).
Differently stated, many firms have adopted a retention/relationship focus
(relationship marketing/management) whose primary goal is to build and
maintain a base of committed customers who are profítable for the organization
(Zeithaml and Bitner, 2000). To achieve this goal, the service firm should focus
on the attraction, retention, and enhancement of customer relationships (Berry,
1983) through the effective satisfaction of customers' requirements, that is,
through the provisión of sufficient service quality.
However, quality is an ambiguous term which can be viewed from many
different points of view. Therefore, Garvín (1988) formulates five approaches to
1
In fact, because of the importance of customer satisfaction to firms and overall quality of life,
many countries have a national índex that measures and tracks customer satisfaction at a macro
level. Many public policymakers believe that these measures could and should be used as tools
for evaluating the health of the nation's economy, along with traditional measures of productivity
and price. These indexes include the quality of economic output, while more traditional economic
indicators tend to focus only on quantity (Zeithaml and Bitner, 2000). Some examples are the
Deutsche Kundenbarometer (DK) in Germany, the Swedish Customer Satisfaction Barometer
(SCSB) in Sweden, or the American Customer Satisfaction Index (ACSI) in United States of
America.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 25
studying quality in the context of tangible products: transcendent (psychology),
product-based (economics), user-based (marketing and operational
management), manufacturing-based (operational management), and valuebased. Kasper et al. (1999) have applied these approaches to the service context.
Thus, the transcendent approach implies that quality cannot always be defíned
and is partly a matter of experience. The product-based approach views quality
as a measurable and objective variable, in terms of the level of services or
features offered. The user-based approach bases on the customer's judgment,
which is largely subjective and leads to perceived service quality. The
manufacturing-based approach views quality as an objective and measurable
term and mainly concerns conformance to requirements (in technical
terms/specifications). Finally, the value-based approach considers quality in
relation to cost and price.
From these different perspectives, we focus on the user-based approach to
service quality. Following Kasper et al. (1999), quality of a service is often a
perceived quality, depending mostly on expectations and the way the service is
received. Therefore, Kasper et al. (1999) define quality as "the extent in which
the service, the service process and the service organization can satisfy the
expectations of the user" (p. 188)2. Putting the customers' satisfaction at the
center of services marketing implies that their subjective evaluations are
decisive in their evaluation of the organization's performance. Therefore,
marketing managers have to pay attention to the expectations of customers (to
properly achieve their requirements) and to their quality perceptions.
In evaluating service quality, customers focus on different attributes that
may differ per service. Consequently, different studies have made an attempt to
come up with a bundle of features that are always (or almost always) present in
customers' evaluations of service quality. The first result in this sense was the
2
In searching for information from a particular service or service organization, customers may
use various characteristics, attributes or qualities of services. Often, three kinds of search
attributes/qualities are distinguished (Zeithaml and Bitner, 2000): search, credence, and
experience attributes. Search attributes are easy to judge by the customer before the service
delivery actually takes place and usually involve tangible aspects. Credence attributes can only be
judged after the actual service delivery and are based on trusting people delivering the service.
Finally, experience attributes are in the middle of the previous two types, being difficult to
evalúate in advance and experienced only after the service delivery. In general, perceived risk will
be greater the less search attributes and the more experience and credence attributes are at stake
(Kasper et al., 1999, p. 156).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
26 • Chapter 2
qualitative study of Parasuraman et al. (1985). The authors identified ten
determinants of service quality (Parasuraman et al., 1985, p. 47): reliability (i.e.,
consistency of performance and dependability), responsiveness (i.e., willingness
or readiness of employees to provide service), competence (i.e., possession of
the required skills and knowledge to perform the service), access (i.e.,
approachability and ease to contact), courtesy (i.e., politeness, respect,
consideration, and friendliness of contact personnel), communication (i.e.,
keeping customers informed in language they can understand and listening to
them), credibility (i.e., trustworthiness, believability, honesty, having the
customer's best interests at heart), security (i.e., the freedom from danger, risk,
or doubt), understanding/knowing the customer (i.e., making the effort to
understand the customer's needs), and tangibles (i.e., physical evidence of the
service).
Later, the authors developed a 22-item instrument (called SERVQUAL),
which recast the ten previous determinants into five specific components (three
original and two combined dimensions): tangibles (i.e., physical facilities,
equipment, and appearance of personnel), reliability (i.e., ability to perform the
promised service dependably and accurately), responsiveness (i.e., willingness
to help customers and provide prompt service), assurance (i.e., knowledge and
courtesy of employees and their ability to inspire trust and confidence), and
empathy (i.e., caring, individualized attention the firm provides its customers)
(Parasuraman et al., 1988:23). These are general dimensions which can be
found in most services and that reflect customers' subjective judgments about
the valué received by service performance (Kasper et al., 1999, p. 213)3.
3
After the publication of the Parasuraman et al.'s (1988) study on SERVQUAL, numerous
studies have examined critically their model, especially the works of Joseph Cronin and Stephen
Taylor (Cronin and Taylor, 1992, 1994). The basic criticisms concern: the need of measuring
expectations, how expectations are measured, the dimensionality of SERVQUAL, or the number
of items in the SERVQUAL scale, among other issues (Kasper et al., 1999, p. 224). Based on
these criticisms, Cronin and Taylor (1992; 1994) developed an altemative method of
operationalizing perceived service quality, the SERVPERF-model. The main difference wíth the
SERVQUAL-model is that they only use questions about performance (i.e., perception), ignoring
the questions about expectations.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 27
In sum, the focus on customer satisfaction logically draws attention to the
management of individual service encounters4 between the ultimate customer
and representatives of the firm (Bitner et al, 1990), sometimes referred to as
'moments of truth' (Carlzon, 1987). It is in these moments of truth, when the
customer interacts with the service firm, that the service quality is most
immediately evident to the final customers. Each service encounter pro vides an
opportunity for the firm to reinforce its commitment to customer satisfaction
and/or to service quality and can potentially be critical in determining customer
satisfaction and loyalty. Additionally, each individual encounter is important in
creating a composite image of the firm in the customer's memory. Over time, it
is likely that múltiple positive (negative) encounters will lead to an overall high
(low) level of satisfaction (Bitner and Hubbert, 1994).
However, as stated before, even the most customer-oriented culture and the
strongest quality program may not entirely eliminate mistakes during service
delivery (Kelley and Davis, 1994). Service encounters can often produce
negative reactions despite the service personnel trying to do their very best
(Zeithaml et al., 1985). These critical encounters can ruin the customer-firm
relationship and drive the customer away, no matter how many or what type of
encounters have occurred in the past (Zeithaml and Bitner, 2000). Therefore,
deeper examination of these 'critical' encounters is needed, providing the global
objective of this dissertation.
2.2 Service failure encounters: defínition and nature
The term we use to define a customer's problem with a service is 'service
failure' (Spreng et al., 1995). Service failures occur when the service experience
falls short of customer's expectations (Bell and Zemke, 1987). Or as stated by
Bitner et al. (1990), service failures are specific events that lead to dissatisfying
service encounters from the customers' point of view.
Marketing research on service failures distinguishes two broad types of
failures or losses customers may experience: core service failures (outcome) and
4
The service encounter has been defíned as that period of time during which the consumer and
service firm interact in person, over the telephone, or through other media (Shostack, 1985).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
28 • Chapter 2
service delivery failures (process) (Bitner et al., 1990; Hoffman et al., 1995;
Keaveney, 1995). Outcome failures refer to actual performance of the basic
service need whereas process failures refer to the customer's experiences while
the service is being performed. Therefore, service encounters, due to their
specific nature, account for both outcome and process dimensions and a failure
could occur along either dimensión. Table 2.1 provides a summary of the main
findings regarding the classification of service failures.
In this dissertation, we analyze both types of service failures. In the first
study (Chapter 4), customers are faced with an outcome service failure, that is, a
delay in their flights. In the next two studies (Chapters 5 and 6), different
service failures are reported, both outcome and process service failures.
Additionally, service failure is assumed to result in dissatisfaction (Zeithaml
and Bitner, 2000), which affects negatively customer retention (e.g., Rust and
Zahorik, 1993), and subsequently has a negative impact on revenue and
profítability (e.g., Fornell and Wernerfelt, 1987, 1988; Rust et al., 1995). As
Hart et al. (1990) stated "mistakes are a critical part of every service, [...] errors
are inevitable [...] but dissatisfíed customers are not" (p. 148). It is the
company's service recovery systems (or the lack of them) that may become a
source of satisfaction or dissatisfaction (Bitner et al., 1990), not necessarily the
mistake or failure itself. However, service providers can not remedy customer's
service failure experiences unless the customer first seeks redress (Blodgett et
al., 1995). This means that complaints lodged directly with the firm are the only
responses that provide the organization with an opportunity to recover
effectively from service failure (Tax et al., 1998).
Specifically, research has shown that firms should invest more resources to
facilitate complaints and encourage dissatisfíed customers to voice their
dissatisfaction through complaints (Fornell and Wernerfelt, 1987). This type of
dissatisfaction management can be an effective tool for customer retention,
particularly in high competitive markets (Fornell and Wernerfelt, 1987, 1988).
Taking into account that today's customers are more demanding, better
informed, and more assertive when service problems arise (Hoffman et al.,
1995, p. 49), many service organizations have developed different strategies to
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 29
deal with service failures, especially in terms of making the complaining
process as simple as possible (e.g., toll-free cali centers, e-mail).
As said, given the importance of service failures for firms, the main goal of
this dissertation is to examine the harmful consequences of service failures in
three different stages of the customer-firm relationship. In the fírst stage
(analyzed in Chapter 4), we examine the consequences of a service failure, in
terms of customer's satisfaction and behavioral intentions (prívate response). At
this stage, the service firm has not an opportunity to respond to the failure
because the customer's response is prívate. In the second stage (analyzed in
Chapter 5), we investígate the fírm's reaction to service failure (service
recovery) but in the specific context of a failed recovery, that is, when the
service firm fails to recover the customer once he/she has complained (voice
response). Finally, in the third stage (analyzed in Chapter 6), we go a step
further and address third-party responses derived from failed recoveries
(following a service failure). Thus, we examine how third-party responses can
damage fírm's performance not only from a customer basis but also from a
fínancial basis (stock prices).
Therefore, our focus in this dissertation is on what actually happens after
customers experience a service failure (service failure encounter and/or failed
recovery encounter). In the rest of this chapter we center on the main
antecedents and consequences of customer (dis)satisfaction. First, however, we
define what we understand by customer (dis)satisfaction in service failure
encounters.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
30 • Chapter 2
Table2.1
Summary of the main findings regarding the classifícation of service failures
Source
Bitneretal. (1990)
Kelley et al. (1993)
Hoffman et al. (1995)
Lewis and
Spyrakopoulos (2001)
Michel (2001)
Hoffman et al. (2003)
First (priman) failure classifícation
• Gl: employee response to service delivery
system failures
• G2: employee response to customer needs
and requests
" G3: unprompted and unsolicited employee
actions
Bitneretal. (1990):
Gl: employee response to service delivery
system failures
G2: employee response to customer needs
and requests
G3: unprompted and unsolicited employee
actions
Bitneretal. (1990):
Gl: employee response to service delivery
system failures
G2: employee response to customer needs
and requests
G3: unprompted and unsolicited employee
actions
G1: Banking procedures
G2: Mistakes
G3: Employee behavior and training
G4: Functional/technical failures
G5: Actions or omissions of the bank that are
against the sense of fair trade
Gl: Advice
G2: Process
G3: Interaction
G4: Documents
G5: Information
G6: Conditions
G7: Systems
G8: 3rd parties
Bitneretal. (1990):
Gl: employee response to service delivery
system failures
G2: employee response to customer needs
and requests
G3: unprompted and unsolicited employee
actions
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Industry
analyzed
Airlines
Hotels
Restaurants
Retail
Restaurants
Retail banking
Retail banking
Hospitality
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 31
2.3 Customer (dis)satisfaction
defmition and nature
in
service
failure
encounters:
Despite extensive research during the last decades, researchers have not yet
developed a consensual defmition of customer (dis)satisfaction (Babin and
Griffin, 1998; Oliver, 1997)5. Marketing literature shows different conceptual
and operational definitions of (dis)satisfaction. This basic defínitional
inconsistency is evident by the debate of whether (dis)satisfaction is a process
or an outcome (Parker and Mathews, 2001; Ruyter et al., 1997; Yi, 1990). Thus,
some (dis)satisfaction definitions have emphasized the (evaluation) process
perspective (e.g., Bearden and Teel, 1983; Day, 1984; Fornell, 1992; Fournier
and Mick, 1999). This approach is based on the expectancy disconfirmation
paradigm (Oliver, 1980), and concentrates on the antecedents to
(dis)satisfaction rather than (dis)satisfaction itself. Thus, (dis)satisfaction is
viewed as an evaluative process derived from the global consumption
experience with unique measures capturing unique components of each stage.
This approach seems to draw more attention to perceptual, evaluative, and
psychological processes that combine to genérate consumer satisfaction (Yi,
1990). Additionally, some (dis)satisfaction definitions view this construct as a
response (outcome) to an evaluative process (e.g., Halstead et al., 1994; Oliver,
1997, 1981). That approach focuses on the nature (not cause) of (dis)satisfaction
and proposes that (dis)satisfaction is a result derived from the evaluation of a
specific consumption experience. We will follow this approach in the present
research.
Given the complex nature of satisfaction, it is difficult to develop a generic
global defmition. Rather, the defmition of satisfaction must be contextually
adapted (Giese and Cote, 2000). Therefore, based on a review of the literature
and their own research, Giese and Cote (2000) identify three general
components present in all definitions of (dis)satisfaction:
•
•
consumer (dis)satisfaction is a response (emotional, cognitive, or both);
the response pertains to a particular focus (expectations, product,
consumption experience, etc.); and,
5
For a recent review of the reported evidence on customer satisfaction, see Szymanski and
Henard (2001).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
32 * Chapter 2
•
the response occurs at a particular time (after consumption, after choice,
based on accumulated experience, etc).
The previous context-specifíc definition refers to service encounter
(dis)satisfaction as opposed to overall service (dis)satisfaction in terms of the
classification made by Bitner and Hubbert (1994)6. Moreover, this definition
captures the complete domain of (dis)satisfaction and is consistent with the
conceptual domain of other researchers (Giese and Cote, 2000). Therefore, the
context-specifíc definition will be our approach in the first two studies carried
out (Chapters 4 and 5) which are the ones that explicitly examine this variable.
In these studies, we adapt the context-specifíc definition to the different
research settings.
Next, we examine the three components identified by Giese and Cote (2000)
in the context of this dissertation.
2.3.1 Customer (dis)satísfaction as a response
Customer (dis)satisfaction has been typically conceptualized following four
maín theoretical approaches: cognitive, affective, contingent, and cognitiveaffective. The cognitive approach is based on the popular view that the
confirmation/disconfirmation of preconsumption product standards is the
essential determinant of (dis)satisfaction (e.g., Bloemer and Kasper, 1995;
Churchill and Surprenant, 1982; Halstead et al., 1994; Tse and Wilton, 1988).
This paradigm posits that confirmed standards lead to modérate satisfaction,
positively disconfirmed (exceeded) standards lead to high satisfaction, and
negatively disconfirmed (underachieved) standards lead to dissatisfaction.
The affective approach views satisfaction as an affective response (e.g.,
Babin and Griffin, 1998; Cadotte et al., 1987; Westbrook, 1980; Woodruff et
al., 1983). Consumer satisfaction, therefore, can be described as an emotion
resulting from appraisals (including positive disconfirmation, perceived
6
Following Bitner and Hubbert (1994), we define service encounter satisfaction as "the
consumer's (dis)satisfaction with a discrete service encounter (p. 76)", and overall service
satisfaction as "the consumer's overall (dis)satisfaction with the organization based on all
encounters and experiences with that particular organization (p. 77)".
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 33
performance, etc.) of a set of experiences (Westbrook, 1980; Woodruff et al.,
1983).
The contingent approach, based on the work of Fournier and Mick (1999),
views satisfaction as a context-dependent process consisting of a multi-model,
multi-modal blend of motivations, cognitions, emotions, and meanings,
embedded in socio-cultural settings, that transforms during progressive and
regressive consumer-product interactions (a dynamic process).
Finally, more recent (dis)satisfaction definitions concede that
(dis)satisfaction has a dual origin, that is, cognitive and affective (e.g., Mano
and Oliver, 1993; Oliver, 1989, 1993, 1997; Westbrook, 1987; Westbrook and
Oliver, 1991; Wirtz and Bateson, 1999; Yi, 1990). Thus, (dis)satisfaction is not
a solely cognitive phenomenon, rather, is likely to comprise an element of affect
or feelings (Yi, 1990, p. 34). This is the approach followed in the first two
studies (Chapters 4 and 5) of this dissertation, which account explicitly for the
specifíc negative emotion of anger and for the customer's (dis)satisfaction
response. In the third study (Chapter 6), we do not investígate explicitly these
variables.
2.3.2 The focus of the customer (dis)satisfaction response
The focus of the (dis)satisfaction response identifies the object of a
consumer's (dis)satisfaction and usually entails comparing performance to some
standard (Giese and Cote, 2000). Several comparison standards, from very
specifíc to more general, have been used in past research, which can be
summarized as follows (Yi, 1990): (1) expectation-disconfirmation paradigm
(e.g., Oliver, 1980, 1981, 1997); (2) comparison level theory (e.g., Swan and
Martin, 1980); (3) equity theory (e.g., Fisk and Young, 1985); (4) norms as
comparison standards (e.g., Woodruff et al., 1983); and (5) value-percept
disparity theory (e.g., Westbrook and Reilly, 1983).
There are often múltiple foci to which these various standards are directed
including the product (e.g., Tse and Wilton, 1988), the consumption experience
(e.g., Bearden and Teel, 1983), a salesperson (e.g., Oliver and Swan, 1989), or a
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
34 • Chapter 2
store/acquisition (e.g., Oliver, 1981), among others. This variety of foci is
confirmed in the work of Giese and Cote (2000).
In this dissertatíon, the focus of the (dis)satisfaction response will be the
service failure experienced. In the fírst study (Chapter 4), the focus (i.e., service
failure) will be a flight delay. In the next two studies (Chapters 5 and 6), the
focus (i.e., service failure) will be the bank's response (or lack of response) to
the problem experienced by the customer.
2.3.3 The timing of the customer (dis)satisfaction response
Following Giese and Cote (2000), (dis)satisfaction can be determined at
various points in time, although it is generally accepted that it is a post-purchase
phenomenon (Yi, 1990; e.g., Churchill and Surprenant, 1982; Fornell, 1992;
Oliver, 1981; Tse and Wilton, 1988; Westbrook and Oliver, 1991).
Additionally, Cote et al. (1989) argüe that satisfaction can vary dramatically
over time and that it is only determined at the time the evaluation occurs.
Therefore, Giese and Cote (2000) suggest that researchers should select the
point of determination most relevant for the research questions.
We consider that customer (dis)satisfaction is determined after service
failure occurs, which in the specific context of this dissertation means after a
flight delay in the fírst study (Chapter 4) and after a failed recovery in the other
two studies (Chapters 5 and 6)7.
In the following section, we review the specific factors present in the two
studies carried out in this dissertation which explicitly investigate/examine in
their proposed models customer (dis)satisfaction in service failure encounters
(Chapters 4 and 5). We will just review the influence of those factors on
customer (dis)satisfaction with service failures, letting their specific
contextualization to the corresponding chapter.
7
We recognize that, since services are a process, the service is also evaluated during the process
of service delivery. Henee, (dis)satisfaction may also oceur during that whole process and not
only at the end. However, our measures of customer (dis)satisfaction have been developed,
adopting the proposals of Giese and Cote (2000), focusing on the specific moment at which the
service failure takes place.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 35
2.4 Antecedents/Determinants of customer (dis)satisfaction in service
failure encounters
Since the early 1970s, one of the major developments in customer
(dis)satisfaction research has been a focus on the theoretical determinants of
(dis)satisfaction. A fairly consistent pattern emerged in that most models were
variations of Oliver's (1980) expectancy-disconfírmation model. Oliver's
(1980) model proposes that customer satisfaction is a positive function of
customer expectations (i.e., pre-purchase customer beliefs about anticipated
product performance) and disconfirmation beliefs (i.e., post-purchase customer
beliefs about the extent to which product performance met expectations).
Later modifications to the expectancy-disconfírmation model of satisfaction
were primarily in the form of adding new predictors in an attempt to provide
greater explanatory power. These variables included, among others, alternative
comparison standards, product performance, causal attributions, affective
response, and equity (see Oliver and DeSarbo, 1988; Oliver and Swan, 1989;
Tse and Wilton, 1988; Westbrook, 1987; Zeithaml et al, 1993).
Specifícally, research on consumer (dis)satisfaction and complaining
behavior has identified individual and contextual characteristics associated with
active responses to dissatisfaction (see Fornell and Wernefelt, 1987; Singh,
1990). Higher income and education (Warland et al., 1975), professional
occupational status (Andreasen, 1985), younger age (Morganosky and Buckley,
1986),
self-confidence
(Gronhaug
and
Zaltman,
1981),
assertiveness/aggressiveness (Richins, 1983), and knowledge and experience
with the service firm (Singh, 1990) are among the individual characteristics
associated with dissatisfaction responses (Dubé and Maute, 1996). Product,
market and contextual variables such as problem severity, blame attributions,
amount of effort to complain (Richins, 1983), expected complaining
consequences (Singh, 1990) and redress environment characteristics (Bolfing,
1989) have also been shown to influence behavioral responses to dissatisfaction.
Other variables that have been analyzed regarding their effect on customer
(dis)satisfaction are performance/expectation ambiguity (Nyer, 1996), product
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
36 • Chapter 2
involvement or importance (Oliver and Bearden, 1981), or consumption valúes
(Oliver, 1996).
However, in this section we will only examine the different predictors of
customer (dis)satisfaction that have been employed in the different empirical
applications present in this dissertation. Specifically, we will address the
following antecedents: service features (failure magnitude and perceived
waiting time), causal attributions, customer emotions, and perceived justice.
2.4.1 Service features
Customer (dis)satisfaction with a service is influenced signifícantly by the
customer's evaluation of service features. For a service such as an airline, an
important feature would be punctuality (Taylor, 1994). For a service such as a
retail bank, important features might include the (competitive) interest rates
(Laroche and Taylor, 1988), or the (convenient) bank location (Levesque and
McDougall, 1996). In any case, research has shown that customers will make
trade-offs among different service features depending, among other issues, on
the criticality of the service (Ostrom and Iacobucci, 1995).
In this dissertation, the two service features that have been explicitly
examined are the failure magnitude and the perceived waiting time. Regarding
the failure magnitude, Hirschman (1970) was the fírst to assess that consumers
would be more likely to voice their complaints when dissatisfied with an
'important' product. After that, researchers have focused on two major
dimensions of product relevance. The fírst is the traditional notion of
instrumental or utilitarian performance whereby the product is seen as
performing a useful function. The second dimensión is that of hedonic or
aesthetic performance (Hirschman and Holbrook, 1982) whereby producís are
valued for their intrinsically pleasing properties. This two-dimensional approach
is frequently typified as one of thinking versus feeling8.
Another conceptualization of product relevance is that of involvement (Zaichkowsky, 1985)
which reflecte the inherent need fulfillment, valué expression, or interest that consumer has in the
product. Involvement's influence on consumption experiences is best illustrated by the
psychological consequences evoked by a product's heightened relevance to the consumer. These
consequences are known to include higher motivation (Bloch et al., 1986), heightened arousal
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 37
In the services field, many researchers have examined and demonstrated the
effects of magnitude of the failure, complaint seriousness/intensity, or severity
of the dissatisfaction problem on customer responses (Bearden and Oliver,
1985; Gilly and Gelb, 1982; Hoffman et al., 1995; Maute and Forrester, 1993;
Richins, 1987; Smith et al., 1999). It is reasonable to assume that consumers
who are dissatisfied with producís they feel are important (or 'worthwhile', see
Singh, 1990) will experience higher levéis of negative emotions than consumers
who are dissatisfied with products of lesser importance. Because of their
negative emotions (e.g., anger), consumers who are dissatisfied with an
'important' product are likely to want to 'hurt' the offending retailer (Folkes,
1984), possibly by engaging in negative word-of-mouth behavior and by
vowing never to repatronize the retailer (Singh, 1990).
The failure magnitude is examined in the first two studies (Chapters 4 and
5). In the context of Chapter 4, the failure magnitude is related to the
punctuality importance. In the context of Chapter 5, the failure magnitude refers
to the importance/severity of the critical incident reported by the bank customer,
without focusing on a specific service attribute. In the third study (Chapter 6),
this service feature is not examined.
The second service feature analyzed in this dissertation is the perceived
waiting time. Waiting is a pervasive element of many purchase situations
(Taylor, 1994), and a negative experience for many customers (Scotland, 1991).
Thus, speed of service is increasingly becoming a very important service
attribute (Katz et al., 1991) that needs to be controlled by service managers.
Researchers have argued that service waits can be controlled by two techniques:
operations management or perceptions management (Katz et al., 1991). The
difficulty of obtaining 'zero defects' in a service setting and thus the limited
success of operations management in these settings has led to increasing interest
in managing the perceptions of the wait experience. Thus, the manager maybe
can not control the actual wait duration but he/she can try to control the
customer's perception of it. To do this effectively, it is important to understand
(Mitchell, 1980), and increases in cognitive elaborations (Petty et al., 1983). Thus, in tenns of
relevance, valué, interest, or need, involvement is a major element in defining and assessing
dimensions of product evaluation (Mano and Oliver, 1993).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
38 • Chapter 2
the variables that influence the perception of the wait and the possible impact of
the wait on service evaluations. However, empirical research in this área is
rather limited (e.g., Dubé-Rioux et al., 1988; Katz et al., 1991; Taylor 1994McDougall and Levesque, 1999; Pruyn and Smidts, 1998).
Therefore, to understand the waiting experience, one must understand what
is meant by wait for service. Several factors can be used to define different
types of wait. First, it is the point of time at which the wait is initiated
Customers can wait before, during, or after a transaction; that is, pre-process
(e.g., waiting at the dentist's waiting room prior to being attended to) inprocess (e.g., waiting at the dentist's stretcher during a visit), and post-process
(e.g., waiting at the dentist's desk prior to receiving and paying the bilí) (DubéRioux et al., 1988, p. 56). It has been shown that pre-process waits are more
unpleasant than in-process waits (Dubé-Rioux et al., 1988), and even that pre
process waits should be the main objective of management (Venkatesan and
Anderson, 1985). Therefore, pre-process waits are categorized further into three
general types: pre-schedule waits, delays (that is, post-schedule waits) and
queue waits. Pre-schedule waits include those in which a customer waits
because he or she has arrived early for a scheduled event. Delay is the wait from
the first moment at which the service was scheduled to begin The queue wait
occurs when appointments or scheduled commencement times are not used For
a queue wait, service usually is provided on afirst-come-first-servedbasis for
which customers must line up to receive the desired service (though a phvskal
queue is not necessary) (Taylor, 1994).
Studies have consistently found a negative relationship between actual or
perceived time spent waiting and service quality evaluations, with longer delays
resulting in lower service evaluation (Katz et al., 1991; Taylor 1994)
Therefore, as the perception of waiting time increases, customer satisfaction
tends to decrease (Katz et al., 1991). Additionally, previous research has also
found that delay is likely to cause anger because the service provider has failed
to dehver service at a specified time (Taylor, 1994).
The perceived waiting time is relevant in the context of the first study
(Chapter 4), but is not specifically addressed in the rest of the dissertation In
that study, we focus on only one type of wait, a pre-process delay, which refers
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 39
to the time from which a customer is ready to receive the service until the time
the service begins.
2.4.2 Causal attributions
Causal attributions are the second group of antecedents of customer
(dis)satisfaction with service failure that we focus on.
The concept of causal explanation was first proposed in a comprehensive
manner by Fritz Heider in his book "The Psychology of Interpersonal
Relations" (1958). Other conceptualizations followed this one in the sixties and
seventies, among which we must highlight the ones developed by Kelley and
Weiner and his colleagues which have served as a basis for numerous studies
about the attribution process (for a review see Harvey and Weary, 1984).
Bernard Weiner's work has been one of the most influential in the numerous
extensions in research and applications in different áreas (e.g., Weiner, 1980,
1985). The author centers on three attributional dimensions: internal/external
(locus), controllability, and stability. In the first dimensión, locus, the outcome
is attributed either to something inherent to the individual or to an external
agent. In the second dimensión, controllability, the author proposes that the
causes of an outcome could be modified either by the individual or by an
external agent. Finally, with respect to stability, the basic proposition is that
some causes of the outcome can be permanent in time (stable and predictable)
while others are highly variable.
Applied to consumer contexts, Weiner's framework explains adequately the
main attribution mechanisms used by consumers to describe purchase outcomes.
In fact, two thorough reviews of attribution theory in marketing are available.
The first covers work up through the late seventies, where interest focused on
consumer attributions for marketing strategies such as advertising content and
for self-reactions to marketing strategies (Mizersky et al., 1979). The second
one was carried out by Folkes (1988) and focuses on the impact of consumer
attributions on post-purchase processes. In Une with this second approach, much
of the attribution work in consumer behavior has examined the effect of
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
40 • Chapter 2
attribution on a number of mix consequences (e.g., repurchase intentions,
redress seeking).
In general, attribution theory views people as information rationale's
processors which actions are influenced by their causal inferences. In the
context of service failure, this means that people try to determine why the
service has failed and the type of inferred reason influences their response to
that failure (Bettman, 1979). In fact, service failure is the type of negative and
unexpected outcome which leads to causal search (Weiner, 1985). Thus, it has
been proposed that attribution search is more likely following failure
(dissatisfaction) rather than success (satisfaction) (Weiner, 2000). McFarland
and Ross (1982) showed that attributions for failure appeared to have more
pervasive affective consequences than did attributions for success, given that
failure induced greater thought than success. Consistently, Wong and Weiner
(1981) obtained evidence that subjects genérate more questions about the causes
of failures than of success.
The causal attributions that are relevant in the context of this dissertation are
the attributions of control and the attributions of stability. Specifícally, in the
first study (Chapter 4), we focus on control and stability attributions. In the
second study (Chapter 5), we focus on control attributions. In both studies, we
go deeper in the relationship between attributions, emotions, (dis)satisfaction,
and post-purchase behaviors, in different service failure encounters. In the third
study (Chapter 6), these dimensions are not examined.
2.4.3 Customer emotions
Customer emotions are the third group of antecedents of customer
(dis)satisfaction with service failure that we focus on.
Firstly though, it is important to reflect on the use of terminology related to
emotions in marketing. Specifícally, a distinction needs to be made on the terms
affect, emotion and mood. Following Bagozzi et al. (1999, pp. 184-185):
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations « 41
"affect might be considered a general categoryfor mental feeling processes,
rather than a particular psychological process, per se "
"by emotion, we mean a mental state ofreadiness that arises from cognitive
appraisals ofevents or thoughts; has a phenomenological tone; is accompanied
by physiological processes; is often expressed physically; and may result ¡n
specific actions to affirm or cope with the emotion, depending on its nature and
meaningfor the person having it; [...] and is typically intentional (i. e., it has an
object or referent) "
"by convention mood is conceived to be longer lasting (from afew hours up
to days) and lower in intensity than an emotion [...]; are generally
nonintentional and global or diffused (Frijda 1993); [...] and are not as
directly coupled with action tendencies and explicit actions as are many
emotions ".
Based on the previous definitions, we can conclude that our interest in this
dissertation refers to emotions, as a mental state that arises in response to
customers' appraisals9 for specific situations of relevance to them (such as a
service failure encounter and/or the subsequent recovery efforts).
Thus, following Price et al. (1995), service delivery has the potential to
evoke a variety of emotional responses (anger, regret, pleasure, affection,
confidence), because the customer is more actively engaged in a service
encounter, and has more investment in that interaction than in many product
purchases. In other words, the interactive nature of services is likely to enhance
the importance of the internal state of customers in the way they evalúate
service encounters (Bateson and Hoffman, 1999). This is particularly true in the
case of 'memorable' service encounters, especially the unsatisfactory ones.
Therefore, an emotional approach will be used to determine consumers'
emotional responses to service encounters involving failure and/or recovery
(Andreassen, 1999, 2000; Smith and Bolton, 2002).
The study of emotions in the context of consumer experiences began with
the work by Holbrook and colleagues (e.g., Holbrook and Batra, 1987;
Holbrook and O'Shaughnessy, 1984), who showed that distinct primary affect
9
Following Bagozzi et al. (1999), we define appraisal as "an evaluative judgment and
interpretation thereof' (p. 185).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
42 • Chapter 2
dimensions were prominent in advertisement evaluations. Using the emotion
typology of Plutchik (1980), Holbrook confirmed the existence of positive and
negative affect as responses to advertising exposure. Since then, researchers
have tried to show how emotion relates to (dis)satisfaction, investigating jointly
emotion and (dis)satisfaction (e.g., Westbrook, 1987; Westbrook and Oliver,
1991; Mano and Oliver, 1993; Oliver, 1993; Price et al., 1995). Thus, Oliver
(1993) suggests that emotions can have a dual role in (dis)satisfaction
judgments. First, emotions can act as a mediator between cognitive evaluations
and, second, they can also contribute independently to (dís)satisfaction (Muller
et al., 1991; Westbrook, 1987). Consequently, emotions are not only an
important dimensión of the service experience, but also a determinant of
customer (dis)satisfaction (Westbrook, 1987; Westbrook and Oliver, 1991).
Based on a recent review of the literature on emotion and consumer behavior
carried out by Bagozzi et al. (1999), we can state that there are basically two
forms of modeling interactions between emotions and (dis)satisfaction (and
behaviors/intentions related to (dis)satisfaction): a valence-based approach and
a specific-emotion approach.
The valence-based approach entails a summation of the positivity and
negativity of the different emotions that customers experience to arrive at an
overall judgment of (dis)satisfaction (Zeelenberg and Pieters, 2004). Thus,
negative emotions are expected to lead to more dissatisfaction, whereas positive
emotions are expected to lead to more satisfaction. The net (weighted) valence
of the affective experience then is the balance between positive and negative
emotions. The behaviors that follow are then supposed to be fully driven by this
overall positivity or negativity. The main advantage of this approach is its
parsimony because it allows for combining all sorts of emotions and other
constructs to be expressed in one single currency, namely, customer
(dis)satisfaction. The main disadvantage is to ignore all the specific elements
that are present in the different emotions, which are not easily expressed by
valence alone (Zeelenberg and Pieters, 2004).
The specific-emotion approach is in line with recent literature that "goes
beyond mere valence and focuses on the idiosyncratíc elements of specific
emotions" (Zeelenberg and Pieters, 2004, p. 446). According to this approach,
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 43
different emotions with the same valence (e.g., anger and frustration) may
differentially impact (dis)satisfaction and henee, more insight into the specifíc
antecedents, phenomenology and consequences of different emotions is needed.
The specifíc-emotion approach is strongly based on appraisal theories (for a
recent review see Scherer et al., 2001; and Bagozzi et al., 1999). A distinctive
feature of appraisal theories (e.g., Roseman, 1991) is their specifícation of the
conditions leading to discrete emotional responses. Thus, emotions are said to
have a specifíc referent (e.g., a customer gets angry with a flight delay), and
arise in response to appraisals one makes for something of relevance to one's
well-being (Bagozzi et al., 1999). By appraisal, we mean an evaluative
judgment and interpretation thereof. By something of relevance, we mean an
incident or episode that happens to oneself (e.g., a flight delay or a banking
failure). Therefore, different people can have different emotional reactions (or
no emotional reactions at all) to the same event or happening (Bagozzi et al.,
1999).
Specifícally, Nyer (1997) addresses appraisal theories and their role in postconsumption responses and he fínds that such post-consumption responses (e.g.,
repurchase intentions) are best predicted by using measures of (dis)satisfaction
plus measures of other emotions. Moreover, previous research has examined the
role of specifíc emotions in customer (dis)satisfaction formation (e.g., Dubé and
Maute, 1996; Folkes et al., 1987; Taylor, 1994).
Given that our goal is to analyze specifíc evaluations and behaviors that
customers may form or take in response to an adverse experience, we choose
the specifíc-emotion approach for the analyses carried out in the first two
studies (Chapters 4 and 5). This variable is not addressed in the third study
(Chapter 6).
Additionally, from the wide range of specifíc emotions that can be related to
service encounters, in this dissertation we will focus on anger, as the most
frequent emotional reaction provoked by service failures (Folkes et al., 1987;
Weiner, 2000). In fact, Zeelenberg and Pieters (2004) addressed that the study
of anger may be very relevant in the context of service failures as an emotion
experienced often in response to failed service encounters.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
44 • Chapter 2
2.4.4 Perceptions of justice
The perceptions of justice are the final group of antecedents of customer
(dis)satisfaction with service failure we center on.
The concept of justice (fairness) is based on social psychology and has been
widely and successfully employed to explain individuáis' reactions to a variety
of conflict situations (Blodgett et al., 1997). Fairness principies have been
applied in the context of pay raises (Folger and Konovsky, 1989), hiring and
promotions decisions, labor relations (see Greenberg, 1990), legal settings
(Thibaut and Walker, 1975), and more recently in buyer/seller transactions
(Clemmer, 1993; Oliver and Swan, 1989). These studies have demonstrated that
fairness has both psychological (e.g., (dis)satisfaction) and behavioral outcomes
(e.g., repurchase intentions).
Building upon the literature in social psychology and organizational
behavior, Clemmer and Schneider (1996) suggest that there are three
dimensions ofjustice: the perceived fairness of the tangible outcome or decisión
(i.e., distributive justice; Homans, 1974), the perceived fairness of the
procedures used in arriving at that outcome (i.e., procedural justice; Thibaut and
Walker, 1975), and the perceived fairness of the manner in which the customer
was treated throughout the conflict resolution process (i.e., interactional justice;
Bies and Shapiro, 1987).
In the context of service failure and recovery encounters, customers want
justice and fairness in handüng their complaints. Thus, customers weigh their
inputs against their outputs when forming recovery evaluations. Inputs can be
described by the costs associated with the service failure including economic,
time, energy, and psychic (cognitive) costs. The sum of the inputs is compared
to the sum of the outputs which includes the specifíc recovery tactic (e.g., cash
refund, apology, replacement, etc.), the manners of personnel, the service
policies developed to handle such situations, and the image associated with
responsive organizations (Kotler, 1997). The perceived justice component of
equity theory would then lead customers to ascertain whether the recovery
strategy offered was fair or just (Sheppard et al., 1992). Henee, service failure
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 45
and recovery researchers have documented three specific types of justice that
customers are looking for following their complaints: outcome/distributive
fairness, procedural fairness, and interactional fairness (e.g., Tax et al., 1998;
Tax and Brown, 1998).
Therefore, in a customer complaint context, the outcome/distributive
component refers to the perceived fairness of the redress offered by the service
provider and whether this outcome/output (redress) offsets the inputs (costs) of
the service failure (Greenberg, 1990). Procedural fairness refers to customers'
expectations of fairness in terms of policies, rules, and timeliness of the
complaint process (Kelley et al., 1993). Finally, interactional fairness
encompasses the manner in which the retailer responds to the customer's
complaint (Blodgett et al, 1993).
Perceptions of equity or fairness are addressed exclusively in the second
study (Chapter 5), in relation to failed service recovery encounters after service
failure (i.e., double deviation contexts). Additionally, based on a review of the
existing literature (e.g., Chebat and Slusarczyk, 2005; Smith and Bolton, 2002)
and following affect control (Heise, 1979, 1989a,b; MacKinnon, 1994) and
appraisal (Folkman and Lazarus, 1985) theories, we will center only on the
distributive component of justice. In Chapter 5, we will thoroughly justify that
distributive justice is the more relevant and appropriate justice dimensión for
examining the joint effects of justice and the specific emotion of anger on
(dis)satisfaction formation in double deviation contexts (the specific study
context).
Next, we examine the outcomes of customer (dis)satisfaction in service
failure encounters that are adapted to the specific context of the three empirical
applications carried out in this dissertation.
2.5 Outcomes of customer (dis)satisfaction in service failure encounters
As we have proposed in the introduction chapter, when a service failure
occurs, customers may respond in a variety of ways. They may start a process
which begins when they have evaluated a consumption experience and ends
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
46 « Chapter 2
when they have completed all behavioral and/or non-behavioral responses to the
experience (Day, 1980).
Following the framework proposed by Singh (1988) to analyze the ways
used by customers to express dissatisfaction after a service failure, three types
of responses are likely to occur (see Figure 2.1): prívate, voice, and third-party
response. We employ this framework to contextualize the three empirical
applications carried out in this dissertation in a clear and comprehensible
manner. Each of these applications will study different aspects of the
dissatisfaction response. Additionally, this categorization has been used and
supported by other researchers in different contexts (e.g., Maute and Forrester,
1993; Harrison-Walker, 2001).
Figure 2.1
Dissertation contextualization adapted from Singh's (1988) taxonomy
<
Dissatisfaction occurs
SERVICE FAILURE
PRÍVATE
RESPONSES
(e.g., repurchase
intentíons)
^
y
VOICE
RESPONSES
(e.g., redress seeking)
J
Specific context of
Chapter 4
K.
-y
J
Specifíc context of
Chapter 5
THIRD PARTY
RESPONSES
(e.g., complain to a
public agency)
^
y-
J
Specifíc context of
Chapter 6
Amount ofeffort involved in complaining
+
In the following subsections, we examine the framework proposed in the
specifíc and corresponding application's context, that is, the studies presented in
Chapter 4, Chapter 5, and Chapter 6.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 47
2 5.1 Prívate responses (Chapter 4 context)
Thefirstelement of Singh's (1988) taxonomy, private responses, defines the
context of the study carried out in Chapter 4.
Those consumers choosing to take some action as a result of an
unsatisfactory experience tend to take actions of a personal or private nature. As
a result, service managers typically grossly underestimate both the frequency of
customer dissatisfaction and the extent of the detrimental effects it triggers (Day
et al., 1981). Especially private boycott or switching behavior, an option that
typically requires little effort on the part of the consumer, can have a damaging
effect. Private complaining (negative word-of-mouth) works in an indirect way
but can also have serious effects, especially for producís that consumers
perceive as risky in a physical, social, or economic sense (Day et al., 1981).
Consumers who have heard about negative experiences are more likely to be
wary of certain products/services the next time they evalúate purchase
alternatives (Broadbridge and Marshall, 1995).
There are three main reasons for customers' reluctance to complain formally
and engagement in private complaining: (1) with purchases of relatively minor
valué, customers feel that complaining simply is not worth their time or trouble;
(2) they feel their complaints will not result in a favorable outcome; and (3) they
often do not know where and how to complain (Bearden and Teel, 1983; Day et
al., 1981).
In sum, private responses can mask problems that need attention and should
be corrected (Bearden and Teel, 1983). Specifically, they limit the utility of
complaint data as market performance indicators and as pre-purchase
information source, masking áreas in which public politics' actions are needed
(Bearden and Teel, 1983). When complaints are encouraged, then the seller has
the opportunity to remedy them (at least, the justified ones) and recover
customers that would have lost otherwise, and additionally, negative word-ofmouth is avoided or limited (Richins, 1983).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
48 • Chapter 2
In Chapter 4, we focus on two different 'prívate' responses: repurchase
intention and complaint intention. Repurchase intention can be viewed as a
proxy of effective boycott or switching behavior of the brand or product. That
is, we assume that customer's expressed intention to repurchase reflects his/her
future behavior. Complaint intention is not exactly a prívate response in the
terms expressed by Singh (1988). Rather, it reflects a propensity to lodge a
complaint to the service firm and thus, give the firm the opportunity to recover
the dissatisfíed customer. However, as long as this variable is measured as an
intention and not as a behavior, we consider it as a 'prívate option'.
2.5.2 Voice responses (Chapter 5 context)
The second element of Singh's (1988) taxonomy, voice responses, defines
the context of the study carried out in Chapter 5. Specifically, we center on the
voice response of redress seeking.
Redress seeking is one of the forms of response to dissatisfaction which are
more familiar to retailers and manufacturers than the less visible and indirect
prívate responses. In fací, for the great majority of marketing firms, the only
information on consumer responses to dissatisfaction available to management
is data on redress-seeking attempts and complaint letters (phones, or e-mails).
However, only a small percentage of dissatisfíed customers ever communicate
with the store (Day et al., 1981). Additionally, not all businesses display equal
interest in responding to consumer complaints (Estelami, 2000). Many firms
make the process diffícult which affects the customer's perceptions of the
organization's responsiveness (Mitchell, 1993) and, probably, encourage prívate
responses. This behavior has been shown to be contrary to the own interest of
the company because sometimes the mere act of complaining, even if it does not
resolve the customer's problem, could significantly increase product loyalty
(Scaglione, 1988). In any case, when a customer seeks redress, the company
may adopt a range of different responses (including no action). This procedure
or range of actions is known as service recovery.
The term 'recovery' originated with Donald Porter of British Airways who
explained:
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 49
'Recovery' was the term we coined to describe a very frequently repeated
concern: Ifsomething goes wrong, as it often does, will anybody make a special
effort to set it right? Will someone go out ofhis or her way to make amenas to
the customer? Does anybody make an effort to offset the negative effects ofa
screw-up? Does anybody even know where, when, or how to deliver a simple
apology? (Zemke and Schaaf, 1989: 22)
Therefore, service recovery is deñned as the actions a service provider takes
in response to a service failure (Gronroos, 1988) or to the activities that are
performed because of customer perceptions of initial service delivery falling
below the customer's 'zone of tolerance' (Zeithaml et al., 1993). This zone
represents the difference between the customer's desired level of service quality
and the adequate level from his/her point of view10.
Since customers are likely to react strongly to service failures, it is critical
that organizations' efforts to recover from failures be equally strong and
effective. Consequently, effective recovery from service failures should be
looked upon as a strategically critical part of a fírm's overall customer
orientation (Bell and Zemke, 1987). Even more, service recovery can be an
opportunity to communicate commitment to customers and strengthen their
loyalty to the fírm (Berry and Parasuraman, 1991), leading to a phenomenon
known as the 'service recovery paradox'. This paradox proposes that customers
whose service failures have been satisfactorily remedied are more satisfied,
more likely to remain loyal, and more likely to engage in favorable word-ofmouth about the company than customers who have never experienced a failure
(Hart et al., 1990; McCollough and Bharadwaj, 1992).
Service recovery is a much encompassing activity than complaint handling, because it includes
situations in which a service failure occurs but no complaint is lodged by the customer (e.g., when
front-line service personnel recognizes and/or acknowledges the failure). Therefore, the 'trigger'
for service recovery should properly be viewed as the failure incident, not the formal complaint.
In other words, a complaint is a sufficient but not necessary condition to actívate a service
recovery effort (Smith, 1997, p. 5). Despite this, most recent studies that have investigated the
impact ofa firm's complaint handling efforts on customer evaluations are based solely on those
failure/recovery situations in which customers fíled a formal complaint with the organization
(e.g., Blodgett et al., 1993). This is actually the approach we follow in this dissertation.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
50 • Chapter 2
Additionally, a good service recovery system facilitates the tracking of
failures and the development of databases to gain insight into failures in order to
deal with them and try to prevent them from happening again (Lewis and
Spyrakopoulos, 2001). Other advantages of a good service recovery system are
the increased opportunities for cross-selling to retained customers (Armistead et
al., 1995), the reduction of perceived risk for new customers, and the
enhancement of company image of both employees and customers. Probably
though, the most important benefít of effective service recovery is the
prevention of customer defection to other providers (Lewis and Spyrakopoulos,
2001). Moreover, firms can actually evalúate their complaint-handling programs
as profít centers by weighting the costs of soliciting and handling complaints
against either the long-term valué of loyal customers or the marketing costs of
winning new ones (Scaglione, 1988). Thus, Fornell and Wernerfelt (1987) show
that defensive marketing (e.g., complaint management) can lower the total
marketing expenditure by substantially reducing the cost of offensive marketing
(e.g., advertising). The savings in offensive marketing are often high enough to
offset the additional costs associated with compensating complaining
customers, therefore, the authors suggest that complaints from dissatisfíed
customers should be maximized subject to certain cost restrictions.
However, the management of service recovery efforts receives little attention
in most firms and is often completely ignored (Heskett et al., 1990), even when
excellent recovery management, on the one hand, has a disproportionately
significant positive impact on customers' perceptions of service quality (Berry
and Parasuraman, 1991) and satisfaction (e.g., Smith et al., 1999); and, on the
other hand, has been proved to affect service organizations' profíts in a positive
way (Reichheld and Sasser, 1990).
Henee, not having service recovery strategies or having them but working
inefficiently, can have serious consequences. A poor recovery following a
service failure can lead dissatisfíed customers to become 'terrorists', looking
actively opportunities to openly criticize the company involved (Tax and
Brown, 1998). In a study on customer switching behavior in a wide variety of
service industries, Keaveney (1995) fínds that service failures and failed
recoveries account for almost sixty percent of the critical behaviors by service
providers that lead directly to customer switching. Additionally, previous
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Service failures: theoretical considerations • 51
research has shown that poor service recoveries exacérbate already low
customer evaluations following a failure, producing a 'double deviation' effect
(Bitner et al., 1990; Hart et al., 1990; Johnston and Fern, 1999; Mattila, 2001b).
These 'double deviations' have been shown to occur even in múltiple service
failures and recovery efforts (Maxham III and Netemeyer, 2002). Hart et al.
(1990) show that of the customers that do complain, more than half of attempted
recovery efforts only reinforce dissatisfaction (Hart et al., 1990). We think that
this is a sufficiently high number to justify a deeper analysis of the underlying
mechanisms which contribute to (dis)satisfaction formation in double deviation
scenarios.
Therefore, our main goal with the study in Chapter 5 is to add to these
previous works that illustrate the importance of an efficient recovery process for
companies. However, we propose a new approach to this issue by examining the
harmful consequences of failed recoveries after service failures, that is, by
examining the customer (dis)satisfaction formation in double deviation
scenarios.
2.5.3 Third-party responses (Chapter 6 context)
The third element of Singh's (1988) taxonomy, third-party responses,
defines the context of the study carried out in Chapter 6.
Third-party responses involve, for example, complaining to the media,
registering a complaint with a consumer association, or bringing legal action
(Day and Landon, 1977; Singh, 1988). Previous studies show that only a small
percentage of complaints are lodged with third parties (Day and Landon, 1977)
and additionally, many outlets for third party complaints, such as the Better
Business Bureau, require that a customer first files his/her complaint with the
provider before filing with the third party11.
11
This is also true for the complaints lodged to the Bank of Spain's Complaints Service
(the third party in the context of Chapter 6).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Volver al Indice/Tornar a l'índex
Chapter 2
In general, third-party comnla"
Ímp0rtant and
to marketers since they represen, T ^ ^
troublesome
hl
friends, family, t h e s a l J
™ ? 8her-order action than complaining to
and involvement associated with htH C ° m P a n y ( F d c k ' 1 9 8 7 ) " T h e h i § h e f f o r t
P a ñ y com la
2-1) normally indicates a de
P ining behavior (see Figure
dissatisfacti
unresponsiveness or related ¿ 1 1- T*™"
™> 'ompany
C M SCVerely t h r e a t e n
relationshipsandeffectiveness In 1
marketing
0n third partv
extensive legal costs, r e ^ t " ,
'
complaints can result in
g
l n t e
(Tipper, 1997).
^
™ ^ and corporate reputation problems
The study carried out in Cha
Party complaints on corran *
I
e
mpirical]y analyzes the impact of third-
attention to be given to ttoM-niT
*' j ° i n i n g t h e d e m a " d s for greater
G pUánbl
made b
Fisheretal., 1999; Singh 1 9 8 9 ^ 7 ^
y various authors (e.g.,
Y
P r P S e t h a t t h e relea
mformation about third-partv caJT
' ^
° °
se of the
eCOnomicall
market, reducing the stock price of ^
*
y «leva* to the stock
mS Í n v o l v e d in
Complaints published by the B .
the Annual Report on
Additi
explanatory p 0 W er of the fol.n
^
°nally, we examine the
lanches, quality corporate i m Z Z
Previous chapters, where the m S
were customer-related, these thre,
^f^'
T
"
Conc
^
° f C ° m p l a i n t s Per
« i o n . Contrary to the
m V 1 V e d ¡n t h e
°
different
models
investor's perspective adopted ' n
"*
> « une with the
not new in the dissatisfaction l i t e r J ^ n * * USe ° f firm-sPecific variables is
effect of market structure variabTe
u
" S I e s e a r c h h a s investígate! the
l0 Se mono ol
and Andreasen, 1977) and indusm, ,
^ °
P y conditions (Best
TARP, 1979) on d i s s a t í s l á c t i ^ r C e ^ Q n ( F ° m e 1 1 a n d R ° b i n s 0 n > 1 9 8 *
n reSponses
(°ubé a n d Maute, 1996).
fírm s p e c i f í c
Now we have examined the
•
ÍSSUCS r e a r d i n
Judies carried out in this d i s s e r t a Z " ^ T * ^
§
8 the three
Chapter w e
summary 0 f them. Each will e l a h l J ' "
^
P r ™de * global
C n t e x t s ecifíc
based upon this general overview
" ^ °" * * °
P
variables,
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Chapter 3
Summary and description of the empirical applications
The main objective of this short chapter is to provide a quick but complete
view on the three empirical applications carried out in the next chapters. Table
3.1 summarizes the variables/measures employed, whereas Table 3.2 provides a
summary of the three empirical applications, in terms of subject, theoretical
background, data sources, etc.
Thus, this dissertation deals with the important issue of service failures from
three different perspectives in three different empirical applications. The fírst
study (Chapter 4) analyzes the specifíc service failure of a flight delay and the
subsequent cognitive and affective reactions which affect customer
(dis)satisfaction and subsequent behavioral intentions.
The second study (Chapter 5) examines the specifíc service failure of a
failed recovery in the banking industry. Again, subsequent cognitive and
affective reactions are analyzed as well as their impact on customer postrecovery (dis)satisfaction.
Finally, the third application (Chapter 6) investigates what happens once
recovery instruments failed to recover customers and they decide to go to third
parties to express their dissatisfaction. We show that this has an effect on
market valué (stock prices) and we try to explain this effect through different
fírm and market variables, such as number of complaints received, corporate
reputation, and market concentration. With this last application, we also
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
54 • Chapter 3
contribute to the research that tries to link marketing and finance in order to
better understand the consequences of firm's strategies and consumer behavior.
Chapter 7 finally provides a summary of the main theoretical and managerial
contributions, limitations, and directions for future research, of the three
empirical applications presented in the previous three chapters.
Table 3.1
Summary of the variables/measures employed in the three empirical
applications
Chapter 4
Chapter 5
•
s
Service features
Product relevance
Waiting time/delay
Causal attributions
Controllability
Stability
Customer emotions: anger
Perceptions of justice: distributive
justice
Customer (dis)satisfaction
Prívate responses
Repurchase intentions
Complaint intention
Voice responses: redress seeking
Recovery strategies
Third-party responses
Number of complaints per branches
Quality corporate image
Target market concentration
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
V
S
V
V
Chapter 6
•/
• /
•/
V
• /
•
V
•/
•
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Volver al Indice/Tornar a l'índex
Summary and description ofthe empirical applications • 55
Table 3.2
Summary ofthe topics discussed in this dissertation
Chapter 4
Chapter 5
Chapter 6
Snbject
Service customers
Service customers
Investors
Typeof service
encounter
Service failure
encounter
Service failure and
recovery encounter
Service failure and
recovery encounter
Type of service
failure
Flight delays (airline)
Banking failures
Banking failures
Type of response
according to Singh's
(1988) classification
Prívate
Voice
Third-party
Timing ofthe
(dis)satisfaction
response
After service failure
After service
recovery
Theoretical
background
Services marketing
literature
Attribution theory
Emotion literature
Satisfaction literature
Services marketing
literature
Attribution theory
Justice literature
Emotion literature
Satisfaction literature
Services marketing
literature
Financial economics
literature (portfolio
theory)
Signaling theory
Data sources
Customer survey
Customer survey
Company data
Stock market data
Service setting
Airline industry
Banking industry
Banking industry
Analysis technique
Structural equation
modeling
Structural equation
modeling
Event study
methodology
(Moderated)
Regression analysis
(OLS)
Bootstrap analysis
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Chapter 4
The consumer's reaction to delays in service12
4.1 Introduction
Waiting for service is a negative experience that arises in many purchase
situations and seems to be a decisive element, not only with regard to the
customer's level of satisfaction with the service but also regarding his loyalty to the
service company (Pruyn and Smidts, 1998). As society becomes more complex
and busy, the majority of people have less time. Therefore, the speed at which
service is dispensed is becoming one of the service's main attributes (Katz et al.,
1991).
We define the wait for service as the time from which a customer is ready to
receive the service until the time it begins (Taylor, 1994). Focusing on the point of
time at which the wait begins, Dubé-Rioux et al. (1988) suggest that the 'preprocess' wait, which occurs before the transaction of the service (e.g., before one
orders at a restaurant), is more unpleasant than the wait one experiences 'inprocess' (e.g., after having placed the order), or than the 'post-process' wait (e.g.,
waiting to pay the bilí). Within the pre-process stage, Taylor (1994) considers the
so-called 'delay' concept (post-schedule waits13), that is, the wait from the fírst
moment at which the service was scheduled to begin. This type of wait (i.e., delay)
is the focus of the present research.
12
This chapter is based on the study of Casado-Díaz and Mas-Ruiz (2002).
13
Other possibilities would be the "pre-scheduled" waits (a customer waits because he/she has
arrived early for a scheduled event) and the "queue" waits (there are no appointments or
scheduled commencement times but the service is províded on a "first-come-first-served" basis).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
The consumer's reaction to delays in service • 57
Henee, previous marketing research focusing on delays has analyzed the
variables that influence the perception of delays and the impact of delays on the
customer evaluations of a service. In this sense, the contribution made by the
attribution theory, which suggests a basic formula, or an 'attribution-affectbehavior' sequence14, has been noteworthy. Thus, when a consumer is faced
with a disconfirmation of expectations situation with the way a service is
provided, he/she diagnoses the underlying causes and, depending on their
nature, assumes a certain level of emotion which affeets his/her subsequent
behavioral intention regarding the service company.
Nevertheless, Price et al. (1995) indícate that the emotional responses to
services have not been thoroughly researched in these patterns and, when they
are, above all they lump together a large number of emotions in a single global
dimensión (e.g., Taylor and Claxton, 1994) or they distinguish satisfaction as a
consequence of other globally considered emotional responses (Oliver, 1989,
1994; Yi, 1990). Weiner (2000) goes further in his reflections concerning
emotional responses and suggests that in situations of negative service results
where negligence is attributed to the service provider (control attributed to the
causes of the failure), the emotional reaction will not be a simple dissatisfaction.
Probably, anger, a negative emotional reaction related with moral judgment
(moral outrage), will be demonstrated due to attributing to the company the
causal control of the negative service result (Weiner, 2000). Consequently, the
study of anger is especially interesting as a central element of social behavior.
Despite this, we have not found any empirical research about service delay
which examines simultaneously both anger and satisfaction with service.
Therefore, the goal of this article is to examine the relationships that exist
among the attributions, anger, satisfaction with service, and behavioral
intentions of customers who experience a service delay. The empirical
application was carried out on a sample of passengers who suffer delays in their
flight schedules at Alicante International Airport (Spain) during 1998.
14
Although the causal ordering of the cognitive component and affect is a controversial topic
(Bagozzi, 1996), an important stream of research (Arnold, 1960; Lazarus, 1982; Lazaros et al.,
1980; Weiner et al., 1979, among others) argüe that affect is the product of cognitive processes.
Basically, affect originates as a function of the individual's evaluation of the intention, the causes,
the consequences and the personal implications of a particular incentive (Westbrook, 1987).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
58 • Chapter 4
The following section reviews the literature on delays in service and presents
the hypotheses of this research. Next, we describe the methodology. The results
obtained are discussed in the fourth section and, finally, we present the main
conclusions.
4.2 The modeling of the service delay evaluations and the hypotheses
Weiner's (1985) attribution theory provides a useful framework to examine
the reactions of customers who suffer service delays. It suggests that the
customer tends to attribute the delay to different reasons that can be classified
according to the two clearly different dimensions proposed in our study:
controllability and stability. The former dimensión reflects the level of control
(total-nuil) over the causes of the delay that the consumer usually attributes to
the company. It includes the actions taken by the firm to correct the delay
(service failure). Stability, on the other hand, refers to the degree of permanence
in time (transitory-durable) that is attributed to the perceived cause of the delay.
In general, literature in consumer behavior (Folkes et al., 1987; Oliver and
DeSarbo, 1988) and psychology (Weiner, 1980) suggest a basic formula
'attribution-affect-behavior' instead of the sequence 'affect-attributionbehavior' implied in the original research on satisfaction (Folkes, 1984;
Krishnan and Valle, 1979; Valle and Wallendorf, 1977). In this new context,
Bitner (1990) places the concept of attribution in the paradigm of
disconfirmation, situating it after disconfirmation and before satisfaction. Thus,
prior to customers determining their level of satisfaction, they will diagnose the
possible causes of disconfirmation and depending on the perceived nature of
these, customers will modify their level of satisfaction and subsequent behavior
(propensity to complain and repurchase intention).
In fact, and despite the apparent affective nature of satisfaction, it has only been
recently that researchers, in regards to satisfaction, have more carefully considered
the emotional components of this concept and their interrelationship with other
measures of more specific affect (Mano and Oliver, 1993; Szymanski and Henard,
2001). Oliver (1989) points out that generalized affect ('good for me, bad for me')
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
The consumer 's reaction to delays in service • 59
and specific emotions resulting from judgments of attribution derived from
disconfirmation are clearly an antecedent of satisfaction. As expressed by Yi
(1990, p. 70), "satisfaction results from processing affect in an experience of
consumption". However, the majority of previous works examine affect as just one
indicator which encompasses múltiple emotional reactions. Additionally, Weiner
(2000) proposes the need to distinguish anger as a fundamental emotional reaction
which determines social behavior.
In any case, insufficient attention has been given by empirical research into the
concepts of anger and satisfaction with service in the context of the sequence
'attribution-affect-behavior'. Therefore, our main goal is to examine the existing
relationships between these elements as well as their relationships with other
variables such as the importance of punctuality (Folkes et al., 1987; Taylor, 1994)
and the perceived waiting time (Maister, 1985; Taylor, 1994, 1995). This is done
with the aim of better understanding the customers' reactions following service
delays.
In Figure 4.1 an overview of relevant variables and their relationships is given.
Figure 4.1
Proposed model of antecedents and consequences of satisfaction with service
failure (delay)
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
60 • Chapter 4
4.2.1 Attribution theory: attribution of control
Control attribution-¿Anger-¿Satisfaction
with service
Bitner (1990) states that the causal control attributions to disconfírmation
constitute a factor that affects customer satisfaction. In this sense, the customers
experience dissatisfaction when they perceive that the company can control the
causes of the service delay. However, Taylor (1994) indi cates that anger is the
dominant emotional reaction when a negative event is attributable to a
controllable cause (Meyer and Mulherin, 1980; Reisenzein, 1986; Weiner,
1980, 1986). However, Weiner (2000) considers that in situations of service
failure, where negligence is attributed to the service provider, customer's
emotional reaction will not be one of simple dissatisfaction but one of anger,
which reflects the moral outrage generated by this. That is, the negative result is
attributed to something that the company should have controlled and it is
therefore considered that the moral code of conduct has been broken. Therefore,
we propose the following:
Hl. Anger mediates the relationship between control attribution and satisfaction
with service.
Control attribution-¿Anger—¿Propensity to Complain
Curren and Folkes (1987) have observed that when the problems with the
service and their solution are attributed to controllable causes customers want to
complain and, in this way, demónstrate their punishment to the company as well as
their desire to prompt the necessary efforts to sol ve the problem. Folkes et al.
(1987) further propose that the attributions of control over the service delay
directly determine the customer's complaint. Nevertheless, Weiner's (2000)
proposal goes directly against the previous direct effect. He considers that in a
service failure context, the customer starts an attributional process with a sequence
'think-feel-act', which highlights the mediating role of emotions. First, the
individual tries to detect attributions of control over the causes of the failure
('think'). Second, if these are demonstrated, he/she will show a feeling of anger
('feeP), which will subsequently have an effect on behavioral intentions, such as
propensity to complain ('act'). In other words, the control attribution has an
indirect effect on the propensity to complain that is influenced by anger (Folkes et
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
The consumer 's reaction to delays in service • 61
al., 1987; Richins, 1980), in the sense that "anger implies that the individual will
dedícate a greater effort to confront a particular situation or problem" (Smith and
Ellsworth, 1985, p. 815). Therefore, we propose:
H2. Anger mediates the relationship between control attribution andpropensity to
complain.
Control attribution—>Anger—>Repurchase Intentions
Following Brickman et al. (1982) and Hamilton (1980), the customer might
penalize the fírm by no longer using its services once he/she has perceived that the
delay was controllable, or when he/she believes that the company could have made
decisions that would have solved the problem. Consistent with this perspective,
Folkes et al. (1987) assume that the perceived control has a direct effect on
repurchase intentions. However, from the perspective of 'think-feel-act' (Folkes,
1984; Weiner, 1985, 2000), the control attribution can have an indirect impact on
repurchase intentions through the mediation of an emotional reaction such as
anger. Therefore, the customer's controllability perception regarding the control
the company had on the service delay could cause an increase in the customer's
anger. Then, since a typical result of anger is the denial or removal of some benefít
enjoyed from the company involved, a lower desire to repurchase would be the
result. Therefore, we assume:
H3. Anger mediates the relationship between control attribution and repurchase
intentions.
4.2.2 Attribution theory: attribution of stability
Stability attribution—>Satisfaction with service
Causal stability involves perceptions of a cause's constancy over time. In
this sense, the attributions of stability for the perceived causes of the delay play
an important role in the judgment of the customers who suffer a service failure
(Smith and Bolton, 1998). Originally, Weiner et al. (1982) and Folkes et al.
(1987) found that stable causes increase anger more than do temporary causes in
an achievement situation. In this sense, stability may influence anger toward the
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
62 • Chapter 4
fírm. However, Weiner (2000) has modified his theory regarding the
relationship between stability and anger. Basically, the customers who attribute
the failures to permanent and stable causes feel surer that identical results will
be produced again in the future (Weiner, 1986). Consequently, Weiner (2000)
indicates that stability attribution generates within the individual other types of
emotional reactions more related with expectations, such as the hope or fear that
identical results will or will not be produced again. Bitner (1990) and Smith and
Bolton (1998) also assume that the stability in the cause of the failure
determines (dis)satisfaction. Therefore, we propose:
H4. The customer will experience lower satisfaction when the cause ofthe delay
isperceivedas stable (le. likely to recur).
Additionally, previous literature has examined diverse influences on the abovementioned relationships in delays in service, such as the perceived waiting time
and the importance of successful service performance. Next, we examine the role
of these variables in our proposed model.
4.2.3 Perceived waiting time and importance of successful service
performance
Following Pruyn and Smidts (1998), the waiting time is an important factor
which affects the customers' evaluation of the wait. These authors also make a
distinction between the real or objective waiting time and the perceived or
subjective duration ofthe delay. They suggest that the customer's response is more
affected by the subjectively perceived time than by the real or objective time.
Therefore, we will center on the perceived waiting time. One implication of the
perceived waiting time is its link with emotional reactions (Hui et al., 1998) such
as anger.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
The consumer 's reaction to delays in service • 63
Perceivedwaiting time->Anger
We propose that the (perceived) duration of the delay could lead to customers'
anger15. Anger arises not only because customers assume that the service provider
has failed in executing a relatively implicit promise, regarding the specified time of
the service (Taylor, 1994), but also because the delay could causefinancialcosts or
social inconveniences (Becker, 1965; Maister, 1985). Furthermore, the injustices
perceived during the wait, such as the probable violation of the principie of 'first
come-first serve', contributes to an increase in anger (Maister, 1985). Henee, we
hypothesize the following:
H5. An increase in perceived waiting time will result in an increase in anger.
Regarding the importance of successful service performance, punctuality is a
decisive element in the context of service delays (Folkes et al., 1987).
Consequently, the relationships that we examine next take as their starting point
tiie importance of punctuality.
Punctuality importance-¿Anger
Some authors suggest that the importance given to a service's successful
performance may influence affective reactions (Brown and Weiner, 1984; Folkes
et al., 1987). Basically, anger due to a service failure (e.g., a flight delay) increases
relative to the importance placed on its successful performance (e.g., punctuality
importance) (Folkes et al., 1987; Taylor, 1994). Therefore, we propose:
H6. The higher the importance of punctuality for a customer, the greater the
feelings of anger with delays.
15
Despite this, Pruyn and Smidts (1998) demónstrate that perceived waiting time generates a
negative evaluation of the wait (judgments about the shortness or length of this), but do not affect the
emotional response. Even Baker and Cameron (1996) propose an inverse relationship where affect
has an effect on the perceived waiting time.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
64
• Chapter 4
Punctuality importance—¿Perceived waiting time—>Anger
Previous literatura suggests that there may not be one-to-one correspondence
between perceived waiting time and affective response (Hui et al., 1998). Due to
the influence of various environmental16, situational, and personal variables,
perceived waiting time may not be able to explain all the effects of a delay on
service evaluations. The customer may accurately estímate or even underestimate
the length of the wait, yet his/her affective responses to the wait could be
extremely negative. Conversely, the customer may grossly overestimate the length
of a wait but may still exhibit positive affective responses to the wait. Therefore,
Hui et al. (1998) propose that the perceived waiting time is a mediator of a delay's
impact on service evaluation. Following this proposal, the perceived waiting time
could medíate the relationship between punctuality importance and anger. That is,
when punctuality is significantly important to someone, perceived waiting time
can genérate higher levéis of anger. Specifically, if someone is under time pressure
(i.e. importance of punctuality) then the waiting time might be seen longer, which
may subsequently increase anger. But when punctuality is not critical (and it will
not be for all people in all circumstances -e.g., some leisure travelers) then the link
between perceived waiting time and anger would be weaker. Thus, we hypothesize
the following:
H7. The perceived waiting time mediates the relationship between importance of
punctuality and anger.
16
The service environment can influence the 'perceived waiting time-»anger' relationship which,
in the framework of services marketing (Booms and Bitner, 1981), represents a distraction during
waiting time (Baker and Cameron, 1996; Taylor, 1994, 1995). On the one hand, the service
environment (measured in terms of facilities and entertainment) can reduce anger when it induces
distraction in the consumer and 'filis' their waiting time (Baker and Cameron, 1996; Maister,
1985). On the other hand, the service environment can affect the perceived waiting time by means
of distraction, in the sense that an 'empty' waiting time, without distractions, would seem longer
than it would were it 'filled' with something interesting (Baker and Cameron, 1996). However, a
previous analysis demonstrated that these relationships are not attained in the particular case of
our work, which can be explained by the impact of other unexamined factors. Also, Berlyne
(1971) indicates that the service environment only captures the attention of the individual and
positively influences the feeling about and perception of the waiting time when the facilities are
well valued and induce some modérate excitement. If the level of excitement reached is high or
the consequent evaluation proves to be controversial or offensive, this could produce adverse
resulte for these facilities.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
The consumer 's reaction to delays in service • 65
4.2.4 Anger
Anger represents the customer's most frequent emotional reaction to a service
failure (Taylor, 1994). This dimensión has been recently examined by the literature
on service delays, linking it to (dis)satisfaction and complaint.
Anger—¿Satisfaction with service
Previous research has examine the role of emotions in satisfaction judgments
(e.g., Mano and Oliver, 1993; Westbrook, 1987; Westbrook and Oliver, 1991)17.
Thus, it seems that the emotions demonstrated during consumption leave affective
traces in the memory; traces that are considered by the customers to be
subsequently integrated into their evaluations of satisfaction (Westbrook and
Oliver, 1991). Additionally, Weiner (1986) suggests that the affect can be
dependent on the attribution (Oliver, 1993; Oliver and DeSarbo, 1988). The
attributions can evoke a specific emotion depending on whether the result of the
consumption experience is a success or a failure. Therefore, the positive/negative
affect is presented as another component of the post-purchase experience which
has a positive/negative impact on evaluations of satisfaction. In short, these
theories imply a relationship between emotions and satisfaction.
In the context of service waits, it has been suggested that the customer's
evaluation of the wait influences his/her level of (dis)satisfaction with the service
process (Hui and Tse, 1996; Pruyn and Smidts, 1993; Taylor, 1994, 1995).
Specifícally, Pruyn and Smidts (1998) fínd that the affective evaluation of the wait
has an effect on customer's satisfaction with the service. Therefore, we propose:
H8. Anger will be negatively related to satisfaction with service.
17
See Szymanski and Henard (2001) for a recent review of research linking affect and
satisfaction.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
66 • Chapter 4
Anger->Propensity to Complain
Previous research has shown that negative affective responses (e.g., anger) have
a direct impact on complaint behavior18 (Westbrook, 1987). More specifically,
anger can genérate a broad range of anti-company reactions. Thus, customers
complain with the aim of being compensated for their purchase (Weiner, 2000).
That is, anger implies that the customer makes an effort to confront the problem
(Smith and Ellsworth, 1985). Henee, we assume:
H9. Anger will be positively related to propensity to complain.
Anger—>Propensity to Complain is not entirely mediated by saíisfaction
with service
One of the most studied consequences of customer (dis)satisfaction is
complaining behavior in an attempt to remedy dissatisfaction (Prakash, 1991;
Yi, 1990). Although dissatisfaction is a necessary condition for complaining
behavior, it only explains a small percentage of complaints (Bearden and Teel,
1983; Blodgett et al., 1993; Day and Landon, 1976). In fact, not all dissatisfíed
customers complain (Warland et al., 1975) and conversely, satisfied customers
might complain in order to record minor problems (Jacoby and Jaccard, 1981).
Consequently, dissatisfaction with a service does not constitute a necessary
prerequisite for expressing complaints (Westbrook, 1987).
Therefore, the motivating source of complaints might not be dissatisfaction but
other negative emotions (Day, 1984) such as anger as previously indicated in the
Hypothesis H9. In any case, the existence of both alternatives suggests that
dissatisfaction with service could mediate the influence of the negative affective
response on complaining behavior since the negative emotional state precedes
dissatisfaction. Nevertheless, Westbrook (1987) points out that the outcomes of a
service can cause negative affect and give rise to complaints although the net
impact of such negative affect on dissatisfaction can be under-estimated by the
presence of other factors (such as positive affect and expectations, among others).
Therefore, we propose the following hypothesis:
18
"Post-purchase complaining behavior comprises consumer-initiated Communications to
marketers, their seller members or public agencies to obtain remedy or restitution for purchase- or
usage-related problems in particular market transactions" (Westbrook, 1987, p. 260).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
The consumer 's reaction to delays in service • 67
¡110. The relationship between anger and propensity to complain is not totally
mediated byjudgments of satisfaction/dissatisfaction with service.
Anger-¿Repurchase Intentions
Anger leads to a broad range of anti-company reactions with which the
customer intends to penalize the company such as lower repurchase intentions
(Weiner, 2000). Such a reaction might even become generalized towards other
services provided by the offending company. Therefore, we propose:
Hll. Anger will be negatively related to repurchase intentions.
4.2.5 Satisfaction with service
Customer satisfaction is derived from a given experience with a service
encounter and, subsequently, from the comparison of such an experience with a
given standard about which consensus does not exist: what was expected (Oliver,
1980), norms based on experience (Woodruff et al., 1983; Cadotte et al., 1987), or
what was desired (Spreng et al., 1996). Regarding the evaluation of the service
delay, we propose the following relationship:
Satisfaction with service—¿Repurchase Intentions
In general, satisfaction is thought to explain the likelihood that consumers will
purchase the service again (Szymanski and Henard, 2001). Oliver (1997) discusses
loyalty as a result of customer satisfaction and proposes three phases of satisfaction
(cognitive, emotional, and conative) which end in action loyalty. Therefore,
following Hirschman's (1970) exit-voice theory, higher customer satisfaction can
lead to higher customer loyalty (Fornell and Wernerfelt, 1987). Moreover, this
positive relationship between satisfaction and repurchase has been demonstrated
by Bearden and Teel (1983) and Fornell et al. (1996), among others. Henee, we
propose:
H12. Satisfaction with service is positively related to repurchase intentions.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
68 « Chapter 4
4.3 Methodology
In order to test the previous hypotheses, we conducted an empirical study. Path
analysis with LISREL 8.3 (Jóreskog and Sorbom, 1996) was used to test the
proposed model. We now discuss the sample and data collection procedure and the
measurement of the variables.
4.3.1 Sample and data collection
This study was conducted in the airline service because airline travel
represents a service in which failure is common (McCollough et al., 2000;
Taylor, 1994). Airline flights suffer delays of different length quite frequently
and for different causes (Taylor, 1994). This is an interesting example for the
analysis of the outlined objectives because the method used to collect the data
(supported by field work directed at a specifíc sample of passengers) allows us
to examine the reactions of customers to delays in service, thus increasing its
external validity. Additionally, using an airport served by numerous airlines also
eliminated any possible brand bias.
Data were collected from passengers waiting to board delayed flights of
different airline companies operating at the Altet International Airport
(Alicante)19. Individuáis were approached by interviewers before the new time
of departure was announced. Interviewers identified themselves as researchers
from a local university. A total of 171 passengers from 23 different delayed
flights accepted to particípate. The twenty-three delayed flights belonged to
eight different airline companies (forty companies were operating at the airport
when the survey was conducted). A 'delayed flight' was defíned as one whose
departure took place more than fifteen minutes later than was original ly
scheduled.
19
The study was conducted during the period 14* July to 11* August, 1998.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
The consumer 's reaction to delays in service • 69
4.3.2 Development of measures
We collected data from passengers' perceptions of the waiting time, the
attributions to the delay, the importance of punctuality, the level of anger, the
degree of (dis)satisfaction, and the behavioral intentions. The majority of variables
were measured with single-item scales instead of using multi-item scales as would
be desirable (Churchill, 1979). The reasons for this approach were the following
(Taylor, 1994): i) to obtain a questionnaire sufficiently short so that it could be
carried out in just a few minutes and to therefore, avoid possible annoyances to the
interviewees who were waiting for their flight departure; and ii) to facilítate the
completion of múltiple interviews in such a short waiting period (with a big risk
that the immediate departure of their flight could be announced during the carrying
out of the interview). In any case, measures used in previous research were
applied.
The attributions to the delay in service were classified according to two
perspectives: control and stability, in line with the proposals of Folkes et al. (1987)
and Taylor (1994). The control attribution, Xu (not only of the cause of the delay
but also of the solution adopted by the company), was indicated on the following
5-point semantic differential scales: i) Degree of control over the cause of the
delay by the airline company, in which a valué of 1 meant "totally uncontrollable"
and 5 "totally controllable"; and ii) control regarding the measures adopted by the
company to shorten the delay, with 1 meaning "it could not have done anything",
and 5 "it could have acted". Attribution of stability, X2, was measured with a oneitem scale and was intended to discover the perceived frequency of the delays of
the airline concerned, on a 5-point scale, where 1 meant "very unusual" and 5
"very frequent". The importance of punctuality in flight arrivals (X3), was
measured on a 5-point scale (Folkes et al., 1987), in which 1 meant "not at all
important" and 5 "very important".
The perceived waiting time (Yi) was established with the following open
question, measured in hours and minutes: How long do you thinkyou have been
waiting for since the indicated departure time?. The traveler's anger was measured
with a single-item 5-point scale (in line with the proposal of Folkes et al., 1987)
that indicated the degree of anger that was felt about the delay (Y2), in which 1
meant "not at all angry" and 5 "very angry". Concerning satisfaction, given the
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
70 • Chapter 4
lack of consensus in the existing literature about the number of experiences and the
standards for comparison on which this concept is based, we chose to consider
satisfaction with service failure. This is done within the traditional paradigm of
disconfirmation of expectations (about whose impact on satisfaction there is
considerable empirical evidence; e.g., Yi, 1990) which limits the experience to a
current situation of the purchasing (Woodruff et al., 1983) of the flight ticket and
its use. We employed a previously used scale (Spreng and Mackoy, 1996) which
measured the degree of general satisfaction, Y3, in an attempt to discover how the
traveller feels about the delay, on a scale of five points, where 1 means "not at all
satisfied" and 5 "very satisfied".
The behavioral intention is reflected by the two following 5-point scales
(Folkes et al., 1987): i) inclination to complain to the airline (Y4); and ii)
repurchase intention (Y5), defined as "not flying with the company again if one
had the choice". In the first scale, 1 indicated "very inclined to complain" and 5
"not at all", while in the other scale, 1 meant "I would not fly again" and 5 "I
would".
Finally, several variables, not directly associated with the hypotheses being
tested, were included in the study. These variables were: age, sex, frequency of
travel, destination, type of trip and flight, causes of the delay and whether or not
these causes were explained by the airline personnel.
The final set of items used to measure the components of the conceptual model
is shown in Appendix 4.1.
4.4 Results and discussion
In this section, we first address the sample characteristics and then present
and discuss the main results obtained.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
The consumer 's reaction to delays in service • 71
4.4.1 Sample characteristics
The profiles of the 171 valid interviewees were previously examined and in
their categorization we found a prevalence of men (57.9%), passengers who were
travelling accompanied (60.8%), and those who were flying on regular commercial
flights (64.3%). In most of the cases (45%), the passengers' ages were between 53
and 36 years, followed by those of between 18 and 35 (42.7%), those over 54
(9.4%) and finally, those who were younger than 18 (2.9%). As for frequency of
travel, 36.8% of the participants travelled frequently (more than six trips per year),
35.7% flew once a year, and 27.5% travelled moderately (two to fíve times a year).
The main reasons for traveling were for vacation (82.4%), business (11.7%), to
visit relatives/friends (3.5%), and for study (2.3%). The passengers, according to
this typology, did not show any significant differences in their control attributions
(ANOVA: F=1.90, p=0.\3) or their stability attributions (ANOVA: F=2.24,
¿?=0.08). However, we found significant differences in anger (ANOVA: F=4.92,
p=0.00). In fact, the 'holiday' passengers adopted a neutral attitude in anger
(mean=3.19; t=2.34; p=0.02) while 'business' passengers demonstrated greater
anger (mean=2.23; t=-3.62;/>=0.00).
Additionally, the countries of residence were: Spain (67.3%), Germany
(12.9%), Great Britain (11.7%), France (4.1%), and others (4.1%). In terms of this
variable (i.e., country of residence), the passengers show significant differences in
control and stability attributions (ANOVAs: F=3.89, p=0.00; F=5.02, p=0.00,
respectively), but not in anger (ANOVA: F=0.51,/?=0.73). The Spanish attributed
greater stability in the causes of the delay (mean=2.40; t=2.78; p=0.0l), and the
Germans less stability (mean =3.57; t=-4.48; p=0.0Q); while the British attribute
less control to the company (mean=7.13; t=-3.78; /?=0.00), and the French greater
control (mean =4.05; t=2.24;p=0.06).
Passengers blamed the delays on the following probable causes: the delay of a
previous flight (41.5%), mechanical problems (19.3%), airline employees fault
(5.8%), other passengers (0.6%), bad weather (0.6%), other causes (13.5%) and
"no comment" (18.7%). The most common reason mentioned (the delay of a
previous flight) was suggested by 78.3% of the passengers. In general, the
assumed reasons were weakly related to the specifically selected sample of flights.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
72 • Chapter 4
For example, at least one different reason was given for the delay of each flight.
Additionally, 33.3% of those interviewed stated that they had been informed about
the reason for the delay by the airline's employees.
Finally, passengers were classified by the different causes assigned to the
delay (see Appendix 4.2). Results showed significant differences in stability
attribution but not in control attribution over the actions of the company. Thus,
the passengers that argued "previous flight delayed" and "other problems"
attributed a greater degree of stability or permanence in the causes of the delay.
However, passengers who pointed out "mechanical-technical problems",
attributed a slightly transitory character to the delay. Additionally, those
passengers who perceived the previous flight as being the cause of the delay,
showed a certain attribution of control to the company, but maintained a neutral
position with respect to whether the airline might have taken some action to
prevent the delay. That is, they did not clearly perceive (even know) whether
the company could have taken such action that the problem could have been
rectified in time so that the flight could have finally departed closer to its
scheduled time (a situation that Hui et al. (1998) cali 'corrective delay').
In Table 4.1 we present the correlations and summary statistics of the
measures.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
The consumer 's reaction to delays in service
• 73
Table 4.1
Torrelations and summary statistics of measures
Variable
Controllab.
(X,)
Stability (X2)
Punctuality
importance
(X3)
Pere. waiting
time(Yi)
Anger(Y2)
Satísfactíon
X]
1.00
X2
X3
Y,
Yj
0.09
1.00
0.31*** 0.24*** 1.00
0.05
0.18**
0.15
1.00
0.32*** 0.21*** 0.42*** -0.01
0.25*** 0.07
0.32*** 0.05
(Y3)
Complaint
0.17** 0.15
Intention (Y4)
019*** 0.11
Repurchase
Intention (Y5)
5.26
2.57
Mean
2.44
1.18
SD
**/j<0.05;***p<0.01.
1.00
0.48*** 1.00
0.16**
0.03
0.38*** 0.23***
1.00
0.09
-0.09
0.31*** 0.18**
0.46***
1.87
1.30
62.43
48.38
3.02
1.46
2.27
1.29
4.33
1.27
1.00
4.13
1.37
4.4.2 Testing the proposed model
To test the hypotheses H r Hi 2 , we analyzed the existing relations among the
variables of attributions (control and stability), importance of punctuality, the
perceived waiting time, anger, satisfaction with service and behavioral response
(propensity to complain and repurchase intentions). Regarding the dimensión of
control attribution (measured with two items), we used one construct: the sum of
both 5-point semantic differential scales20. Therefore, all variables were singleítem measures.
We used path analysis to test the model presented in Figure 4.1. Therefore, the
model was estimated using observed variables with LISREL 8.3 (Joreskog and
Sorbom, 1996). Given the non-fulfillment of the hypothesis of multivariate
normality, we adopted the Weighted Least Squares (WLS) estimation procedure.
The weight matrix required for such analysis is the inverse of the estimated
20
Foliowing Churchill (1979), we detected that both variables "attribution of control over the
cause of delay" and "over the solution adopted by the company" show a certain level of
correlation (0.51; p=0.00). Additionally, the corresponding Cronbach alpha is 0.7, a level that is
considered acceptable.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
74 » Chapter 4
asymptotic covariance matrix Wof the polychoric correlations (Bollen, 1989). The
asymptotic covariance matrix as well as the matrix of polychoric correlations were
obtained with PRELIS. The results are presented in Table 4.2 and in a graphic
form in Figure 4.2.
Figure 4.2
Significant results of the conceptual model of antecedents and consequences of
satisfaction with service failure (delay)
Control
Attribution
Repurchase
Intentions
Note: dotted arrows denote non-significant (hypothesized) relationships. Indirect effects
not showed for clarity purposes.
***p<0.0l.
In general, the fit of the model is acceptable (x2=32.33, 16 d.f., p<0.01;
GFI=0.98; RMSEA=0.08, p=0.11; AGFI=0.96; NFI=0.96; PNFI=0.55).
Results show that the attribution of control and the importance of punctuality
have an effect on anger, which in turn affects behavioral intentions. Therefore, we
fínd a causal sequence of 'attribution-affect-behavior', in line with the proposals of
Weiner (1980), Folkes et al. (1987), Oliver and DeSarbo (1988) and Bitner (1990).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
The consumer 's reaction to delays in service • 75
Table 4.3
Pirect, indirect, and total effects (standard errors in brackets")
Path
Control A ttribution
Stability Attribution
Direct
Effect
0.23***
(0.07)
Indirect
Effect
-»Sat. with service
"
-0.15***
(0.05)
->Propensity to Complain
"
—»Repurchase Intentions
"
(0.06)
-0.16***
(0.06)
-0.04
(0.08)
"
-»Anger
—>Sat. with service
—>Repurchase Intentions
Punctuality Importance• ->Perceived waiting time
->Anger
"
-0.26***
(0.09)
0.56***
(0.09)
->Sat. with service
*
->Propensity to Complain
"
—»Repurchase Intentions
"
0.13
(0.08)
Perceived waiting time —> Anger
Anger
->Sat. with service
"
-»Propensity to Complain
"
-»Repurchase Intentions
"
->Satisfaction with service
->Propensity to Complain
—>Repurchase Intentions
Satisfaction with serWce->Repurchase Intentions
-0.68***
(0.08)
0.77***
(0.07)
-0.67***
(0.10)
0.06
(0.11)
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
-
-0.00
(0.01)
"
-0.03
(0.03)
-0.35***
(0.07)
0.40***
(0.07)
-0.37***
(0.07)
"
-0.09
(0.06)
0.10
(0.06)
-0.09
(0.06)
"
"
-0.04
(0.08)
-
Total
Effect
0.23***
(0.07)
-0.15***
(0.05)
0.17***
(0.06)
-0.16***
(0.06)
-0.04
(0.08)
-0.00
(0.01)
-0.26***
(0.09)
0.52***
(0.08)
-0.35***
(0.07)
0.40***
(0.07)
-0.37***
(0.07)
0.13
(0.08)
-0.09
(0.06)
0.10
(0.06)
-0.09
(0.06)
-0.68***
(0.08)
0.77***
(0.07)
-0.71***
(0.09)
0.06
(0.11)
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
76 » Chapter 4
Table 4.3
Direct, indirect, and total effects (standard errors in brackets)(cont.)
Squared Múltiple Correlations for
Structural Equations (R2)
Perceived waiting time
0.06
Anger
0.47
Satisfaction with service
0.43
Propensity to complain
0.53
Repurchase intentions
0.46
Model Goodness-of-fit statistics
Chi-Square (X2) £dfj
Root Mean Square Error of Approximation
(RMSEA) (p-value)
32.33*** [16]
0.08(0.11)
Goodness of fit index (GFI)
0.98
Adjusted goodness of fit index (AGFI)
0.96
Normed fit index (NFI)
0.96
Parsimony normal fit index (PNFI)
0.55
Note: indirect effects are obtained by multiplying the standardized coefficients of the
implied direct effects and the total effect adding the direct and indirect effects.
***p<0.0l.
Specifically, the control attribution has an indirect, negative and significant
effect on satisfaction mediated by anger (-0.15, p<0.0l), which supports Hi.
This result is in line with Weiner's thoughts (2000). He proposes that in situations
of negative result of a service, where negligence in the productive process is
attributed to the service provider, the reaction of the customer will not be one of
simple dissatisfaction but one of anger, which reflects moral outrage. Therefore,
anger is the dominant emotional reaction when a negative event is attributable to a
controllable cause. Moreover, the control attribution has a positive and signifícant
indirect effect on complaint intentions mediated by anger (0.17, p<0.0\),
consistent with H2. This result is in line with Weiner's proposal (2000) and with
the results achieved by Folkes et al. (1987). Consequently, the more controllable
the cause of the failure and the angrier the passenger, the greater his/her inclination
to complain will be. The data also support the predicted mediating role of anger in
the relationship between control attributions and repurchase intentions (-0.16,
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
The consumer 's reaction to delays in service • 77
/K0.01), consistent with Hypothesis H3. This result is in line with the proposals of
Folkes (1984) and Weiner (1985,2000).
Additionally, stability attributions do not influence satisfaction with service, so
Hypothesis H4 is not supported. It appears that stability attribution generates within
the individual other types of emotional reaction more related with expectations,
such as the hope or fear that identical results will or will not be produced again
(Weiner, 2000).
It also appears that perceived waiting time has no impact on anger (H5). This
result can be explained, in opinión of Hui et al. (1998), by the influence of various
environmental, situational and personal factors. Pruyn and Smidts (1998)
demónstrate that the perceived waiting time has a stronger effect on the judgments
about the delay's shortness/length than on the affect.
Furthermore, the importance of flight punctuality has a positive signifícant
effect on anger, supporting Hypothesis H6. Therefore, the delays for which
punctuality is of great importance increase anger towards the airline (Folkes et al.,
1987).
We do not detect a signifícant effect between the importance of punctuality21
and anger mediated by the perceived waiting time, so Hypothesis H7 is not
supported. This result could be explained by the influence of other variables such
as environmental and situational factors (Hui et al., 1998).
As predicted in Hypothesis H8, greater levéis of customers' anger result in less
satisfaction with the service (Szymanski and Henard, 2001). Furthermore, greater
customer anger results in more propensity to complain and thus Hypothesis H9 is
supported. This result corroborates the ones obtained by Folkes et al. (1987) and
Westbrook (1987). More specifícally, anger in itself can genérate a broad range of
With the aim of testing the possible existence of a moderator effect of the importance of
punctuality in the 'perceived waiting time-anger' relationship, we apply Moderated Regression
Analysis (MRA) (Sharma et al., 1981). However, the results show the absence of a moderator
effect.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
78 » Chapter 4
anti-company reactions in the sense that the consumer makes complaints with the
aim of being compensated for his/her purchase22 (Weiner, 2000).
For testing if the link between anger and propensíty to complain is not
completely mediated by satisfaction with service (HI0), we compare two causal
patterns. One pattern incorporates the path 'satisfaction with service-»propensity
to complain' (model proposed in Figure 4.1) while the other does not, but
maintains the remaining causal relationships. However, the evaluation of the
former model presents a problem of offending estimates which prevenís drawing
conclusions about the mediating role of satisfaction with service.
We also find that anger has a significant negative effect on repurchase
intentions, supporting Hypothesis Hu- Non-repurchase of the service constitutes an
anti-company reaction; a means by which customers attempt to penalize the
company as a response to their anger (Weiner, 2000).
Finally, we do not detect a positive significant effect of satisfaction with service
on repurchase intentions, and thus Hypothesis H12 is not supported. Following
Fornell et al. (1996), in sectors where the price of the service determines the
general customer satisfaction a low relation between the levéis of satisfaction and
customer loyalty is expected. This is especially true for sectors such as
transport/communications in which competition is relatively commodity-based
(producís which have become basic and scarcely present any real differentiation
among them).
In fact, during 1999 (one year after our fíeld work was carried out), and far from palliating the
problem of flight delays, passengers continué being skeptical with regard to receiving any form of
compensation. The consumere' organizations, thus, decide to sue the aírline companies for flight
delays (La Gaceta de los Negocios, 21* September 1999) due to the poor effectiveness of the
claims made directly to the companies, and urge the Government to impose the same system of
compensations that holds for other means of public transpon, such as railways, for instance (El
País, 23* April 1999). The representatives, likewise, request in meetings of the Spanish Congress,
that the Government promote initiatives to facilítate claims for compensations (El Mundo, 20*
May 1999); and the European Commissary for Transport has proposed that airline companies
compénsate their customers for excessive delays in an effort to reduce the current chaotic
situation (El País, 2 December 1999).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
The consumer 's reaction to delays in service • 79
4.5 Conclusions
We have examined the customer's reactions to delays in service through the
analysis of a sample of passengers that suffers delays in their flights at the airport
of Alicante during 1998. Specifically, we have focused on the relationships among
the perceived waiting time, the causal attributions of control and stability, anger,
satisfaction with service, and behavioral intentions (propensity to complain and
repurchase intentions).
Results show the causal sequence 'attribution-affect-behavioral intention' (i.e.,
'think-feel-act'). This causal sequence has been supported by the literature in
Psychology (Weiner, 1980) and of Consumer Behavior (Folkes et al., 1987; Oliver
and DeSarbo, 1988). That is to say, the attribution that the customer makes
regarding the causes of the flight delay precedes his negative affective reaction,
which subsequently has an influence on behavioral intentions.
Specifically, we have found that a passenger's anger in face of a flight delay
mediates the relationship between control attribution and satisfaction with
service. Additionally, anger mediates the relationship between control
attributions and behavioral intentions (propensity to complain and repurchase
intentions). Furthermore, the importance of punctuality leads to greater anger.
Finally, anger generates lower levéis of satisfaction with service, higher
propensity to complain, and lower repurchase intentions.
Therefore, it appears that anger is the dominant emotional reaction which is
in line with Weiner's (2000) reflection. That is, in situations of service failure
(flight delay) where negligence in the productive process is attributed to the
service provider, the emotional reaction will not be one of simple
dissatisfaction. Probably, anger will be shown (Westbrook 1987), a negative
emotional reaction related to a moral judgment, the moral outrage derived from
the fact that the service failure (flight delay) has been attributed to the airline
company.
Moreover, it appears that anger (instead of satisfaction with service) is the
main antecedent of propensity to complain and repurchase intentions, which
supports Westbrook (1987) and Weiner's (2000) proposals.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
80 « Chapter 4
Appendix 4.1
Measures employed in the study
Control attribution
1. Do youfeel the reason for the delay is something the airline has control over? (l=totally
uncontrollable; 5=totally controllable)
2. To what extent do you think there are actions that the airline could take to shorten this
delay but has not takeríl (l=it could not have done anything; 5=it could have acted)
Initial set of measurement Ítems: 1-2
Final set of measurement Ítems adopted in the structural model: land 2
^^
Stability attribution
1. Do you think that the reasonfor the delay is temporary and unusual, or is it something
that happens to thisflight veryfrequently? (anchors: l=very unusual; 5=very frequent)
Initial set of measurement Ítems: 1
Final set of measurement ítems adopted in the structural model: 1
Punctuality importance
Look at the sentence below and decide to what extent you agree or disagree with it:
1. It is importan!for me that myflight arrives on time at its destination (anchors: l=not at
all important; 5=very important)
Initial set of measurement items: 1
Final set of measurement Ítems adopted in the structural model: 1
Perceived waiting time
1. How long have you been waiting since the expected departure time?
minutes
Initial set of measurement items: 1
Final set of measurement items adopted in the structural model: 1
hours;
Anger
Look at the sentence below and decide to what extent you agree or disagree with it:
X.Iam angry with the airlinefor the delay (anchors: l=not at all angry; 5=very angry)
Initial set of measurement items: 1
Final set of measurement items adopted in the structural model: 1
Satisfaction with service failure
1. How satisfiedare you with the delay? (anchors: l=not at all satisfied; 5=very satisfied)
Initial set of measurement items: 1
Final set of measurement items adopted in the structural model: 1 (R)
Propensity to complain
Look at the sentence below and decide to what extent you agree or disagree with it:
1. / think I will complain to the airline personnel about the delay (anchors: l=very inclined
to complain; 5=not at all)
Initial set of measurement items: 1
Final set of measurement items adopted in the structural model: 1 (R)
Repurchase intentions
Look at the sentence below and decide to what extent you agree or disagree with it:
1. / wouldnotfly with this airline again iflhad the chotee (anchors: 1=1 would not fly
again; 5=1 would)
Initial set of measurement items: 1
Final set of measurement items adopted in the structural model: 1
Note: All items measured with 5-point scales. (R) Reverse coded for analysis
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Volver al Indice/Tornar a l'índex
The consumer 's reaction to delays in service • 81
Appendix:4.2
Delay causes and attributions
Perceived
causes for
the flight
delay
Previous
flight
delayed
Mechanical
problems
Airline
personnel
fault
Others
Anova F:
Average control attribution
Average control attribution of
Average stability attribution
of the cause of the delay
the measures adopted
Passenger Passenger
Passenger Passenger
Passenger Passenger
indicates does not
t
indicates does not
t
indicates does not
t
cause
indícate
cause
indícate
cause
indícate
cause
cause
cause
2.31
2.56
1.14
3.18
2.53
2.91***
2.38
2.70
1.73
(1.36)
(1.38)
(1.48)
(1.38)
(1.21)
(1.14)
2.80
(1.31)
2.80
(1.98)
2.37
(1.38)
2.43
(1.33)
2.30
(1.69)
2.48
(1.32)
0.873
1.59
0.56
0.48
2.87
(1.43)
2.60
(1-50)
2.78
(1.46)
2.81
(145)
-0.30
2.47
(1.56)
2.85
(1.44)
2.046
1.14
0.45
3.56
(1.12)
2.10
(1.28)
2.33
(1.06)
2.60
(1.17)
2.18
(1.10)
2.63
(1.18)
6.190***
_
5.86***
1.30
1.72
Note: due to the scarce number of individuáis attributing the delay to "other passengers"
and "bad weather" causes, we could not carry out the respective tests. Standard errors are
shown in brackets.
***/?< 0.01.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Chapter 5
Anger and Distributive Justice in a Double Deviation
Scenario: Explaining (Dis)satisfaction in Service Failure
and Failed Recovery Contexts
5.1 Introduction
The main objective of this research is to propose and empirically test a
model to examine the formation of customer (dis)satisfaction judgments in
double deviation contexts (i.e., failed recoveries following service failures)
(Bitner et al., 1990). In the last years, a growing number of studies in the service
marketing field have investigated how customer and managerial actions
following service failures impact on subsequent consumer decisions (e.g.,
Andreassen, 1999, 2000; Hess et al., 2003; Maxham III and Netemeyer, 2002;
Smith and Bolton, 1998, 2002; Smith et al., 1999; Tax et al., 1998). Previous
research has found that a successful recovery, the successful actions a service
provider takes in response to a service failure (Gronroos, 1988), could mean the
difference between customer retention and defection. Service failure is viewed
as one determinant for customers to switch providers (Roos, 1999) and it has
been proved that customer retention is critical to profitability (Reicheld and
Sasser, 1990). Additionally, appropriate complaint handling systems could
contribute to effectively differentiate the firm from competitors and thus, to
obtain a competitive advantage (Kelley and Davis, 1994). Therefore, service
recovery systems should be looked upon as a strategically critical part of a
fírm's overall customer orientation (Bell and Zemke, 1987).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
84 » Chapter 5
Poor service recoveries exacérbate already low customer evaluations
following a failure, producing a 'double deviation' effect (Bitner et al., 1990).
Bitner et al. (1990) find that "it is not the initial failure to deliver the core
service alone that causes dissatisfaction, but rather the employee's response to
the failure" (p. 80). It is the service recovery system (or the lack of it) which
causes the dissatisfaction and not necessarily the failure itself. In this sense,
various authors have analyzed the recovery strategies that firms should follow
to effectively recover their customers (Bell and Zemke, 1987; Bitner et al.,
1990; Boshoff, 1997; Kelley et al., 1993; Johnston and Fern, 1999). Johnston
and Fern (1999) find that when the firm shows an inappropriate or inadequate
response to the failure (double deviation), there is still an option to restore
satisfaction. Therefore, it seems that recovery strategies and, to a lesser extent,
the initial service failure are the elements that contribute to the formation of the
customer (dis)satisfaction judgments in double deviation scenarios.
Some work on service failure and recovery encounters has tried to explain
satisfaction with service recovery incorporating two main streams of research in
these proposals: justice theory (cognitive antecedent) and emotions theory
(emotional antecedent). In the first case, many studies have found a direct and
significant relationship between the three dimensions of perceived justice
(distributive, interactional and procedural23) and satisfaction with service
recovery (eg., Andreassen, 2000; Davidow, 2000; Smith et al., 1999; Tax et al.,
1998). In the second case, the few articles that examine emotions are centered
on the emotions triggered by the initial service failure (initial negative
emotions) (Andreassen, 2000; Smith and Bolton, 2002). To our knowledge, the
only study which empirically analyzes the emotions triggered by the service
recovery (secondary negative emotions) is the one of Chebat and Slusarczyk
(2005). However, these authors do not examine the impact of the secondary
Following the justice literature, each part of the recovery procedure is subject to fairness
considerations and creates a justice episode (Bies, 1987). Therefore, customers evalúate fairness
with the service recovery by three perceived factors: outcomes, procedural and interaction
(Blodgett et al., 1997; Goodwin and Ross, 1992; Smith et al., 1999; Tax et al., 1998). Distributive
justice refers to the perceived outcome of the firm's recovery effort, procedural fairness involves
the policies and rules by which recovery effort decisions are made, and interactional justice
focuses on the manner in which the service recovery process is implemented (Tax et al., 1998).
Thus, the role of perceived justice has been repeatedly shown to be crucial in determining postrecovery satisfaction in service failure and recovery contexts (e.g., Andreassen, 2000; Davidow,
2003; Hoffman and Kelley, 2000; Maxham III and Netemeyer, 2003; Ruyter and Wetzels, 2000;
Smith et al., 1999; Tax et al., 1998).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Disíríbuiive Justice in a Double Deviation Scenaho
• 85
negative emotions on satisfaction with the service recovery, which is our main
objective.
Therefore, following the main streams of research in the service failure and
recovery encounters field, our conceptual model includes the analysis of the
distributive component of justice (only this dimensión and not the other two)
and the negative emotion of anger. With respect to the choice of the distributive
component of justice, recent research has shown that customers with negative
emotional responses to service failures weigh distributive justice more heavily
than the other two components of justice (Smith and Bolton, 2002). These
authors conclude that when failure produces negative emotions, customers focus
on the outcome itself (i.e., recovery attributes and distributive justice) in stead
of, for instance, on the procedures (e.g., information exchanged) or on the
interactional elements (e.g., courtesy, concern). In fact, Smith and Bolton
(2002) show that the distributive component accounts for 75.6 percent of the
explained variance in the satisfaction with service recovery judgments for
customers showing an emotional response. Therefore, following the previous
reasoning and taking into account that our main goal with this research is to
analyze the joint effects of justice and anger in double deviation contexts, the
choice of the distributive justice component seems appropriate.
From the wide range of specific negative emotions that can be related to
failed service encounters, we focus on anger as the most frequent emotional
reaction elicited by service failures (Folkes et al., 1987; Weiner, 2000;
Zeelenberg and Pieters, 2004). In this sense, the belief that one deserved a better
(more fair) outcome than one had received (e.g., after service or after recovery)
is characteristic of experiences leading up to anger (Roseman, 1991; Oliver,
2000). Even more, the focus on one specific emotion (i.e., anger) is in line with
literature that focuses on the idiosyncratic elements of specific emotions
(Bougie et al., 2003; Tsiros and Mittal, 2000; Zeelenberg and Pieters, 2004).
According to this specific emotions approach (in contrast with the valencebased approach)24, different negative emotions may differently impact on
24
Based on the recent review of the literature on emotion and consumer behavior by Bagozzi et
al. (1999), there are essentially two ways to model the interactions between emotions and
satisfaction (and satisfaction-related behaviors), namely the valence-based approach and the
specific emotions approach. The valence-based approach entails a summation of the positivity and
negativity of the different emotions that customers experience to arrive at an overall judgment of
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
86 • Chapter 5
(dis)satisfaction, and henee more insight into the specifíc antecedents,
phenomenology and consequences of different emotions (such as anger) is
needed (Lings et al., 2004)25.
In sum, our main goal with this article is to contribute to the service failure
and recovery knowledge. Therefore, we adopt an interdisciplinary approach to
develop and empirically test a model of how customers form (dis)satisfaction
judgments in double deviation scenarios (failed recovery after failed service),
and to gain a more detailed understanding of the role of anger and distributive
justice in these situations. In the next section, we develop the conceptual
framework and research hypotheses. The third section justifíes the research
design, indicating the methodology, the sample, data collection and the
measurement of variables. Then, the results obtained are showed in the fourth
section and discussed in the fifth section and, finally, we address the key
conclusions of our study.
5.2 Conceptual framework and research hypotheses
The conceptual model of the present study is presented in Figure 5.1.
Regarding this conceptual model, we would like to pay attention to the
following issues. First, the conceptual framework clearly differentiates between
initial (negative) emotional response triggered by the service failure and the
(negative) emotional response triggered by the service recovery. We propose
that specifíc emotions experienced during and after the service recovery effort
(dis)satisfaction. The net (weighted) valence of the affective experience is the balance between
positive and negative emotions.
23
Recent theoretical approaches (not empirically validated yet), such as the ones of Stewart
(1998) or McColl-Kennedy and Sparks (2003), are aiso useful for basing our conceptual model.
Stewart (1998) proposes that customers end bank relationships after an involving process of
problem(s) effort, emotion and evaluation. Thus, they state that negative emotions may be
occasioned by the company's response to the problem and the repetition of a problem.
Additionally, McColl-Kennedy and Sparks (2003) state that when service providers do not appear
to put proper effort into the service recovery attempt, this is viewed negatively, and this led to the
customer experiencing negative emotions such as anger and subsequently dissatisfaction with the
service recovery attempt.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario • 87
affect satisfaction with service recovery, and focus on this emotional response26.
Second, we distinguish between cognitive (justice) and emotional antecedents
(post-recovery emotions) to customers' satisfaction with service recovery.
Specifically, we focus on the distributive component of perceived justice and
the post-recovery emotion of anger. Third, we distinguish between service
failure-related variables, and service recovery-related variables. We propose
that service failure-related variables have a direct effect on satisfaction with
service recovery, as well as an indirect effect through the cognitive and
emotional antecedents to customers' satisfaction with service failure and
recovery encounters. We also propose that service recovery-related variables
have an indirect effect on satisfaction with service recovery through the
cognitive and emotional components. And fourth, we apply this framework to
the specifíc context of double deviations, that is, failed recoveries following
service failures. In the following sections, we define the double deviation
context, define several key constructs and present the specifíc hypotheses
regarding the relationships shown in Figure 5.1.
26
We would like to note that we have adopted the most common view in marketing literature
regarding the relationship between anger and satisfaction, that ¡s, that specifíc emotions like anger
contribute to customer satisfaction and dissatisfaction (e.g., Mano and Oliver, 1993; Oliver, 1997,
2000; Westbrook, 1987). However, a recent approach suggests that anger could medíate the
relationship between service encounter dissatisfaction and customers' behavioural responses to
service failure as well (Bougie et al., 2003).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
88 • Chapter 5
Figure 5.1
Conceptual model of (dis)satisfaction with service recovery in double deviation
scenarios
Service failure
related variables
(magnitude)
(controllability)
Service recovery
related variables
(apology)
(explanation)
Note: bold arrows denote the relationships that are the focus of this study. Dotted arrows
denote relationships established in prior research.
5.2.1 Service failure and failed recovery: a double deviation scenario
Service recovery is defíned as the actions a service provider takes in
response to a service failure (Grónroos, 1988), or the activities that are
performed because of customer perceptions of initial service delivery falling
below the customer's 'zone of tolerance' (Zeithaml et al., 1993). This zone
represents the difference between the customer's desired level of service quality
and the adequate level from his/her point of view.
Effective recovery from service failures should be looked upon as a
strategically critical part of a firm's overall customer orientation (Bell and
Zemke, 1987). Even more, service recovery can be an opportunity to
communicate commitment to customers and to strengthen their loyalty to the
firm (Berry and Parasuraman, 1991), leading to a phenomenon known as the
'service recovery paradox'. This paradox means that customers whose service
failures have been satisfactorily remedied are found to be more satisfied, more
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario • 89
likely to remain loyal, and more likely to engage in favorable word-of-mouth
about the company than customers who have never experienced a failure (Hart
et al., 1990; Kasper, 1988; McCollough and Bharadwaj, 1992).
However, the management of service recovery efforts receives little attention
in most firms and is often completely ignored (Heskett et al., 1990). This is
especially surprising given that recovery management, on the one hand, has a
disproportionately significant positive impact on customers' perceptions of
service quality (Berry and Parasuraman, 1991, p. 54) and, on the other hand, has
been proved to affect service organizations' profits positively (Reichheld and
Sasser, 1990).
Previous research has shown that more than half of attempted recovery
efforts only reinforce dissatisfaction (Hart et al., 1990). Poor service recoveries
exacérbate already low customer evaluations following a failure, producing a
'double deviation' effect (Bitner et al., 1990; Hart et al., 1990; Johnston and
Fern, 1999; Mattila, 2001b). These 'double deviations' have been shown to
occur even in múltiple service failures and recovery efforts (Maxham III and
Netemeyer, 2002).
Therefore, our main goal with this study is to add to the previous works that
¡Ilústrate the importance of an efficient recovery process for companies.
However, we propose a new approach to this issue by examining the harmful
consequences of failed recoveries after service failures. We analyze the
customer behavior in double deviation scenarios with a specific emotion/justice
approach, where anger and distributive justice play a central role in explaining
customers' post-recovery evaluations (satisfaction with service recovery).
5.2.2 Determinants of (dis)satisfaction in a double deviation context
In this section, we address the main hypotheses implicit in the proposed
model (Figure 5.1). First, we focus on magnitude of service failure and service
failure controllability (service failure-related variables). We propose different
hypotheses regarding the direct and indirect effects of these variables on
distributive justice, anger with service recovery, and (dis)satisfaction with
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
90 • Chapter 5
service recovery. Second, we examine the role of apologies and explanations
(service recovery-related variables). We propose different hypotheses that
centre on the direct and indirect effects of these recovery strategies on
distributive justice, anger with service recovery, and (dis)satisfaction with
service recovery. Finally, we analyze the relationships between distributive
justice, anger with service recovery, and (dis)satisfaction with service recovery.
5.2.2.1 Direct effects and indirect effects, through cognitive
emotional antecedents, of service failure-related variables
and
The fírst group of determinants of customer (dis)satisfaction analyzed are the
service failure-related variables. Specifically, we address the role of magnitude
of service failure and service failure controllability.
Magnitude of service failure
Hirschman (1970) was the fírst to assess that consumers would be more
likely to voice their complaints when dissatisfied with an 'important' product.
After that, many researchers have analyzed the effects of magnitude of the
failure (also called complaint seriousness/intensity, or severity of the
failure/dissatisfaction problem), on a service failure/recovery encounter context
(e.g., Hess et al., 2003; Smith and Bolton, 1998; Smith et al., 1999). In line with
Hess et al. (2003), we define magnitude of service failure as the magnitude of
loss (how important the problem was for the customer) that customers
experience due to the failure.
Consistent with social exchange theory (e.g., Walster et al., 1973), a service
recovery can be considered as an exchange in which the customer experiences a
loss while the service organization tries to make up that loss by a recovery
attempt (Smith et al., 1999). In this sense, as service importance increases so
does the magnitude of the gains and losses experienced by customers. The higher
the magnitude of the service failure, the greater are the customer's expectations of
high recovery efforts to be taken by the organization, and the harder it is for
service organizations to execute an effective recovery (e.g., Mattila, 2001a;
Smith and Bolton, 1998; Smith et al., 1999). Therefore, when customer's
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario
• 91
expectations are not achieved (the customer experiences a failed recovery) we
propose that the magnitude of service failure influence subsequent (post-failure)
customer's satisfaction judgments directly and indirectly through the cognitive
and emotional antecedents.
Previous research suggests a direct relationship between magnitude of service
failure and satisfaction. Thus, prospect theory (Kahneman and Tversky, 1979) and
mental accounting principies (Thaler, 1985) suggest that losses from service
failures will be weighed more heavily than gains received during service recovery
(Smith et al., 1999). As such, the magnitude of service failure will be influential on
the evaluation of a service provider after a service failure (Weun et al., 2004) in
terms of that the higher the magnitude of service failure, the lower the level of
customer satisfaction (e.g., Gilly and Gelb, 1982; Hoffman et al., 1995; Richins,
1987). Moreover, Zeithaml et al. (1996) maintain that a service failure is one of
the major factors that narrow a customer's zone of tolerance. As the service failure
becomes more severe, the customer's tolerance zone becomes narrower, increasing
the potential for customer dissatisfaction (Gilly and Gelb, 1982; Hoffman et al.,
1995). Following the initial reasoning, it seems that as the size of the loss
increases and the customer expectations are not met (failed recovery), the less
likely it is for the customer to be satisfied with the service recovery. Therefore,
we state that:
Hla. The magnitude of the service failure affects the satisfaction with the
service recovery negatively.
We also propose that cognitive (distributive justice) and emotional (anger),
as antecedents of customer satisfaction, play a mediating role in the relationship
between magnitude of service failure and customer satisfaction with service
recovery. These assumptions are based on the following arguments.
With respect to the cognitive antecedent, Blodgett et al. (1993) fínd that
consumers who experience a failure with producís that they feel are important
are likely to perceive the act of complaining as an injustice in and of itself.
Following the previous reasoning based on exchange theory, the higher the
magnitude of service failure, the greater the customers' expectations of high
recovery efforts to be taken by the organization. These expectations result in
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
92 » Chapter 5
lower evaluations of firm's efforts, especially with respect to the outcomes
received (distributive justice). Therefore, the higher the subject's perception of
the magnitude of service failure, the lower his/her perceptions of distributive
justice associated with the service recovery effort. Additionally, given the
widely recognized existence of a signifícant relationship between distributive
justice and satisfaction with service recovery (see section 5.2.2.3), we propose
that the higher the magnitude of service failure, the lower customer's
perceptions of distributive justice associated with the service recovery effort,
and the lower customer's subsequent satisfaction with the failed recovery. In
sum:
Hlb. The magnitude of the service failure affects the perceived distributive
justice negatively.
Hlbb. The perceived distributive justice mediates the relationship between the
magnitude ofthe service failure and the satisfaction with the service recovery.
Regarding the emotional antecedent, previous research suggests that anger due
to service failures increases relative to the importance placed on its successful
performance (Folkes et al., 1987; Taylor, 1994). Based on exchange theory, the
higher the magnitude of service failure, as long as customer's expectations are
not met (unsuccessful recovery), the greater the anger triggered by the failed
recovery. Additionally, following Mano and Oliver's (1993) proposal where
negative affect is viewed as a negative antecedent of satisfaction (see section
5.2.2.3), we propose that the higher the magnitude of service failure, the greater
the anger triggered by the failed recovery, and the lower the subsequent
satisfaction with the failed recovery. That is:
Hlc. The magnitude ofthe service failure affects the degree of anger with the
service recovery positively.
Hice. Anger with service recovery mediates the relationship between the
magnitude ofthe service failure and the satisfaction with the service recovery.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario • 93
Control attributions of service failure ("controllability)
The attribution paradigm used in this study is derived from the work of
Weiner et al. (1971) and Weiner (1985). Attributions are what people perceive
to be the causes behind their own behavior (locus), the behaviors of others
(controllability), or the events they observe (stability) (Bitner, 1990). Previous
research has shown that customers' attributions about a failure influence their
attitudes and behavioral intentions toward the fírm (e.g., Bitner, 1990; Folkes et
al., 1987). Attributional search is, therefore, more likely following unexpected
and negative events (e.g., service failure) rather than after a successful event
(Weiner, 1985, 2000).
In this study, we specifícally address the causal dimensión of controllability.
This attribution involves the customer's belief about whether the service
organization could influence or prevent a failure from occurring (Weiner, 2000).
The logic of this choice is that our model is based on the specifíc emotion
approach, with anger being one of the central elements. In this sense, anger is
thought to be an externally directed or focused emotion (Godwin et al., 1995),
the result of attributing responsibility for what is happening (e.g., service
failure) to someone else (fírm). Or, as stated by Oliver (2000) and Weiner
(2000), anger represents the dominant customer's negative emotional reaction to a
service failure when the negative event (i.e., failure) is attributed to a
controllable cause.
Previous research suggests a direct relationship between service failure
controllability and satisfaction. Attribution theory proposes that the perceived
cause of a failure will influence the subsequent dissatisfied consumer's response
(Bitner, 1990; Folkes, 1984). Additionally, Maxham III and Netemeyer (2002)
state that attributions of blame toward the fírm after a failure significantly
increase the dissatisfaction felt from one failure (i.e., service failure) to the next
(i.e., failed recovery), and that the effect is larger for customers who perceived
an unsatisfactory recovery after the fírst failure. In this line, we propose that:
H2a. The degree of service failure controllability affects the satisfaction with
the service recovery negatively.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
94 « Chapter 5
We propose that cognitive (distributive justice) and emotional (anger)
antecedents of customer satisfaction play a mediating role in the relationship
between service failure controllability and customer satisfaction with service
recovery. These assumptions are based on the following arguments.
As for the cognitive antecedent, the few studies that analyze the
controllability attribution in a service failure and recovery context suggest that
consumere' controllability attributions about service failure might influence
their service recovery perceptions (Hess et al., 2003). Blodgett et al. (1993)
propose that complainants who perceive the problem suffered (service failure)
to have been controllable are likely to perceive a lack of justice during the
recovery process, because the problem should never have occurred in the fírst
place. When a controllable failure occurs, customers perceive that the service
provider should put a greater recovery effort in order to restore equity to the
exchange (Hess et al., 2003). Therefore, recovery strategies seem to be less
effective when failures are perceived to be controllable (Kelley et al., 1993).
Additionally, given the widely recognized existence of a significant relationship
between distributive justice and satisfaction with service recovery (see section
5.2.2.3), the more controllable the failure is perceived, the lower customer's
perceptions of justice associated with the service recovery effort, and the lower
customer's subsequent satisfaction with the failed recovery. Following the
previous arguments, we hypothesize that:
H2b. The degree of service failure controllability affects the perceived
distributive justice negatively.
H2bb. The perceived distributive justice mediates the relationship between the
degree of service failure controllability and the satisfaction with the service
recovery.
Finally, we propose that the customer's perceived controllability over the
cause of the failure influences anger with service recovery. Therefore, the anger
experienced after an unsuccessful recovery episode will be higher the more the
customer has attributed the cause of the failure to the firm. Additionally, in line
with Weiner (2000), we consider that anger with service recovery mediates the
relationship between service failure controllability and (dis)satisfaction with
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario • 95
service recovery. When the service failure is attributed to the service provider,
the emotional reaction of the consumer after the recovery effort will not be one
of simple dissatisfaction but one of anger, which reflects the moral outrage
suffered. Consequently, we propose that:
H2c. The degree of service failure controllability affects the degree of anger
with the service recovery positively.
H2cc. The degree of anger with the service recovery mediates the relationship
between the degree of service failure controllability and the satisfaction with
the service recovery.
5.2.2.2 Indirect effects of service recovery-related variables
cognitive and emotional antecedents
through
From a process perspective, an organization's response to service failure can
be viewed as an exchange situation, as a sequence of events in which a
procedure, beginning with a complaint, generates a process of interaction
through which a decisión and outcome occurs (Tax et al., 1998). A central
element of this process is the action taken by the organization (recovery
strategy) to face the initial service failure.
Recovery strategies comprise, among others, compensation (e.g., refund,
apology), assuming responsibility for the failure, speed and convenience of the
process, being flexible in applying policies, demonstrating politeness, concern,
and honesty, providing explanation for the initial failure, and putting effort into
resolving the problem (Tax et al., 1998). Recently, Davidow (2003) has divided
the organizational responses to service failure into six sepárate dimensions:
timeliness (perceived speed to handle a complaint), facilitation (firm's
procedures to handle complaints), redress (benefits received from the firm in
response to the complaint), apology (acknowledgment by the firm of the
complainant's distress), credibility (explanation for the problem), and
attentiveness
(interpersonal
communication
between
organizational
representative and the customer).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
96 • Chapter 5
Davidow's (2003) classification facilitates to determine the importance of
each recovery attribute to customer evaluations of the organization's recovery
effort. In this study, we center on two of the previous dimensions, apology and
explanation, as two recovery strategies that recent research has claimed that
deserve more attention in relation to the justice aspect, and, specifically, to the
distributive dimensión of justice (Davidow, 2003; Mattila and Patterson,
2004)27.
Previous research has proposed a direct (and positive) effect of apology and
explanation on satisfaction28, although empirical results are contradictory. For
example, apology and explanation are two of the required strategies found by
Johnston and Fern (1999) to satisfy bank customers that experience double
deviation scenarios. However, Boshoff (1997), Davidow (2000), Goodwin and
Roos (1992), and Ruyter and Wetzels (2000) do not fínd a signifícant effect of
apology with respect to satisfaction.
These contradictory results raise the question of whether distributive justice
and anger with service recovery are complete or partial mediators in
determining post-recovery customer satisfaction. Following recent works of
Oliver (1997), McColl-Kennedy and Sparks (2003), and Smith et al. (1999), we
also assume complete mediation. This assumption implies that recovery
strategies do not impact customer satisfaction directly but indirectly through
cognitive and emotional components.
Additionally, we propose that recovery strategies have a direct effect on
customer's perceptions of distributive justice. When a service failure occurs,
customers expect outcomes that match the level of their initial dissatisfaction
(Zeithaml and Bitner, 2000). In this study, this outcome takes the form of an
apology, on the one hand, and an explanation, on the other hand. Therefore, we
propose that both recovery strategies have a direct and positive effect on
customer's perceptions of distributive justice (Davidow, 2003; Hoffman and
Kelley, 2000; Mattila and Patterson, 2004; Tax et al., 1998). Additionally, in
Indeed, Bitner et al. (1990) identify apology and explanation as two of the three key elements
for a successful recovery: "sincere apologies, compensatory actions, and explanations can
dissipate anger and dissatisfaction (p. 81)."
For a review of the main research examining these relationships see Davidow (2003).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario • 97
terms of the equity theory (Adams, 1965), the service failure violates the equity
exchange norm implicit in the purchase, reducing the customer's outcomes to
zero or to a déficit. The recovery effort by the service organization tries to
restore equity by means of different recovery strategies. These recovery
strategies influence individuáis' perceptions of justice, which ultimately
influence satisfaction with the recovery effort (Oliver, 1997; Smith et al., 1999).
In the context of the present study, recovery strategies of apology and
explanation influence individuáis' perceptions of distributive justice which
ultimately influence satisfaction with the recovery effort (indirect effect). This
'complete mediation' sequence is also in line with the post-complaint customer
behavior responses model proposed by Davidow (2003). Thus, we hypothesize
that:
H3a. Offering apologies for service failure affects the perceived distributive
justice positively.
H3aa The perceived distributive justice mediates the relationship between
apology and the satisfaction with the service recovery.
H3b. Offering adequate explanations for service failure affects the perceived
distributive justice positively.
H3bb. The perceived distributive justice mediates the relationship between
explanation and the satisfaction with the service recovery.
Regarding the emotional component of customer evaluations of the
organization's recovery effort, McColl-Kennedy and Sparks (2003) present a
fairness theory-based conceptual framework which allows us to propose a direct
effect of the recovery strategies on anger with service recovery. The authors
argüe that when a poor service recovery occurs, the customer assesses whether
the service provider could and should have done something more in terms of
recovery efforts. When a service provider does not appear to exhibit an
appropriate level of effort, consumers attribute this to the service provider as not
caring. This in turn leads to the customer feeling negative emotions, such as
anger. In our study, the assessment of the recovery effort refers to the level of
apology and explanation felt by the customer.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
98 • Chapter 5
Nguyen and McColl-Kennedy (2003) propose that apology and explanation,
among others, are positively associated with the reduction of customer anger
experienced during the recovery effort. Boshoff and Leong (1998) also state that
an apology can, at least to some extent, defuse the anger felt by a complaining
customer, whereas Bell and Zemke (1987) further suggest that 'annoyed'
customers will be recovered by the provisión of an apology and the correction
of the problem. Additionally, past research in Organization literature suggests
that appropriate explanations can reduce feelings of anger and resentment that
often occur when people experience unfavorable events (Bies and Shapiro,
1987; Folger et al., 1983). Therefore, it seems reasonable to presume that both
apologies and explanations have as their ultímate goal the creation of positive
customer attitudes toward the organization's recovery effort by reducing the
feelings of anger.
Finally, we propose a mediating effect of anger with service recovery in the
relationship between recovery strategies (apology and explanation) and
satisfaction with service recovery. Following equity theory (Adams, 1965) and
the conceptual fairness theory-based model proposed by McColl-Kennedy and
Sparks (2003), recovery effort by the service organization tries to restore equity
by means of different recovery strategies. However, when service providers do
not appear to put proper effort into the service recovery attempt, this is viewed
negatively, which leads to the customer experiencing negative emotions such as
anger and, subsequently, dissatisfaction with the service recovery attempt
(McColl-Kennedy and Sparks, 2003: 262). Therefore, we propose:
H4a. Offering apologies for service failure affects the degree ofanger with the
service recovery negatively.
H4aa. The degree ofanger with the service recovery mediates the relationship
between apology and the satisfaction with the service recovery.
H4b. Offering adequate explanations for service failure affects the degree of
anger with the service recovery negatively.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Jusíice in a Double Deviation Scenario • 99
H4bb. The degree of anger with the service recovery mediates the relationship
between explanation and the satisfaction with the service recovery,
5.2.2.3 Direct and indirect effects ofcognitive andemotional
antecedents
The last group of determinants of customer (dis)satisfaction analyzed are the
distributive justice (cognitive antecedent) and anger with service recovery
(emotional antecedent).
Cognitive antecedent: perceived distributive justice
Fairness or justice dimensions deal with whether an action was fair or not (a
subjective feeling), they do not deal with the actual action itself (recovery
strategy) taken by the organization (Davidow, 2003). This customer's
evaluation of recovery efforts can be explained through equity theory (Adams,
1963). Equity theory is important in situations where an exchange takes place,
weighting the inputs (e.g., sacrifices and investments made by customers)
against outputs received (perceived justice), and comparing them with those of
other similar situations experienced (Greenberg, 1990). Customers evalúate
fairness with the service recovery by three factors: perceived outcomes,
procedural and interaction (Blodgett et al., 1997; Goodwin and Ross, 1992;
Smith et al., 1999; Tax et al., 1998). Distributive justice refers to the perceived
outcome of the firm's recovery effort, procedural fairness involves the policies
and rules by which recovery effort decisions are made, and interactional justice
focuses on the manner in which the service recovery process is implemented
(Taxetal., 1998).
As we have discussed in the introduction, in this research we focus on the
distributive component of justice29. We consider that distributive justice is the
more relevant and appropriate justice dimensión for examining the jointly
29
Distributive justice has its origins in the study of social exchange (Blau, 1964) and equity
theory (Adams, 1965), and refers to the outcomes an individual receives in social exchanges.
Perceptions of distributive justice are thought to result in three outcome components: equity,
equality, or needs (Deutsch, 1985). The most commonly studied of these components, and also
the focus of this research, is equity, which refere to the recipient's perception of whether or not
rewards are proportional or fair given the amount of inputs (Tyler, 1994).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
100 • Chapter5
effects of justice and the specifíc emotion of anger on satisfaction formation in
double deviation contexts. We based this decisión on a review of the existing
literature (e.g., Chebat and Slusarczyk, 2005; Smith and Bolton, 2002), on
affect control theories (Heise, 1979, 1989a,b; MacKinnon, 1994), and on
appraisal theories (Folkman and Lazarus, 1985). Appraisals theories of
emotions maintain that specifíc emotions and their intensity are tied to an
appraisal of the event eliciting the emotional response (Folkman and Lazarus,
1985). Affect control theory proposes that individuáis act in such a way that
their emotions are appropriate to the events they experience (Heise, 1979,
1989a,b; MacKinnon, 1994). Therefore, a perceived lack ofjustice (appraisal of
the event) is expected to produce negative emotions of anger which are
consistent with the negative event experienced (affect control). That is, from a
customer' viewpoint, complaint-related justice is more than a matter of
economic calculus in unbalanced exchanges, they also involve emotions
(Chebat and Slusarczyk, 2005).
Previous research has examined the effects of customer perceived justice on
satisfaction with service recovery (Seiders and Berry, 1998). In fact, justice
theory appears to be the dominant theoretical framework applied to service
recovery (Tax and Brown, 2000). In this sense, we can find many examples in
the literature in which the distributive component of perceived justice (the
justice component focus of this study) is shown to be a signifícant antecedent of
customers' satisfaction with complaint handling (e.g., Mattila, 2001a;
McCollough et al., 2000; Maxham III and Netemeyer, 2003; Smith et al., 1999;
Tax et al., 1998; Wirtz and Mattila, 2004). Therefore, we propose that:
H5a. The perceived distributive justice affects the satisfaction with the service
recovery positively.
Additionally, we propose that the emotional antecedent of customer
satisfaction (i.e., anger with service recovery) play a mediating role in the
relationship between distributive justice and customer satisfaction with service
recovery. These assumptions are based on the following.
Regarding the cognitive antecedent, scarce research has been conducted in
order to examine the role of distributive justice on the elicitation of specifíc
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Jusíice in a Double Deviation Scenario • 101
emotions in the context of service failures and recoveries (Chebat and
Slusarczyk, 2005; Schoefer, 2001). Following the work of Mikula et al. (1998)
we propose injustice as an antecedent (appraisal) dimensión of negative
emotions (i.e., anger with service recovery). Recent research supports this
assumption (e.g., Chebat and Slusarczyk, 2005; Schoefer, 2001). For example,
Schoefer (2001) employs a quasi-experimental design to examine the role of
perceived justice as an antecedent (appraisal) dimensión of emotion, obtaining
that negative emotions elicited by the complaint handling encounter varied
across different levéis of an averaged measure of perceived justice. Chebat and
Slusarczyk (2005) fínd that low levéis of the three dimensions of justice
(interactional, distributive and procedural) enhance negative emotions, although
they do not include anger in their set of negative emotions. However, a number
of studies in social exchanges have shown that anger is by far the most likely
emotional reaction to events perceived as very unjust (e.g., double deviation
scenarios) (Adams, 1965; Clayton, 1992; Homans, 1974; Mikula et al., 1998;
Skarlicki and Folger, 1997). Therefore, we propose that:
HSb. The perceived distributive justice qffects the degree of anger with the
service recovery negatively.
Finally, although justice theory appears to be the dominant theoretical
framework applied to service recovery (Tax and Brown, 2000), we propose that
anger mediates the relationship between distributive justice and satisfaction
with service recovery. This assumption is based on the works of Mikula et al.
(1998) and Chebat and Slusarczyk (2005). Mikula et al. (1998) fínd that anger is
by far the most likely emotional reaction to events perceived as very unjust, and
one of the central mediators of reactions to perceived injustice. Additionally,
Chebat and Slusarczyk (2005) assess that consumers who suffer an
unacceptable service recovery may express their emotions and behave in such a
way that these emotions are appropriate to the situation. Chebat and Slusarczyk
(2005) propose that emotions elicited by the justice of service recovery medíate
the relationship between justice and behavioral responses. However, two main
issues differentiate their work from the present study. First, they measure
negative emotions with two discrete emotions, anxiety and disgust, not
including anger (the focus emotion of this study). Second, they center on the
emotions elicited by the justice of service recovery to explain actual post-
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
102 • Chapter5
recovery (exit) behavior, but they do not exnlain n™,
P
judgments.
P ost - rec °very satisfaction
Given that the literature suggests the existence of a signifícant r^f
uP
between anger and satisfaction with service r e c o v e n Í ,
Hypothesis HA we complete the Chebat and S u « r o l T
*
suggesting that the negative emotions
tíg^^^Z^"^
*
(i-e anger with service recovery) mediatele
r
^
^
~
>
e
jus.ce and the attitudinal consequences of the failed recovery ^ 7 ^ T
with service recovery). Therefore, we propose that:
satisfaction
HSbb. The degree of anger with the service recoverv merlina tu
, •
Emotional anteceden!: a n ^ r w i t l ^ r y j ^ ^ t ^ ^ ^
Research has shown that positive and negativo »«,,-•
la,e, affective jndgmen*
^ ^ m ^ l r T
dem„„sttaw «ng consumpt¡0„remamin !„e^r"of r r
traces; traces that are considerad by the customers tr, h* ¡ h
T
^
T
'
,
with these resnlts, Wirtz and Bateson (1999) have « Z
, I, ' i r ' 8 ' ™ '
a patHy cognitive and partly affective emo, o„a, £ E £ * . T ^ '"
experience. V* anthors also outiine the
I m p a ^ T ^ ^ T T
and emotional antecedents separately.
' g * COgn"'ve
More specifícally, Mano and Oliver MQ91\ ¡ ^ . A
fadure and recovery encounters, and to the specifle eontext of ! M
cev,at,o„s, we p,„p„se that anger with the s i ¡ce re c o " 2 a f L t i l l
satefaction with the service recovery neeativelv Th, f.
. A
faiiure and recove^ contex, , ¿ 3 T £ £ T Z £ t
" t
emotions triggered by the initial service failure SomTof L
! °°
* e effec, of this initia, (negative) emotiola,
responÍ
« I !• 7 ^
JUdgmen, of service recove^ (Andreassen, ,999, £ £ £ £ , £ £ &
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario •
103
Smith and Bolton, 2002). To our knowledge, the only study which empirically
analyzes emotions triggered by a service recovery context is the one of Chebat
and Slusarczyk (2005). However, as it has been explained before, they do not
examine the effect of these emotions on satisfaction with service recovery.
Therefore, the present research is the fírst empirical attempt to examine anger
with service recovery and satisfaction with service recovery in the same
conceptual model. This assumption is based on recent theoretical works, such as
the ones of Stewart (1998) or McColl-Kennedy and Sparks (2003). Henee, we
hypothesize that:
H6. The degree ofanger with the service recovery affects the satisfaction with
the service recovery negatively.
In Figure 5.2, we summarize the main hypotheses presented in this section.
Figure 5.2
Proposed model of (dis)satisfaction with service recovery in a double deviation
scenario
Service recovery-related variables
Service failure-related variables
Magnitude of
service failure
Service failure
con trolla bilí ty
T
Explanation
Apology
y
^
y
/
H3c +
(
Distributive
V^
iustice
\
H5b-
Anger with
service recovery
H5a+\
Note: indirect effeets not showed for clarity purposes.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
\
H6-
—•[
Satisfaction \
with service J
recovery 1
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
104 •
Chapter5
5.3 Methodology
In order to test the aforementioned hypotheses, we conducted an empirical
study. Structural equation analysis with LISREL 8.3 (Joreskog and Sorbom, 1996)
was used to test the proposed model. We now discuss the sample and data
collection procedure, the measurement of the variables, and the data analysis.
5.3.1 Sample and data collection
While the issue of consumer dissatisfaction is of importance to all marketers,
some underlying characteristics of services make the topic especially critical to
services marketers. First, services are, to a greater degree than goods, intangible,
heterogeneous, and simultaneously produced/distributed and consumed
(Zeithaml et al., 1993). Second, in the performance of services both customers
and service personnel play a role (Solomon et al., 1985). These characteristics
increase the likelihood of errors (service failures) in the service área both from
an operational perspective and from the customer's viewpoint and, therefore,
the need for recovery (Brown et al., 1996). Specifically, we select the banking
industry because it is a kind of services industry high in experience and
credence properties, where failures are quite common. Moreover, banking
products are highly diffused in the consumer market (almost all households
have some type of banking product), which means that the probability of
unsatisfactory experiences resulting in complaints is quite high. In fact, the
banking sector receives the greatest number of complaints according to Spanish
consumer organizations (Ortega, 2003). Finally, the probability that customers
rely on their emotional reactions to derive satisfaction judgments is also high
(Alford and Sherrell, 1996; Smith and Bolton, 2002), and this is one of the main
variables of interest in the present research.
The data were collected via a self-reported questionnaire administered to
2,000 households that were members of the regional branch of a consumer
organization (UCE), during a two-month period (November lst 2002 to January
7* 2003). We employed the critical incident technique (CIT), which has been
used previously in numerous marketing and management studies (e.g., Bitner et
al., 1990; Keaveney, 1995; Kelley et al., 1993). Thus, we understand critical
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Jusíice in a Double Deviation Scenario
.
105
incidents as events that deviate significantly, either positively or negatively,
from what is normal or expected which are also called triggers or trigger events
(Gardial et al., 1996). In our case, the interest is on negative incidents and
specifically, we define a critical incident as the most recent problem of special
relevance that a customer has experienced during his/her relationship with
his/her main bank. The information obtained with this methodology allowed us
to detect failed recoveries and thus, to analyze double deviation scenarios30.
Respondents were told to report a critical service incident in dealing with
banks, and then to answer some structured questions about the manner in which
the problem was handled and other issues. Questionnaire packets included a
letter of introduction, a questionnaire booklet, and a postage-paid return
envelope. Reminder cards were mailed approximately two weeks after the
initial mailing.
Four hundred seventy two questionnaires were returned yielding a 24%
response rate. Fifty-nine questionnaires were unusable due to incomplete
responses or incongruence, and two hundred and eleven reported no problem
leaving a total sample size of 202. From these, 181 showed unsuccessful
recovery situations (double deviation scenarios)31. This sample size is adequate
30
In a recent article, Gremler (2004) discusses the strengths and drawbacks of the CIT method
and its contribution to service research. The study assesses that in investigations of service failure
and recovery and customer switching behavior, CIT appears to be a particularly useful method.
The author concludes that the CIT method has been accepted as an appropriate method for use in
service research (for previous examples of this technique in service failure/recovery literature see
for example, Hoffman et al., 2003, and Kelley et al., 1993). Additionally, Liljander and Strandvik
(1997, p. 167) state: "There is a need for a more qualitative approach to emotions in services in
order to genérate an in-depth understanding of emotional dimensions in service episodes and
relationships. For example, studies employing the critical incident technique could easily also
focus on affective responses. When studying critical incidents, researchers so far have only
collected data concerning descriptions of the incidents, overall satisfaction (sometimes with
service recover) and intentions to use the same service providers in the future. Neither the
emotions that the incidents evoke in the customers, ñor the strength of these emotions have been
studied".
31
We employed the following procedural to classify the remaining 202 questionnaires as
representing a double deviation scenario. First, we use a measure of recovery disconfirmation i.e.
the degree to which a customer's expectations about service recovery were met, adopted from
Oliver (1980) and Oliver et al. (1997). Ratings were collected with a 5-point scale ranging from 1
(much worse than expected), 3 (as expected), to 5 (much better than expected). The answers
falling into 4 or 5 were considered successful recoveries (9 of the 202 questionnaires showed this
pattern of response). The answers falling into 1, 2, and 3 points in this scale (193 questionnaires),
were considered for the subsequent detection of the double deviation scenarios. Second, we
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
106 •
Chapter5
given the recommendation of a minimum sample size of 150 (Anderson and
Gerbing, 1988), or 200 (Hair et al., 1999), when testing structural equation
models with LISREL, and it is in line with similar studies (e.g., Blodgett et al.,
1993).
Of the respondents, 60% were men, the average age was 44 years (ranged
between 23 and 81 with a fairly normal spread), and the average household size
was 3.02 people. All levéis of income were represented. With respect to
education, the level of the respondents was quite high, as 36.3% of the
respondents had a degree. The average membership length to the consumer
organization was 6.21 years32.
5.3.2 Development of measures
The majority of the items were taken from the relevant literature, and also
based on the results of our in-depth interviews (qualitative) with 26 members of
the consumer organization. The questionnaire was subjected to the scrutiny of
some experts in marketing to check instructions, layout and length and item
relevance, sequence, wording and diffículty. After some modifícations, a pretest
of the questionnaire (in-depth interviews) was conducted using a sample of
individuáis of the intended population. Based on a descriptive examination of
the pretest data some items were modified. Next, we describe the final set of
measures employed.
employed an opened question to collect 'should' expectations, i.e. what the firm should have done
in order to restore initial satisfaction. Thus, we crossed this qualitative measure with the recovery
disconfirmation one (1, 2, and 3 points only) to assess that a failed recovery had occurred. The
combination of both the quantitative and the qualitative measures confirmed that all
questionnaires with scores 1 or 2 in the recovery disconfirmation scale were representatives of
double deviation scenarios (108 of the 193 questionnaires). Additionally, for the questionnaires
with a neutral score of 3 (85 of the 193 questionnaires) in the recovery disconfirmation scale, only
those that specifically reported the importance of improving recovery activities ('should'
expectations in the opened question) were classified as double deviations (73 of the 193
questionnaires). In brief, we classified 181 questionnaires as double deviation ones.
32
A series of one-way ANOVAs were performed in order to check whether significant
differences in the main variables for the demographic characteristics were present. In general, we
did not find significant differences.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario • 107
Magnitude ofservicefailure. The importance of the problem was developed
from different studies on service failures and recoveries (e.g. Folkes et al., 1987;
Hoffman et al., 1995; Mattila, 2001a,b; Maxham III and Netemeyer, 2002),
although many others (e.g., Smith et al., 1999) have manipulated this variable in
an experimental context (i.e., low/high importance of problem conditions).
Following Mattila (2001b), we assume that customer perceptions of magnitude
of failure are individually based. Participants were asked to rate two items
according to how they viewed the problem suffered. Specifically, we employed
two 5-point items, ranging from 1 (not at all importanf) to 5 {very important),
and 1 (not at all severe) to 5 (very severé).
Service failure controllability. We employed a one-item scale adapted from
Russell (1982) that had previously been used in similar studies (e.g., Taylor,
1994). The 5-point item was if the branch office could have prevented the
failure from occurring, ranging from 1 (not at all likely) to 5 (very likely).
Recovery Strategies: Apology and Explanation. Traditionally, these two
forms of recovery have been manipulated in experimental designs (e.g., Smith
and Bolton, 2002). However, we employ two single direct measures in order to
achieve greater knowledge about their specific contribution in service
failure/recovery encounters. The items were "the employees and/or director
apologized for the problem", for apology, and "the explanations I was given
were adequate", for explanation; both scales ranging from 1 (strongly agree) to
5 (strongly disagreé).
Distributive Justice. The items employed were adapted from Blodgett et al.
(1997) and Tax et al. (1998), and have been used previously (with some
modifícations) in similar studies such those of Maxham III and Netemeyer
(2003) and Smith et al. (1999), among others. Thus, participants were asked
about their evaluations of the bank's handling of the problem. The ratings were
collected with two items, "the outcome I received was fair" and "I got what I
deserved", with both scales ranging from 1 (strongly agree) to 5 (strongly
disagreé).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
108 • Chapter5
Anger with service recovery. Anger with service recovery was made up of
six ítems, "angry", "annoyed", "powerless", "frustrated", "irritated", and
"deceived". The fírst fíve were drawn from the works of Richins (1997), Taylor
and Claxton (1994), and Taylor (1994), and the last one from our preliminary
qualitative study. Taylor (1994) used items "angry", "annoyed", "frustrated",
and "irritated" in her study of delayed flights; Taylor and Claxton (1994) added
the items "bored", "powerless", and "helpless" to the previous ones in a similar
context; whereas Richins (1997) used items "angry", "frustrated", and
"irritated" in her refinement of several emotion-related scales into the
Consumption Emotion Set (CES). In our study, participants were asked to rate
the six items according to how they felt about the service recovery. Ratings
were collected with 5-point scales from 1 (not at all) to 5 (very much).
Satisfaction with service recovery. We use a three-item scale adopted from
Crosby and Stephens (1987), Spreng, MacKenzie and Olshavsky (1996), and
Westbrook and Oliver (1981). A similar scale has been used in previous studies
of service failure and recovery (e.g., Smith et al., 1999; Hess et al., 2003).
Participants were asked to indícate how they felt about the branch office given
its response to the problem suffered (service recovery), with three scales
ranging from 1 (pleased) to 5 (displeased), 1 (satisfied) to 5 (dissatisfled), and 1
(happy) to 5 (unhappy).
Finally, several variables not directly associated with the hypotheses testing
were included in the study. These control variables were based on the
demographic characteristics of the respondents: gender, age, education,
household size, income, and membership length to the consumer organization
(UCE). These variables were used to check possible differences among the main
variables used in the conceptual model, and also provided basic descriptive
information about the sample.
The final set of items used to measure the components of the conceptual
model are shown in Appendix 5.1.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario
•
109
5.3.3 Data analysis
In this section, we address the general data analysis procedure employed to
test the proposed model. We also provide a description of the measurement
jnodels tested.
5.3.3.1. General data analysis procedure
The method used to test the hypothesized model entailed a two-step
procedure suggested by Anderson and Gerbing (1988). First, the quality of the
measures of the constructs, i.e., the components of the conceptual model,
needed to be established. Subsequently, the proposed conceptual model as a
whole needed to be tested. This staged approach allowed us to maximize the
interpretability of both the fíndings for the measures and findings for the
conceptual model as a whole.
Within the context of Simultaneous Equation Modeling (SEM) the
conceptual model or structural equation model comprises two (sub)models
(Leeflang et al., 2000, p. 443). The fírst is called the measurement model, the
second the structural model. The measurement (sub)models relate the observed
or manifest indicator variables to a set of unobserved or latent constructs (see
Appendix 5.2). This part of the model is also called a confirmatory factor model
if it is considered in isolation. The structural (sub)model captures the
relationships between exogenous and endogenous latent constructs.
All the measures of the constructs were measured with fíve-point scales.
Additionally, our sample size was relative small (n=181), and that the majority
of the distributions of the data deviated from normality (which implies that the
necessary assumption of multivariate normality can not be accomplished).
Therefore, we used the Satorra-Bentler scaled Chi-Square (X2SB) statistic, a
statistic corrected for violations of multivariate normality (Satorra and Bentler,
1988)33, as recommended by Curren et al. (1996). The use of X2SB requires the
33
The Satorra-Bentler scaled Chi-Square (X2SB) works by adjusting, usually downward, the
obtained model fit chi-square statistic based on the amount of non-normality in the sample data.
The larger the multivariate kurtosis of the input data, the stronger the applied adjustment to the
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
110 » Chapter5
covariance and asymptotic covariance matrices as input matrices into LISREL.
The method of estimation was Máximum Likelihood.
5.3.3.2 Analysis ofthe measurement models
In this section, we center on the fírst stage of the procedure suggested by
Anderson and Gerbing (1988). We conducted a confírmatory factor analysis
(CFA) using LISREL 8.30 (JSreskog and Sorbom, 1996), which provided
assessment of overall fit with the data, convergent validity, discriminant validity
and construct reliability. First, however, for the construct "anger with service
recovery", we conducted exploratory factor analysis (EFA). The reason is that
we can consider this construct as a new construct which has not been previously
used in the literature (it includes the item "deceived") and, therefore,
preliminary exploratory research previous to the confírmatory one is useful to
assess unidimensionality (Hair et al., 1999). From this exploratory analysis, two
primary factors emerged from the data. The fírst factor included Ítems "angry",
"annoyed", and "irritated", whereas the second factor included items
"powerless" and "frustrated". These two factors were labeled "anger with
service recovery" and "frustration with failed service recovery", respectively,
following Roseman's (1991) appraisal theory of emotions. The sixth item,
"deceived" loaded on both factors, thus it was eliminated for further analysis
(Hair et al., 1999). Apart from this, given that our initial interest was in the
effect of anger in the double deviation context (in line with the specifíc emotion
approach), and for the sake of parsimony (given the small sample size), we
decided to center only on the fírst factor, "anger with service recovery". Thus,
the initial six-item scale was reduced to a three-item scale in the subsequent
confírmatory analysis.
chi-square test statistic. Standard errors for parameter estimates are adjusted upwards in much the
same manner to reduce appropriately the type I error rate for individual parameter estímate tests.
Although the parameter estímate valúes themselves are the same as those from a standard ML
solution, the standard errors are adjusted (typically upward), with the end result being a more
appropriate hypothesis test that the parameter estímate is zero in the population from which the
sample was drawn (Curren et al., 1996).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario • 111
Given that we had so few items per construct (max. 3)34, we could not
conduct confirmatory factor analysis on single construct measurement models,
because these models were under- or exactly identified (Ping, 2004). Therefore,
and following the recommendation of Bagozzi (1994), we computed a full
measurement model to gauge measurement model fít. However, insufficient
sample size existed for this completely inclusive factor analysis (i.e., too many
items to derive asymptotic covariance matrix, needed for non-normality
correction). Following the recommendations of Bentler and Chou (1987), we
specified two sub-models: MM1, for exogenous variables, and MM2, for
endogenous variables (see Table 5.1)35.
In general, we obtained acceptable levéis of model fít after modifícations for
double loading and non-loading items, which led us to the elimination of three
items (one of anger and two of satisfaction with service recovery, see Table
5.1). Following Ding and Hershberger (2002), the content validity can be
operationalized to be the magnitude of the direct structural relation between the
content structure (latent construct) and the observed item. Thus, as evidence of
content validity, each item loaded significantly on its respective construct,
which is also seen as a proof of convergent validity (Anderson and Gerbing,
1988; Bagozzi and Yi, 1988). An examination of the variance extracted
estimates (AVE) shows that all measures meet the norm set (AVE > 0.50;
Fornell and Larcker, 1981), indicating that a substantial amount (at least half) of
the variance in the measures is captured by the latent constructs, and showing
appropriate convergent validity (Fornell and Larcker, 1981). As evidence of
discriminant validity, for each construct, we obtained that the average variance
extracted estímate exceeded shared variance between the construct and all other
variables in the model (Fornell and Larcker, 1981). Finally, according to the
With respect to the use of many single-item measures, we recognize that this has an effect on
the assessing of psychometric reliability. However, we were concerned with the length of the
questionnaire and the desire of collecting information of many different constructs. In this sense,
we refer to the work of Drolet and Morrison (2001), who fínd that "as the number of items grow,
respondents are more likely to engage in mindless response behaviour. Thus the cost of asking the
same question more than once or twice appears to be higher than the cost of survey time only (p.
200)". Additionally, there are several examples of use of single-item scales in service research
(e.g., Bortón, 1998; Rustet al., 1995).
35
Following the decisión rules established by Jarvis et al. (2003) to avoid measurement model
misspecification, all constructs with more than one indicator were modeled as reflective.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
112 • Chapter5
LISREL-based composite (construct) reliabilities (CR), all measures meet the
norm set (CR > 0.60; Bagozzi and Yi, 1988).
Table 5.1
Analysis of measurement models (n=181)
Loading
Construct
Measurement Model 1 (exoe. vbles)
Magnitude of service failure
MAGSFl-important
MAGSF2-severe
Service failure controllability
SFCONTRl-prevent from occurring
Recovery strategies:
Apology
RECSTRl-apology
Explanation
RECSTR2-explanation
Measurement Model 2 (cndoa. vbles)
Distributive justice
DISTJl-outcome was fair
DISTJ2-got what deserved
Anger with service recovery
ANGRESl-angry
ANGRES2-annoyed (*)
ANGRES3-irrítated
Satisfaction with service recovery
SATRESl-pleased (*)
SATRES2-satisfied
SATRES3-happy (*)
Reliability of
Latent construct1
X
CR
AVE
1.000
.833
.897
.804
1.000
-
-
1.000
-
-
1.000
-
-
1.000
.837
.870
.768
1.000
.905
.908
.865
1.000
-
-
Goodness ofd fit
measures
Z2SB = 6.059
(p=0.081)
df=3
CFI=0.986
SRMR=0.025
Z2SB = 6.593
(p=0.253)
df=5
CFI=0.997
SRMR=0.022
-
a. pO.OOl
b. CR = composite reliability of latent construct; AVE = averaged variance extracted:
overall amount of variance in the indicators accounted for by the latent construct
(both measures añer item deletion).
c- X2SB = Satorra-Bentler scaled Chi-Square; df = degrees of freedom; CFI =
Comparative Fit Index; SRMR = Standardized Mean Square Residual.
(*) ítem deleted añer respecifícation of confirmatory model
- not available
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario • 113
5.4 Results
To test the role of anger and distributive justice in a double deviation
scenario, we employ path analysis. FoUowing the two-step procedure of
Anderson and Gerbing (1988), once we have estimated the measurement
models, the second step implies to estimate the structural model. Specifically,
the measurement models allow us to obtain estimations of loadings and error
variances of the fírst order variables' indicators, which are employed to estimate
the full structural equation model using single indicator structural equation
analysis (Ping, 1995, 2004). The single indicator approach implies to declare
each latent variable with only one indicator, providing a favorable ratio between
the number of parameters to be estimated and the sample size, which is needed
for the calculation and use of both the covariance and the asymptotic covariance
matrices (necessary to derive the Satorra-Bentler scaled Chi-Square statistic).
This procedure involves summing the items and then averaging them to provide
a single indicator of the unobserved construct. To avoid identifícation problems
(trying to estimate both a unique and a common factor loading as well as the
variance for one construct using only one indicator), we fix the common and
unique factor loadings at predetermined valúes and estimate only the variance
of the latent variable (Kelloway, 1998).
When we have at least two indicators for a latent variable (i.e., "magnitude
of service failure", "distributive justice", and "anger with service recovery"),
the solution, foUowing Ping's (2004) proposal, is to fíx the common factor
loading (the loading A of the averaged indicator X) to be equal to:
n
where X¡ are the loadings of the construct on x¡ from the measurement model
(exogenous or endogenous); and to fix the unique factor loading (the
measurement error variance of the average indicator X) at a valué equal to:
^=Var(X)(l-p),
where Var(X) is the variance of the averaged indicator X, and p is the latent
construct reliability of X.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
114 • Chapter5
When we have only one indicator for a latent variable (i.e., "servíce failure
controllability", "apology", "explanation", and "satisfaction with service
recovery"), we fíx the common factor loading equal to 1 and the unique factor
loading equal to 0 (assuming perfect reliability in the single item) (Kelloway,
1998, p. 136).
Correlations and summary statistics of the composite measures are presented
in Table 5.2.
Table 5.2
Correlations and summary statistics of composite measures (n=181)36
Variable
Magnitude of
service failure (Xj)
(R)
Service failure
controllability (X 2 )
(R)
Apology (X3)
Explanation (X 4 )
Distributivejustice
(Y,)
Anger with service
recovery (Y2) (R)
Satisfaction with
service recovery
(Y3)
Mean
SD
X,
1.000
-0.026
x2
x3
X4
Y,
Y3
1.000
-0.195***
-0.297***
-0.359***
-0.121
-0.180**
-0.019
1.000
0.473***
0.427***
1.000
0.509***
1.000
0.379***
-0.008
-0.127
-0.239***
-0.158**
-0.325***
-0.156**
0.298***
0.420***
0.433***
4.130
1.277
4.270
1.042
4.146
1.256
1.668"
0.810
Y2
1.614a
1.133
1.000
-0.511***
1.917"
1.211
1.000
4.540
0.801
a. Mean score refers to the reverse coded construct.
(R) Reverse coded for analysis
**p<.05; ***p<M
The relationships hypothesized in Figure 5.2 are tested using LISREL 8.3
(Joreskog and Sorbom, 1996) with the sample covariance and asymptotic
covariance matrices as input matrices. The results are summarized in Table 5.3
and in a graphic form in Figure 5.3.
Given that none of the bivariate correlations was greater than 0.85, we can assume that
multicollinearity is not a problem in our data (Grewal et al., 2004).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario • 115
Table 5.3
Results of the structural equation model of (dis)satisfaction with service
recovery in a double deviation scenario
• ' —
Hy P .
Model
estímate
Hla
0.021
0.081
Path
Stand.
error
Direct Effects
Magnitude of service failure —* Satisfaction with service
recovery (SatSR)
Magnitude of service failure —* Distributive Justice
Magnitude of service failure —* Anger with service recovery
(AngerSR)
Service failure controllability —* SatSR
Service failure controllability —* Distributive Justice
Service failure controllability —» AngerSR
Apology —»Distributive Justice
Apology -> AngerSR
Explanation —•Distributive Justice
Explanation —»AngerSR
Distributive Justice —» SatSR
Distributive Justice —> AngerSR
Anger with service recovery —» SatsR
Hlb
Hlc
-0.354**
0.584***
0.147
0.125
H2a
H2b
H2c
H3a
H4a
H3b
H4b
H5a
H5b
H6
-0.102***
0.074
-0.023
0.242**
-0.009
0.462***
-0.209
0.247***
0.090
-0.322***
0.038
0.060
0.063
0.094
0.086
0.127
0.128
0.050
0.112
0.071
Indirect Effects
Magnitude of service failure—> Distributive justice —* SatSR
Magnitude of service failure—» AngerSR—• SatSR
Service failure controllability —* Distributive justice—• SatSR
Service failure controllability —»AngerSR—> SatSR
Apology —* Distributive justice —* SatSR
Apology —> Anger SR -> SatSR
Explanation —>Distributive justice —> SatSR
Explanation —»AngerSR —» SatSR
Distributive Justice —»AngerSR—> SatSR
Hlbb
Hice
H2bb
H2cc
H3aa
H4aa
H3bb
H4bb
H5bb
-0.087**
-0.188***
0.018
0.007
0.060**
0.003
0.114***
0.067
-0.029
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
0.036
Squared Múltiple Correlations for Structural Equations (R2)
Distributive Justice
Anger with service recovery
Satisfaction with service recovery
Model Goodness-of-Fit Statistics
Satorra-Bentler Scaled Chi-Square (X2SB) [dfj (p-value)
Root Mean Square Error of Approximation (RMSEA) (pvalue)
Comparative Fit Index (CFI)
Root Mean Square Residual (RMR)
n.a. not available. **/K.05; ***/?<.01
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
0.414
0.203
0.465
1.791 [2] (0.408)
0.000 (0.544)
0.997
0.013
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
116 • Chapter5
Figure 5.3
Signifícant results of the proposed model of (dis)satisfaction with service
recovery in a double deviation scenario
Service recovery-related variables
Service failure-related variables
Magnitude of
service failure
Service failure
controllability
Apology
^ÑT
Z
'0.24*
-0.10***
/
Explanation
0.46*í
-0.35*
0.58*
*
f
\.
y
*\.
Distributive
iustice
0.25***^^
A'
\ - 0 . 3 2 * * * ' * / Satisfaction
Anger with
service recovery
with service
recovery
)——n
Note: indirect effects not showed for clarity purposes. Dotted arrows denote nonsignifícant (hypothesized) relationships.
Model fit statistics in Table 5.3 collectively indícate that the proposed model
fíts the data quite well (X 2 SB=1-791, p=0.408, df=2; RMSEA =0.000, p=0.544;
CFI=0.997; RMR=0.013). Most path coeffícients are as hypothesized.
5.4.1 Direct effects and indirect effects, through cognitive and
emotional antecedents, of service failure-related variables
In this section, we address the main results regarding the effects of the
service failure-related variables examined: magnitude of service failure and
service failure controllability.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario
•
117
Magnitude of service failure
We do not find evidence of the direct and negative effect of magnitude of
service failure on satisfaction with service recovery (Hla). However, we find
that magnitude of service failure affects the perceptions of distributive justice
negatively, consistent with Hypothesis Hlb. Additionally, we also find that
magnitude of service failure affects anger with service recovery positively,
congruent with Hypothesis Hlc.
The data also support the predicted indirect effect of magnitude of service
failure on satisfaction with service recovery through distributive justice (Hlbb: 0.087, p<0.05), and through anger with service recovery (Hice: -0.188,
p<0.01)37. It appears that anger has a stronger indirect effect (in absolute terms)
than distributive justice in the relationship between magnitude of service failure
and satisfaction with service recovery. In fact, results in Table 5.3 show that the
effect of magnitude of service failure is higher on anger with service recovery
than on distributive justice (P = 0.584 vs. p = -0.354). A Satorra-Bentler Scaled
Chi-Square difference test on the equality of the parameters confirms this (x2sBd
(1)= 16.359;/? = 0.000)38.
LISREL 8.3 provides only the total indirect effect of magnitude of service failure on
satisfaction (-0.280, p<0.01), without distinguishing the two ¡nherent and different mediation
effeets of anger and distributive justice. Therefore, for testing separately the specific mediating
effeets, Barón and Kenny (1986) provide a test of ab which is a modification of a test originally
proposed by Sobel (1982). Being, a = raw (unstandardized) path coeffícient for the association
between the independent variable and the mediator, s¡¡ = standard error of a, b = raw coeffícient
for the association between the mediator and the dependent variable (when the independent
variable is also a predictor of the dependent variable), and sb = standard error of b, and following
Goodman (1960), the standard error of ab can be shown to equal approximately the square root of
b2sa2 + a2sb2 - sa2sb2. The test of the indirect effect is given by di viding ab by the square root of the
above variance and treating the ratio as a Z test. Sobel (1982) presents the above formula without
the last term. Barón and Kenny (1986) present the above formula with the last term added, not
subtracted. That formula is a population formula whereas the Goodman formula is the samplebased estímate. The reported p-values are drawn from the unit normal distribution under the
assumption of a two-tailed z-test of the hypothesis that the mediated effect equals zero in the
population. No differences in the three versions of the test were found.
38
A scaled chi-square cannot be used for chi-square difference testing of nested models because a
difference between two scaled chi-squares for nested models is not distributed as chi-square
(Satorra and Bentler, 2001). To deal with this, Satorra and Bentler (2001) show how to conduct a
chi-square difference test of nested models using the scaled chi-square. Specifically, when fitting
models M0 and Mi, we obtain the unscaled and scaled goodness-of-fít tests, that is T0 and
T0 when fitting M0, and T and 7] when fitting Mi. Let r0 and r, be the associated degrees of
freedom of the goodness-of-fít test statistics. Then we compute the scaling corrections
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
118 • Chapter5
Service failure controllability
As predicted in Hypothesis H2a, service failure controllability has a negative
direct impact on satisfaction with service recovery. However, the paths from
service failure controllability to distributive justice (H2b), and to anger with
service recovery (H2c), as well as the indirect effects of distributive justice
(H2bb) and anger (H2cc) in the relationship between service failure
controllability and satisfaction with service recovery, are not significant.
5.4.2 Indirect effects, through cognitive and emotional antecedents,
of service recovery-related variables
Regarding the recovery strategies, first, we find that the paths from apologies
and explanations to distributive justice are positive and significant, thus
providing support for Hypotheses H3a and H3b. Additionally, we detect a
significant indirect effect of distributive justice in the relationship between
recovery strategies and satisfaction with service recovery (H3aa, H3bb). We
also tested an alternative model to examine the direct effects of recovery
strategies on satisfaction with service recovery, maintaining the rest of the
relationships hypothesized unaltered, which represented the partial mediation
model. The model showed globally a good fit, and all parameters that were
significant in the final model showed in Table 5.3 remain significant and with
the same sign. However, none of the paths relating recovery strategies with
satisfaction with service recovery were significant. These results confirm that
distributive justice is a complete mediator in determining post-recovery
customer satisfaction.
Finally, we do not detect a direct effect of apology and explanation on anger
with service recovery, thus rejecting Hypotheses H4a and H4b. Logically, we
neither find a significant indirect effect of anger with service recovery
(emotional antecedent) in the relationship between recovery strategies and
c 0 = T0/T0 and c¡ = 7 j / 7 ¡ , and the usual chi-square difference Td = T0 - T]. The SB scaled
difference test can thus be computed as Td = Td/cd , where cd = (r0c0 - >j<5i)/(>b —>\)
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario
satisfaction with service recovery (H4aa, H4bb). Further tests to examine partial
tnediation make, therefore, no sense.
5.4.3 Direct and indirect effects
antecedents
of cognitive and
emotional
In this section, we address the main results regarding the effects of
distributive justice (cognitive antecedent) and anger with service recovery
(emotional antecedent).
Cognitive antecedent: distributive justice
We fmd a signifícant and positive effect of distributive justice on satisfaction
with service recovery, consistent with Hypothesis H5a. However, we do not
find evidence of the direct and negative effect of distributive justice on anger
with service recovery, thus rejecting Hypothesis H5b. At first, it appears that
distributive justice parallels anger with service recovery for complementary
impact on satisfaction, that is, that they coexist as sepárate dimensions in
consumers' satisfaction judgments. We will discuss this result later on.
Additionally, contrary to our predictions, we do not fmd a signifícant indirect
effect of anger with service recovery in the relationship between distributive
justice and the satisfaction construct (H5bb). As long as the direct relationship
between distributive justice and anger is not signifícant (H5b), this result could
be expected.
Emotional antecedent: anger with service recovery
Finally, the data support the predicted direct and negative effect of anger
with service recovery on satisfaction with service recovery (H6). In fact, our
results show that the emotional component is a stronger predictor of satisfaction
with service recovery than the cognitive component ((3 = -0.322 vs. p = 0.247).
A Satorra-Bentler Scaled Chi-Square difference test on the equality of the
parameters confirms this (X2SB<I (1) = 30.575; p = 0.000) (Satorra and Bentler,
2001).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
120 » Chapter5
5.5 Discussion
We have not found evidence of a direct and negative effect of magnitude of
service failure on satisfaction with service recovery (Hla), contrary to previous
research on how customers respond to service failures (e.g., Gilly and Gelb,
1982; Hoffman et al., 1995; Richins, 1987). However, the results show that
magnitude of service failure affects the perceptions of distributive justice
negatively (Hlb), in Une with Blodgett et al. (1993). That is, consumers who
experience a failure with products they feel are important are likely to perceive
the act of complaining as an injustice in and of itself. Additionally, magnitude
of service failure affects anger with service recovery positively (Hlc).
Consistent with social exchange theory (e.g., Walster et al., 1973), as service
importance increases, the greater are the customer's expectations of high recovery
efforts to be taken by the organization. Therefore, when customer's expectations
are not achieved (the customer experiences a failed recovery), the magnitude of
service failure enhances anger with service recovery. This result is similar to the
one obtained by Smith and Bolton (2002) for initial negative emotions (they do
not measure post-recovery emotions).
With respect to the indirect effects of magnitude of service failure, the
results show a significant indirect effect on satisfaction with service recovery
through distributive justice (Hlbb), and through anger with service recovery
(Hice). Thus, the magnitude of the service failure reduces the perceptions of
distributive justice associated with the recovery effort (Hlb) and enhances anger
experienced during the recovery (Hlc), and both the higher anger and the lesser
distributive justice diminish the subsequent satisfaction with the failed service
recovery. Additionally, anger has a stronger mediating effect (in absolute terms)
than distributive justice in the relationship between magnitude of service failure
and satisfaction with service recovery. All this together evidence the main role
of anger vs. distributive justice in this context. That is, when the magnitude of
service failure is high, as long as customer's expectations are not met
(unsuccessful recovery), the effect will be greater on the anger triggered by the
failed recovery than on the distributive justice associated to the recovery effort.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario • 121
Regarding service failure controllability, we have only found evidence of a
negative direct impact on satisfaction with service recovery (H2a), in line with
proposals of Maxham III and Netemeyer (2002). The direct effects on
distributive justice (H2b) and anger with service recovery (H2c) are not
significant. Neither are the indirect effects of distributive justice (H2bb) and
anger (H2cc) in the relationship between service failure controllability and
satisfaction with service recovery. A possible explanation for these results could
be found on the work of Hess et al. (2003). These authors consider that
customers may revise their attributions after the service recovery in an
attribution-updating process which is more probably affecting post-recovery
customer perceptions than the initial attributions (immediately after failure). As
long as we do not have measured post-recovery attributions, we can not check
this assumption.
As for service recovery-related variables, we have found support for
Hypotheses H3a and H3b. First, apology affects positively customer's
perceptions of distributive justice. This result is in line with Hoffman and
Kelley (2000), Mattila (2001a), Mattila and Patterson (2004), Tax et al. (1998),
or Tax and Brown (2000). Second, explanations influence the perceived fairness
of outcomes (distributive justice), in line with Shaw et al. (2003), and the
proposal of Mattila and Patterson (2004). We have also detected a significant
indirect effect of distributive justice in the relationship between recovery
strategies and satisfaction with service recovery (H3aa, H3bb), in line with the
proposals of Oliver (1997) and Smith et al. (1999). These authors state that
recovery strategies opérate indirectly through perceived justice to influence
customer satisfaction. The test of alternative models has confirmed that
distributive justice is a complete mediator in determining post-recovery
customer satisfaction, in line with Oliver's (1997) proposal.
However, we have not found evidence for hypotheses H4a, H4aa, H4b and
H4bb (direct and indirect effect of anger with service recovery in the
relationship between recovery strategies and satisfaction with service recovery).
These results are contrary to the fairness theory-based conceptual framework
proposed by McColl-Kennedy and Sparks (2003), and support the proposals of
Oliver (1997) and Smith et al. (1999) which state that recovery strategies
opérate through perceived justice to influence customer satisfaction. All this
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
122 •
Chapter5
together evidence the main role of distributive justice vs. anger as a complete
mediator in determining post-recovery customer satisfaction in double deviation
contexts. That is, service recovery-related variables influence customer postrecovery satisfaction indirectly through distributive justice, but have no effect
on post-recovery emotions.
Results also have shown a significant and positive effect of distributive
justice on satisfaction with service recovery (H5a), in line with previous
research (e.g., Matóla, 2001a; Smith and Bolton, 2002; Smith et al., 1999; Tax
et al., 1998). In fact, Smith and Bolton (2002) show that the distributive
component accounts for 75.6 percent of the explained variance in the
satisfaction with service recovery judgments for customers showing an
emotional response. As stated previously, distributive justice has its origins in
the study of social exchange (Blau, 1964) and equity theory (Adams, 1965), and
refers to the outcomes an individual receives in social exchanges. Thus, we find
that distributive justice, defíned as the equity or fairness of rewards with respect
to the bank inputs (recovery strategies of apology and explanation), affects
positively satisfaction with service recovery.
Interestingly, our data do not support the direct and negative effect of
distributive justice on anger with service recovery (H5b). Logically, the
hypothesized indirect effect of distributive justice on satisfaction with service
recovery through anger with service recovery is also rejected (H5bb). At fírst, it
appears that distributive justice parallels anger with service recovery for
complementary impact on satisfaction, that is, that they coexist as sepárate
dimensions in consumers' satisfaction judgments. This result is in line with
Dubé-Rioux (1990) who found that affective responses were independent and
more powerful indicators of customer satisfaction than cognitive evaluations.
Although this result seems contrary to the findings of Chebat and Slusarczyk
(2005), it is important to note that they do not measure anger as negative
emotion but anxiety and disgust, and therefore, comparisons are difficult. Given
that distributive justice and anger with failed recovery constitute two main
elements in our research, this result deserves a deeper discussion.
A possible explanation for this result is that there can be other appraisal
dimensions different from distributive justice that covary with the evaluation of
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distribuíive Justice in a Double Deviation Scenario • 123
distributive justice to form the emotional response. Justice theory generally
conceptualizes injustice as an instance of inconsistency, dissonance, or
refutation of existing expectations that elicits some kind of distress in the
perceiver (Mikula et al., 1998). Theory and research point to a wide range of
possible consequences of injustice that should be reflected in the appraisal of
the situation (e.g., Adams, 1965; Mikula, 1984; Reis, 1984; Walster et al,
1978). According to this body of knowledge, unjust situations can be expected
to be appraised, among other things, as less expected, more aversive or
unpleasant, more of a hindrance to goal achievement, and more detrimental to
the self-concept, as compared with situations that are not regarded as unjust. In
this Une, Mikula et al. (1998) find, for the specifíc emotion of anger, strong and
highly signifícant unfairness main effects for goal hindrance and immorality
(i.e., events perceived as unfair are seen as much more goal hindering and
immoral than events not seen as unfair). Therefore, the authors conclude that
although perceived injustice may increase the likelihood of an anger reaction in
terms of an a priori response tendency, the exact nature of the emotional
reaction would be determined by the combined outcome of the appraisal on
several dimensions- such as the pertinence of the consequences, the perceived
causal agent, the estimated coping ability, and so forth (Mikula et al., 1998, p.
779). Thus, as long as we have not measured other a priori signifícant appraisal
antecedents of injustice when examining its role as anger-antecedent, this could
explain the lack of signifícance of the relationship between distributive justice
and anger with service recovery.
Additionally, there could be differences in affective reactions to situations
perceived as unjust because of the existence of interrelations between perceived
injustice and the appraisal variables above mentioned, such as the attributions of
responsibility. This perspective is present in the work of Montada and his
coworkers (e.g., Montada, 1994; Montada and Schneider, 1989; Reichle and
Montada, 1994), which focus on the mediating (and/or moderating) role of
evaluations of injustice and attributions of responsibility in the elicitation of
specifíc emotions, which, in turn, affect the way in which people experience and
cope with a given situation. They state that in many cases, the judgment of
injustice may be conditional on the ability to assign agency to an external factor,
particularly a person or a group, and thus, most unfairness appraisals are
probably conditioned by assigning the responsibility for a consequence to
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
124 • Chapter 5
someone else (see Mikula, 1993). In our context of double deviation, if
consumers feel that the service provider could have done more during the
recovery process and did not, then they assess this negatively and experienced
more intense negative emotions such as more anger. This sequence corresponds
to the consumer model of appraisal of service failure/recovery events developed
by McColl-Kennedy and Sparks (2003) by which customers make assessments
of the service recovery attempt following a service failure in terms of
accountability (perceived volitional control over actions taken by the provider)
and counterfactual thinking (contrasting what is perceived to be with what
might have been). Thus, in evaluating accountability (and therefore fairness) for
an event (e.g., poor service recovery), McColl-Kennedy and Sparks (2003)
propose that a customer engages in three contrastive actions: what could have
occurred, what should have occurred, and how he/she would have felt had
alternative action been taken, in terms of interactive, procedural, and
distributive justice. The answers to these questions enable the customer to then
assign accountability for the negative event (poor service recovery). Such an
assessment results in an emotional response (e.g., less or more angry as the time
of the service failure). In any case, the absence of these accountability and
counterfactual elements in our model could explain the lack of significance and
the contrary sign of the relationship between distributive justice and anger with
service recovery.
Finally, our fíndings show the existence of a direct negative effect of anger
with service recovery on satisfaction with service recovery, thus supporting
Hypothesis H6. That is, when service providers do not appear to put proper
effort into the service recovery attempt, this is viewed negatively, and this led to
the customer experiencing negative emotions such as anger (McColl-Kennedy
and Sparks, 2003). This result confírms also the Mano and Oliver's (1993)
framework, which identifies negative affect as a negative antecedent to
satisfaction, in the context of double deviation scenarios (service failure and
failed recovery encounters). In fact, our results show that the direct effect of
anger with service recovery on satisfaction is higher (in absolute terms) than the
direct effect of distributive justice. These results evidence a main role of anger
vs. distributive justice as antecedent of satisfaction with service recovery. This
is in line with Dubé-Rioux (1990) who found that affective responses were
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Jusíice in a Double Deviation Scenario • 125
independent and more powerful indicators of customer satisfaction than
cognitive evaluations.
5.6 Conclusions
Financial institutions in general and the banking sector in particular, are
among the service organizations that face huge competition all over the world.
This competition has enabled customers to act in a more demanding way in
their interaction with service providers due to the increased abundance of
choice. Obviously, service failures or mistakes are unavoidable, even for the
best service company, and therefore, the effective management of consumer
responses to service failure becomes very important in these highly competitive
markets (Hart et al., 1990). But the critical test comes when the company fails
in the recovery process, that is, when the customer is faced with a double
deviation. Has the service provider still a possibility to satisfy this customer?.
This work has proposed and empirically analyzed a model centered on
double deviation scenarios (i.e., failed recovery following a service failure),
where different service failure- and service recovery-related variables impact on
satisfaction with service recovery through the cognitive antecedent of
distributive justice and the emotional antecedent of anger with service recovery.
Our main goal was to broad the knowledge about the type of variables and the
magnitude of their effect that contribute to the formation of satisfaction with
service recovery judgments in a double deviation scenario, integrating two main
theories in our conceptual framework, the justice theory and the emotions
theory.
Our field study based on a cross-sectional sample of 181 dissatisfíed banking
customers has shown the following results. First, magnitude of service failure
directly affects anger with service recovery and distributive justice, and
indirectly affects satisfaction with service recovery through distributive justice
and anger (with a main role of anger vs. distributive justice as a mediating
variable). Second, service failure controllability affects satisfaction with service
recovery. Additionally, apologies and explanations significantly affect
distributive justice. Fourth, distributive justice mediates the relationship
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
126 •
Chapter5
between apologies and explanations and satisfaction with service recovery, and
also has a direct, positive, and signifícant effect on the later variable. And,
finally, anger with service recovery is found to directly affect satisfaction with
service recovery.
Whereas justice theory appears to be the dominant theoretical framework
applied to service recovery (Tax and Brown, 2000; Tax et al., 1998), this study
suggests that a specific emotion approach should also be considered when
dealing with double deviation scenarios. In fact, in this study we fínd that
distributive justice play a dominant role in explaining customer satisfaction with
service recovery when dealing with service recovery-related variables, whereas
the specific emotion of anger play a main role in explaining customer
satisfaction with service recovery when dealing with a service failure-related
variable such as magnitude of service failure. Additionally, our results evidence
a main role of anger vs. distributive justice as a direct antecedent of satisfaction
with service recovery. That is, when customer's expectations are not met
(unsuccessful recovery), the effect of the anger triggered by the failed recovery
on post-recovery judgments (satisfaction with service recovery) will be higher
than the effect of distributive justice associated to the recovery effort. Thus,
emotions have a greater direct impact on customer's service evaluations than
cognitions in double deviation contexts. This finding supports the importance of
understanding the antecedents to anger triggered by the failed recovery39.
39
In fact, the squared múltiple correlation for the structural equations, that indícate the proportion
of variance in the endogenous variables accounted for by the variables in structural equations, that
IS !-
{
V'l/, . , ., show that there are antecedents of anger with failed service recovery that we
/M»,)J
are not taking into account (we explain only a 20.3% of the variance of this construct with our
model).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Anger and Distributive Justice in a Double Deviation Scenario • 127
Appendix 5.1
Measures employed in the study
Magnitude of service failure
Based on all ofyour experiences with banks, how didyou view thisproblem? (anchors:
1. l=Not at all important /5=Very important
2. l=Not at all severe/5=Very severe
Initial set of measurement items: 1-2
Final set of measurement items adopted in the structural model: land 2 (R)
Service failure controllability
1. From your point ofview, could the branch office have prevented this problem from
occurring? (anchors: l=not at all likely; 5=very likely)
Initial set of measurement items: 1
Final set of measurement items adopted in the structural model: 1 (R)
Recovery strategies
Apology
1. The employees and/or the director apologized for my problem (anchors: l=strongly agree;
5=strongly disagree)
Initial set of measurement items: 1
Final set of measurement items adopted in the structural model: 1
Explanation
1. The explanations I was given were adequate (anchors: l=strongly agree; 5=strongly
disagree)
Initial set of measurement items: 1
Final set of measurement items adopted in the structural model: 1
Distributive Justice Thinking about the bank's handling of the problem (anchors: l=strongly
agree; 5=strongly disagree):
1. The outcome I received wasfair
2.1 got what I deserved
Initial set of measurement items: 1-2
Final set of measurement items adopted in the structural model: 1 and 2
Anger with service recovery
On that moment, thinking about the bank's handling ofthe problem, to what extent didyou
feelyourself. (anchors: l=not at all; 5=very)
1. Angry?
2. Annoyed?
3. Powerless?
4. Frustrated?
5. Irritated?
6. Deceived?
Initial set of measurement items: 1-6
Final set of measurement items adopted in the structural model: 1 and 5 (R)
Satisfaction with service recovery
On that moment, thinking about the bank 's handling ofthe problem, how didyou feel about
the branch office'! (anchors:
1. l=Pleased/5=Displeased
2. l=Satisfied/ 5=Dissat¡sfied
3. l=Happy/ 5=Unhappy
Initial set of measurement items: 1-3
Final set of measurement items adopted in the structural model: 2
Note: all items measured with 5-point scales. (R) Reverse coded for analysis
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
128 •
Chapter5
Appendix 5.2
Formulation of measurement and structural models
The measurement models for the endogenous and the exogenous indicator variables
are formulated in a general form respectively as (Bollen, 1989; Hair et al., 1999;
Leeflang et al., 2000):
y = AyTj + s
x = AJ + S
where:
y = a \p x l) vector of manifest endogenous indicator variables;
r¡ = a ( m x l j vector containing the latent endogenous constructs (i.e., the variables
that are explained within the model);
Ay = the \p x m) matrix of loadings, showing which manifest indicator variable
loads on which latent endogenous construct;
£ = a vector of random error terms with expectation zero, and uncorrelated with TJ.
Each £¡ is homoscedastic and nonautocorrelated (see explanation in the vector of
disturbances of the structural model, below);
x = a \q x 1) vector of manifest exogenous indicator variables;
S, = a \n x 1 j vector of latent exogenous constructs (i.e., variables that explain the
model);
Ax =the \qx n) matrix of loadings, showing which manifest indicator variable loads
on which latent exogenous construct; and
8 = a vector of random error terms with expectation zero, and uncorrelated with £.
Each St is homoscedastic and nonautocorrelated (see explanation in the vector of
disturbances of the structural model, below).
The structural model is formulated in a general form as (Bollen, 1989; Hair et al.,
1999; Leeflang et al., 2000):
where:
B = a (mxm) matrix specifying the relationships among the m latent endogenous
constructs;
r = a (m x n) matrix that captures the effects of the n exogenous constructs on the
m latent endogenous constructs;
g= a vector of disturbances with expectation zero and uncorrelated with the
endogenous and exogenous variables. Each gi is homoscedastic and nonautocorrelated.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Volver al Indice/Tornar a l'índex
Anger and Distributive Justice in a Double Deviation Scenario • 129
The homoscedasticity assumption is that theVARyg^
is constant across cases [i.e.,
E\Cik)= VAR\g¡) for all k]. The no autocorrelation assumption means that ga for all
jfc and /, where k^l (i.e., COVygilc, gu ) = 0 for k^ /)• The homoscedasticity and no
autocorrelation assumptions do not mean that the disturbances from two different
equations need be uncorrelated ñor that they need have the same variance.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Chapter 6
Third-party Complaints and Banking Market Valué: The
Moderating Effects of Quality Corporate Image and
Market Concentration
6.1 Introduction
The main objective of this research is to analyze the impact of third-party
responses on company performance. Until now, little attention has been paid to
third-party complaints, where the customer takes a concern to a government
agency, consumer protection group, Better Business Bureau, or some formal
party external to the original marketing exchange (McAlister and Erffmeyer,
2003).
According to Singh's (1988) classifícation, third-party complaints are one of
the possible responses customers might employ to cope with unsatisfactory
service experiences. Moreover, third-party actions are most likely to occur when
customers (Duhaime and Ash, 1979; Singh, 1989; Tipper, 1997; Ursic, 1985):
perceive that the company's initial remedy was inadequate; have good access to
the legal system and other formal agencies; believe that all other complaining
options have been unsuccessful; experience high anxiety levéis about the
complaint situation, and have generally negative attitudes toward business
practices.
Therefore, third-party complaints are especially important and troublesome
to marketers since they represent a higher-order action than complaining to
friends, family, the salesperson or the company (Feick, 1987). The effort and
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
132 » Chapteró
involvement associated with third-party complaining behavior normally
indicates a degree of consumer dissatisfaction, company unresponsiveness or
related factors that can severely threaten marketing relationships and
effectiveness. In addition, third-party complaints can result in extensive legal
costs, regulatory intervention and corporate reputation problems (Tipper, 1997).
However, despite the importance of third-party complaints, we have not
found studies that empirically analyze their impact on company performance.
Previous research has demonstrated that a defensive marketing strategy
(retaining customers through complaint handling systems) brings about an
increase in market share and profits, as well as reduces the costs of offensive
marketing (obtaining additional customers, encouraging brand switching and
increasing buying frequency) (Fornell and Wernerfelt, 1987, 1988).
Nevertheless, empirical studies in this área focus only on the impact of
complaint management on intention to repurchase not on actual firm
performance (Andreassen, 1999; Fornell, 1992; Gilly and Gelb, 1982; TARP,
1979, 1981).
This study is, therefore, an attempt to fill this gap in marketing literature
joining the demands for greater attention to be given to third-party complaints
made by various authors (e.g., Fisher et al., 1999; Singh, 1989). Specifically, we
examine complaints from the Bank of Spain's Complaints Service (the third
party), which publishes an Annual Report on Complaints to Spanish banks.
Thus, we propose that the reléase of this information about third-party
complaints (Annual Reports on Complaints) is economically relevant to the
stock market. Therefore, the methodology applied is based on the event study to
estímate the variation in bank share returns resulting from the pubhcation of
these annual reports between 1992 and 2002.
Consequently, our research also contributes to the empirical literature by
applying the event study's methodology on stock market share prices, which
solves the following operational problems derived from the way in which
company performance is usually measured (Myers, 1972; Ramírez-Alesón and
Espitia-Escuer, 2001). First, the majority of studies use fínancial information
from accounting records to measure performance but their conclusions are
largely incomparable due to disparities between the variables used. Second,
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and banking market valué • 133
accounting data can be insufficient as it does not incorpórate expectations on
future profíts and can lead to confusión due to the defíciencies inherent in its
dependence on different conventions (e.g., rate of depreciation). Third, they do
not allow the adjustment of differences in performance for differences in the
risk supported by the firms. Fourth, they do not reflect all the opportunity costs
held by the fírm. And, fifth, accounting measurements are inappropriate in
certain contexts, such as that of complaints, quality and dissatisfaction, whose
returns are manifested over long periods of time (Anderson et al., 1994;
Lambert, 1998).
Finally, we also examine the effect of the number of complaints and the
moderating effects of quality corporate image and market concentration in this
context. For testing these relationships we use regression analysis and the
moderated regression analysis (MRA).
6.2 Conceptual framework and research hypotheses
In the following subsections, we review the Hterature that supports the
relationships proposed and present the hypotheses.
6.2.1 Relationship between the firm's appearance in the Annual
Report on Complaints and the firm's performance
As indicated previously, third-party complaints suggest: i) customer
dissatisfaction or related factors which threaten marketing relationships; ii) cost
increases that reduce productivity; and iii) corporate reputation problems.
With regard to the fírst aspect, research into emotions holds that negative
emotional responses (such as anger or disappointment) to a service failure
precede complaint behavior and less intention to repurchase (Weiner, 2000;
Zeelenberg et al., 1998, 2000). On the other hand, the Hterature on
dissatisfaction considers a complaint to be a negative response to dissatisfaction
which precedes customer defection (Richins, 1983; Zeithaml et al., 1996).
Essentially, and despite their differences, both perspectives suggest that third-
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
134 • Chapter 6
party complaints negatively affect company performance, as complaints precede
customer defection. Firstly, less customer loyalty implies a lower probability of
repurchase from the same provider (Fornell, 1992), which will be reflected by
lower company returns, as less loyalty jeopardizes a steady stream of future
cash fiow (Reichheld and Sasser, 1990; Rust and Zahorik, 1993; Rust et al.,
1994, 1995). Secondly, lower customer retention should increase a company's
future transaction costs as it will no longer benefít from the purchase of other
goods and services (offered by the firm) by satisfled customers, or the price
premiums that satisfied customers are willing to pay (Reichheld and Sasser,
1990). Moreover, the company needs to spend heavily (advertising, promotions
and sales costs) to gain new customers (Zeithaml et al., 1996). Additionally,
dissatisfied customers are most likely to engage in negative word-of-mouth
(Anderson, 1994; Reichheld and Sasser, 1990), which could reduce the
effectiveness of advertising and the attractiveness of warranties (Anderson et
al., 1994).
Apart from the effect on performance from less customer retention, thirdparty complaints reflect product/service failures. Therefore, authors such as
Crosby (1987) and Garvín (1988) propose that companies should allocate a high
level of resources to the handling and management of complaints, as well as
rework any defective producís, which worsen productivity. Henee, Anderson et
al. (1994) and Anderson et al. (1997), following an economic approach, suggest
that product failure cost increases and productivity reduction will diminish
company performance.
Finally, third-party complaints negatively influence company performance
as a result of the loss of company reputation. The dissatisfaction of the
complaining customer would indícate a worsening company reputation. Lower
company reputation does not aid in introducing new produets given the instant
awareness of the existence of complaints and the increase of risk of trial for the
buyer (Robertson and Gatignon, 1986). Additionally, lower corporate reputation
is not beneficial for maintaining and establishing relationships with suppliers,
distributors and potential allies (Anderson and Weitz, 1989) ñor does it facilitate
the building of other company assets such as brand equity (Aaker, 1992).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and bankingmnr^t „„;„,, .
135
In this study, third-party complaints refer to the complaints that appear in the
Annual Report on Complaints that the Complaints Service of the Bank of Spain
publishes every year. Thus, we propose that there is a relationship between the
firm's appearance in the complaint report and its performance. This relationship
is based theoretically on the relationship between third-party complaints and
corporate reputation. Therefore, we follow the signaling theory to assume that
investors may respond to fírm-level news (like the firm's appearance in a report
of complaints to a third-party agency) by selling the stock (Asquith and Mulitas,
1986), given that these negative announcements constitute signáis of lower
future profitability. A sudden negative event exposed in the media to the public
could damage corporate reputation, which would genérate negative abnormal
returns (Rose and Thomsen, 2004) because investors disapprove of firm
managerial decisions (Lee, 2001).
Specifically, this information about third-party complaints appears to be
economically relevant to the stock market. In fací, the reléase of customer
satisfaction measures provide new or incremental information to the stock
market in financial services industries (Anderson et al., 1997) since
improvements in áreas such as quality, customer or employee satisfaction, and
innovation represent investments in firm-specific assets (Ittner and Larker,
1998) or market-based assets (Srivastava et al., 1998). Additionally, Anderson
et al. (1994) and Ittner and Larcker (1998) show that stock market
measurements (stock prices) can be of great interest as indicators of
performance derived from annual indexes of expectations, quality and customer
satisfaction (American Customer Satisfaction Index,- ACSI; Swedish Customer
Satisfaction Barometer, SCSB) as can be derived from the third-party
complaints report considered in our study.
In virtue of all the above, we can assume that the firm's appearance in the
third-party complaints report (negative signal) has a negative impact on future
company profits. Specifically, we posit that the firm's presence in the
complaints report will lead to lower firm's expected future cash flows. In other
words, the stock market will penalize firms which appear in the complaint
report made by a third-party agency giving them lower potential valué.
Therefore, we propose the following hypothesis:
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
136 •
Chapter6
Hl. The publication of the Annual Complaints Service Report by an external
agency is associated with a reduction in the performance ofthefirm involved.
6.2.2 Moderating effect of quality corporate image on the
relationship between the firm's appearance in the Annual Report on
Complaints and the firm's performance
One of the current priorities of quality research is to examine the relationship
between the quality of goods and services and company performance (Greising,
1994; Rust et al., 1995), although the net effect is difficult to predict (Bharadwaj
and Menon, 1993). Quality and company performance have been found to be
negatively related. Superior quality could require the use of non-standardized
procedures, as well as an increase in the costs of investments in resources and
training (Juran and Gryna, 1980; Shycon, 1992) or an increase in sales forcé
spending to support the high levéis of customer service which come along with
better quality (Phillips et al., 1983). High quality could be more expensive to
produce and unless the cost increases can be passed on to customers, profít
margins would be squeezed. However, there is also evidence showing that
quality and company performance could be positively related. This positive
relationship could origínate from the following factors: i) a strategy of superior
quality can reduce customer price sensitivity, making price increases possible
without harming sales (Buzzell and Gale, 1987; Buzzell et al., 1975); ii)
superior quality can also protect a company from competitive forces (e.g.,
buyers bargaining power) which reduce price-cost margins and service costs
(e.g., cost of recovery and warranties) (Garvin, 1988; Thompson et al., 1985);
and iii) Crosby (1987) holds that the costs associated with quality (investments
in human resources and materials) are always lower than those implied by lack
of quality (costs derived from complaint management and reworking producís);
thus showing that better quality increases receipts and reduces costs (Spitzer,
1993).
A possible explanation for the disagreement concerning the nature of the
relationship between quality and performance could be found in the existence of
a moderating effect of satisfaction in the above relationship. In fact, Storbacka
et al. (1994) propose that a dissatisfied customer can be faced with a
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and banking market valué » 137
paradoxical situation in which service is judged to be of high quality. In these
situations, Taylor and Baker (1994) and Taylor (1997) explain customer loyalty
through the service quality-satisfaction interaction effect, in such a way that the
accumulation of positive service experiences tends to créate a self-sustaining
cycle of satisfaction and quality which has a positive effect on loyalty.
Alternatively, we propose a negative moderating effect of quality on the
relationship between the fírm's appearance in a report of complaints to a thirdparty agency and its performance. However, instead of considering the concept
of quality, we centered on the notion of quality corporate image. The reason is
that, unlike search goods, in the case of experience producís such as services
(e.g., banks), consumers cannot infer through simple inspection whether a
product is of high or low quality (Bharadwaj and Menon, 1993). In this context,
Klein and Leffler (1981) point out that a fírm's reputation/image40 is
determinant to sell high-quality experience products. This implies that
reputation/image is a proxy for quality and, therefore, is another key buying
criterion for customers. In fact, Shostack (1977) suggests that since services
contain a great amount of intangibility, it is often difficult to depict them in
clear and meaningful ways to customers. Thus, Bharadwaj and Menon (1993)
and Berry et al. (1988) propose fírm's symbols to make tangible its intangible
aspects. Following this proposal, our study employs obtaining a prestigious
quality award as a symbol that makes reputation/image "tangible", in line with
Soteriou and Zenios (2000).
In general, the existing literature suggests a positive moderating role of
quality corporate image in the above relationship, which can be explained in the
following way.
Firstly, the attribution theory (Kelley, 1967) considers the attributions that
consumers might make about the source of the complaints. Thus, as long as the
third-party complaints are not consistent with other experiential data the
consumer might have (as to the quality of the brand), he/she is required to make
some kind of attribution regarding the source of this difference. Therefore, the
customer will think about whether the brand is actually of lesser quality than
40
Keller (1993) defines corporate image as the perceptions of an organization reflected in the
associations held in consumer memory.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
138 » Chapter 6
he/she thought or the complainer is reporting on experiences that are
idiosyncratic and not relevant to him or her. In this sense, external quality cues
(e.g., quality awards) can act to alter customers' attributions and thus modérate
positively the impact of the complaints on purchase intentions. That is, the
customer-fírm relationship may continué even when the customer makes a
third-party complaint because he/she is aware of the quality award. Secondly,
Andreassen (1999) proposes that the loyalty of dissatisfied customers is
moderated by corporate image in such a way that one or two unsatisfactory
experiences will not cause the customer to exit a company when it has a good
corporate image. Ruyter and Bloemer (1999) also fínd that "in those occasions
in which satisfaction reaches a sub-optimum, valué attainment and positive
mood may function as a buffer for diminishing loyalty (p. 331)." Consequently,
for dissatisfied complaining customers, corporate image is positively correlated
with customer loyalty.
Alternatively, our study proposes that quality corporate image (i.e., obtaining
a quality award) moderates negatively the relationship between the firm's
appearance in the Annual Report of Complaints to a third-party agency and its
performance. This assumption is supported as follows. The market signaling
perspective (from Economics literature) highlights mechanisms by which
information about fírm identity can be passed on the investors (Asquith and
Mullins, 1986). In particular, the signaling perspective sheds light on the
strength of signáis (e.g., third-party complaints) from firms that also announce
other strategies (e.g., obtaining quality awards) that are inconsistent with the
previous ones. This signáis' inconsistency should be reflected in a stock prices
change (Lee, 2001).
Additionally, signáis need to be costly to be credible (Milgrom and Roberts,
1986). Thus, as we will explain later, the procedure of obtaining quality
certifications is costly. Moreover, the strength of the signal can be
amplifíed/reduced by other costly firm actions. For example, corporate actions
in search of customer satisfaction are consistent with a service quality strategy
and thus, give additional credibility to the signal of quality award. This implies
that quality award announcements that are accompanied by other consistent
strategies reinforce a commitment in service quality because the incremental
cost of signaling is larger (Williamson, 1983). Therefore, this stronger message
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and banking market valué • 139
should be followed by stronger and positive reactions in stock prices (Lee,
2001). But instead, corporate actions that genérate customer dissatisfaction and
third-party complaints are inconsistent with quality awards announcements, i.e.
they do not give credibility to the signal of quality awards. Consequently, this
contradictory message should be followed by negative reactions in stock prices.
That is, investors' reactions are more negative when firms incur in substantial
costs in signaling changes that are in opposite strategic directions. As states by
Lee (2001), the market is not easily 'fooled' by cosmetic changes in quality
corporate image. In this sense, Rose and Thomsen (2004) suggest that a sudden
negative event appearing in the media could have a negative effect on fírm's
stock price if the fírm is not able to convince investors that they can manage
firture threats to corporate reputation.
In virtue of all the above, we propose the following hypothesis:
H2. For firms appearing in the Annual Report of Complaints to a third-party
agency, quality corporate image moderates negativelyfirm performance.
6.2.3 Relationship between the number of third-party complaints
and the fírm's performance
In general, the literature suggests a negative relationship between the number
of complaints and fírm performance due to the following aspects. First, for
firms with complaint handling systems, a possible explanation would be the so
called "vicious circle of complaints" (Fornell and Westbrook, 1984). According
to it, as a bank receives more complaints it becomes less responsive so that
instead of making good use of its complaints procedure system the company
"behaves dysfunctionally" (Fornell, 1992, p. 15). Therefore, an increasing
number of complaints makes customers more prone to leave the firm and this
has a negative impact on fírm performance. Basically, behind this "vicious
circle of complaints" there is the implementation of an emphasis on costs by the
fírm. In a dynamic context, this emphasis on costs represents firings and loss of
benefits and perks, which lowers morale among employees who opérate at the
market interface. This, in turn, may lower customer service, loyalty and sales,
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
140 • Chapter 6
which leads to further cost cutting and thus to a "vicious circle" (Grónroos,
1984; Rustet al., 2002).
Second, the position achieved by a company in terms of customer
complaints to an external agency is of great strategic importance as it attests to
its vulnerability or effectiveness in defending its current customers and future
profits. In particular, a company's ranking in terms of complaints received
allows a distinction to be made in a competitive context between the losers and
the winners over the time period analyzed. This represents, respectively, a lower
or higher valué provided to customers and lower or higher potential valué for a
shareholder who would be less or more motivated to get involved in an
investment relationship with the firm (Ngobo, 1999).
Alternatively, our study proposes a positive relationship between the number
of complaints and firm performance. This proposal is based on the following
aspects. First, the stream of research that analyzes the defensive marketing
strategies based on customer complaint management (Fornell, 1992; Fornell and
Wernerfelt, 1987, 1988) proposes that firms should encourage dissatisfied
customers to complain because these complaints give the firm a chance to
recover otherwise lost customers. Therefore, the firm should invest to facilítate
complaints and should compénsate generously. This type of complaint
management can be an effective tool for customer retention, because it can
increase the consumer's expected utility from the purchase. Thus, the
establishment of complaint management and the increases in complaint volume
may well go hand in hand and this has a positive impact on firm performance
(Fornell and Wernerfelt, 1988)41. Additionally, the economic perspective
considers that the pursuit of costs decreases and thereby increases in
productivity and performance would implicitly cause customer dissatisfaction
and a greater amount of complaints. In economics, the relationship between
productivity and customer satisfaction is generally viewed as negative.
Customer satisfaction (utility) is modeled as a function of product attributes.
41
The problem is that such increases in complaint volume are not often viewed as favorable by
the personnel responsible for their handling. In fact, Ross and Gardner (1985) found that firms
often draw incorrect inferences from changes in complaint volume. Instead of considering the
opportunity cost of not receiving a complaint, an increase in complaints is perceived as negative
and sometimes is reflected in management performance evaluation and lower executive bonus
compensation. Moreover, Fornell and Wernerfelt (1988) indícate that such as "myopic policy"
can be very costly.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and banking market valué
•
141
Reducing the level of utility (e.g., worse raw material, features elimination, etc.)
requires reducing the level of product attributes and, therefore, costs (Griliches,
1971; Lancaster, 1979). Thus, it is common to expect increasing returns to such
efforts (Anderson et al., 1997). All this would imply that the fírm has an
emphasis on costs, as mentioned earlier (Rust et al., 2002).
In the specifíc case of Spanish banking, Freixas (1996) and Tortosa-Ausina
(2002) fínd a steady increase in their efficiency in the nineties, which can also
be viewed as an indicator of productivity earnings. Additionally, in retail
banking, the customer does not necessarily need customized services (Storbacka
et al., 1994). Therefore, following the economic approach, we assume a positive
relationship between the number of complaints (meaning customer
dissatisfaction that reduces costs and increases productivity) and performance:
H3. A greater number of complaints reported in the Annual Complaints Service
Report is associatedpositively with the performance ofthefirm involved.
6.2.4 Moderating effect of market concentration on the relationship
between the number of third-party complaints and the fírm's
performance
Traditionally, literature proposes a positive relationship between market
concentration and fírm performance. The assumption is based on the collusion
hypothesis (Bain, 1951) as suggested by economics of industrial organization,
which posits that the divergences in performance origínate in collusive
behaviors. These behaviors arise in sectors in which fírm concentration is high
and the existing entry barriers allow more cooperative long-term behaviors, as
well as extraordinary returns.
However, the negative impact of concentration on performance has been
found repeatedly in the existing literature (e.g., Gual and Vives, 1991; Kwoka
and Ravenscraft, 1986; Ravenscraft, 1983). The explanation is based on
different oligopoly theories (e.g., dominant fírm, price leadership, limit pricing,
strategic groups), which suggest important differences between leading and
non-leading firms depending on different factors. These factors affect firms'
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
142 •
Chapteró
environment, generating a sufficient rivalry that can drive margins down to
competitive levéis (Kwoka and Ravenscraft, 1986). Among the factors, Kwoka
and Ravenscraft (1986) emphasize the scale economies. That is, when
economies are great, the price preferences of large leading fírms may compress
non-leaders' price-cost margins since the latter suffer cost disadvantages.
Therefore, the performance of a market with two fírms competing in a Bertrand
way (in prices) could be equal to that in perfect competition despite the high
level of concentration (Gual and Vives, 1991).
Moreover, this study proposes a moderating effect of market concentration
on the relationship between the number of complaints received in the Annual
Report on Complaints and fírm performance.
Previous literature suggests a positive moderating effect, which can be
explained in the following way. Storbacka et al. (1994) report that, in the
customer-firm relationship context, an unsuccessful critical episode (what
Bitner et al., 1990 cali a negative 'critical incident') represents a cost for the
fírm that affects negatively its performance, as it may end the relationship
abruptly even when it has been preceded by years of satisfactory episodes.
However, a stream of research in consumer behavior defended by authors such
as De Ruyter, Moorman, and Lemmink (2001), Fornell (1992), Rowley and
Dawes (1999), Sharma and Patterson (2000), and Storbacka et al. (1994),
propose that the customer-firm relationship may continué even when the
customer does not desire to remain loyal, because of lack of perceived
alternatives. In other words, the customer may remain loyal for a longer time
when he/she perceives that there are few alternatives in the market (especially
when they have a comparable offer as in banking), which is quite common in
concentrated monopolistic or oligopolistic markets (Storbacka et al., 1994).
Alternatively, our study proposes that market concentration moderates
negatively the relationship between the number of complaints received in the
Annual Report on Complaints and fírm performance. Our arguments for this
approach are the following. From an economic perspective, as indicated in the
previous section, a firm's cost emphasis focuses on efficiency improvement and
cost reduction to obtain higher benefits. Customer satisfaction improvements
are sought only indirectly with such strategies by means of lower prices
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and banking market valué » 143
(Anderson et al., 1997; Rust et al., 2002). Therefore, the pursuit of customer
dissatisfaction (a greater amount of complaints) reduces costs and thereby
increases productivity and performance. However, Rust et al. (2002) propose
that in highly competitive industries, prices will be competed down to levéis
that make subsequent cost reductions less attractive. That is, an emphasis in
costs (e.g., customer dissatisfaction and implicitly a higher volume of
complaints) in a rivalry context leads to subsequent cost reductions, which are
less attractive because they have a negative impact on performance. Thus,
taking into account that market concentration is an indicator of the existing
competitive rivalry in a market (Fuentelsaz et al., 2002) and also that there is
evidence of a negative relationship between market concentration and
performance and high rivalry for the Spanish banking (e.g., Gual and Vives,
1991), we propose the following hypothesis:
H4. The flrm's target market concentration moderates negatively the
relationship between the number of third-party complaints and firm
performance.
6.3 Methodology
In this section, we describe the sample, the analysis procedures, the data
collection and measures, and the specific third-party complaint procedure
analyzed.
6.3.1 Sample
The design of the study has been developed for the particular case of the
Spanish banking sector as a source of data on third-party complaints, which is
an appropriate choice for our objectives due to the following aspects. Firstly, the
banking sector is subject to constant supervisión by fínancial authorities, which
have an obligation to guarantee the honesty of banks (Cañáis, 1992) given the
strategic valué of the sector to a country's economy. Essentially, the behavior of
banks transcends the banking sector in a more signifícant manner than is the
case in other sectors. This is because banking institutions live on the confídence
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
144 » Chapter 6
and prudence they inspire in economic agents. Lack of trust in a particular bank
could spread to the banking system as a whole and result in a massive
withdrawal of funds which banks would not be able to handle, causing a
consequent paralysis of the economy. Also, the actions of both public and
prívate consumer organizations constitute an element which has strengthened
control over banking practices in Spain in recent years. Consequently, public
organizations have been developing a system of supervisión for the banking
sector, which includes the Complaints Service created by the Bank of Spain (see
Consumer Complaint Procedure section).
Secondly, banking producís are highly diffused in the consumer market
(almost all households have some type of banking product), which means that
the probability of unsatisfactory experiences resulting in complaints is quite
high. In fact, the banking sector is one of those which receive the greatest
number of complaints to Spanish consumer organizations (Ortega, 2003).
Finally, there is the availability of data on complaints from the Annual
Complaints Service Report of the Bank of Spain, which can be consulted by any
economic agent (consumers, companies and government organizations).
Electronic versions of the most recent reports can be found on the Bank of
Spain Web site (www.bce.es).
In the sample of banks selected, we include all the banks quoted on the
Madrid stock market which have received complaints through the Complaints
Service of the Bank of Spain. These eleven banks represent 75.3% of the
sector's total assets in 2002, according to information from the Banking Control
Council.
6.3.2 Analysis procedures
In a first stage, the 'event study' technique is used to estimate company
return variations due to publication of annual reports on complaints by an
external agency (Bank of Spain) (Hypothesis Hi). This method is based, firstly,
on the portfolio theory of financial economics on the premise that stock markets
are efficient and, secondly, on the idea that company share prices better reflect
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and banking market valué • 145
strategy (e.g. of quality and customer satisfaction) than financial records
(Brown and Warner, 1985). In an efficient stock market, share prices reflect all
the available information on a company; in fact, any information received by
the market (e.g. the publication of the Annual Complaints Service Report) will
be instantly incorporated by investors into its share price. Likewise, any change
to a company's share price will reflect, without bias, alterations to its future
cash flows. Because of this and faced with the introduction of new information
on customer complaints from the annual report of an external agency, the
examination of share price behavior allows us to explicitly analyze underlying
changes to unbiased market predictions on the future returns of a company due
to its inclusión in the report. This allows us to isolate returns derived from thirdparty complaints by eliminating the impact of other events, which avoids this
problem aroused when using accounting data. Furthermore, stock market
measures overeóme other limitations resulting from the use of accounting data
as they allow us to consider the risk supported by the firm and the capitalized
valué (future expectation) of the benefits (Lang and Stulz, 1994). They also
make possible to impute the equilibrium profits and to minimize the distortions
resulting from tax laws and accounting standards (Wernerfelt and Montgomery,
1988).
In particular, an event study measures the impact of unanticipated events on
share prices, being based on the estimation of a market model for each company
event and on the subsequent calculation of abnormal returns. The returns on the
share price of a company i on day / (Rtl), are expressed as:
&„=<*,+P,Rm,+eu
(1)
where ai are returns on company shares which are independent of the market;
Rm, is the rate of returns of the market portfolio; J3f is the returns sensitivity of
share i to variations to market returns; and s¡, is the random disturbance.
The estimation of equation (1) allows us to calcúlate daily abnormal returns
(AR) for news on company i:
ARu^Ril-{ál+PiRml)
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
(2)
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
146 • Chapter 6
where ái and f3i represent the OLS parameter estimates obtained in the
regressions (1) for the period Tpreceding the event. Abnormal returns are those
obtained by a company once investors have adjusted for 'normal' returns and
returns on shares are adjusted by subtracting expected returns from actual
returns with any signifícant difference being considered abnormal.
To analyze the effect of the publication of the Annual Complaints Service
Reports by an external agency (Bank of Spain) on the share prices of the entities
concerned (Hypothesis Hi) we test:
1) The signifícance of the abnormal returns, using the parametric test
proposed by Jaffe (1974):
N
1
|~Ñ
Ñ
'
(3)
where N is the number of news items, AR,0 are the abnormal returns on day 0 of
the event, and <J2e and cree represent, respectively, the variance and
covariance of shares /' and y obtained in equation (1) in the period of estimation.
The choice of this test is justified by the potential presence of contemporary
correlation problems in the chosen sample, which can come from the existence
of overlapping periods in any of the news items on different shares and from the
fact that the companies analyzed are from the same industry (Bernard, 1987;
Collins and Dent, 1984).
Additionally, to avoid possible problems derived from lack of normality in
the returns, we also use the non-parametric test of Corrado (1989):
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and banking market valué • 147
i
N
-S
Klo-~(T
+ l)
u =•
1
ir£?
N
Ku--(?
+ !)
(4)
where £;o is the range occupied by abnormal returns ARit in the temporal series
of abnormal returns estimated for share / and T is the total number of days
analyzed. This test uses ordinal information on returns on the event day and is
not affected by variance in distribution.
2) The significance of the average standardized cumulative abnormal returns
over k days (event window for N news items):
ACAR,= — *
I.CAR*
i=l
(T-2),
\T-A\
YI
(5)
where T is the period before the estimation plus the event window, and being:
CAR^^-j^SAR,
S¡=S¡
11 + 1+
(^-^)
1=1
S¡ the standard deviation of the residuals of the regression estimated before
publication; T the period before the estimation plus the event window; and Rm is
the mean return on the market portfolio in the period of estimation.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
148 • Chapter 6
To test whether the average standardized cumulative abnormal returns is
significantly distinct from zero we use the test Z = ACAR, N'/', which follows a
standardized normal distribution for large N.
In a second stage, we analyze the determinant factors of return variation on
the share of the banks involved in terms of the moderating effect of quality
corporate image in the relationship between the spreading of the complaints
report and firm performance (Hypothesis H2), the direct effect of the number of
complaints per branch on firm performance (Hypothesis H3), and the
moderating effect of market concentration in the relationship between the
number of complaints and firm performance (Hypothesis H4). To test
Hypotheses H2 and H3, we employ regression analysis. Specifically, to test the
moderating role of quality corporate image (Hypothesis H2), we follow the
proposal of Andreassen (1999). That is, for a sample of firms appearing in the
complaints report of a third-party agency, we examine the direct effect of
quality corporate image on performance. To test Hypothesis H4, we follow the
moderated regression analysis (MRA) methodology proposed by Sharma et al.
(1981). In our study, this method requires estimating three models in terms of
the average cumulative abnormal returns (ACAR), with the following predictor
variables: i) Model 1, independent variable (number of complaints/branches,
NCB); ii) Model 2, independent variable (NCB) and moderator variable (market
concentration, MC); and, iii) Model 3, independent variable (NCB), moderator
variable (MC), and interaction term (NCBxMC). Following Sharma et al.
(1981), if Models 2 and 3 are not significantly different, the predicted moderator
variable is not a moderator but simply an independent predictor variable. If
Models 1 and 2 are not significantly different but are different from Model 3,
then the predicted moderator variable is a puré moderator. Finally, if Models 1,
2, and 3 are different from each other, then the predicted moderator variable is
classified as a quasi moderator. We also examine the amount of incremental
explained variance due to additional predictors. In order to minimize the effects
of multicollinearity, the continuous independent variables, NCB and MC, are
mean-centred (by sustracting the corresponding variable mean from each valué)
as suggested by Aiken and West (1991). Additionally, to test that the results
obtained are not biased by small sample size, we also perform a bootstrap
analysis of the above regressions. Thus, we calcúlate the finite-sample critical
valúes using a general non-parametric bootstrap procedure detailed in Horowitz
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and banking market valué •
149
(1997). Specifically, the bootstrapping procedure we use amounts to sampling
10,000 (yJC) pairs from the original data with and without replacement. We
estímate b*, s*, and t* for each bootstrap sample, where b* and s* are the
ordinary least squares coefficient and standard error estimates, and t* or equal,
(¿*-P)/s*, where (3 is the original ordinary least squares estímate. The empirical
distribution of the t-statistic is used to determine the bootstrap p-values of the
traditional t-statistic obtained in the original regression.
6.3.3 Data collection and measurement
The application of the proposed methodology is based on the following data
collection process (MeWilliams and Siegel, 1997). In the first stage, we
consider all banks quoted on the Madrid stock market, which have complaints
registered in the Complaints Service of the Bank of Spain between 1992 and
2002. These are: the Banco Bilbao-Vizcaya, Banco Bilbao-Vizcaya-Argentaria
(a result of a merger in 2000 of the banks of Bilbao-Vizcaya and Argentaría),
Banco de Santander, Banco Central Hispano, Banco Santander-Central-Hispano
(formed in 1999 from the merger of the Banco de Santander and the Banco
Central Hispano), Banco Popular, Bankinter, Banco Español de Crédito
(Banesto), Banco Zaragozano, Banco Atlántico and Banco Pastor.
In the second stage, the first news items about the 11 Annual Reports of the
Complaints Service of the Bank of Spain between 1992 and 2002 are identified.
To this end, the necessary search was made in the BARATZ datábase, which
provides information published in 28 different newspapers of national or
regional coverage. The event date is defined as the first day on which the news
is disclosed in any of the publications included in the datábase. In this sense, we
detect 121 impaets derived from the publication of the annual reports on
complaints related to the eleven banks analyzed.
In the third stage, in order to identify any abnormal behavior in the returns
on these companies' shares, we select the k length of the 'event window'. In
other words, we consider the five days before and after the publication date (-5,
+5) due to the fact that, although the majority of information on complaints is
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
150 « Chapteré
usually quickly incorporated into share prices, it can occasionally be leaked out
before its formal publication or its effect on the share price can be delayed.
In the fourth stage, we reject the impact of publication of annual reports on
complaints on banks whose 'event window' coincides with the publication of
other events fínancially relevant to the banks themselves. In our case, relevant
events include public share offers, dividend declarations, profíts
announcements, and large-scale share acquisitions. This allows for the exclusive
measurement of the effect of third-party complaints and eliminates the
possibility of including confusing effects. This rigorous procedure reduces the
sample from 121 to 51 news releases about annual reports on complaints
relating to the 11 banks, allowing us to isolate its effect on the market valué of
the firm.
In the fifth stage, we collect data on the daily share returns (Z?„) of the eleven
banks. We, therefore, use stock market data, as opposed to accounting data, to
measure company performance. The historical data sample selected is of the
daily returns of the 11 banks quoted on the Madrid Stock Exchange in the
period from the 2nd of January 1992 to the 31st of December 2002. This
temporal period is defmed by the availability of daily market information. As a
subrógate variable of the true return on the market portfolio (Rml), we use the
IBEX-35 index, which is representative of the Spanish stock market. This
information is obtained from the Stock Exchange Information System.
Finally, in order to analyze the determinants of excess returns derived from
complaints, we collect information on the following variables for each news
item:
1. Quality corporate image. This is measured with a dummy variable, where
1 means that the bank has a certifícate of quality (a quality award), and 0 that it
does not; information taken from the annual reports of the banks. In general,
receiving a quality award from a prestigious external organization is an
indicator of good quality company practice. This indicator is used in various
studies (e.g., Hendricks and Singhal, 1997; Soteriou and Zenios, 2000;
Terziovski et al., 1997) and is different to other, more subjective, measurements
of service quality (Soteriou and Zenios, 2000). Specifícally, we use the
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and banking markeí valué •
151
certification of quality. This kind of certification of quality bears witness to the
fact that the entity in question reaches the mínimum standards set. In Spain,
certificates of quality are awarded by independent companies, which have to be
recognized by the National Company for Accreditation (ENAC), a dependent
entity of the Ministry of Industry and Energy, before they can carry out
Normalizaron and Certification activities.
2. Number of complaints/branches. The volume of complaints about each
entity, corrected by their size, denotes the position reached by a bank as regards
customer complaints to an external agency. Information relative to the volume
of complaints received by each entity is found in the Annual Complaints
Service Reports of the Bank of Spain, while the number of branches per bank
comes from the Commercial Performance Information Bureau of the Bank of
Spain.
3. Target market concentration. It is measured with a Herfindahl Index and
computed through the following steps (Fuentelsaz et al., 2002): first, we
calcúlate a province Herfindahl index by summing the squares of the market
shares of all firms operating in the market (province). As a proxy of market
share we employ the number of branches. The use of the number of branches
instead of bank deposits is due to the availability of individual information for
each bank in each province. Additionally, service competence may be related to
branch establishment, as better service is implied by being cióse to customers
and, thus, offering a better service (Gual and Vives, 1991). Second, the core
market Herfindahl index is calculated by multiplying each single province
Herfindahl index in which the entity was operating in the year for the news item
by the relative importance of the province for the entity under observation (the
number of branches was again used to measure the importance of the province
in the whole activity of the entity). This measure of target market concentration
is firm specific, in the sense that it takes different valúes for any two different
firms.
As a control variable, we use the natural logarithm of company size,
measured by assets in the event year, which is found in the Commercial
Performance Information Bureau of the Bank of Spain. It is used to control
economies and diseconomies of scale at a corporate level. In fact, this measure
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
152 • Chapteró
is an established way of accounting for differences in firm size when examining
organizational outcomes, and has been used in other bank-related studies
(Hopkins and Hopkins, 1997, p. 642).
6.3.4 Consumer complaint procedure
The Complaints Service of the Bank of Spain was created in 1987 in order to
receive and process complaints by banking customers about acts which possibly
break rules of discipline or good practice. For a complaint to be accepted by the
Service, it is essential to show that a written complaint was fírst made to the
customer complaints manager or equivalent office of the bank, when available
(75% of Spanish banks have a Customer Rights Offícer or similar). In the
period from 1992-2002 the Complaints Service received an annual average of
20 complaints per bank.
Once a complaint is accepted, the entity concerned has the opportunity to put
its case forward. As a rule, the process concludes with a report which states
whether the entity has complied, or not, with good banking practices. This
report is sent to the customer and the bank concerned. Although the reports
made by the Service have no official power, they are largely complied with by
banks.
The Complaints Service publishes an annual report which includes a
statistical summary of proceedings taken in the previous year: a summary of
complaints (presented, accumulated and in process), the location of the
presentation of the complaint, the entities receiving most complaints and the
type of complaint, or the material object of complaint (active operations,
passive operations, credit cards etc.). These reports are also available on the
Internet (www.bce.es).
In summary, the essential nature of this external agency coincides with the
following dimensions, which Singh (1988) used to describe third-party
responses: (a) there is a contact with the companies which have become the
object of the complaints so that they are aware of the customer's problem; and
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and banking market valué • 153
(b) there is a provisión of advice to other customers regarding the problems
found in companies' services.
6.4 Results and discussion
In this section, we present and discuss the main results obtained.
6.4.1 Estimation of return variation resulting from the firm's
appearance in the Annual Report on Complaints
Taking the event study as a starting point, this section estimates the
parameters of the market model (see equation 1) for a period T of 145 days
(from t-150 to t-5 days relative to the event date t=0), which is a time period
often used in this type of study (McWilliams and Siegel 1997). The parameters
estimated allow us to calcúlate abnormal returns for the event date (see equation
2) derived from the publication of the Complaints Service Report by the Bank
of Spain.
Table 6.1 presents the estimations of the average abnormal returns (average
AR) and the average standardized abnormal returns (average SAR) on each of
the days comprising the event windows for the 51 event impacts on the banks.
The results obtained show that, on average, the publication of annual reports on
complaints is associated with negative abnormal returns: average abnormal
returns are of-0.898% on the event date, -0.877% on the day after the event and
-0.857% on t=3. They are all significant using the parametric test of Jaffe
(1974) and the non parametric test of Corrado (1989). On day t=0 we find the
greatest abnormal losses or negative excess returns for the period ±5 days
around the event. This indicates that, on average, entities mentioned in the
annual reports of the Complaints Service of the Bank of Spain, suffer a loss of0.898% of abnormal returns on day t=0.
Alternatively, a standard event study practice is to examine cumulative
abnormal returns for various windows around the event date. The analysis of the
abnormal returns which surround the day of publication allows us to take
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
154 « Chapter 6
account for any uncertainty around the real event date and fínd the accumulative
effect of an event. Also, the effect can be felt over various days around the
event, given the gradual availability of information and, therefore, of
interpretation of the event's impact on future company profits. In this respect,
we show (see Table 6.1) that the average standardized cumulative abnormal
returns (ACAR) in the pre-event windows {(-5, -1), (-4, -3), (-4, -2), (-3, -1), (.
2, 0) and (-1, 0)} are not significant, which seems to show that information is
not divulged before publication (e.g. press conferences).
However, the results change when we look at the post-event windows. Three
temporal periods: {(0, +1), (0, +2), and (+1, +3)} show significant ACAR,
which indicates that some investors react on the same day and others later. To
be precise, the global 3-day window (0,+2) shows the greatest loss of ACAR
with a valué of -1.86%. This reaction delay seems plausible as news of the
publication of the annual reports on complaints could motivate investors to reexamine company product positioning and marketing strategies. It seems,
therefore, that they make a negative valuation of strategies and react
accordingly (Mathur and Mathur, 1995).
Apart from the statistical significance of the results, their economic
importance is illustrated as follows (Nayyar, 1995): i) an ACAR of-1.86% over
three days is the equivalent of annual returns of-223.2 %; and, ii) an ACAR of1.86% for an average sample market valué (product of the number of shares by
the share price) of Euro 89,892.83 million on t=0, implies a loss in valué of
Euro 1672.01 million in three days.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and banking market valué •
155
Table 6.1
Variation of returns due to publication of Annual Complaints Service Reports
Event
day
" -5
-4
-3
-2
-1
0
1
2
3
4
5
Average
AR (%)
Average
SAR (%)
f
/"
Event
window
-0.656
0.068
-0.076
-0.106
-0.075
-0.898
-0.877
-0.085
-0.857
0.242
-0.309
-0.999
-0.086
-0.364
-0.261
0.031
-1.486
-1.540
-0.131
-1.238
0.308
-0.097
-1.768
0.184
-0.203
-0.285
-0.202
-2.420***
-2.365***
-0.229
-2.311***
0.653
-0.833
-1.521
0.400
-0.280
-0.240
-1.400
-2.321***
-2.641***
-0.840
-1.960**
-1.120
0.840
(-5,-1)
(-4,-3)
(-4,-2)
(-3,-1)
(-2,0)
(-1,0)
(0.+D
(0.+2)
(+1.+3)
(+2,+4)
(+3,+5)
Average Stand.
Cumulative A R
(%)
-0.844
-0.007
-0.113
-0.256
-1.079
-0.973
-1.775
-1.860
-1.820
-0.700
-0.924
Zc
-1.017
-0.014
-0.176
-0.399
-1.678
-1.854
-3.383***
-2.895***
-2.832***
-1.089
-1.439
Note: AR = Abnormal Return; SAR = Standardized Abnormal Returns
a. Statistic / based on the parametric contrast of Jaffe (1974)
b. Statistic / based on the nonparametric contrast of Corrado (1989)
c. Z=ACARtN1/2, statistical test which follows a standardized normal distribution for
large N
***p<0M.
In summary, the evidence of significant negative ARs on the event date
(t=0), on t=l and on t=3, along with the negative ACARs for windows (0, +1),
(0, +2), and (+1, +3), allow us to accept Hypothesis Hi. Therefore, the
publication of the Annual Reports of the Complaints Service of the Bank of
Spain has a negative effect on the performance of the banks involved. In line
with Fornell (1992), this result implies that the publication of the Annual Report
on Complaints is an index which provides share holders and investors with
useful information. Thus, investors give lower potential valué to entities which
appear in the report as they assume that this represents a negative event that
damages corporate reputation (Rose and Thomsen, 2004).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
156 » Chapter6
6.4.2 Determinants of return variation
Once we have proved that the publication of the annual reports of the
Complaints Service of the Bank of Spain is associated with negative changes to
the returns on the shares of the banks involved, we examine the determinants of
these returns. Table 6.2 shows the descriptive statistics of the variables for the
sample of 51 impacts of the publication of the annual reports.
Table 6.2
Means, standard deviations, and correlations
X|
Quality corporate image (X\)
Number of complaints per branches
1.000
» . ,„
(X2)
X2
X3
X4
1.000
0.033
1.000
Target market concentration (X3)
Ln(Size)(X4)
0.237
0.132
-0.022
0.876***
Mean
SD
***/K0.01.
0.549
0.502
0.077
0.032
0.818
0.951
1.000
16.932
1.190
6.4.2.1 Moderating effect ofquality corporate image and direct effect of
the number of third-party complaints
For testing the moderating effect of quality corporate image, as well as the
direct effect of the number of complaints, we employ regression analysis (see
equation 1 in Table 6.3). By using normality tests, such as that of Jarque-Bera,
the assumption that residuals are normally distributed cannot be rejected. The
test of Durbin-Watson rejects the existence of autocorrelation of order 1
residuals and the test of Breusch-Godfrey rejects order 2 autocorrelation. With
respect to homoscedasticity, White and Breusch-Pagan tests accept the nuil
hypothesis of equality of residual variance. Consequently, the proposed
relationships are examined with OLS regressions. The potential for collinearity
among variables is assessed by calculating the variance inflation factor (VIF)
for each of the regression coefficients. The valúes obtained are very low,
indicating that it is unlikely that any predictor is a linear combination of other
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and banking market valué •
157
independent variables (the valúes range from a low of 1.02 to a high of 1.04,
well below the cutoff figure of 10 recommended by Neter et al., 1985).
The significance tests of the individual parameters show that quality
corporate image influences the bank's abnormal returns, as its coefficient is
significant at a level below 5% (see equation 1 in Table 6.3). The negative sign
of quality corporate image indicates that the possession of a quality award (a
certifícate of quality) in firms which have appeared in the complaints report
published by the Bank of Spain (third-party), is associated with negative
abnormal returns. This supports Hypothesis H2 which proposes a negative
moderating effect of quality corporate image in the relationship between a
firm's appearance in the annual report of complaints and its performance, in line
with the economic approach of market signaling (Asquith and Mullins, 1986;
Lee, 2001). That is, obtaining a quality award is inconsistent with the
appearance of a firm in the complaints report of a third-party agency;
contradictory messages that are penalized by investors (Lee, 2001).
Additionally, results show that the number of complaints per branch has an
influence on the banks' abnormal returns as its coefficient is significant at a
level below 5% (see equation 1 in Table 6.3). The positive sign of the variable
number of complaints per branch shows that a greater valué for this ratio is
associated with greater abnormal returns, which leads us to accept Hypothesis
H3 according to the economic approach (Anderson et al., 1997; Rust et al.,
2002). That is, firms that follow the cost emphasis are not interesting in
customer satisfaction, and therefore, a greater volume of complaints reflects that
the firm is trying to reduce costs to reach higher levéis of productivity and
performance.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
158 « Chapteró
Table 6.3
Estimated regression models (standard errors in brackets)
Hypotheses
Hypothesis 4
2 and 3
Independent Variables
Intercepta
Control variable: Ln(Size)*
Main effects
Quality corporate image (QCI)
Number of complaints per
branch (NCB)
Target market concentration
(MC)
Interaction effect
NCB x MC
R2
Adjusted R2
F statistic
AR2
Equation 3
Equation 4
-0.057
(0.034)
0.002
(0.002)
Equation
2
-0.053
(0.035)
0.002
(0.002)
-0.158**
(0.071)
0.008
(0.004)
-0.148**
(0.071)
0.008
(0.004)
-0.008**
(0.005)
0.178**
(0.075)
0.158**
(0.076)
0.172**
(0.075)
-0.009**
(0.005)
0.146**
(0.077)
-0.006
(0.006)
Equation 1
0.153
0.099
2.833**
0.099
0.062
2.647
-0.120
(0.095)
0.151
0.179
0.108
0.097
2.780**
2.512
0.052
0.028
F(l,46)=2.817 F(l,45)=1.535
F(2,45)=4.385**
NOTE: Dependent Variable = Average Standardized Cumulative Abnormal Returns,
ACAR (0,+2). One-tailed signifícance tests, unless the contrary is indicated.
a. Two-tailed test.
**p<0.05; ***p<0.0\.
6.4.2.2 Moderating effect of target market concentration
For testing the moderating role of target market concentration, we employ
moderated regression analysis (see equations 2-4 in Table 6.3). Again, we test
the normality of the residuals, the absence of autocorrelation, and the presence
of homocedasticity, therefore, allowing us to apply OLS for estimating the
model. The variance inflation factors (VIF) range from a low of 1.01 to a high
of 4.35, well below the cutoff figure of 10 recommended by Neter et al. (1985).
The inclusión of the interaction term "number of complaints per branch x target
market concentration" in Equation 4 does not increase significantly the
proportion of explained variance (F(l,45)=1.535, ns), ñor the global
signifícance with respect to Equation 3 (see Table 6.3). Additionally, the
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Third-party complaints and banking market valué •
159
interaction term is not signifícant. This implies that the conditions specified by
Sharma et al. (1981) for identifying moderators are not met and thus we reject
Hypothesis H4, which proposes a negative moderating effect of target market
concentration in the relationship between number of complaints per branch and
abnormal returns.
It should be pointed out that the inclusión of the interaction term "number of
complaints per branch x target market concentration" changes the significance
levéis of the main effect of target market concentration, although the negative
sign remains unaltered. Despite the lack of robustness of this result, the negative
sign of the target market concentration seems to reflect a tendency that a greater
market concentration is associated with negative abnormal returns. This result is
similar to that obtained by Gual and Vives (1991) in the Spanish banking over
the period 1984-1994, confirming the existence of a high rivalry in the Spanish
banking market, which leads to a decrease in returns despite the high
concentration.
Finally, we carry out a bootstrap analysis of the regressions in Table 6.3 to
verify the robustness of the results (see Appendix 6.1). We fínd that the
significance of the regression coefficients does not vary with respect to the
results obtained with the traditional contrasts. Therefore, we conclude that the
results obtained previously are robust42.
6.5 Conclusions
The suggestion that company performance might be explained through the
spreading of the complaints report by a third-party agency and the number of
complaints as well as through the moderating effects of quality corporate image
and market concentration, has led us to analyze these phenomena in the Spanish
context of the publication of annual reports by the Complaints Service of the
Bank of Spain between 1992 and 2002.
42
Additionally, following Efron and Tibshirani (1993, pp. 111-112), we replícate the results by
applying a method called "bootstrapping the residuals" obtaining identical results (see Appendix
6.1).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
160 •
Chapter 6
The methodology of analysis is based on an event study, starting with the
theory of portfolio of financial economics with the premise that stock markets
are efficient and that company share prices better reflect their quality strategy
and their customers' satisfaction than their accounting structure. To be precise,
this method measures the impact of a sample of events (publication of Annual
Complaints Service Reports by the Bank of Spain) on share prices, estimating a
market model for each news item referring to a company and calculating the
abnormal returns deriving from each event. Additionally, regression analysis
and moderated regression analysis (MRA) help us to examine the impact of
quality corporate image (measured through certificates of quality), the number
of complaints per branch, and market concentration on excess returns. The
methodology is shown to be particularly useful for analyzing the impact of
complaints on performance and the moderating effects of quality corporate
image and market concentration avoiding some inherent problems of using
accounting data.
The empirical application carried out on the sample finds signifícant
negative returns on the event date (t=0), on the day after the event (t=l) and on
t=3, as well as negative accumulated returns in different post-event windows,
which suggests that the stock market reacts negatively to the publication of the
annual reports of the Complaints Service of an external agency. Likewise, the
model proposed shows that abnormal returns depend on the number of
complaints per branch and on quality corporate image. Specifically, we find a
negative moderating role of quality corporate image in the relationship between
the spreading of the complaints report and firm performance as well as a
positive relationship between performance and the number of complaints per
branch.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Volver al Indice/Tornar a l'índex
Third-party complaints and banking market valué » 161
Appendix 6.1
P-values obtained with bootstrap estimation
Equation 1.
Bootstrap type
Initial ¿-valué
Initial p-value
Residuals
Pairs (with replacement)
Pairs (without replac.)
Equation 2.
Bootstrap type
Initial í-value
Initial p-value
Residuals
Pairs (with replacement)
Pairs (without replac.)
Equation 3.
Bootstrap type
Initial /-valué
Initial p-value
Residuals
Pairs (with replacement)
Pairs (without replac.)
Equation 4.
Bootstrap type
Initial í-value
Initial p-value
Residuals
Pairs (with replacement)
Pairs (without replac.)
Intercept
-1.6759
0.1004
0.0968
0.1598
0.0602
Ln(Size)
1.257
0.2150
0.2132
0.2906
0.17
_QC¿_
-1.7281
0.0905
0.0816
0.0854
0.0286
Intercept
-1.545
0.1289
0.1176
0.1882
0.0722
Intercept
-2.2353
0.0302
0.0268
0.0104
0.0004
Intercept
-2.0935
0.0418
0.041
0.01
0.0006
Ln(Size)
1.0102
0.3175
0.3128
0.3988
0.2562
Ln(Size)
1.9733
0.0544
0.0536
0.0228
0.0014
Ln(Size)
1.8334
0.0732
0.073
0.0222
0.0018
NCB
2.089
0.0420
0.0442
0.0366
0.0032
NCB
2.2998
0.0259
0.0206
0.0136
0.0012
NCB
1.8864
0.0656
0.0658
0.0834
0.0176
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
NCB
2.3724
0.0218
0.021
0.0172
0.001
MC
-1.0544
0.2972
0.2948
0.159
0.0714
MC
-1.6862
0.0984
0.101
0.0742
0.0106
NCBxMC
-1.2657
0.2120
0.2202
0.1926
0.0704
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Chapter 7
Conclusión: summary, implications, limitations and future
research
7.1 Synopsis
Acquiring a new customer is expensive, particularly in mature markets, such
as the airline or the banking industries where competition is fierce, product
differentiation is low, and promotional costs have substantially increased. Thus,
recent articles on loyalty-based management, offensive versus defensive
marketing, and complaint management attest the importance of nurturing one's
loyal base of customers and understanding and anticipating their post-purchase
experienees and behaviors (e.g., Fornell and Wernerfelt, 1987).
Henee, understanding the different elements that affect (dis)satisfaction after
service failure and subsequent behaviors derived from this (dis)satisfaction is a
key factor in the strategic management of a service firm. Therefore, the global
objective of this dissertation was to contribute to the theoretical and empirical
evolution of service failure research toward a better understanding of their
management's importance.
Following the framework proposed by Singh (1988) to classify responses to
dissatisfaction (prívate, voice and third-party responses), we have carried out
three empirical applications in which there is a progression (from less to more)
of the amount of effort involved in complaining (Chapters 4, 5, and 6).
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
164 •
Chapter7
In Chapter 4, the objectives were: 1) to develop and empirically test a
comprehensive conceptual framework grounded in several research fields that
identifies the antecedents and consequences of the (dis)satisfaction with the
service failure (i.e., the flight delay), 2) to examine the impact of the specific
negative emotion of anger on the previous framework, and 3) to explore the
effects of different service-failure related variables on (dis) satis faction with
service failure and on behavioral and complaining intentions, directly and
indirectly through anger with service failure and (dis)satisfaction with service
failure. Thus, we analyzed the impact of the initial negative emotion of anger
on the initial (dis)satisfaction judgment and subsequent behavioral and
complaining intentions.
The findings show the causal sequence "attribution-affect-behavioral
intention" (i.e., think-feel-act). That is, attributions customers make regarding
the causes of flight delays (think) have an impact on anger (feel), which in turn
affects behavioral intentions (act). Furthermore, anger is found to be a mediator
between controllability and (dis)satisfaction with service failure. Moreover, the
importance of punctuality has an effect on anger. Finally, anger has a negative
effect on satisfaction with service failure and repurchase intentions and a
positive effect on propensity to complain.
In Chapter 5, we go a step further with respect to Chapter 4 and analyze the
impact of a secondary negative emotion (anger with service recovery) on
secondary satisfaction judgments (satisfaction with service recovery), in the
specific context of double deviation scenarios (voice response). The objectives
of this chapter, which was focused on the banking industry, were: 1) to develop
and empirically test a comprehensive conceptual framework grounded in
several research fields that identifies the antecedents of the (dis)satisfaction
with service recovery in the specific context of double deviation scenarios (i.e.,
failed recoveries after servicefailures), 2) to examine the role ofthe secondary
emotion of anger (i.e., anger with service recovery) and the distributive
component ofjustice on the previous framework, and 3) to explore the effects of
service failure- and service recovery-related variables on (dis)satisfaction with
service recovery judgments, directly and indirectly through the secondary
emotion of anger with service recovery and the distributive justice component.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Conclusión • 165
To begin with, our results report a significant positive effect of magnitude of
service failure on anger with service recovery as well as an indirect effect on
satisfaction with service recovery through anger with service recovery. In
addition, the magnitude of service failure negatively affects distributive justice
and also has an indirect effect on satisfaction with service recovery through
distributive justice. Furthermore, it appears that service failure controllability
has a significant negative effect on satisfaction with service recovery.
Moreover, apologies and explanations (recovery strategies) affect distributive
justice positively and also affect satisfaction with service recovery indirectly
through distributive justice. Results also show that distributive justice affects
satisfaction with service recovery positively. Finally, our findings reveal that
anger with service recovery has a significant negative effect on satisfaction with
service recovery. This has been the first attempt to model the effect of specific
secondary emotions on secondary (dis)satisfaction. This has also been the first
attempt to empirically test a model of (dis)satisfaction with service recovery in
double deviation scenarios.
In Chapter 6, we examine the impact of third-party complaints on company
performance. Specifically, we analyze the complaints from the Bank of Spain's
Complaints Service (the third party), which publishes an Annual Report on
Complaints to Spanish banks. We propose that the reléase of this information
about third-party complaints is economically relevant to the stock market. The
objectives of this chapter were: 1) to determine the economic impact for the
banks involved, in terms ofvariation in stock prices, of appearing on the Annual
Report on Complaints of the Bank of Spain 's Complaint Service, and 2) to
examine to what extent the variation in stock prices can be explained through
the number of complaints received by the bank in the Annual Report, its quality
corporate image, and its market concentraron.
Our results report that the stock market reacts significantly and negatively to
the publication of the Annual Reports of the Bank's of Spain Complaint
Service. In addition, there appear to be a significant moderating effect of quality
corporate image in the relationship between the reléase of the complaint report
and firm performance. Finally, we have found a positive relationship between
firm performance and the number of complaints per branch. Until now, the
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
166 • Chapter7
influence of customer's third-party responses had been analyzed from a
customer perspective but not on the basis of its impact on firm performance.
In Table 7.1, we provide a general overview of the fmdings that have been
discussed in the sepárate chapters.
Table 7.1
Summary of the main results for the different variables/measures employed in
the three empirical applications
Role/Impact of
Control attributions:
• indirect effect on
SATSF mediated by
Causal
attributions
Chapter 5
(voice response)
Chapter 4
(prívate response)
ANGSF (Hl)
• indirect effect on
complaint intentions
mediated by ANG S F
(H2)
• indirect effect on
repurchase intentions
mediated by ANG S F
Control attributions:
• direct and negative
effect on SATSR
(H2a)
• direct and negative
effect on DJ (H2b:
n.s.)
• indirect effect on
SATSR mediated by
DJ (H2bb: n.s.)
• direct and negative
(H3)
Stability attributions:
• direct effect on SATSF
(H4: n.s.)
Perceived
waiting time
effect on ANG S R
(H2c: n.s.)
• indirect effect on
SATSR mediated by
ANGSR (H2cc: n.s.)
Perceived waiting time
(PWT):
• direct effect on
ANGSF(H5:n.s.)
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Chapter 6
(third-party
response)
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Conclusión • 167
Role/Impact of
Product
relevance
Chapter 4
(prívate rcsponse)
Punctuality importance:
• direct and positive
effect on ANGSF (H6)
• indirect effect on
ANGSF mediated by
PWT (H7: n.s.)
Chapter 5
(voice response)
Magnitude ofservice
failure:
• direct and negative
effect on SATSR
(Hla: n.s.)
• direct and negative
effect onDJ(H Ib)
• indirect effect on
SATSR mediated by
DJ(Hlbb)
• direct and positive
effect on ANG S R
(Hlc)
• indirect and negative
effect on SATSR
mediated by ANG S R
(Hice)
Apologies:
• direct and positive
effect on DJ (H3a)
• indirect effect on
SAT S R mediated by
DJ (H3aa)
• direct and negative
effect on ANG S R
Recovery
strategies
(H4a: n.s.)
• indirect effect on
SATSR mediated by
ANGSR (H4aa: n.s.)
Explanations:
• direct and positive
effect on DJ (H3b)
• indirect effect on
SAT S R mediated by
DJ (H3bb)
• direct and negative
effect on ANG S R
(H4b: n.s.)
• indirect effect on
SAT S R mediated by
ANGSR (H4bb: n.s.)
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Chapter 6
(third-party
response)
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
168 • Chapter 7
Role/Impact of
Chapter 4
(prívate response)
Distributive
justice
Chapter 5
(voice response)
Chapter 6
(third-party
response)
Distributive justice
(DJ):
• direct and positive
effect on SATSR
(H5a)
• direct and negative
effect on ANGSR
(H5b: n.s.)
• indirect effect on
SAT S R mediated by
ANGSR (H5bb: n.s.)
Anger
Anger with service failure
(AngSF):
• direct and negative
effect on SATSF (H8)
• direct and positive
effect on complaint
intentions (H9)
• indirect effect on
complaint intentions
Anger with service
recovery (ANGSR):
• direct and negative
effect on SAT S R
(H6)
mediated by SAT S F
(H10: notpossible)
• direct and negative
effect on repurchase
intentions (Hll)
Satisfaction with service
(Dis)satisfaction failure (SATSF):
• direct and positive
effect on repurchase
intentions (H12: n.s.)
Satisfaction with service
recovery (SATSR)
Annual report
on complaints
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
—
(Publication of)
annual reports ofthe
Complaints Service
ofthe Bank of Spain:
• direct and negative
effect on firm
performance (Hl)
~
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Conclusión • 169
Role/Impact of
Chapter 4
(prívate response)
Chapter 5
(voice response)
Chapter 6
(third-party
response)
Quality
corporate image
Moderating effect in
the relationship btw.
a firm's appearance
¡n the annual report
and its performance
(H2)
Number of
complaints
/branches
Direct and positive
effect on bank's
abnormal returns
(H3)
Target market
concentration
Moderating effect in
the relationship btw.
the number of
complaints/branches
and the bank's
abnormal returns
(H4: n.s.)
Note: n.s. = non significant
In the remainder of this chapter, we present the specifíc conclusions and
contributions of the three empirical applications and discuss their limitations
and the directions for future research they have opened.
7.2 The consumer's reaction to delays in service (Chapter 4)
7.2.1 Main results and conclusions
The main goal of this study is to provide empirical evidence regarding the
relationships among the perceived waiting time, the attributions of the delay 's
cause, anger, satisfaction with service, and consumer's behavioral intention
(propensity to complain and repurchase intentions). Moreover, we have tested
whether the causal sequence "attribution-affect-behavioral intention" suggested by
Weiner (2000) was present in our data.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
170 •
Chapter7
For examining these relationships, we conducted an empirical study with a
sample of passengers that suffer delays in their flight at the airport of Alicante
during 1998. Results show that a passenger's anger in face of a flight delay is a
mediator of the relationship between the attribution of control and the
(dis)satisfaction with service, as well as between the attribution of control and
the behavioral intention (propensity to complain and repurchase intentions).
Additionally, the importance of punctuality leads to greater anger. Anger, in
tum, causes lower satisfaction with the service and lower repurchase intentions
as well as higher behavioral intentions to complain. We have found support for
the causal sequence "attribution-affect-behavioral intention" (i.e., "think-feel-act"),
which means that the attribution that the consumer makes regarding the causes of
the delay of a flight is an antecedent to his/her negative affective reaction, which,
in turn, has an influence on his/her behavioral intention.
In short, these results seem to demónstrate (in line with Weiner's (2000)
reflection) that anger is the dominant emotional reaction in situations of
negative result of a service where negligence in the productive process is
attributed to the provider (attribution control over the causes of the failure).
Moreover, anger, instead of satisfaction with the service, is the main
determining factor of the propensity to complain and repurchase intentions,
which supports Westbrook's (1987) and Weiner's (2000) proposals.
7.2.2 Managerial implications
Our results also suggest a number of managerial implications. First, the
important role of anger in the formation of satisfaction judgments and behavioral
intentions, as well as a mediator between attributions of control and behavioral
intention, has several consequences for the firm's management of flight delays.
Specifically, managers should oriéntate the complaint-handling actions towards the
lessening of anger by reducing the perceived delay time and improving the
perceived control attributions. Additionally, the role of anger as an antecedent of
complaint intention (in a last attempt to obtain compensation for the inconvenience
caused by the delays) should lead to efficient management of complaints and
claims which would facilitate compensation for these customers; which in turn
would give the service provider an opportunity to recover the customer.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Conclusión •
171
7.2.3 Limitations and future research
Our model is based on several assumptions that need to be taken into
consideration. Although these assumptions may be perceived as limiting the
ability to make generalizations of our results, they also provide rich
opportunities for further research in this área.
First, the measurement of some dimensions of the study with single-item
scales has been necessary given the specifíc nature of the service failure being
analyzed (an unexpected flight delay) (Antón, 1996; Sivadas and Baker-Prewitt,
2000). However, the literature points towards the superiority of multi-item
scales (Churchill, 1979; Peter, 1981; Szymanski and Henard, 2001). Therefore,
future studies should verify our findings using multi-item scales. Second,
although constraining the study to a single industry eliminates the problems
associated with industry differences (Hartline and Jones, 1996), future research
has to ascertain whether the results can be generalized to other settings. Finally,
it would be interesting to measure these constructs at more periods in time.
Thus, future research should make more use of longitudinal rather than crosssectional studies to examine the causal relationship between the different
proposed dimensions.
7.3 Anger and distributive justice ¡n a double deviation scenario
(Chapter 5)
7.3.1 Main results and conclusions
Financial institutions in general and the banking sector in particular are
among the service organizations that face huge competition all over the world.
This competition has enabled customers to act in a more demanding way in
their interaction with service providers due to the increased abundance of
choices. Obviously, service failures or mistakes are not completely unavoidable
even for the best service company and therefore the effective management of
consumer responses to service failure becomes very important in these highly
competitive markets (Hart et al., 1990). However, company can fail in the
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
172 •
Chapter7
recovery process and the customers are faced with a double deviation. The
crucial question is whether the service provider still has an opportunity to
satisfy these customers.
This work has proposed and empirically analyzed a model centered on
double deviation scenarios (i.e., failed recovery following a service failure). We
have examined the direct and indirect impact of different service failure and
service recovery-related variables on satisfaction with service recovery. Our
main goal is to broaden the knowledge about the type of variables and the
magnitude of their effect that contribute to the formation of satisfaction with
service recovery judgments in a double deviation scenario by integrating two
main theories, justice and emotions theories, in our conceptual framework.
Our field study based on a cross-sectional sample of 181 dissatisfied banking
customers has shown the following results. First, magnitude of service failure
directly affects anger with service recovery and distributive justice, and
indirectly affects satisfaction with service recovery through distributive justice
and anger. Second, service failure controllability affects satisfaction with
service recovery. Third, apologies and explanations signifícantly affect
distributive justice. Fourth, distributive justice mediates the relationship
between apologies and explanations and satisfaction with service recovery, and
also has a significant positive effect on satisfaction with service recovery.
Finally, anger with service recovery is found to directly affect satisfaction with
service recovery.
Whereas justice theory appears to be the dominant theoretical framework
applied to service recovery (Tax and Brown, 2000; Tax et al., 1998), this study
suggests that a specific emotion approach should also be considered when
dealing with double deviation scenarios.
7.3.2 Managerial implications
Double deviation events result in the magnifícation of negative evaluations
by customers. There is evidence that these negative evaluations by customers
prompt behavioral responses that transíate directly into losses for service firms.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Conclusión •
173
In a recent study on customer switching behavior in a wide variety of service
industries (Keaveney, 1995), service failures and failed recoveries accounted for
almost 60 percent of the critical behaviors by service providers that led directly
to customer switching. Of the 60 percent, 45 percent of these behaviors were
cited as the solé reason for the customer switching to another service provider.
In terms of customer defection, these results provide compelling evidence of the
potentially damaging impact of service failures followed by ineffective or nonexistent service recoveries. Henee, the service provider who is faced with this
critical situation should have information for taking decisions in two main
directions: to avoid/diminish the effect of the double deviation scenario and to
act on the explanatory variables to try to recover the customer that has
experienced an unsuccessful recovery.
One of the main findings of this study is that emotional responses derived
from failed service recovery (anger triggered by the failed recovery) influence
satisfaction judgments after accounting for cognitive antecedents of satisfaction
(distributive justice). We fínd that a failed recovery after a service failure
arouses negative emotions such as anger, which have a direct impact on
satisfaction with the recovery encounter since the problems for customers only
increase. Consequently, these incremental emotion-based effeets should be
avoided.
Additionally, apologies and explanations seem to be particularly effective in
driving customers' fairness perceptions, restoring justice, and indirectly
contributing to enhance satisfaction with service recovery. Given that there is
little incremental cost to the firm of providing explanations and apologies, such
practice should be built into all recovery programs.
All together, these results point out an interesting implication for
management in terms of the training programs directed to deal with customers'
responses to service failures. We suggest that these training programs should be
oriented to aspeets different than the technical or tangible ones. As said before,
customers judge the service received in terms of different elements that can be
summarized in the following five dimensions: tangibles, reliability,
responsiveness, assurance and empathy (Parasuraman et al., 1988). Therefore,
employees should be trained to deal with the specifíc emotions that arise in
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
174 » Chapter7
service failures and subsequent encounters (e.g., failed recoveries) by using
specifíc ('social') elements, such as empathy, that in fact are usually present in
customers' evaluations of services. Moreover, Mattila and Enz (2002) propose
that frontline staff members could correct service failures in real time if they
were able to process the customer's nonverbal signáis such as facial expressions
and were trained to respond to these forms of immediate feedback. When
feeling angry, customers tend to clench their jaws and narrow their eyebrows
downward, and by identifying these cues, frontline employees can adapt their
recovery styles to fít the individual customer (Menon and Dubé, 2000).
According to Smith and Bolton (2002), frontline employees should be trained
"to decode emotional cues [...] and to offer customized recovery efforts to
customers who exhibit negative emotional cues" (p. 19). Additionally, frontline
employees should be aware of customer expectations regarding emotional
responses (Menon and Dubé, 2000) and because anger is one of the most
contagious emotions (Tavris, 1989), they should also be trained to monitor their
intuitive reactions to upset customers. Finally, as McColl-Kennedy and Sparks
(2003) suggest, these communication/interaction skills should ensure that
recovery strategies are not perceived as avoiding responsibility or using
excuses, which would have a contrary effect.
In sum, this study adds to previous works that ¡Ilústrate the importance of
efficient recovery processes for fírms but with a new approach. By examining
the harmful consequences of failed recoveries (i.e. double deviation scenarios
with a specifíc emotion/justice approach), this research aims to encourage
service fírms to improve the design and execution of defensive strategies that
focuses on rebuilding the relationship with customers. Furthermore, this study is
based on the analysis of real service failures and recovery strategies, which we
think will help to explain the phenomena under analysis in a more realistic
fashion than most of the published research based on failure scenarios presented
to customers (or students).
7.3.3 Limitations and future research
Several limitations of this study must be recognized. First, we limit our
analysis to the negative emotion of anger and the distributive dimensión of
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Conclusión •
175
justice. Therefore, future research should try to determine whether different
specific negative emotions (e.g., frustration) and the interactional and
procedural components of justice affect post-recovery judgments in double
deviation scenarios in a different way, also in different service settings.
Second, we have not found a signifícant relationship between distributive
justice and anger with service recovery. More research is needed to examine the
effects of accountability (perceived volitional control over actions taken by the
provider) and counterfactual elements, as long as other appraisal antecedents of
injustice (e.g., goal hindrance and immorality), to identify how they influence
the relationships above and subsequent customers' satisfaction judgments.
Third, our results show that double deviation scenarios are especially
troublesome in a high competitive and mature market such as the banking
industry. Future research should also explore whether the pattern of responses
found in this research at the individual level is influenced by industry
characteristics (e.g., high vs. low level of competition) and market conditions
(e.g., mature markets such as fínancial industries vs. growing markets such as
the mobile industry).
Fourth, this study relies on self and retrospective reports (critical incident
technique, CIT); therefore, recall bias could have influenced the results (Michel,
2001). Similarly, the CIT method may result in other undesirable biases, such as
consistency factors or memory lapses (Singh and Wilkes, 1996), or
reinterpretation of the incident (Johnston, 1995). Future research could employ
controlled experiments to avoid disadvantages associated to the critical incident
technique, although external validity would then be an issue.
Fifth, some of our measures of the dependent variables are represented using
single-item scales, also due to the limitations in questionnaire length. Thus, the
measurement unreliability introduced by single items might have attenuated
some relationships. Although Drolet and Morrison (2001) argüe that single-item
scales have the advantage of avoiding the problems of incremental information
and potential greater error term correlations associated with multi-item
measures, future studies should consider multiple-item operationalizations.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
176 •
Chapter7
Finally, the research setting involves a single service category. In a strict
sense, the results pertain only to the respondents and generalizations to a wider
population or industry should be handled with caution. Thus, future research in
other service categories is needed to broaden our understanding of the role of
negative emotions and justice in double deviation contexts.
In sum, we hope that further conceptual and empirical development will
enrich research and practice concerned with the effects of specifíc emotions and
justice on post-consumption judgments in 'extreme' scenarios, that is, a failed
recovery following a service failure.
7.4 Third-party complaints and banking market valué (Chapter 6)
7.4.1 Main results and conclusions
From a theoretical perspective, this study aims to provide insights into the
impact of marketing strategies on financial performance by examining the
impact of actual customer complaint behavior to third parties (a sign of failure
on recovery efforts, a failed recovery strategy) on company performance.
Specifically, we have analyzed how the stock market (i.e., the investors) reacts
to the Annual Report on Complaints to Spanish banks published by the Bank of
Spain's Complaints Service (the third-party). The period of time examined
comprises the publication of Annual Reports between 1992 and 2002. The
methodology of analysis is based on the event study technique. This technique is
based on the theory of portfolio of financial economics which involves the
premise that stock markets are efficient and that company share prices better
reflect their quality strategy and their customers' satisfaction than their
accounting structure. Additionally, regression analysis and moderated
regression analysis (MRA) are used to examine the impact of quality corporate
image (measured through certificates of quality), the number of complaints per
branch, and market concentration on excess returns.
Results show signiñcant negative returns on the event date (t=0), on the day
after the event (t=l) and on t=3, as well as negative accumulated returns in
different post-event Windows, which suggests that the stock market reacts
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Conclusión •
177
negatively to the publication of the annual reports of the Complaints Service of
an external agency. Additionally, we find that the abnormal returns depend on
the number of complaints per branch and on quality corporate image.
Specifically, we find a negative moderating role of quality corporate image in
the relationship between the spreading of the complaints report and firm
performance, as well as a positive relationship between performance and the
number of complaints per branch.
7.4.2 Managerial implications
These findings are very interesting for management. First, they suggest that
firms should follow a dual emphasis according to the work of Rust et al. (2002).
The dual emphasis brings together the emphasis on costs and the emphasis on
revenues. In our study context, the emphasis on costs is showed by the positive
relationship between the number of complaints per branches and firm
performance. This result implicitly means that firms are reducing cost by
putting less economic resources in complaint management (henee the higher
number of complaints received) and that these cost decreases are viewed
favorably by investors (henee the significant and positive relationship). The
emphasis on revenues is showed by the negative moderating effect of quality
corporate image and the firm's appearance in the complaints report. The fact
that the firm's appearance in the complaints report of a third-party agency has a
negative impact on abnormal returns, reflects that this information damages
corporate reputation (Rose and Thomsen, 2004) and that investors disapprove of
a firm's management (Lee, 2001). This result implicitly means that investors
penalize firms with inconsistent signáis, such as the appearance in the
complaints report and having a quality award (henee the significant and
negative relationship). Therefore, firms should invest in improving complaint
management systems in order to avoid future complaints to reach the third-party
level. Complaining customers should be rewarded adequately using appropriate
complaint handling systems thereby avoiding subsequent complaints to thirdparty agencies. The employment of these complaint handling systems represents
the implementation of an emphasis on revenues by the firm, which focuses on
improvements in customer retention.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
178 •
Chapter7
The dual emphasis, therefore, allows the fírm to reach, according to
shareholders, acceptable fínancial returns. Thus, the implementation of a dual
emphasis implies the development of organizational structures that bring
together customers and cost áreas within the fírm, as well as reward structures
to incentive long-term evaluations.
Second, results point out the importance of effective recovery systems
showing that failures in recovery can have an impact on fírm performance. We
fínd that quality corporate image moderates negatively the relationship between
the publication of the complaints report and fírm performance. This result
reflects that investors penalize more the firms appearing in the complaints
report when they have a quality award than when they have not. That is,
investors penalize firms with inconsistent signáis (Lee, 2001). Therefore, firms
should avoid these inconsistent signáis. However, given that the quality awards
are signáis that should be communicated to the market, the only way of
achieving "consistency" with regard to third-party complaints would be
preventing them from reaching the public/market. This could be achieved with a
revenue emphasis that could improve customer satisfaction and avoid
complaints to third-party agencies.
7.4.3 Limitations and future research
Our study has several limitations, which may stimulate future research. First,
the selected banks that are quoted on the Madrid Stock Market are large
companies which restricts generalization of the conclusions. Future research is
needed to validate our findings across a wider sample base (e.g., crossnationally).
Second, the study described in this article is conducted in the banking
industry. Therefore, more research is needed to examine the effects of thirdparty complaints on performance in different industry contexts.
Third, our study does not account for every important predictor variables of
abnormal returns. Thus, the potential for an omitted variable bias exists.
Although our study assesses some of the predictors of the abnormal returns (i.e.,
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Conclusión • 179
quality corporate image, number of complaints per branch, and market
concentration), future studies are needed to assess the effects of other variables
on abnormal returns in relation to the effects of third-party complaints. In
addition, the lack of availability of information on other delimitating variables
of quality corporate image prevenís a more precise analysis of this dimensión.
In summary, for consumer behavior theory it is important to gain further
insights into the link between marketing constructs (e.g., complaint
management) and fínancial performance. Specifícally, the results obtained in
this study may encourage organizations to improve service delivery and
recovery systems and to implement better training systems to inspire customer
service employees to reduce the negative effects customers' problems. This
would lead to fewer complaints made to third-party agencies, an objective to
achieve in order to preserve fínancial performance from the encountered
negative effects of this type of consumer complaint behavior.
7.5 Final conclusión
Finally, we can give answer to the general research questions formulated in
the introduction of this dissertation.
The first two questions have been addressed in the first two empirical
applications (Chapters 4 and 5), whereas the last question has been examined in
the third empirical application (Chapter 6).
Research Question 1. Which are the main variables that affect specific
customers' responses after a service failure (Le., prívate and voice responses)
and subsequentjudgments and/or behaviors?
Results show that in the specific service settings considered, airline and
banking industries, different variables impact on the proposed models
differently. In Chapter 4, the service failure is a flight delay. We have obtained
that control attributions and punctuality importance affect anger with service
failure which subsequently affects satisfaction with service failure and
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
180 • Chapter7
behavioral intentions (propensity to complain and repurchase intentions)
Additionally, punctuality importance has an effect on perceived waiting time.
As a new element in the existing literature, we have jointly examined anger
(emotional reaction) and (dis)satisfaction with service failure (cognitive and
emotional evaluation). Thus, we have analyzed the impact of the initial negative
emotion of anger on the initial (dis)satisfaction judgment and subsequent
behavioral and complaining intentions.
In Chapter 5, the service failure is a banking failure and we examine double
deviation scenarios (i.e., failed recoveries that follow initial service failures).
Results show that magnitude of service failure affects distributive justice and
anger with service recovery directly, and satisfaction with service recovery
indirectly (through distributive justice and anger with service recovery). Service
failure controllability impacts on satisfaction with service recovery. Apologies
and explanations (recovery strategies) have a direct impact on distributive
justice and an indirect impact on satisfaction with service recovery through
distributive justice. Finally, distributive justice and anger with service recovery
affect satisfaction with service recovery.
To our knowledge, this has been the first attempt in the existing literature to
model the effect of specifíc secondary emotions on secondary (dis)satisfaction.
It has also been the first attempt to empirically test a model of (dis)satisfaction
with service recovery in double deviation scenarios.
All together, these results suggest an important implication to both theory
and practice in terms of the development and use of customer satisfaction
surveys. Ratings of customer satisfaction surveys provide a formal feedback to
the fírms and are usually used to evalúate the performance of company
employees, to enhance sales management and training programs, or to obtain
msights into competitors, among other Utilities. However, Peterson and Wilson
(1992) propose that to be able to interpret and effectively utilize customer
satisfaction ratings, it is necessary to understand what determines them as well
as know what variables and/or factors relate to them. The authors state that
"attempts should be made to explicitly control for variables like emotions, either
through the research designs employed or post-hoc statistical analyses (e.g.,
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Conclusión • 181
analysis of covariance)" (Peterson and Wilson, 1992, p. 69). Therefore, our
results contribute to this stream of research by explicitly controlling for the
negative emotion of anger in different service settings and conceptual models.
Research Question 2. Which is the role played by negative emotions (le.,
anger) vs. cognitive evaluations in customers' judgments and/or behaviors
following a service failure?
In Chapter 4, data support the main role of anger vs. satisfaction with service
recovery in affecting behavioral intentions (i.e., propensity to complain and
repurchase intentions). In Chapter 5, results show that anger with service
recovery has a greater impact on satisfaction with service recovery than
distributive justice. Additionally, anger with service recovery has a stronger
mediating effect than distributive justice in the relationship between magnitude
of service failure and satisfaction with service recovery.
All this together evidence the main role of emotions vs. cognitive
evaluations in the specifíc contexts examined. Therefore, our studies join the
existing services marketing literature that proposes that "emotions should
conceptually be included [into service satisfaction models] and combined with
cognitive evaluations of service" (Liljander and Strandvik, 1997, p. 168).
Research Question 3. Do customers' third-party responses affect Jirm
performance?
Finally, our results show that customers' third-party responses have an
impact on company performance through the investors' reaction. We examine
how the stock market (i.e., investors) reacts to the publication of the Annual
Report on Complaints to Spanish Banks published annually by the Bank of
Spain's Complaint Service (the third party). Following the signaling theory, we
fínd that investors respond to firm-level negative news (the fírm's appearance in
the Annual Report on Complaints) negatively by selling the stock. The results
also show that the number of complaints per branches has a positive impact on
the abnormal returns and that quality corporate image moderates negatively the
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Volver al Indice/Tornar a l'índex
182 « Chapter7
relationship between the spreading of the complaints report and the number of
complaints per branches.
Until now, the influence of customer's third-party responses has been
analyzed from a customer perspective but not on the basis of its impact on firm
performance. To our knowledge, this has been the fírst attempt to empirically
test the existence of a relationship between third-party responses and firm
performance.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References
A
Aaker, David A. (1992), "The valué of brand equity," Journal of Business Strategy, 13
(July/August), 27-32.
Aaker, David A. and Robert Jacobson (1994), "The financial information contení of
perceived quality," Journal o/Marketing Research, 31 (2), 191-201.
Adams, J. Stacey (1963), "Toward an understanding of inequity," Journal ofAbnormal
and Social Psychology, 67 (5), 422-436.
Adams, J. Stacey (1965), "Inequity in social exchange," in Advances in Experimental
Social Psychology, Vol. 2, L. Berkowitz, Ed. New York: Academic Press, 267-299.
Aiken, Leona S. and Stephen G. West (1991), Múltiple regression: testing and
interpreting interactions. London: Sage.
Alford, Bruce L. and Daniel L. Sherrell (1996), "The role of affect in consumer
satisfaction judgments of credence-based services," Journal of Business Research,
37(1), 71-84.
Anderson, Erin and Barton Weitz (1989), "Determinants of continuity in conventional
industrial channel dyads," Marketing Science, 8 (4), 310-323.
Anderson, Eugene W. (1994), "Post-purchase evaluations and word of mouth,"
Working Paper. National Quality Research Center, University of Michigan Business
School.
Anderson, Eugene W., Claes Fornell, and Donald R. Lehmann (1994), "Customer
satisfaclion, market share, and profitability: findings From Sweden," Journal of
Marketing, 58 (3), 53-66.
Anderson, Eugene W., Claes Fornell, and Roland T. Rust (1997), "Customer
satisfaclion, productivity and profitability: differences between goods and services,"
Marketing Science, 16 (2), 129-145.
Anderson, James C. and David W. Gerbing (1988), "Structural equation modeling in
practice: a review and recommended two-step approach," Psychological Bulletin,
103 (May), 411-423.
Andreasen, Alan R. (1985), "Consumer responses to dissatisfaction in loóse
monopolies," Journal of Consumer Research, 12 (September), 135-141.
Andreassen, Tor W. (1999), "What drives customer loyalty with complaint resolution?,"
Journal of Service Research, 1 (4), May, 324-332.
Andreassen, Tor W. (2000), "Antecedents to satisfaction with service recovery,"
European Journal of Marketing, 34 (1/2), 156-175.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
184 • References
Antón, J. (1996), Cusíomer relationship management. Making hand decisions with soft
numbers. New York: Prentice Hall.
Armistead, C.G., G. Clark, and P. Stanley (1995), Managing service recovery.
Cranfield: Cranfield School of Management.
Arnold, M. (1960), "Emotions and personality," in Psychological Aspects, Vol. 1. New
York: Columbia University Press.
Asquith, Paul and David W. Mullins (1986), "Signaling with dividends, stock
repurchases, and equity issues," Financial Management, 15 (3), 27-44.
Averill, J.R. (1983), "Studies on anger and aggression," American Psychologist, 38,
1145-1159.
B
Babin, Barry J. and Mitch Gríffín (1998), "The nature of satisfactíon: an updated
examination and analysis," Journal ofBusiness Research, 41,127-136.
Bagozzi, Richard P. (1994), "Structural equation models in marketing research: basic
principies," in Principies of Marketing Research, R.P. Bagozzi, Ed. Oxford,
England: Blackwell Publishers, 317-385.
Bagozzi, Richard P. (1996), "The role of arousal in the creation and control of the halo
effect in attitude models," Psychology and Marketing, 13 (3), May, 235-264.
Bagozzi, Richard P., Mahesh Gopinath, and Prashanth U. Nyer (1999), "The role of
emotions in marketing," Journal ofthe Academy of Marketing Science, 27 (2), 184206.
Bagozzi, Richard P. and Youjae Yi (1988), "On the evaluation of structural equation
models," Journal ofthe Academy ofMarketing Science, 16 (1), 74-99.
Bain, Joe S. (1951), "Relation of profit rate to industry concentration: American
manufacturing, 1936-1940," Quarterly Journal of Economics, 65 (3), August, 293324.
Baker, Julie and Michaelle Cameron (1996), "The effects ofthe service environment on
affect and consumer perception of waiting time: an integrative review and research
propositions," Journal ofthe Academy ofMarketing Science, 24 (4), 338-349.
Barón, Reuben M. and David A. Kenney (1986), "The moderator-mediator variable
distinction in social psychological research: conceptual, strategic, and statistical
considerations," Journal ofPersonality and Social Psychology, 51, 1173-1182.
Bateson, J.E.G. and D.K. Hoffman (1999), Managing services marketing: text and
readings. New York: The Dryden Press.
Bearden, William O. and Richard L. Oliver (1985), "The role of public and private
complaining in satisfaction with problem resolution," The Journal of Consumer
Affairs, 19 (2), 222-240.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References • 185
Bearden, William O. and Jesse E. Teel (1983), "Selected determinants of consumer
satisfaction and complaint reports," Journal of Marketing Research, 20 (February),
21-28.
Becker, G. (1965), "A theory of the allocation of time," The Economic Journal, 75, 493517.
Bell, Chip R. and R.E. Zemke (1987), "Service breakdown: the road to recovery,"
Management Review, October, 32-35.
Bentler, P.M. and C. Chou (1987), "Practical issues in structural equation modeling,"
Sociological Methods and Research, 16 (August), 78-117.
Berlyne, D. (1971), Aesthetics and Psychobiology. New York: Appleton-CenturyCrofts.
Bernard, Victor L. (1987), "Cross-sectional dependence and problems in inference in
marked-based accounting research," Journal ofAccounting Research, 25 (1), Spring,
1-48.
Berry, Leonard L. (1993), "Relationship marketing," in Emerging perspectives on
services marketing, L.Berry, G. Shostack, and G. Upah (eds.). Chicago: American
Marketing Association.
Berry, Leonard L. and A. Parasuraman (1991), Marketing services: competing through
quality. New York: The Free Press.
Berry, Leonard, Edwin F. Lefkowith, and Terry Clark (1988), "In services, what's in a
ñame?," Harvard Business Review, 66 (September/October), 28-30.
Best, Arthur and Alan R. Andeasen (1977), "Consumer response to unsatisfactory
purchases: a survey of perceiving defects, voicing complaints, and obtaining
redress," Law andSociety Review, 11 (Spring), 701-42.
Bettman, James R. (1979), An information processing theory of consumer choice.
Reading, MA: Addison-Wesley.
Bharadwaj, Sundar G. and Añil Menon (1993), "Determinants of success in services
industries," Journal of Services Marketing, 7, 19-40.
Bies, Robert J. (1987), "The predicament of injustice: the management of moral
outrage," Research in Organizational Behavior, 9, 289-319.
Bies, Robert J. and D.L. Shapiro (1987), "Interactional fairness judgments: the influence
of causal accounts," Social Justice Research, 1, 199-218.
Bitner, Mary Jo (1990), "Evaluating service encounters: the effects of physical
surroundings and employee responses," Journal of Marketing, 54 (April), 69-82.
Bitner, Mary Jo (1991), "The evolution of the services marketing mix and its
relationship to service quality," in Service Quality: A Multidisciplinar and
Multtnational Perspectives. S. Brown, E. Gumemsson, B. Edvardsson and B.
Gustavsson (eds.). New York, Lexington Books: 23-37.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
186 • References
Bitner, Mary Jo (1992), "Servicescapes: the impact of physical surroundings on
customers and employees," Journal of Marketing, 56 (2), 57-71.
Bitner, Mary Jo, Bernard H. Booms, and Mary Stanfield Tetreault (1990), "The service
encounter: diagnosíng favorable and unfavorable incidents," Journal of Marketing,
54 (January), 71-84.
Bitner, Mary Jo and Amy R. Hubbert (1994), "Encounter satisfaction versus overall
satisfaction versus quality: the customer's voice," in Service quality: new directions
in theory andpractice, R.T. Rust and R.L. Oliver, Eds. Newbury Park, CA: Sage
Publications, 72-94.
Blau, Peter M. (1964), Exchange andpower in social Ufe. New York: Wiley.
Bloch, Peter H., Daniel L. Sherrell, and Nancy M. Ridgway (1986), "Consumer search:
an extended framework," Journal of Consumer Research, 13 (June), 119-126.
Blodgett, Jeffrey G., Donald H. Granbois, and Rockney G. Walters (1993), "The effects
of perceived justice on complainants' negative word-of-mouth behavior and
repatronage intentions," Journal ofRetailing, 69 (4), Winter, 399-428.
Blodgett, Jeffrey G., Donna J. Hill, and Stephen S. Tax (1997), "The effects of
distributive, procedural, and interactional justice on postcomplaint behavior,"
Journal ofRetailing, 73 (2), 185-210.
Blodgett, Jeffrey G., Kirk L. Wakefíeld, and James H. Barnes (1995), "The effects of
customer service on consumer complaining behavior," Journal of Services
Marketing, 9,31-42.
Bloemer, Josée and Hans Kasper (1995), "The complex relationship between consumer
satisfaction and brand loyalty," Journal ofEconomic Psychology, 16, 311-329.
Bolfing, Claire P. (1989), "How do customers express dissatisfaction and what can
service marketers do about it?," The Journal ofServices Marketing, 3 (2), 5-23.
Bollen, Kenneth A. (1989), Structural equations with latent variables. New York: John
Wiley & Sons, Inc.
Bolton, Ruth N. (1998), "A dynamic model of the duration of the customer's
relationship with a continuous service provider: the role of satisfaction," Marketing
Science, 17(1), 45-65.
Booms, Bernard H. and Mary Jo Bitner (1981), "Marketing strategies and organization
structures for service firms," in Marketing of Services, J. Donnelly and W. George,
Eds. Chicago: A.M.A.
Boshoff, Christo (1997), "An experimental study of service recovery options,"
International Journal ofService Industry Management, 8 (2), 110-130.
Boshoff, Christo and Jason Leong (1998), "Empowerment, attribution and apologizing
as dimensions of service recovery. An experimental study," International Journal of
Service Industry Management, 9 (1), 24-47.
Bougie, Roger, Rik Pieters, and Marcel Zeelenberg (2003), "Angry customers don't
come back, they get back: the experience and behavioral implications of anger and
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References • 187
dissatisfaction in services," Journal of the Academy of Marketing Science, 31 (4),
377-393.
Brown, Stephen W., Deborah L. Cowles, and Tracy L. Tuten (1996), "Service recovery:
its valué and limitations as a retail strategy," International Journal of Service
Industry Management, 7 (5), 32-46.
Brickman, P., V.C. Rabinowitz, J. Karuza, D. Coates, E. Cohn, and L. Kidder (1982),
"Models of helping and coping," American Psychologist, 37 (4), 368-384.
Broadbridge, Adelina and Julie Marshall (1995), "Consumer complaint behavior: the
case of electrical goods," International Journal of Retail and Distribution
Management, 23 (9), 8-18.
Brown, Stephen J. and Jerold B. Warner (1985), "Using daily stock returns: the case of
event studies," Journal of Financial Economics, 14 (1), March, 3-31.
Brown, J. and Bernard Weiner (1984), "Affective consequences of ability versus effort
ascriptions: controversies, resolutions and quandaries," Journal of Education
Psychology, 76 (1), 145-158.
Buzzell, Robert D. and Bradley T. Gale (1987), The PIMSprincipies: Linking strategy
to performance. New York: The Free Press.
Buzzell, Robert D., Bradley T. Gale, and Ralph G.M. Sultán (1975), "Market share - a
key to profitability," Harvard Business Review (January-February), 97-106.
c
Cadotte, Ernest R., Robert B. Woodruff, and Roger L. Jenkins (1987), "Expectations
and norms in models of consumer satisfaction," Journal of Marketing Research, 24,
August, 305-314.
Cañáis, Jordi (1992), Estrategias del sector bancario en Europa. Madrid: Ariel
Economía.
Carlzon, Jan (1987), Moments oftruth. Cambridge, MA: Ballinger Publishing.
Casado-Díaz, Ana B. and Francisco J. Mas-Ruiz (2002), "The consumer's reaction to
delays in service," International Journal of Service Industry Management, 13 (2),
118-140.
Chebat, Jean-Charles and Witold Slusarczyk (2005), "How emotions medíate the effects
of perceived justice on loyalty in service recovery situations: an empirical study,"
Journal of Business Research, 58, 664-673.
Churchill, Gilbert A., Jr. (1979), "A paradigm for developing better measures of
marketing constructs," Journal of Marketing Research, 64-73.
Churchill, Gilbert A. Jr. and Carol Surprenant (1982), "An investigation into the
determinants of customer satisfaction," Journal of Marketing Research, 19
(November), 491-504.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
188 • References
Clayton, S.D. (1992), "The experience of injustice: some characteristics and correlates,"
SocialJustice Research, 5, 71-91.
Clemmer, E. C. (1993) An investigation into the relationship offairness and customer
satisfacíion with services. Hillsdale N.J.: Lawrence Erlbaum Associates.
Clemmer, Elizabeth C. and Benjamín Schneider (1996), "Fair servíce," in Advances in
Services Marketing and Management, Vol. 5, T.A. Swartz, D.E. Bowen and S.W.
Brown, Eds. Greenwich, CT: JAI Press, 109-126.
Collins, Daniel W. and Warren T. Dent (1984), "A comparison of alternative testing
methodologies used in capital market research," Journal ofAccounting Research, 22
(1), Spring, 48-84.
Cote, Joseph A., Ellen R. Foxman, and Bob D. Cutler (1989), "Selecting an appropriate
standard of comparison for post-purchase evaluations," Advances in Consumer
Research, 16, 502-506.
Corrado, Charles J. (1989), "A nonparametric test for abnormal security-price
performance in event studies," Journal of Financial Economics, 23 (2), August,
385-395.
Cronin, J. Joseph Jr. and Steven A. Taylor (1992), "Measuring service quality: a reexamination and extensión," Journal of Marketing, 56 (July), 55-68.
Cronin, J. Joseph Jr. and Steven A. Taylor (1994), "SERVPERF versus SERVQUAL:
reconciling performance-based and perceptions-minus-expectations measurement of
service quality," Journal ofMarketing, 58 (January), 125-131.
Crosby, Lawrence A. and Nancy Stephens (1987), "Effects of relationship marketing on
satisfaction, retention, and prices in the life insurance industry," Journal of
Marketing Research, 24 (November), 404-411.
Crosby, Philip B. (1987), Quality isfree. New York: McGraw-Hill.
Curren, Patrick J., Stephen G. West, and John F. Finch (1996), "The robustness of test
statistics to nonnormality and specification error in confirmatory factor analysis,"
Psychological Methods, 1 (1), 16-29.
Curren, Mary T. and Valerie S. Folkes (1987), "Attributional influences on consumer's
desires to communicate about producís," Psychology and Marketing, 4, Spring, 3145.
D
Davidow, Moshe (2000), "The bottom line impact of organizational responses to
customer complaints," Journal of Hospitality and Tourism Research, 24 (4), 473490.
Davidow, Moshe (2003), "Organizational responses to customer complaints: what
works and what doesn't," Journal ofService Research, 5 (3), 225-250.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References • 189
Day, Ralph L. (1980), "Research perspectives on consumer complaining behavior," in
Theoretical Developments in Marketing, C. Lamb and P. Dunne, Eds. Chicago:
American Marketing Association, 211-215.
Day, Ralph L. (1984), "Modeling choices among alternative responses to
dissatisfaction," in Advances in Consumer Research, 11, 496-499.
Day, Ralph L., Klaus Grabicke, Thomas Schaetzle, and Fritz Staubach (1981), "The
hidden agenda of consumer complaining," Journal ofRetailing, 57 (3), Fall, 88-106.
Day, Ralph L. and E. Laird Landon, Jr. (1976), "Collecting comprehensive complaint
data by survey research," Advances in Consumer Research, 5,263-268.
Day, Ralph L. and E. Laird Landon, Jr. (1977), "Toward a theory of consumer
complaining behavior," Consumer and Industrial Buyer Behavior, 425-437.
Deutsch, Morton (1985), Distributive justice: a social-psychological perspective. New
Haven, Ct.: Yale University Press.
Ding, Cody S. and Scott L. Hershberger (2002), "Assessing content validity and content
equivalence using structural equation modeling," Structural Equation Modeling, 9
(2), 283-297.
Drolet, Aimee L. and Donald G. Morrison (2001), "Do we really need multiple-item
measures in service research?," Journal of Service Research, 3 (3), February, 196204.
Dubé-Rioux, Laurette (1990), "The power of affective reports in predicting satisfaction
judgments," Advances in Consumer Research, 17, 571-576.
Dubé-Rioux, Laurette and Manfred F. Maute (1996), "The antecedents of brand
switching, brand loyalty and verbal responses to service failure," Advances in
Services Marketing and Management, 5, 127-151.
Dubé-Rioux, Laurette, B. Schmitt, and F. Leclerc (1988), "Consumer's reactions to
waiting: when delays affect the perception of service quality," Advances in
Consumer Research, 16, 59-63.
Duhaime, Carole P. and Stephen B. Ash (1979), "Satisfaction, dissatisfaction, and
complaining behavior: a comparison of male and female consumers," in Refining
Concepts of Consumer Satisfaction and Complaining Behaviors, H.K. Hunt and
R.L. Day, Eds. Bloomington (IN): Bureau of Business Research, 102-111.
E
Efron, Bradley and Robert J. Tibshirani (1993), An introduction to the bootstrap. New
York: Chapman & Hall.
El Mundo, "La Ley permite reclamar a los aeropuertos por los retrasos," May 20* 1999.
El País, "Caos en el tráfico aéreo, los consumidores solicitan que las compañías
indemnicen como Renfe cuando haya retrasos," April 23rd 1999.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
190 • References
El País, "Bruselas pide que las compañías aéreas indemnicen a los usuarios por los
retrasos," December 20th 1999.
Estelami, Hooman (2000), "Competitive and procedural determinants of delight and
disappointment in consumer complaint outcomes," Journal of Services Research, 2
(3), 285-300.
F
Feick, Lawrence F. (1987), "Latent class models for the analysis of behavioral
hierarchies," Journal ofMarketing Research, 24 (May), 174-186.
Fisher, James E., Dennis Garrett, Andy Cannon, and Jeri M. Beggs (1999), "Problem
businesses: consumer complaints, the better business bureau, and ethical business
practices," in Marketing and Public Policy Conference Proceedings, Vol. 9, G.T.
Gundlach, W.L. Wilkie, and P.E. Murphy, Eds. Chicago (IL): American Marketing
Association, 69-72.
Fisk, Raymond P. and Clifford E. Young (1985), "Disconfirmation of Equity
Expectations: Effects of Consumer Satisfaction with Services," Advances in
Consumer Research, 12,340-345.
Folger, R. and M.A. Konovsky (1989) "Effects of procedural and distributive justice on
reactions to pay decisions ', Academy ofManagement Journal, 32, 115-130
Folger, R., D. Rosenfield, and T. Robinson (1983), "Relative deprivation and procedural
justifications," Journal ofPersonality and Social Psychology, 45, 268-273.
Folkes, Valerie S. (1984), "Consumer reactions to product failures: an attributional
approach," Journal of Consumer Research, 10 (March), 398-409.
Folkes, Valerie S. (1988), "Recent attribution research in consumer behavior: a review
and new directions," Journal of Consumer Research, 14 (March), 548-65.
Folkes, Valerie S., Susan Koletsky and John L. Graham (1987), "A field study of causal
inferences and consumer reaction: the view from the airport," Journal of Consumer
Research, 13 (March), 534-539.
Folkes, Valerie S., and Barbara Kotsos (1986), "Buyers' and sellers' explanations for
product failure: who done it," Journal ofMarketing, 50 (April), 74-80.
Folkman, Susan and Richard S. Lazarus (1985), "If it changes, it must be a process:
study of emotion and coping during three stages of college examination," Journal of
Personality and Social Psychology, 49 (1), 150-170.
Fornell, Claes (1992), "A national customer satisfaction barometer: the Swedish
experience," Journal ofMarketing, 56 (January), 6-21.
Fornell, Claes, Michael D. Johnson, Eugene W. Anderson, Jaesung Cha, and Barbara
Everitt Bryant (1996), "The American customer satisfaction index: nature, purpose
and findings," Journal ofMarketing, 60 (October), 7-18.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References • 191
Fornell, Claes and David F. Larcker (1981), "Evaluating structural equation models
with unobservable variables and measurement error," Journal of Marketing
Research, 18 (February), 39-50.
Fornell, Claes and W. Robinson (1983), "Industrial organization and consumer
satisfaction/dissatisfaction," Journal of Consumer Marketing, 9,403-428.
Fornell, Claes and Birger Wernerfelt (1987), "Defensive marketing strategy by
customer complaint management: a theoretical analysis," Journal of Marketing
Research, 24 (November), 337-346.
Fornell, Claes and Birger Wernerfelt (1988), "A model for customer complaint
management," Marketing Science, 7 (3), Summer, 287-298.
Fornell, Claes and Robert A. Westbrook (1984), "The vicious circle of consumer
complaints," Journal of Marketing, 48 (Summer), 68-78.
Fournier, Susan and David Glen Mick (1999), "Rediscovering satisfaction," Journal of
Marketing, 63 (October), 5-23.
Freixas, Xavier (1996), Los límites de la competencia en la banca española. Bilbao:
Fundación BBV.
Fuentelsaz, Lucio, Jaime Gómez, and Yolanda Polo (2002), "Followers' entry timing:
evidence from the Spanish banking sector after deregulation," Strategic
Management Journal, 23 (3), March, 245-264.
G
Gardial, Sarah Fisher, Daniel J. Flint, and Robert B. Woodruff (1996), "Trigger events:
exploring the relationships between critical events and consumers' evaluations,
standards, emotions, valúes and behavior," Journal of Consumer Satisfaction,
Dissatisfaction and Complaining Behavior, 9, 35-51.
Garvín, David A. (1988), Managing quality: the strategic and competitive edge. New
York: The Free Press.
Giese, Joan L. and Joseph A. Cote (2000), "Defining consumer satisfaction," Academy
of
Marketing
Science
Review
(online),
00
01.
http://www.amsreview.org/articles/gieseQ 1 -2000.pdf.
Gilly, Mary C. and Betsy D. Gelb (1982), "Post-purchase consumer processes and the
complaining consumer," Journal of Consumer Research, 9 (December), 323-328.
Godwin, Beth F., Paul G. Patterson, and Lester W. Johnson (1995), "Emotion, coping
and complaining propensity following a dissatisfactory service encounter," Journal
of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 8, 155-163.
Goodman, Leo A. (1960), "On the exact variance of producís," Journal ofthe American
StatisticalAssociation, 55, 708-713.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
192 • References
Goodwin, Cathy and Ivan Ross (1992), "Consumer responses to service failures:
influence of procedural and interactional fairness perceptions," Journal of Business
Research, 25 (2), 149-163.
Greenberg, Jerald (1990), "Organizational justice: yesterday, today, and tomorrow,"
Journal ofManagement, 16 (2), 399-432.
Greising, David (1994), "Quality: how to make it pay," Business Week, 8 (August), 5459.
Gremler, Dwayne D. (2004), "The critical incident technique ín service research,"
Journal ofService Research, 1 (1), 65-89.
Grewal, Rajdeep, Joseph A. Cote, and Hans Baumgartner (2004), "Multicollinearity and
measurement error in structural equation models: Implications for theory testing,"
Marketing Science, 23 (4), 519-530.
Griliches, Zvi (1971), Price Índices and quality change. Cambridge, MA: Harvard
University Press.
Gronhaug, K. and G. Zaltman (1981), "Complainers and noncomplainers revisited:
another look at the data," Journal ofEconomic Psychology, 1, 121-134.
Gronroos, Christian (1984), Strategic management and marketing in the service sector.
Cambridge: Marketing Science Institute.
GrSnroos, Christian (1988), "Service quality: the six criteria of good perceived service
quality," Review ofBusiness, 9 (Winter), 10-13.
Gual, J., and X. Vives (1991), Ensayos sobre el sector bancario español. Madrid:
Fedea.
H
Hair, Joseph F. Jr., Rolph E. Anderson, Ronald L. Tatham, and William C. Black
(1999), Análisis multivariante, 5a edición. Madrid: Prentice Hall Iberia.
Halstead D., D. Hartman, and S. L. Schmidt (1994), "Multisource effects on the
satisfaction formation process," Journal of the Academy of Marketing Science, 22
(2), 114-129.
Hamilton, V. (1980), "Intuitive psychologist or intuitive lawyer? Altemative Models of
the Attribution Process," Journal ofPersonality and Social Psychology, 39 (5), 767772.
Harrison-Walker, L. Jean (2001), "E-complaining: a contení analysis of an Internet
complaint forum," Journal of Services Marketing, 15 (5), 397-412.
Hart, Christopher W.L., James L. Heskett and W. Earl Jr. Sasser (1990), "The profítable
arí of service recovery," Harvard Business Review, 68 (July-Augusí), 148-57.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References • 193
Hartline, M.D. and K.C. Jones (1996), "Employee performance cues in a hotel service
environment: influence on perceived service quality, valué, and word-of-mouth
intentions," Journal of Business Research, 35 (3), 207-215.
Harvey, John H. and Gifford Weary (1984), "Current issues in attribution theory and
research," Annual Reviews Psychology, 427-459.
Heider, Fritz (1958), The psychology of interpersonal relations. Hülsdale, NJ: Erlbaum.
Heise, David R. (1979), Understanding events: ajfect and the construction of social
action. New York: Cambridge University Press.
Heise, David R. (1989a), "Effects of emotion displays on the assessment of character,"
Social Psychology Quarterly, 52, 10-21.
Heise, David R. (1989b), "Modeling event structures," Journal of Mathematical
Sociology, 14, 139-169.
Hendricks, Kevin B. and Vinod R. Singhal (1997), "Does implementing an effective
TQM program actually improve operating performance? Empirical evidence from
firms that have won quality awards," Management Science, 43 (9), September,
1258-1274.
Heskett, James L., W. Earl Jr. Sasser and Christopher W.L. (1990), Service
breakthroughs: changing the rules ofthe gante. New York: Free Press.
Heskett, James L., T.O. Jones, G.W. Loveman, W. Earl Jr. Sasser, and L.A. Schlesinger
(1997), The Service Proflt Chain. New York: The Free Press
Hess, Ronald L., Jr., Shankar Ganesan, and Noreen M. Klein (2003), "Service failure
and recovery: the impact of relationship factors on customer satisfaction," Journal of
the Academy ofMarketing Science, 31 (2), 127-145.
Hirschman, Albert O. (1970), Exit, voice, and loyalty: responses to decline in firms,
organizations andstates. Cambridge, MA: Harvard University Press.
Hirschman, Elizabeth C. and Morris B. Holbrook (1982), "Hedonic consumption:
emerging concepts, methods and propositions," Journal ofMarketing, 46 (Summer),
92-101.
Hoffman, K. Douglas and Scott W. Kelley (2000), "Perceived justice needs and
recovery evaluation: a contingency approach," European Journal of Marketing, 34
(3/4), 418-432.
Hoffman, K. Douglas, Scott W. Kelley and Beth C. Chung (2003), "A CIT investigation
of servicescape failures and associated recovery strategies," Journal of Services
Marketing, 17 (4), 322-340.
Hoffman, K. Douglas, Scott W. Kelley and Holly M. Rotalsky (1995), "Tracking
service failures and employee recovery efforts," Journal of Services Marketing, 9
(2), 49-61.
Holbrook, Morris B. and Rajeev Batra (1987), "Assessing the role of emotions as
mediators of consumer responses to advertising," Journal ofConsumer Research, 14
(December), 404-420.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
194 • References
Holbrook, Morris B. and John O'Shaughnessy (1984), "The role of emotion ¡ n
advertising," Psychology and Marketing, 1,45-64.
Homans, George C. (1974), Social behavior: its elementaryforms. New York: Harcourt
Brace Jovanovich.
Hopkins, Willie E. and Shirley A. Hopkins (1997), "Strategic planning-financial
performance relationships in banks: a causal examination," Strategic Management
Journal, 18 (8), 635-652.
Horowitz, Joel L. (1997), "Bootstrap methods in econometrics: theory and numerical
performance," in Advances in economics and econometrics: theory and applications
Vol. 3, D.M. Kreps and K.F. Wallis, Eds. Cambridge: Cambridge University Press'
188-222.
HUÍ, M.( M. Thakor, and R. Gilí (1998), "The effect of delay type and service stage on
consumers' reactions to waiting," Journal of Consumer Research, 24, March, 469479.
HUÍ, Michael K. and David K. Tse (1996), "What to tell consumers in waits of different
lengths: an integrative model of service evaluation," Journal of Marketing, 60
April, 81-90.
I
Ittner, Chrístopher and David F. Larcker (1998), "Are nonfinancial measures leading
indicators of financial performance? An analysis of customer satisfaction," Journal
of'Accounting Research, 36 (Supplement), 1-35.
J
_____
Jacoby, Jacob and James J. Jaccard (1981), "The sources, meaning and validity of
consumer complaint behavior: a psychological analysis," Journal of Retailing, 57
Fall, 4-24.
Jaffe, Jeffrey F. (1974), "Special information and insider trading," Journal of Business,
47 (3), 410-428.
Jarvis, Cheryl Burke, Scott B. Mackenzie and Philip M. Podsakoff (2003), "A critical
review of construct indicators and measurement model misspecification in
marketing and consumer research," Journal of Consumer Research, 30 (September),
199-218.
Johnston, Robert (1995), "The zone of tolerance. Exploring the relationship between
service transactions and satisfaction with the overall service," International Journal
ofService Industry Management, 6 (2), 46-61.
Johnston, Robert and Adrián Fern (1999), "Service recovery strategies for single and
double deviation scenarios," The Service Industries Journal, 19 (2), April, 69-82.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References • 195
Johnston, Timothy C. and Molly A. Hewa (1997), "Fixing service failures," Industrial
Marketing Management, 26,467-473.
JSreskog, Karl G. and Dag Sorbom (1996), LISREL 8: user's reference guide. Chicago:
Scientific Software International, Inc.
Juran, Joseph M. and Frank M. Gryna (1980), Quality planning and analysis. New
York: McGraw-Hill.
K
Kasper, Hans (1988), "On problem perception, dissatisfaction and brand loyalty,"
Journal ofEconomic Psychology, 9, 387-397.
Kasper, Hans, Piet van Helsdingen, and Wouter de Vries jr (1999), Services marketing
management. An internationalperspective. Chichester, West Sussex, England: John
Wiley & Sons Ltd.
Katz, K, Larson, B. and Larson, R. (1991), "Prescription for the waiting in Une blues:
entertain, enlighten and engage," Shan Management Review, Winter, 44-53.
Keaveney, Susan M. (1995), "Customer switching behavior in service industries: an
exploratory study," Journal of Marketing, 59 (April), 71-82.
Kahneman, Daniel and Amos Tversky (1979), "Prospect theory: an analysis of decisión
under risk," Econometrica, 47, 263-291.
Keller, Kevin Lañe (1993), "Conceptualizing, measuring and managing customer-based
brand equity," Journal of Marketing, 57 (1), January, 1-22.
Kelley, H. (1967), "Attribution theory in social psychology," in Nebraska Symposium of
Motivation, Vol. 15, D.L. Lincoln, Ed. University of Nebraska Press, 192-238.
Kelley, Scott W. and Mark A. Davis (1994), "Antecedents of customer expectations for
service recovery," Journal ofthe Academy of Marketing Science, 22 (1), 52-61.
Kelley, Scott W., K. Douglas Hoffman, and Mark A. Davis (1993), "A typology of
retail failures and recoveries," Journal ofRetailing, 69 (Winter), 429-52.
Kelloway, E. Kevin (1998), Using LISREL for structural equation modeling. A
researcher 's guide. Thousand Oaks, California: SAGE Publications, Inc.
Klein, Benjamín and Keith B. Leffler (1981), "The role of market forces in assuring
contractual performance," Journal ofPolitical Economy, 89 (4), August, 615-641.
Kotler, Phillip (1997), Marketing management (9th ed.). New Jersey: Prentice Hall
Krishnan, S. and Valerie A. Valle (1979), "Dissatisfaction attributions and consumer
complaint behavior," Advances in Consumer Research, 6,445-449.
Kwoka, John E. Jr. and David J. Ravenscraft (1986), "Cooperation v. rivalry: price-cost
margins by Une of business," Económica, 53, August, 351-363.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
196 • References
L
La Gaceta de los Negocios, "La OCU ampliará la demanda por retrasos," 21 septiembre
1999.
Lancaster, Kelvin (1979), Variety, equity, and efficiency. New York: Columbia
University Press.
Lang, Larry HP and Rene Stulz (1994), "Tobin's Q, corporate diversifícation, and firm
performance," Journal ofPolitical Economy, 102(6), 1248-1280.
Lambert, R.A. (1998), "Customer satisfaction and future fínancial performance.
Discussion of: Are nonfinancial measures leading indicators of fínancial
performance? An analysis of customer satisfaction," Journal of Accounting
Research, 36, 37-46.
Laroche, M. and T. Taylor (1988), "An empirical study of major segmentation issues in
retail banking," International Journal ofBank Marketing, 6 (1), 31-48.
Lazarus, Richard S. (1982), "Thoughts on the relations between emotion and
cognition," American Psychologist, 37, 1019-1024.
Lazarus, Richard S., Alien D. Kanner, and Susan Folkman (1980), "Emotions: a
cognitive phenomenological analysis," in Emotion: Theory, Research and
Experience, 1: Theories of Emotion, R. Plutchnik and H. Kellerman, Eds. New
York: Academic Press, 189-217.
Lee, Peggy M. (2001), "What's in a name.com?: The effects of 'xom' ñame changes on
stock prices and trading activity," Strategic Management Journal, 22, 793-804.
Leeflang, Peter S.H., Dick R. Wittink, and Michel Wedel (2000), Building modelsfor
marketing decisions. Boston: Kluwer Academic Publishers.
Levesque, Terrence and Gordon H.G. McDougall (1996), "Determinants of customer
satisfaction in retail banking," International Journal ofBank Marketing, 14 (7), 1220.
Lewis, Barbara R. and Sotiris Spyrakopoulos (2001), "Service failures and recovery in
retail banking: the customers' perspective," International Journal of Bank
Marketing, 19(1), 37-47.
Liljander, Verónica and Tore Strandvik (1997), "Emotions in service satisfaction,"
International Journal ofService Industry Management, 8 (2), 148-169.
Lind, E.A. and T.R. Tyler (1988), The social psychology of procedural justice. New
York: Plenum.
Lings, Ian, Jos Lemmink, and Günther Botschen (2004), "Editorial. An agenda for
research into services management. Añerword to the special edition on European
research in service marketing," Journal ofBusiness Research, 57, 456-458.
Louro, Maria J., Rik Pieters, and Marcel Zeelenberg (2005), "Negative returns on
positive emotions: the influence of pride and self-regulatory goals on repurchase
decisions," Journal ofConsumer Research, 31 (March), 833-840.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References • 197
M
MacKinnon, Neil J. (1994), Symbolic interactionism as affect control. Albany (NY):
State University of New York Press.
Maister, D. (1985), "The psychology of waiting lines," in The Service Encounter,
Czepiel, J., Solomon, M. and Surprenant, C , Eds. Lexington: Lexington Books,
113-123.
Mano, Haim and Richard L. Oliver (1993), "Assessing the dimensionality and structure
of the consumption experience: evaluation, feeling and satisfaction," Journal of
Consumer Research, 20 (December), 451-466.
Marquis, M, Laurette Dubé, and Jean-Charles Chebat (1994), "Consumers" responses to
waiting time: new segmentation bases are required for service industries," Advances
in Consumer Research, 21, 548-553.
Mathur, Lynette Knowles and Ike Mathur (1995), "The effect of advertising slogan
changes on the market valúes of firms," Journal of Advertising Research, 35 (1),
January/February, 59-65.
Mattila, Anna S. (2001a), "The effectiveness of service recovery in a multi-industry
setting," Journal of Services Marketing, 15 (7), 583-596.
Mattila, Anna S. (2001b), "The impact of relationship type on customer loyalty in a
context of service failures," Journal ofService Research, 4 (2), November, 91-101.
Mattila, Anna S. and Cathy A. Enz (2002), "The role of emotions in service
encounters," Journal ofService Research, 4 (4) 268-277.
Mattila, Anna S. and Paul G. Patterson (2004), "Service recovery and fairness
perceptions in collectivist and individualist contexts," Journal of Service Research,
6 (4), 336-346.
Maute, Manfred F. and William R. Forrester, Jr. (1993), "The structure and
determinants of consumer complaint intentions and behavior," Journal ofEconomic
Psychology, 14,219-247.
Maxham III, James G. and Richard G. Netemeyer (2002), "A longitudinal study of
complaining customers' evaluations of múltiple service failures and recovery
efforts," Journal of Marketing, 66 (October), 57-71.
Maxham III, James G. and Richard G. Netemeyer (2003), "Firms reap what they sow:
the effects of shared valúes and perceived organizational justice on customers'
evaluations of complaint handling," Journal of Marketing, 67, January, 46-62.
McAlister, Debbie Thorne and Robert C. Erffmeyer (2003), "A content analysis of
outcomes and responsibilities for consumer complaints to third-party organizations."
Journal of Business Research, 56, 341-351.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
198 • References
McColl-Kennedy, Janet R. and Beveriy A. Sparks (2003), "Application of fairness
theory to servíce failures and service recovery," Journal of Service Research, 5 (3)
251-266.
McCollough, Michael A., Leonard L. Berry, and Manjit S. Yadav (2000), "An empirical
investigation of customer satisfaction after service failure and recovery," Journal of
Service Research, 3 (2), November, 121-137.
McCollough, Michael A. and Sundar G. Bharadwaj (1992), "The recovery paradox: an
examination of consumer satisfaction in relation to disconfirmation, service quality,
and attribution based theories," in Marketing theory andapplications, Chris T. Alien
et al., Eds. Winter Educators' Conference, Chicago, IL: American Marketing
Association, 119.
McDougall, Gordon H.G. and Terrence J. Levesque (1999), "Waiting for service: the
effectiveness of recovery strategies," International Journal of Contemporary
Hospitality Management, 11 (1), 6-15.
McFarland, Cathy and Michael Ross (1982), "Impact of causal attributions on affective
reactions to success and failure," Journal of Personality and Social Psychology, 43
(5), 937-946.
MeWilliams, Abigail and Donald Siegel (1997), "Event studies in management
research: theoretical and empirical issues," Academy of Management Journal, 40
(3), 626-657.
Menon, Kalyani and Laurette Dubé-Rioux (2000), "Ensuring greater satisfaction by
engineering salesperson response to customer emotions," Journal of Retailing, 76
(3), 285-307.
Meyer, John and Anne Mulherin (1980), "From attribution to helping: an analysis of the
mediating effeets of affect and expectaney," Journal of Personality and Social
Psychology, 39 (2), 201-210.
Michel, Stefan (2001), "Analyzing service failures and recoveries: a process approach,"
International Journal of Service Industry Management, 12 (1), 20-33.
Mikula, Gerold (1993), "On the experience of injustice," in European Review of Social
Psychology, Vol. 4, W. Stroebe and M. Hewstone, Eds. Chichester, UK: John Wiley
and Sons, 223-244.
Mikula, Gerold, Klaus R. Scherer, and Úrsula Athenstaedt (1998), "The role of injustice
in the elicitation of different emotional reactions," Personality and Social
Psychology Bulletin, 24 (7), 769-783.
Milgrom, Paul and John Roberts (1986), "Price and advertising signáis of product
quality," Journal ofPolitical Economy, 94 (4), 796-821.
Mitchell, Andrew A. (1980), "The use of an information processing approach to
understanding advertising effeets," Advances in Consumer Research, 8,193-196.
Mitchell, Adrián C. and Michael D. Richard (1994), "Consumer evaluation of the
complaint resolution process," Midwestern Journal of Business and Economics, 9
(1), 1-20.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References • 199
Mittal, Vikas, Pankai Kumar, and Michael Tsiros (1999), "Attribute-level performance,
satisfaction, and behavioral intentions," Journal ofMarketing, 63 (April), 88-101.
Mizerski, Richard W., Linda L. Golden, and Jerome B. Kernan (1979), "The attribution
process in consumer decisión making," Journal of Consumer Research, 6
(September), 123-140.
Montada, Leo (1994), "Injustice in harm and loss," Social Jusíice Research, 7, 5-28.
Montada, Leo and A. Schneider (1989), "Justice and emotional reactions to the
disadvantaged," Social Justice Research, 3,313-344.
Morganosky M. and M. Buckley (1986), "Complaint behaviour: analysis by
demographics, lifestyle and consumer valúes," Advances in Consumer Research, 14,
223-226.
Muller, T.E., D.K. Tse, and R. Venkatasubramaniam (1991), "Post-consumption
emotions: exploring their emergence and determinants," Journal of Consumer
Satisfaction, Dissatisfaction and Complaining Behavior, Vol. 4, pp. 13-20.
Myers, Stewart C. (1972), "The application of finance theory to public utility rate
cases," The Bell Journal ofEconomics and Management Science, 3(1), 58-97.
N
Nayyar, Praveen R. (1995), "Stock market reactions to customer service changes,"
Strategic Management Journal, 16 (1), January, 39-53.
Neter, J., W. Wasserman, and M. H. Kutner (1985), Applied linear statistical models:
regression, analysis ofvariance, and experimental design. Homewood, IL: Irwin.
Ngobo, Paul-Valentin (1999), "Industry versus firm effects: what drives fírm
performance in customer satisfaction," Working Paper, University of Montpellier II,
France.
Nguyen, Doan T. and Janet R. McColl-Kennedy (2003), "Diffusing customer anger in
service recovery: a conceptual framework," Australasian Marketing Journal, 11 (2),
46-55.
Nyer, Prashanth U. (1996), "The determinants of satisfaction: an experimental
verification of the moderating role of ambiguity," Advances in Consumer Research,
23, 255-259.
Nyer, Prashanth, U. (1997), "A study of the relationships between cognitive appraisals
and consumption emotions," Journal of the Academy of Marketing Science, 25, 4,
296-304.
O
Oliver, Richard L. (1980), "A cognitive model of the antecedents and consequences of
satisfaction decisions," Journal ofMarketing Research, 17 (November), 460-469.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
200 • References
Oliver, Richard L. (1981), "Measurement and evaluation of satisfaction processes in
retail settings," Journal ofRetailing, 57 (3), 25-48.
Oliver, Richard L. (1989), "Processing of the satisfaction response in consumption: a
suggested framework and research propositions," Journal of Satisfaction,
Dissatisfaction and Complaining Behavior, 2, 1-16.
Oliver, Richard L. (1993), "Cognitive, affective, and atributive bases of the satisfaction
response," Journal of Consumer Research, 20, December, 418-430.
Oliver, Richard L. (1994), "Conceptual issues in the structural analysis of consumption
emotion, satisfaction, and quality: evidence in a service setting," Advances in
Consumer Research, 21, 16-22.
Oliver, Richard L. (1996), "Varieties of valué in the consumption satisfaction
response," Advances in Consumer Research, 23,143-147.
Oliver, Richard L. (1997), Satisfaction: a behavioral perspective on the consumer. New
York: McGraw-Hill.
Oliver, Richard L. (2000), "Customer satisfaction with service," in Handbook of
Services Marketing and Management, T.A. Swartz and D. lacobucci, Eds. Thousand
Oaks, CA: Sage, 247-254.
Oliver, Richard L. and William O. Bearden (1981), "The role of involvement in
satisfaction processes," Advances in Consumer Research, 8 (1), 250-255.
Oliver, Richard L. and Wayne S. DeSarbo (1988), "Response determinants in
satisfaction judgements," Journal of Consumer Research, 14, March, 495-507.
Oliver, Richard L., Roland T. Rust and Sajeev Varki (1997), "Customer delight:
foundations, findings, and managerial insight," Journal ofRetailing, 73 (3), 311336.
Oliver, Richard L. and John E. Swan (1989), "Consumer perceptions of interpersonal
equity and satisfaction in transactions: a field survey approach," Journal of
Marketing, 53, April, 21-35.
Ortega, Patricia (2003). "El calvario de los consumidores," El País, 30* November, 2830.
Ostrom, Amy and Dwan lacobucci (1995), "Consumer trade-offs and the evaluation of
services," Journal of Marketing, 59 (January), 17-28.
P
Parasuraman, A., Valarie Zeithaml, and Leonard L. Berry (1985), "A conceptual model
of service quality and its implications for future research," Journal of Marketing, 49
(Fall), 41-50.
Parasuraman, A., Valarie Zeithaml, and Leonard L. Berry (1988), "SERVQUAL: a
multiple-item scale for measuring consumer perceptions of service quality," Journal
ofRetailing, 64 (1), Spring, 12-40.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References • 201
Parker, Cathy and Brian P. Mathews (2001), "Customer satisfaction: contrasting
academic and consumers' interpretations," Marketing Intelligence and Planning, 19
(1), 38-44.
Peter, J. (1981), "Construct validity: a review of basic issues and marketing practices,"
Journal of Marketing Research, 18, May, 133-145.
Peterson, Robert A. and William R. Wilson (1992), "Measuring customer satisfaction:
fact and artifact," Journal ofthe Academy ofMarketing Science, 20 (1), 61-71.
Petty, Richard E., John T. Cacioppo, and David T. Schumann (1983), "Central and
peripheral routes to advertisement effectiveness: the moderating effect of
involvement," Journal ofConsurner Research, 10 (September), 135-146.
Phillips, Lynn W., Dae R. Chang, and Robert D. Buzzell (1983), "Product quality, cost
position and business performance: a test of some key hypotheses," Journal of
Marketing, 47 (2), Spring, 26-43.
Ping, Robert A., Jr. (1995), "A parsimonious estimating technique for interaction and
quadratic latent variables," Journal of Marketing Research, 32 (3), 336-347.
Ping, Robert A., Jr. (2004), "On assuring valid measures for theoretical models using
survey data," Journal ofBusiness Research, 57,125-141.
Plutchik, Robert (1980), Emotion: a psychoevolutionary synthesis. New York: Harper &
Row.
Prakash, Ved (1991), "Intensity of dissatisfaction and consumer complaint behaviors,"
Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 4,
110-122.
Price, Linda L., Eric J. Arnould, and Sheila L. Deibler (1995), "Consumer's emotional
responses to service encounters. The influence of the service provider,"
International Journal of Service lndustry Management, 6 (3), 34-63.
Pruyn, Ad and Ale Smidts (1993), "Customers1 evaluations of queues: three exploratory
studies," European Advances in Consumer Research, 1,371-382.
Pruyn, Ad and Ale Smidts (1998), "Effects of waiting on the satisfaction with the
service: beyond objective time measures," International Journal of Research in
Marketing, 15, 321-334.
R
Ramírez-Alesón, Marisa and Manuel A. Espitia-Escuer (2001), "The effect of
international diversification strategy on the performance of Spanish-based firms
during the period 1991-1995," Management International Review, 41 (3), 211-315.
Ravenscraft, David J. (1983), "Structure-profit relationships at the Une of business and
industry level," Review ofEconomics andStatistics, 65 (1), February, 122-131.
Reichheld, Frederick F. and W. Earl Sasser, Jr. (1990), "Zero defections: quality comes
to services," Harvard Business Review, 68 (September-October), 105-11.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
202 • References
Reichle, Barbara and Leo Montada (1994), "Problems with the transition to parenthood:
perceived responsibility for restrictions and losses and the experience of injustice,"
in Entitlemení and the Affectional Bond: Justice in Cióse Relationships, M.J. Lerner
and G. Mikula, Eds. New York: Plenum, 205-228.
Reis, Harry T. (1984), "The multidimensionality of justice," in The Sense of Injustice:
Social Psychological Perspectives, R. Folger, Ed. New York: Plenum, 25-61.
Reisenzein, Rainer (1986), "A structural equation analysis of weiner's attribution-affect
model of helping behavior," Journal of Personality and Social Psychology, 50 (6),
1123-33.
Richins, Marsha L. (1980), "Consumer perceptions of costs and benefits associated with
complaining," in Refining Concepts and Measures of Consumer Satisfaction and
Complaining Behaviors, H. Hunt and R. Day, Eds. Bloomington: Indiana University
Press, 50-53.
Richins, Marsha L. (1983), "Negative word-of-mouth by dissatisfied consumers: a pilot
study," Journal ofMarketing, Al (Winter), 68-78.
Richins, Marsha L. (1987), "A multivariate analysis of responses to dissatisfaction,"
Journal ofthe Academy of Marketing Science, 15 (3), Fall, 24-31.
Richins, Marsha L. (1997), "Measuring emotions in the consumption experience,"
Journal of Consumer Research, 24, 127-146.
Robertson, Thomas S. and Hubert Gatignon (1986), "Competitive effects on technology
diffusion," Journal ofMarketing, 50 (July), 233-255.
Roos, Inger (1999), "Switching processes in customer relationships," Journal of Service
Research, 2 (1), 68-85.
Rose, Caspar and Steen Thomsen (2004), "The impact of corporate reputation on
performance: some Danish evidence," European Management Journal, 22 (2), 201210.
Roseman, Ira J. (1991), "Appraisal determinants of discrete emotions," Cognition and
Emotion, 5, 161-200.
Ross, J. and K. Gardner (1985),"The use of consumer initiated communication as
marketing research data," Working Paper, School of Management, University of
Minnesota.
Rowley, Jennifer and Jillian Dawes (1999), "Customer loyalty -a relevant concept for
libraries?," Library Management, 20 (6), 345-351.
Russell, Daniel (1982), "The causal dimensión scale: a measure of how individuáis
perceive causes," Journal of Personality and Social Psychology, 42 (June), 11371145.
Rust, Roland T., Christine Moorman, and Peter R. Dickson (2002), "Getting return on
quality: revenue expansión, cost reduction, or both?," Journal ofMarketing, 66 (4),
7-24.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References • 203
Rust, Roland T. and Anthony J. Zahorik (1993), "Customer satisfaction, customer
retention and market share," Journal ofRetailing, 69 (2), Summer, 193-215.
Rust, Roland T., Anthony J. Zahorik, and Timothy L. Keiningham (1994), Return on
quality: measuring the flnancial impací of your company 's quest for quality.
Chicago, IL: Probus.
Rust, Roland T., Anthony J. Zahorik, and Timothy L. Keiningham (1995), "Return on
quality (ROQ): making service quality financially accountable," Journal of
Marketing, 59 (April), 58-70.
Ruyter, Ko de and Josée Bloemer (1999), "Customer loyalty in extended service
settings. The interaction between satisfaction, valué attainment and positive mood,"
International Journal of Service Industry Management, 10 (3), 320-336.
Ruyter, Ko de, Luci Moorman, and Jos Lemmink (2001), "Antecedents of commitment
and trust in customer-supplier relationships in high technology markets," Industrial
Marketing Management, 30, 271-286.
Ruyter, Ko de and Martin Wetzels (2000), "Customer equity considerations in service
recovery: a cross industry perspective," International Journal of Service Industry
Management, 11 (1), 91-108.
s
Satorra, Albert and Peter M. Bentler (1988), "Scaling corrections for chi-square
statistics in covariance structure analysis," Proceedings of the Business and
Economics Statistics Section ofthe American Statistical Association, 308-313.
Satorra, Albert and Peter M. Bentler (2001), "A scaled difference chi-square test
statistic for moment structure analysis," Psychometrika, 66, 507-514.
Scaglione, Fred (1988), "Two-way communication. Tapping into gripes and profits,"
Management Review, September, 51-53.
Scherer K.R., R. Banse, and H.G. Walbott (2001), "Emotion inferences from vocal
expression correlate across languages and cultures," Journal of Cross-Cultural
Psychology, 32, 76-92.
Schoefer, Klaus (2001), "The role of perceived justice in the elicitation of differential
emotions following service complaint experiences: an exploratory study,"
Proceedings ofthe 30th EMAC Conference, Bergen, Norway, May 8-11.
Scotland, Randy (1991), "Customer service: a waiting game," Marketing, March 11, 13.
Seiders, Kathleen and Leonard L. Berry (1998), "Service fairness: what it is and why it
matters," Academy of Management Executive, 12 (2), 8-20.
Sharma, Neeru and Paul G. Patterson (2000), "Switching costs, alternative
attractiveness and experience as moderators of relationship commitment in
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
204 • References
professional, consumer services," International Journal of Service lndustry
Management, 11 (5), 470-490.
Sharma, Subhash, Richard M. Durand, and Oded Gur-Arie (1981), "Identification and
analysis of moderator variables," Journal ofMarketing Research, 18 (August), 291300.
Shaw, John, Eric Wild, and Jason Colquitt (2003), "To justify or excuse? A metaanalytic review ofthe effects f explanations," Journal ojAppliedPsychology, 88 (3),
444-458.
Sheppard, B.H., R.J. Lewicki, and J.W. Minton (1992), Organizational justice: the
searchforfairness in the workplace. New York: Lexington Books.
Shostack, G. (1977), "Breaking free from product marketing," Journal of Marketing,
41,73-80.
Shostack, G. Lynn (1985), "Planning the service encounter," in The service encounter,
J.A. Czepiel, M.R. Solomon, and C.F. Surprenant, Eds. Lexington: Lexington
Books, 243-254.
Shycon, H. N. (1992), "Improved customer service: measuring the payofF," The Journal
of Business Strategy, 13-17.
Singh, Jagdip (1988), "Consumer complaint intentions and behaviour: definitional and
taxonomical issues," Journal of Marketing, 52 (January), 93-107.
Singh, Jagdip (1989), "Determinants of consumers' decisions to seek third party
redress: an empirical study of dissatisfied patients," The Journal of Consumer
Affairs, 23 (Winter), 329-363.
Singh, Jagdip (1990), "Voice, exit, and negative word-of-mouth behaviors: an
investigaron across three service categories," Journal ofthe Academy ofMarketing
Science, 18(1), 1-15.
Singh, Jagdip and Roy D. Howell (1985), "Consumer complaining behavior: a review
and prospectus," in Consumer Satisfaction, Dissatisfaction and Complaining
Behavior, H.K. Hunt and R.L. Day, Eds. Bloomington, IN: Indiana University, 5966.
Singh, Jagdip and Robert E. Wilkes (1996), "When consumer complain: a path analysis
of the key antecedents of consumer complaint response estimates," Journal of the
Academy of Marketing Science, 24 (4), 350-365.
Sivadas, Eugene and Jamie L. Baker-Prewitt (2000), "An examination of the
relationship between service quality, customer satisfaction, and store loyalty,"
International Journal ofRetail & Distribution Management, 28 (2), 73-82.
Skarlicki, Daniel P. and Robert G. Folger (1997), "Retaliation in the workplace: the
roles of distributive, procedural, and interactional justice," Journal of Applied
Psychology, 82 (3), 434-443.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References • 205
Smith, Amy K. (1997), "Customer satisfaction with service encounters involving failure
and recovery: an integrative model of exchange," Doctoral Dissertation, University
of Maryland.
Smith, Amy K. and Ruth N. Bolton (1998), "An experimental investigation of customer
reactions to service failure and recovery encounters: paradox or peril?," Journal of
Service Research, 1 (1), August, 65-81.
Smith, Amy K. and Ruth N. Bolton (2002), "The effect of customers' emotional
responses to service failures on their recovery effort evaluations and satisfaction
judgments," Journal ofthe Academy of MarketingScience, 30 (1), 5-23.
Smith, Amy K., Ruth N. Bolton and Janet Wagner (1999), "A model of customer
satisfaction with service encounters involving failure and recovery," Journal of
Marketing Research, 36 (3), 356-372.
Smith, C. and P. Ellsworth (1985), "Patterns of cognitive appraisal in emotions,"
Journal ofPersonality and Social Psychology, 48 (4), 813-838.
Sobel, Michael E. (1982), "Asymptotic confidence intervals for indirect effects in
structural equation models," in Sociológica! Methodology, S. Leinhardt, Ed. San
Francisco: Jossey-Bass, 290-312.
Solomon, Michael R., Carol Surprenant, John A. Czepiel, and Evelyn G. Gutman
(1985), "A role theory perspective on dyadic interactions: the service encounter,"
Journal ofMarketing, 49,99-111.
Soteriou, Andreas C. and Stavros A. Zenios (2000), "Searching for the valué of quality
in financial services," Working Paper 00-39, Philadelphia, P.A.: The Wharton
Financial Institutions Center, University of Pennsylvania.
Spitzer, R. D. (1993), "Valuing TQM through rigorous financial analysis," Quality
Progress, July, 49-54.
Spreng, Richard A., G. Harrell, and R. Mackoy (1995), "Service recovery: impact on
satisfaction and intentions," Journal of Services Marketing, 9 (1), 15-23.
Spreng, Richard A., Scott B. MacKenzie and Richard W. Olshavsky (1996), "A
reexaminaron ofthe determinants of consumer satisfaction," Journal of Marketing,
60 (July), 15-32.
Spreng, Richard A. and Robert D. Mackoy (1996), "An empirical examination of a
model of perceived service quality and satisfaction," Journal of Retailing, 72 (2),
201-214.
Srivastava, Rajendra K., Tasadduq A. Shervani, and Liam Fahey (1998), "Market-based
assets and shareholder valué: a framework for analysis," Journal of Marketing, 62
(1), January, 2-18.
Stewart, Kate (1998), "An exploration of customer exit in retail banking," International
Journal of Bank Marketing, 16 (1), 6-14.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
206 • References
Storbacka, Kaj, Tore Strandvik, and Christian Gronroos (1994), "Managing customer
relationships for profit: the dynamics of relationship quality," International Journal
ofService Industry Management, 5 (5), 21-38.
Swan, John E. and Warren S. Martin (1980), "Testing comparison level and predictive
expectations models of satisfaction," ^í/rawcey in Consumer Research, 8, 77-82.
Swartz, Teresa A. and Dawn Iacobucci (2000), Handbook of services marketing and
management. Thousand Oaks, CA: Sage Publications, Ltd.
Szymanski, David M. and David H. Henard (2001), "Customer satisfaction: a metaanalysis of the empirical evidence," Journal ofthe Academy of Marketing Science,
29 (1), 16-35.
T
TARP/Technical Assistance Research Program (1979), Consumer complaint handling
in america: an update study. Washington, DC: White House Office of Consumer
Affairs.
TARP/Technical Assistance Research Program (1981), Measuring the grapevine:
consumer response and word-of-mouth. Atlanta, GA: The Coca-Cola Co.
Tavris, Carol Anne (1989), Anger: the misunderstood emotion (rev. ed..). New York:
Simón & Schuster / Touchstone.
Tax, Stephen S. and Stephen W. Brown (1998), "Recovering and learning from service
failure," Shan Management Review, Fall, 75-88.
Tax, Stephen S. and Stephen W. Brown (2000), "Service Recovery: Research Insights
and Practices," in Handbook ofServices Marketing and Management, T. Swartz and
D. Iacobucci, Eds. Thousand Oaks, CA: Sage, 271-286.
Tax, Stephen S., Stephen W. Brown and Murali Chandrashekaran (1998), "Customer
evaluations of service complaint experiences: implications for relationship
marketing," Journal ofMarketing, 62 (April), 60-76.
Taylor, Shirley (1994), "Waiting for service: the relationship between delays and
evaluations of service," Journal of Marketing, 58 (April), 56-69.
Taylor, Shirley (1995), "The effects of filled waiting time and service provider control
over the delay on evaluations of service," Journal of the Academy of Marketing
Science, 23 (1), 38-48.
Taylor, Shirley and John D. Claxton (1994), "Delays and the dynamics of service
evaluations," Journal ofthe Academy ofMarketing Science, 22 (3), 254-264.
Taylor, Steven A. (1997), "Assessing regression-based importance weights for quality
perceptions and satisfaction judgements in the presence of higher order and/or
interaction effects," Journal ofRetailing, 73 (1), Spring, 135-159.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References • 207
Taylor, Steven A. and Thomas L. Baker (1994), "An assessment of the relationship
between service quality and customer satisfaction ¡n the formation of consumer's
purchase intention," Journal ofRetailing, 70 (2), Summer, 163-178.
Terziovski, Mile, D. Samson, and D. Dow (1997), "The business valué of quality
management system certification. evidence from Australia and New Zealand,"
Journal qf Operations Management, 15 (1), 1-18.
Thaler, Richard (1985), "Mental accounting and consumer choice," Marketing Science,
4 (3), Summer, 199-214.
Thibaut, J., and L. Walker (1975), Procedural justice: A psychological analysis.
Hillsdale, NJ: Erlbaum.
Thompson, Phillip, Glenn DeSouza, and Bradley T. Gale (1985), "The strategic
management of service quality," Quality Progress, June, 20-25.
Tipper, R.H. (1997), "Characteristics of consumers who seek third party redress,"
Consumers Interests Annual, 43, 222-226.
Tortosa-Ausina, Emili (2002), "Bank cost efficiency and output specification," Journal
ofProductivity Analysis, 18, 199-222.
Tse, David K. and Peter C. Wilton (1988), "Models of consumer satisfaction formation:
an extensión," Journal of Marketing Research, 25 (May), 204-212.
Tsiros, Michael and Vikas Mittal (2000), "Regret: a model of its antecedents and
consequences in consumer decisión making," Journal of Consumer Research, 26
(March), 401-417.
Tyler, Tom R. (1994), "The psychological models of the justice motive: antecedents of
distributive and procedural justice," Journal of Personality and Social Psychology,
67 (November), 850-863.
U
Ursic, Michael L. (1985), "A model of the consumer decisión to seek legal redress,"
Journal of Consumers Affairs, Summer, 20-36.
y
Valle, Valerie and Melanie Wallendorf (1977), "Consumers' attributions of the cause of
their product satisfaction and dissatisfaction," in Consumer Satisfaction,
Dissatisfaction, and Complaining Behavior, R. Day (ed.). Bloomington: Indiana
University, 26-30.
Venkatesan, M. and Beverlee B. Anderson (1985), "Time budgets and consumer
services," in Services Marketing in a Changing Environment, T.M. Bloch, G.D.
Upah, and V. Zeithaml, Eds. Chicago: American Marketing Association, 52-55.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
208 • References
w
Walster, Elaine, Ellen Berscheid and G. William Walster (1973), "New directions in
equity research," Journal of Personality and Social Psychology, 25 (2), 151-176.
Walster, Elaine, G. William Walster, and Ellen Berscheid (1978), Equity: theory and
research. Boston: Allyn & Bacon.
Warland, Rex H., Robert O. Herrmann, and Jane Willits (1975), "Dissatisfied
consumers: who gets upset and who takes action," Journal of Consumer Affairs, 9,
Winter, 148-163.
Weiner, Bernard (1980), "A cognitive (attribution)-emotion-action model of motivated
behavior: an analysis of judgements of help-giving," Journal of Personality and
Social Psychology, 39 (2), 186-200.
Weiner, Bernard (1985), "An attributional theory of achievement motivation and
emotion," Psychological Review, 92 (4), 548-573.
Weiner, Bernard (1986), An attributional theory of motivation and emotion. New York:
Springer-Verlag.
Weiner, Bernard (1995), Judgments ofresponsibility: afoundationfor a theory of social
conduct. New York: Guilford Press.
Weiner, Bernard (2000), "Attributional thoughts about consumer behavior," Journal of
Consumer Research, 27 (December), 382-387.
Weiner, Bernard, Sandra Graham, and Carla Chandler (1982), "Pity, anger, and guilt: an
attributional analysis," Personality and Social Psychology Bulletin, 8 (2), 226- 232.
Weiner, Bernard, Irene Frieze, Andy Kukla, Linda Reed, Stanley Rest, and Robert M.
Rosenbaum (1971), Perceiving the causes of success andfailure. Morristown, NJ:
General Learning Press.
Weiner, Bernard, Dan Russell, and David Lerman (1979), "The cognition-emotion
process in achievement-related contexts," Journal of Personality and Social
Psychology, 37 (7), 1211-1220.
Welch, Patrick J. (1983), "Concentration in local commercial banking markets: a study
of the eight federal reserve district," Federal Reserve Bank ofSt. Louis, October, 1521.
Wernerfelt, Birger and Cynthia A. Montgomery (1988), "Tobin's Q and the importance
of focus in firm performance," American Economic Review, 78 (1), 246-250.
Westbrook, Robert A. (1987), "Product/consumption-based affective responses and
postpurchase processes," Journal ofMarketing Research, 24 (August), 258-270.
Westbrook, Robert A. (1980), "Intrapersonal affective influences upon consumer
satisfaction with producís," Journal of Consumer Research, 1 (June), 49-54.
Westbrook, Robert A. and Richard L. Oliver (1981), "Developing better measures of
consumer satisfaction: some preliminary results," Advances in Consumer Research,
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
References • 209
Vol. 8, K.B. Monroe, Ed. Ann Arbor, MI: Association for Consumer Research, 9499.
Westbrook, Robert A. and Richard L. Oliver (1991), "The dimensionality of
consumption emotion patterns and consumer satisfaction," Journal of Consumer
Research, 18 (June), 84-91.
Westbrook, Robert A. and Michael D. Reilly (1983), "Value-percept disparity: an
alternative to the disconfirmation of expectations theory on consumer satisfaction,"
Advances in Consumer Research, 10, 256-261.
Weun, Seungoog, Sharon E. Beatty, and Michael A. Jones (2004), "The impact of
service failure severity on service recovery evaluations and post-recovery
relationships," Journal of Services Marketing, 18 (2), 133-146.
Williamson, Oliver E. (1983), "Credible commitments: using hostages to support
exchange," American Economic Review, 73 (4), 519-540.
Wirtz, Jochen and John E.G. Bateson (1999), "Consumer satisfaction with services:
integrating the environment perspective in services marketing into the traditional
disconfirmation paradigm," Journal of Business Research, AA, 55-66.
Wirtz, Jochen and Anna S. Mattila (2004), "Consumer responses to compensation,
speed of recovery and apology after a service failure," International Journal of
Service Industry Management, 15 (2), 150-166.
Woodruff, Robert B„ Ernest R. Cadotte, and Roger L. Jenkins (1983), "Modeling
consumer satisfaction processes using experience-based norms," Journal of
Marketing Research, 20, August, 296-304.
Wólfl, Anita (2005), "The service economy in OECD countries," STI Working Paper
2005/3, Statistical Analysis of Science, Technology and Industry.
Wong, Paul T. P. and Bernard Weiner (1981), "When people ask 'why' questions, and
the heuristics of attributional search," Journal of Personality and Social Psychology,
40 (April), 650-63.
Y
Yi, Youjae (1990), "A critical review of consumer satisfaction," in Review of
Marketing, V. Zeithaml, Ed. Chicago: A.M.A., 68-123.
z
Zaichkowsky, Judith Lynne (1985), "Measuring the involvement construct," Journal of
Consumer Research, 12 (December), 341-352.
Zeelenberg, Marcel, W.W. Van Dijk, A.S.R. Manstead, and J. Van der Pligt (1998),
"The experience of regret and disappointment," Cognition and Emotion, 12 (2), 221230.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Volver al Indice/Tornar a l'índex
210 • References
Zeelenberg, Marcel, W.W. Van Dijk, A.S.R. Manstead, and J. Van der Pligt (2000)
"On bad decisions and disconfírmed expectancies: the psychology of regret and
disappointment," Cognition and Emotion, 14 (4), 521-541.
Zeelenberg, Marcel and Rik Pieters (1999), "Comparing service delivery to what might
have been. Behavioral responses to regret and disappointment," Journal of Service
Research, 2 (1), 86-97.
Zeelenberg, Marcel and Rik Pieters (2004), "Beyond valence in customer
dissatisfaction. a review and new findings on behavioral responses to regret and
disappointment in failed services," Journal ofBusiness Research, 57, 445-455.
Zeithaml, Valarie A., Leonard L. Berry, and A. Parasuraman (1993), "The nature and
determinants of customer expectations of service," Journal of the Academy of
Marketing Science, 21 (1), Winter, 1-12.
Zeithaml, Valarie A., Leonard L. Berry, and A. Parasuraman (1996), "The behavioral
consequences of service quality," Journal ofMarketing, 60 (2), 31-46.
Zeithaml Valarie A., A. Parasuraman, and Leonard L. Berry (1985), "Problems and
strategies in services marketing," Journal ofMarketing, 49, 33-46.
Zeithaml, Valarie A. and Mary Jo Bitner (2000), Services marketing: integrating
customer focus across the firm (2nd edition). USA: The McGraw-Hill Companies,
Inc.
Zemke, Ron and Dick Schaaf (1989), The service edge: 101 companies that profit from
customer care. New York: New American Library.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Resumen de la tesis doctoral
En este apartado se presenta un resumen en castellano de la tesis doctoral.
Dicho resumen contiene los siguientes apartados: introducción general,
objetivos de la investigación, metodología, resultados obtenidos, discusión de
los resultados (implicaciones para la gestión, limitaciones y futura
investigación) y conclusión final.
Introducción general
Comprender las consecuencias de los fracasos en los servicios es un
elemento clave en la gestión de las empresas de servicios. De hecho, ni las
empresas con una cultura orientada al cliente y programas de calidad rigurosos
pueden evitar los problemas durante la prestación del servicio (Kelley y Davis,
1994). Es decir, a pesar de todos los recursos invertidos, algunas cosas
simplemente pueden ir mal, especialmente teniendo en cuenta que la prestación
de servicios se apoya en la interacción humana.
Durante los últimos años las empresas de servicios han hecho numerosos
esfuerzos en desarrollar diferentes estrategias para abordar los fracasos en los
servicios (ej., entrenamiento de personal, departamentos de atención al cliente).
El objetivo de dichos esfuerzos ha sido doble: prevenir que el mismo problema
sucediera de nuevo y tratar de restaurar la satisfacción de los clientes que se
quejan. Así pues, la comprensión de los diversos elementos que afectan la
(in)satisfacción tras un fracaso en el servicio y los comportamientos que se
derivan de dicha (in)satisfacción puede ser muy útil para los gestores de las
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
212 • Resumen de la tesis doctoral
empresas de servicios en términos de reducir el impacto de dichos fracasos en el
resultado empresarial.
Si bien es cierto que la (in)satisfacción del consumidor es importante para
todos los gestores de marketing en general, algunas características subyacentes
en los servicios hacen que sea especialmente crítica para los gestores de las
empresas de servicios en particular. Primero, los servicios son en mayor grado
que los bienes, intangibles, heterogéneos, y simultáneamente producidos,
distribuidos y consumidos (Zeithaml et al., 1993). Segundo, el personal de
servicios juega un papel fundamental en la prestación de los mismos (Solomon
et al., 1985). Estas características incrementan la probabilidad de errores
(fracasos en los servicios) en el área de los servicios desde dos puntos de vista,
el operativo y el del cliente y, por lo tanto, incrementan la necesidad de
recuperación (Brown et al., 1996).
En este sentido, el fracaso en el servicio es definido como un problema
sufrido por un cliente con un servicio (Spreng et al., 1995) y que ocurre cuando
la experiencia del servicio queda por debajo de las expectativas previas del
cliente (Bell and Zemke, 1987). Tradicionalmente, la literatura de marketing de
servicios considera que los fracasos en los servicios son inevitables, o como
sostienen Hart et al. (1990) "los errores son una parte crítica de cada servicio"
(p. 148).
La literatura sobre la (in)satisfacción del cliente después de un fracaso en el
servicio puede dividirse en tres grandes grupos: (1) estudios sobre los
principales antecedentes y consecuencias de la (in)satisfacción después del
fracaso en el servicio (ej., Oliver, 1997; Westbrook, 1987); (2) estudios sobre
los principales antecedentes y consecuencias de la (in)satisfacción después del
fracaso en el servicio y en la recuperación del servicio (ej., Smith et al., 1999;
Tax et al., 1998); y (3) estudios sobre los principales antecedentes y
consecuencias de los estilos de respuesta a la (in)satisfacción del cliente (ej.,
Singh, 1988). Detrás de estos estudios existen multitud de teorías como el
paradigma de desconfirmación de expectativas (Oliver, 1981), la teoría de la
equidad (Clemmer y Schneider, 1996), la teoría de emociones/afecto (ver
Bagozzi et al., 1999 para una revisión) o la teoría de la atribución (Weiner,
1985), entre otras.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Resumen de la tesis doctoral • 213
Cuando sucede un fracaso en el servicio los clientes pueden responder de
multitud de formas, desde no hacer nada en absoluto hasta demandar a la
empresa por millones de euros. Este proceso comienza cuando el cliente ha
evaluado la experiencia de consumo y termina cuando completa todas las
respuestas de comportamiento y/o de no comportamiento (ej., no hacer nada)
que se derivan de la experiencia sufrida (Day, 1980).
Así, se han propuesto diferentes taxonomías para analizar las formas
empleadas por los clientes para expresar su insatisfacción (ej., Day, 1980; Day y
Landon, 1977; Richins, 1983 1987; Singh, 1988). En esta tesis, nos centramos
en el esquema propuesto por Singh (1988) que englobaría las tres aplicaciones
empíricas llevadas a cabo. Singh (1988) propone y evalúa empíricamente una
taxonomía de respuestas a la insatisfacción tras un fracaso en el servicio. Los
servicios analizados por el autor incluyen reparación de automóviles y servicios
financieros, entre otros. Según el esquema propuesto, cuando ocurre un fracaso
en el servicio tres tipos de respuestas son susceptibles de ocurrir: respuesta
privada (ej., intención de recompra), respuesta de voz (ej., pedir compensación
al vendedor), y respuesta a terceras partes (ej., quejarse a una asociación de
consumidores). Cada una de las tres respuestas recogidas en la clasificación de
Singh (1988) será objeto de atención en la presente tesis doctoral, de cara a
abordar los objetivos de investigación.
La tesis se organiza en siete capítulos. Los primeros tres capítulos (Capítulos
1 a 3), revisan los fundamentos teóricos que subyacen en las aplicaciones
teóricas presentadas, buscando la conexión entre dichas aplicaciones y la teoría
existente en marketing de servicios. Los siguientes tres capítulos (Capítulos 4 a
6), contienen los tres estudios empíricos que conforman la presente tesis
doctoral. Finalmente, el Capítulo 7 recoge los principales resultados,
contribuciones, implicaciones para la gestión, limitaciones y áreas de
investigación futura que se derivan de las tres aplicaciones empíricas
desarrolladas.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
214 • Resumen de la tesis doctoral
Objetivos de la investigación
La motivación general existente en la elaboración de esta tesis doctoral
radica en la importancia de los fracasos en los servicios para la vida diaria, no
sólo de las empresas sino también de los clientes. Por lo tanto, el objetivo de
esta tesis doctoral es contribuir a la evolución teórica y empírica de la
investigación acerca de los fracasos en los servicios a través de una mejor
comprensión de sus características y, consecuentemente, de sus implicaciones
para la gestión empresarial.
Este objetivo general se puede descomponer en tres objetivos de
investigación generales. Los dos primeros objetivos se examinan en las dos
aplicaciones empíricas iniciales (Capítulos 4 y 5 de la tesis doctoral), mientras
que el tercer objetivo se analiza en la tercera aplicación empírica (Capítulo 6 de
la tesis doctoral).
Objetivo de investigación 1. ¿Cuáles son las principales variables que afectan
a las respuestas específicas tras un fracaso en el servicio (i.e., respuestas
privada y de voz) y a los juicios y/o comportamientos subsiguientes?
Para responder a este objetivo, usamos datos empíricos para examinar los
principales antecedentes y consecuencias de una experiencia insatisfactoria que
sigue a un fracaso en el servicio. En la primera aplicación empírica (Capítulo 4
de la tesis doctoral), nuestro contexto de servicio es la industria aérea y el
fracaso en el servicio es un retraso en el vuelo. El segundo estudio (Capítulo 5
de la tesis doctoral) se lleva a cabo en la industria bancada y se centra en
distintos tipos de problemas con la entidad correspondiente. Tomando como
partida la literatura de una variedad de disciplinas, como el marketing y la
psicología social, primero formulamos un esquema conceptual para cada
estudio. A continuación, se analizan las principales relaciones implícitas en los
esquemas formulados.
Objetivo de investigación 2. ¿Cuál es el papel que juegan las emociones
específicas (i.e., enfado) frente a las evaluaciones cognitivas en los juicios y/o
comportamientos de los clientes tras un fracaso en el servicio?
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Resumen de la tesis doctoral • 215
En las dos primeras aplicaciones empíricas (Capítulos 4 y 5 de la tesis
doctoral), subrayamos la importancia de estudiar el papel de las emociones
específicas en la formación de los juicios de (insatisfacción en el contexto de
los fracasos en los servicios. Específicamente, nos centramos en la emoción
negativa del enfado. Este enfoque, apoyado en una emoción específica (i.e.,
enfado) está en línea con la literatura reciente que se centra en los elementos
idiosincrásicos de las emociones específicas (Bougie et al., 2003; Louro et al.,
2005; Zeelenberg y Pieters, 2004). Según estos estudios, se necesita profundizar
en los antecedentes, fenomenología y consecuencias específicos de las
diferentes emociones (como el enfado) (Lings et al., 2004). Sin embargo, se ha
prestado poca atención al estudio del enfado como la reacción emocional más
frecuente tras un fracaso en el servicio, así como, su influencia en las respuestas
de los clientes privadas (ej., intención de recompra) y de voz (ej., pedir
compensación al vendedor). Específicamente, examinamos el papel del enfado
frente a diferentes elementos cognitivos presentes en los dos modelos
propuestos.
Objetivo de investigación 3. ¿Influyen las respuestas a terceras partes en el
resultado empresarial?
En la tercera aplicación empírica (Capítulo 6 de la tesis doctoral),
examinamos cómo afectan las respuestas a terceras partes derivadas del fracaso
en el servicio al resultado empresarial. En este estudio, nos centramos en la
perspectiva del inversor incorporando el punto de vista financiero a la visión
tradicional del marketing. Específicamente, usamos literatura de varias
disciplinas, como el marketing, la economía financiera o la teoría de señales,
para formular un esquema conceptual. Después, evaluamos empíricamente las
relaciones propuestas.
Estos tres objetivos generales se pueden concretar en una serie de objetivos
específicos para cada uno de los estudios empíricos analizados. Dichos
objetivos específicos así como los resultados obtenidos se comentan en la
sección de resultados.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
216 • Resumen de la tesis doctoral
Metodología
En la primera aplicación empírica (Capítulo 4 de la tesis doctoral),
empleamos encuestas personales dirigidas a pasajeros que están esperando en la
sala de embarque tras anunciarse un retraso en su vuelo. Así, proponemos un
modelo que relaciona las percepciones del cliente acerca de las atribuciones de
control y estabilidad sobre la causa del retraso, la importancia de la puntualidad
del vuelo, el tiempo de espera, el enfado, la satisfacción con el servicio, y las
intenciones de queja y recompra futuras. El modelo propuesto es evaluado a
través de un análisis path con LISREL 8.3 (Joreskog y Sorbom, 1996). La
respuesta analizada en este estudio, siguiendo la clasificación de Singh (1988)
es la respuesta privada (i.e., intención de queja y de recompra futura).
En la segunda aplicación empírica (Capítulo 5 de la tesis doctoral),
empleamos una encuesta postal con cuestionarios auto-administrados en el
domicilio de los clientes. A través de la técnica del incidente crítico,
recolectamos información acerca de diversas variables referidas a un problema
sufrido por el cliente con su entidad bancaria principal. Específicamente, nos
centramos en los incidentes que se refieren a una estrategia de recuperación
fracasada tras el fracaso inicial del servicio, lo que se conoce como escenario de
"doble desviación" (Bitner et al., 1990). Así, proponemos un modelo que
relaciona las atribuciones de control, la magnitud del problema sufrido, las
estrategias de recuperación de disculpas y explicaciones, la justicia distributiva
percibida, el enfado derivado del proceso de recuperación, y la satisfacción con
el proceso de recuperación del servicio. El modelo propuesto es evaluado a
través de un sistema de ecuaciones estructurales con LISREL 8.3 (Joreskog y
Sorbom, 1996). La respuesta analizada en este estudio, siguiendo la
clasificación de Singh (1988), es la respuesta de voz (i.e., el cliente se queja a la
entidad bancaria buscando una solución al problema sufrido).
En la tercera aplicación empírica (Capítulo 6 de la tesis doctoral), utilizamos
datos del mercado de valores y de los balances de los bancos que cotizan en la
Bolsa de Madrid desde 1992 hasta 2002. A través de la técnica del estudio de
evento ("event study") analizamos el impacto de la publicación de los informes
anuales elaborados por el Servicio de Reclamaciones del Banco de España
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Resumen de la tesis doctoral « 217
sobre las rentabilidades de los bancos involucrados en dicho informe.
Adicionalmente, examinamos la capacidad explicativa del número de quejas
recibidas por oficinas, la concentración del mercado objetivo y la imagen
corporativa de calidad de la entidad bancaria, con respecto a las rentabilidades
anormales observadas. Para ello, utilizamos análisis de regresión y análisis de
regresión jerárquica moderadora. La respuesta analizada en este estudio,
siguiendo la clasificación de Singh (1988), es la respuesta a terceras partes (i.e.,
la tercera parte es el Banco de España que a través del Servicio de
Reclamaciones gestiona/recibe las quejas de los clientes bancarios españoles).
Resultados obtenidos
Adquirir un cliente nuevo es caro, particularmente en mercados maduros
como el de las compañías aéreas o la industria bancaria donde la competencia es
elevada, la diferenciación del producto es baja, y los costes promocionales se
han incrementado sustancialmente. Artículos recientes acerca de la gestión
basada en la lealtad, el marketing defensivo frente al marketing ofensivo, y la
gestión de las quejas, reflejan la importancia de nutrir la base de clientes leales
y de comprender y anticiparse a sus experiencias y comportamientos postcompra (ej., Fornell y Wernerfelt, 1987).
Así, una cuestión clave en la gestión estratégica de una empresa de servicios
es comprender los diferentes elementos que afectan la (in)satisfacción después
de un fracaso en el servicio y los comportamientos subsiguientes que se derivan
de esta (in)satisfacción. Por lo tanto, el objetivo global de esta tesis doctoral ha
sido contribuir a la evolución teórica y empírica de la investigación en los
fracasos de servicios poniendo de manifiesto su importancia para la gestión.
Las tres aplicaciones empíricas presentadas pretendían dar respuesta a los
siguientes objetivos específicos:
En el Capítulo 4, los objetivos eran: 1) desarrollar y evaluar empíricamente
un esquema conceptual basado en diferentes áreas de investigación que
identificara los antecedentes y las consecuencias de la (insatisfacción con el
fracaso del servicio (i.e., el retraso en el vuelo), 2) examinar el impacto de la
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
218 • Resumen de la tesis doctoral
emoción específica de enfado en el esquema anterior, y 3) explorar los efectos
de diferentes variables relacionadas con el servicio fracasado en la
(insatisfacción con el fracaso del servicio y en las intenciones de queja y de
comportamiento, directa e indirectamente a través del enfado con el servicio
fracasado y de la (insatisfacción con el servicio fracasado. Por lo tanto,
analizamos el impacto de la emoción inicial negativa del enfado en el juicio
inicial de (in)satisfacción y en las intenciones de comportamiento y queja
posteriores.
Los resultados obtenidos muestran la secuencia causal "atribución-afectointención de comportamiento" (i.e., pensar-sentir-actuar). Esto es, las
atribuciones que los clientes hacen con respecto a las causas de los retrasos de
los vuelos (pensar) tienen un impacto en el enfado sentido (sentir), que a su vez
afecta a las intenciones de comportamiento (actuar). Más aún, encontramos que
el enfado es un mediador entre las atribuciones de control y la (in)satisfacción
con el servicio fracasado. Además, la importancia de la puntualidad tiene un
efecto positivo en el enfado. Finalmente, el enfado tiene un efecto negativo en
la satisfacción con el servicio y en las intenciones de recompra y un efecto
positivo en la propensión a quejarse. En la tabla 1 se resumen los principales
resultados obtenidos en este capítulo.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Resumen de la tesis doctoral • 219
Tabla 1
Resumen de los principales resultados obtenidos en el Capítulo 4
Papel/Impacto de
Atribuciones causales
Capítulo 4 (respuesta privada)
Atribuciones de control:
• Efecto indirecto en SATSF mediado por ANGSF (Hl)
• Efecto indirecto en intenciones de queja mediado por ANG S F
(H2)
• Efecto indirecto en intenciones de recompra mediado por
ANGSF (H3)
Atribuciones de estabilidad:
• Efecto directo en SATSF (H4: n.s.)
Tiempo percibido de espera
Tiempo percibido de espera (PWT):
• Efecto directo en ANGSF (H5: n.s.)
Relevancia del producto
Importancia de la puntualidad:
• Efecto directo y positivo en ANGSF (H6)
• Efecto indirecto en ANGSF mediado por PWT (H7: n.s.)
Enfado
Enfado con el fracaso del servicio (AngSF):
• Efecto directo y negativo en SATSF (H8)
• Efecto directo y positivo en intenciones de queja (H9)
• Efecto indirecto en intenciones de queja mediado por SATSF
(H10: estimación infractora)
• Efecto directo y negativo en intenciones de recompra (Hl 1)
(Insatisfacción
Satisfacción con elfracaso del servicio (SATSf):
• Efecto directo y positivo en intenciones de recompra (H12:
n.s.)
Nota: n.s. = no significativo
En el Capítulo 5 analizamos el impacto de emociones negativas secundarias
(enfado con la recuperación del servicio) en los juicios de satisfacción
secundarios (satisfacción con la recuperación del servicio), en el contexto
específico de escenarios de desviación doble. Los objetivos de este capítulo
eran: 1) desarrollar y evaluar empíricamente un esquema conceptual
comprensivo basado en diferentes campos de investigación que identificara los
antecedentes de la (insatisfacción con la recuperación del servicio en el
contexto específico de los escenarios de desviación doble (le., recuperaciones
fracasadas tras fracasos en los servicios); 2) examinar el papel de las
emociones secundarias de enfado (i.e., enfado con la recuperación del servicio)
y el componente distributivo de la justicia percibida en el esquema previo, y 3)
explorar los efectos de variables relacionadas con el fracaso del servicio y con
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
220 • Resumen de la tesis doctoral
el fracaso de la recuperación en los juicios de (insatisfacción con la
recuperación del servicio, directa e indirectamente a través de las emociones
secundarias de enfado y el componente distributivo de la justicia.
Los resultados muestran un efecto significativo y positivo de la magnitud del
fracaso del servicio en el enfado con la recuperación del servicio, así como un
efecto indirecto en la satisfacción con la recuperación del servicio a través del
enfado. Además, la magnitud del fracaso del servicio afecta negativamente a la
justicia distributiva y tiene también un efecto indirecto en la satisfacción con la
recuperación del servicio a través de esta última variable. Por otro lado, la
atribución de control sobre el fracaso del servicio tiene un efecto negativo en la
satisfacción con la recuperación del servicio. También encontramos que las
disculpas y las explicaciones (las estrategias de recuperación) afectan a la
justicia distributiva positivamente y también a la satisfacción con la
recuperación del servicio indirectamente a través de la justicia distributiva.
Finalmente, los resultados muestran un efecto positivo de la justicia distributiva
y un efecto negativo del enfado en la satisfacción con la recuperación del
servicio. Éste ha sido el primer intento de desarrollar un modelo para examinar
el efecto de emociones secundarias específicas en la (in)satisfacción secundaria,
así como de evaluar empíricamente un modelo de (in)satisfacción con la
recuperación del servicio en escenarios de desviación doble. En la Tabla 2 se
muestra un resumen de los principales resultados obtenidos.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Resumen de la tesis doctoral • 221
Tabla 2
Resumen de los principales resultados obtenidos en el Capítulo 5
Papel/Impacto de
Capítulo 5 (respuesta de voz)
Atribuciones causales
Atribuciones de control:
Efecto directo y negativo en SAT S R (H2a)
Efecto directo y negativo en DJ (H2b: n.s.)
Efecto indirecto en SATSR mediado por DJ (H2bb: n.s.)
Efecto directo y negativo en ANG S R (H2C: n.s.)
Efecto indirecto en SAT S R mediado por ANG S R (H2CC: n.s.)
Relevancia del producto
Magnitud del fracaso del servicio:
Efecto directo y negativo en SAT S R (Hla: n.s.)
Efecto directo y negativo en DJ (Hlb)
Efecto indirecto en SAT S R mediado por DJ (Hlbb)
Efecto directo y positivo en ANG S R (Hlc)
Efecto indirecto en SATSR mediado por ANG S R (Hice)
Estrategias de recuperación Disculpas:
Efecto directo y positivo en DJ (H3a)
Efecto indirecto en SATSR mediado por DJ (H3aa)
Efecto directo y negativo en ANGSR (H4a: n.s.)
Efecto indirecto en SAT S R mediado por ANGSR (H4aa: n.s.)
Explicaciones:
Efecto directo y positivo en DJ (H3b)
Efecto indirecto en SATSR mediado por DJ (H3bb)
Efecto directo y negativo en ANGSR (H4b: n.s.)
Efecto indirecto en SATSR mediado por ANG S R (H4bb: n.s.)
Justicia distributiva
Justicia distributiva (DJ):
Efecto directo y positivo en SATSR (H5a)
Efecto directo y negativo en ANGSR (H5b: n.s.)
Efecto indirecto en SATSR mediado por ANGSR (H5bb: n.s.)
Enfado
Enfado con la recuperación del servicio (ANGSn):
• Efecto directo y negativo en SATSR (H6)
Satisfacción con la recuperación del servicio (SATSR)
(Insatisfacción
Nota: n.s. = no significativo
Finalmente, en el Capítulo 6 examinamos el impacto de las quejas a terceras
partes en el resultado empresarial. Específicamente, analizamos las quejas del
Servicio de Reclamaciones del Banco de España (la tercera parte), que publica
anualmente un Informe de Quejas de los bancos españoles. Proponemos que la
divulgación de esta información sobre quejas a terceras partes es
económicamente relevante para el mercado de valores. Los objetivos de este
capítulo eran: 1) determinar el impacto económico para los bancos
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
222 • Resumen de la tesis doctoral
involucrados en el informe anual, en términos de la variación en los precios de
las acciones, de aparecer en el Informe Anual de Quejas del Servicio de
Reclamaciones del Banco de España, y 2) examinar hasta qué punto la
variación en los precios de las acciones puede ser explicada a través del
número de quejas recibidas por el banco en el informe anual, su imagen
corporativa de calidad y la concentración de su mercado objetivo.
Los resultados muestran que el Mercado de Valores reacciona de forma
significativa y negativa a la publicación de los Informes Anuales del Servicio de
Reclamaciones del Banco de España. Además, obtenemos un efecto moderador
de la imagen corporativa de calidad en la relación entre la divulgación del
informe de quejas y el resultado empresarial. Finalmente, encontramos un
efecto positivo del número de quejas por oficinas en el resultado empresarial.
Hasta la fecha, la influencia de las respuestas de los clientes a terceras partes se
había analizado desde una perspectiva del cliente pero nunca con respecto a su
impacto en el resultado empresarial. En la Tabla 3 se muestra un resumen de los
resultados obtenidos en este capítulo.
Tabla 3
Resumen de los principales resultados obtenidos en el Capítulo 6
Papel/Impacto de
Capítulo 6 (respuesta a terceras partes)
Informe Anual de quejas
(Publicación de) informes anuales del Servicio de
Reclamaciones del Banco de España:
• Efecto directo y negativo en el resultado empresarial (Hl)
Imagen corporativa de
calidad
Efecto moderador en la relación entre la aparición de la empresa
en el informe anual y su resultado empresarial (H2)
Número de quejas/oficinas
Efecto directo y positivo en las rentabilidades anormales del
banco (H3)
Concentración del mercado
objetivo
Efecto moderador en la relación entre el número de
quejas/oficinas y las rentabilidades anormales del banco (H4:
n.s.)
Nota: n.s. = no significativo
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Resumen de la tesis doctoral • 223
Discusión de los resultados
La reacción del consumidor a los retrasos en el servicio (Capítulo 4)
El principal objetivo de este estudio es suministrar evidencia empírica con
respecto a las relaciones entre el tiempo percibido de espera, las atribuciones
sobre la causa del retraso, el enfado, la satisfacción con el servicio, y las
intenciones de comportamiento (propensión a la queja e intenciones de
recompra). Más aún, hemos evaluado si la secuencia causal "atribución-afectointenciones de conducta" sugerida por Weiner (2000) estaba presente en
nuestros datos.
Para examinar estas relaciones, llevamos a cabo un estudio empírico con una
muestra de pasajeros que sufren retrasos en sus vuelos en el aeropuerto de
Alicante durante 1998. Los resultados comentados en el apartado anterior
apoyan la existencia de la secuencia causal "atribución-afecto-intención de
conducta" (i.e., "pensar-sentir-actuar"), que implica que las atribuciones que los
clientes hacen acerca de las causas del retraso de su vuelo son un antecedente de
su reacción afectiva negativa que, a su vez, influye en sus intenciones de
conducta. En suma, estos resultados parecen demostrar (en línea con las
reflexiones de Weiner, 2000) que el enfado es la reacción emocional dominante
en situaciones de resultados negativos con un servicio donde la negligencia en
el proceso productivo se atribuye al proveedor. Más aún, el enfado en lugar de
la satisfacción con el servicio es el principal determinante de la propensión a la
queja y de las intenciones de recompra, lo que apoya las propuestas de
Westbrook (1987) y de Weiner (2000).
Basándonos en estos resultados, los gestores deberían orientar las acciones de
gestión de las quejas hacia la reducción del enfado reduciendo el tiempo percibido
de espera y mejorando las atribuciones de control percibidas. Además, el papel del
enfado como antecedente de las intenciones de queja (en un último intento por
parte del cliente de obtener una compensación por los inconvenientes causados por
el retraso) debería llevar a una gestión más eficiente de las quejas que facilitase la
compensación de estos clientes lo que, a su vez, le daría al proveedor del servicio
una oportunidad para recuperar al cliente insatisfecho.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
224 • Resumen de la tesis doctoral
Nuestro modelo está basado en diversas decisiones que deben tenerse en
cuenta. Aunque dichas decisiones pueden ser vistas como limitaciones para
hacer generalizaciones de nuestros resultados, también proporcionan
oportunidades para futuras investigaciones en el área. En primer lugar, la
medida de la mayoría de las dimensiones empleadas en el estudio se ha llevado
a cabo con escalas de un solo ítem. Ello ha sido motivado por la naturaleza
específica de los fracasos de servicio que estábamos analizando (un retraso en
un vuelo que no podía ser previsto) (Antón, 1996; Sivadas y Baker-Prewitt,
2000). Sin embargo, la literatura señala la superioridad de escalas de múltiples
ítems (Churchill, 1979; Meter, 1981; Szymanski y Henard, 2001). Por lo tanto,
estudios futuros deberían verificar nuestros resultados usando escalas de varios
ítems. En segundo lugar, aunque restringir nuestro estudio a una sola industria
elimina los problemas asociados a las diferencias entre industrias (Hartline y
Jones, 1996), la investigación futura debería investigar si nuestros resultados se
pueden generalizar a otras industrias. Finalmente, sería interesante medir estos
constructos en distintos periodos temporales. Así, la investigación futura
debería hacer uso de estudios longitudinales en lugar de estudios de sección
cruzada para examinar las relaciones causales entre las diferentes dimensiones
propuestas.
El enfado y la justicia distributiva en un escenario de desviación doble:
explicando la (insatisfacción en contextos de fracaso de servicio y
recuperación fracasada (Capítulo 5)
Las instituciones financieras en general y el sector bancario en particular se
encuentran entre las organizaciones de servicios que afrontan una gran
competencia por todo el mundo. Esta competencia ha permitido a los clientes
actuar de una forma más exigente en su interacción con las entidades bancarias
debido al incremento de opciones para elegir. Obviamente, los fracasos en los
servicios o errores no son completamente evitables ni siquiera para la mejor
empresa de servicios, por lo tanto la gestión efectiva de las respuestas de los
clientes a los fracasos en los servicios se convierte en una estrategia muy
importante en este contexto tan competitivo (Hart et al., 1990). Sin embargo, la
entidad puede fallar en el proceso de recuperación y es en ese momento cuando
los clientes se enfrentan a un escenario de desviación doble. La cuestión crucial
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Resumen de la tesis doctoral • 225
es si el proveedor del servicio tiene todavía una oportunidad de satisfacer a esos
clientes.
Este trabajo ha propuesto y ha analizado empíricamente un modelo centrado
en los escenarios de desviación doble (i.e., recuperación fracasada tras un
fracaso en el servicio). Hemos examinado los impactos directos e indirectos
sobre la satisfacción con la recuperación de diferentes variables relacionadas
tanto con el fracaso del servicio como con el fracaso de la recuperación. Nuestro
objetivo principal ha sido ampliar el conocimiento acerca del tipo de variables
que influyen (y en qué grado lo hacen) en la formación de los juicios de
satisfacción en un escenario de desviación doble. Para ello, hemos integrado dos
teorías principales en nuestro esquema conceptual, la de emociones y la de
justicia.
En función de los resultados comentados en la sección anterior, podemos
concluir que aunque la teoría de la justicia parece ser el esquema teórico
dominante aplicado a la recuperación del servicio (Tax y Brown, 2000; Tax et
al., 1998), nuestro estudio sugiere que un enfoque de emociones específicas
también debería considerarse cuando se trabaja en escenarios de doble
desviación.
Uno de los resultados principales de este estudio es que las respuestas
emocionales derivadas de una recuperación fracasada (el enfado que surge
debido a la recuperación fracasada) influyen en los juicios de satisfacción
después de tener en cuenta los antecedentes cognitivos (justicia distributiva).
Adicionalmente, las disculpas y las explicaciones parecen ser particularmente
explicativas en la formación de los juicios de justicia distributiva. Estos
resultados ponen de manifiesto una implicación muy interesante para los
gestores de marketing desde el punto de vista de los programas de
entrenamiento dirigidos a tratar con las respuestas de los clientes a los fracasos
en los servicios. Nuestra propuesta es que dichos programas deberían orientarse
a aspectos diferentes de los técnicos o tangibles. Como se ha comentado
previamente, los clientes juzgan los servicios recibidos en términos de
diferentes elementos que pueden ser resumidos en las siguientes cinco
dimensiones: tangibles, fiabilidad, responsabilidad, garantía y empatia
(Parasuraman et al., 1988). Los resultados obtenidos en esta tesis sugieren que
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
226 • Resumen de la tesis doctoral
los empleados deberían ser entrenados para tratar con las emociones específicas
que surgen en los fracasos de servicio y los encuentros subsiguientes (ej.,
recuperaciones fracasadas) usando elementos específicos ("sociales"), como la
empatia, que de hecho están normalmente presentes en las evaluaciones de los
servicios que realizan los clientes. Más aún, Mattila y Enz (2002) proponen que
el personal "de primera línea" podría corregir los fracasos en los servicios en
tiempo real si fueran capaces de procesar las señales no verbales de los clientes
como las expresiones faciales y estuviesen entrenados para responder a estas
expresiones que suponen un feedback inmediato. De acuerdo con Smith y
Bolton (2002), los empleados deberían estar entrenados para "decodificar las
claves emocionales [...] y ofrecer esfuerzos de recuperación adaptados a
aquellos clientes que muestran claves emocionales" (p. 19). Asimismo, los
empleados deberían ser conscientes de las expectativas de los clientes con
respecto a las respuestas emocionales (Menon y Dubé, 2000) y dado que el
enfado es una de las emociones más contagiosas (Tavris, 1989), deberían estar
entrenados para controlar sus reacciones intuitivas hacia los clientes
disgustados. Finalmente, tal y como sugieren McColl-Kennedy y Sparks (2003),
estas habilidades de comunicación o interacción con los clientes deberían
asegurar que las estrategias de recuperación no son percibidas como un intento
de eludir responsabilidades o usar excusas lo que tendría un efecto contrario y
nocivo.
Para terminar con la discusión de los resultados de este estudio, debemos
reconocer la existencia de algunas limitaciones que, por otro lado, abren el
camino a futuras investigaciones. Primero, hemos limitado nuestro análisis a la
emoción negativa del enfado y al componente distributivo de la justicia
percibida. Por lo tanto, la investigación futura debería tratar de determinar si
emociones específicas diferentes (ej., la frustración) y los otros componentes de
la justicia percibida (ej., interactiva y de procedimiento) afectan de modo
diferente los juicios de satisfacción post-recuperación en escenarios de
desviación doble, así como, en diferentes industrias de servicios.
Segundo, no hemos encontrado una relación significativa entre la justicia
distributiva y el enfado con la recuperación del servicio. Se necesitan más
investigaciones para examinar los efectos de otros elementos, como el control
percibido sobre las acciones tomadas por el proveedor del servicio después del
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Resumen de la tesis doctoral • 227
fracaso inicial, así como de otros antecedentes de la injusticia (ej., la
inmoralidad), para identificar cómo influyen en las relaciones planteadas y los
juicios de satisfacción subsiguientes.
Tercero, nuestros resultados muestran que los escenarios de desviación doble
son especialmente problemáticos en un mercado maduro y altamente
competitivo como el de la industria bancaria. La investigación futura debería
explorar también si los patrones de respuestas encontrados en este estudio a
nivel individual están influenciados por características de la industria (ej., nivel
de competencia alto vs. bajo) y por las condiciones del mercado (ej., mercados
maduros como la industria financiera vs. mercados en crecimiento como el de la
industria de la telefonía móvil).
Cuarto, este estudio se apoya en historias retrospectivas (técnica del
incidente crítico, CIT), por lo tanto un sesgo de recuerdo podría haber influido
en los resultados (Michel, 2001). De forma similar, la técnica del incidente
crítico puede tener otros sesgos no deseados, como factores de consistencia
(Singh y Wilkes, 1996), o de reinterpretación del incidente recordado (Johnston,
1995). La investigación futura podría emplear experimentos controlados para
evitar las desventajas asociadas al CIT, aunque la validez externa se podría
entonces ver comprometida.
Quinto, algunas de las medidas empleadas están representadas por escalas de
un solo ítem, en parte debido a limitaciones en la longitud del cuestionario.
Aunque Drolet y Morrison (2001) argumentan que las escalas de un solo ítem
tienen la ventaja de evitar los problemas de información incremental y de
correlaciones en los términos de error potencialmente mayores asociadas a las
escalas de múltiples ítems, las investigaciones futuras deberían considerar el
empleo de medidas de múltiples ítems.
Finalmente, esta investigación se ha desarrollado en una única industria de
servicios. En sentido estricto, los resultados pertenecen sólo a los individuos
que han participado en este estudio y las generalizaciones a poblaciones
mayores o a otras industrias deberían tratarse con cuidado. Así, se necesitan
estudios adicionales en otras categorías de servicios para ampliar nuestro
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
228 • Resumen de la tesis doctoral
conocimiento del papel de las emociones negativas y de la justicia percibida en
contextos de desviación doble.
Quejas a terceras partes y valor de mercado de la banca: efectos moderadores
de la imagen corporativa de calidad y la concentración de mercado (Capítulo
6)
Desde un punto de vista teórico, este estudio tiene como objetivo suministrar
nuevas perspectivas acerca del impacto de las estrategias de marketing en el
resultado empresarial examinando el impacto del comportamiento de queja
actual a terceras partes (un signo de fracaso en los esfuerzos de recuperación) en
el resultado de la empresa.
Los resultados comentados en la sección anterior son muy interesantes para
la gestión. En primer lugar, sugieren que las empresas deberían seguir un
énfasis dual de acuerdo con el trabajo de Rust et al. (2002). El énfasis dual
recoge el énfasis en costes y el énfasis en ingresos. En el contexto de nuestro
estudio, el énfasis en costes se muestra por la relación positiva entre el número
de quejas por oficinas y el resultado empresarial. Este resultado implícitamente
significa que las empresas están reduciendo costes a través de la disminución de
recursos económicos en la gestión de las quejas (de ahí el elevado número de
quejas recibidas) y que estas disminuciones de costes son percibidas de manera
positiva por los inversores (de ahí la relación significativa y positiva). El énfasis
en ingresos es demostrado por la aparición de la empresa en el informe de
quejas y por el efecto moderador negativo de la imagen corporativa de calidad.
El hecho de que la aparición de la empresa en el informe de quejas de una
tercera parte tiene un impacto negativo en los precios de las acciones refleja que
esta información daña la reputación corporativa (Rose y Thomsen, 2004) y que
los inversores desaprueban la gestión de la empresa (Lee, 2001). Además, los
inversores penalizan a las empresas que muestran señales inconsistentes, como
la aparición en el informe de quejas y la tenencia de un premio de calidad (de
ahí la relación moderadora significativa y negativa). Por lo tanto, las empresas
deberían invertir en mejorar los sistemas de gestión de quejas con el objetivo de
evitar que en el futuro las quejas de los clientes lleguen a las terceras partes. Los
clientes que se quejan deberían ser recompensados adecuadamente usando
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Resumen de la tesis doctoral • 229
sistemas de gestión de quejas apropiados que evitarían que las quejas llegasen a
las terceras partes. El empleo de estos sistemas representa un énfasis en ingresos
al centrarse en mejoras en la retención de los clientes.
El énfasis dual, por lo tanto, permite a la empresa alcanzar, de acuerdo con
los accionistas, resultados financieros aceptables. Así, la implantación de un
énfasis dual implica el desarrollo de estructuras organizativas que combinan las
áreas de clientes y de costes dentro de la empresa.
En segundo lugar, nuestros resultados señalan la importancia de contar con
sistemas de recuperación efectivos al mostrar que los fracasos en la
recuperación tienen un impacto en el resultado empresarial.
Finalmente, nuestro estudio presenta ciertas limitaciones que pueden
estimular la investigación futura. En primer lugar, los bancos seleccionados que
cotizan en la Bolsa de Madrid pueden ser clasificados como grandes empresas,
lo que restringe la generalización de las conclusiones obtenidas. Se necesitan
más estudios para validar nuestros resultados en muestras más amplias (ej., en
varios países).
En segundo lugar, el estudio se ha llevado a cabo en la industria bancaria.
Por lo tanto, la investigación futura debería examinar los efectos de las quejas a
terceras partes en el resultado empresarial en diferentes contextos industriales.
En tercer lugar, el estudio no tiene en cuenta otras variables que podrían ser
importantes para predecir los rendimientos anormales. Así, existe potencial de
sesgo de variables omitidas. Aunque nuestra investigación tiene en cuenta
algunos de estas variables explicativas (ej., imagen corporativa de calidad,
concentración del mercado objetivo), se necesitan más estudios que demuestren
los efectos de otras variables sobre las rentabilidades anormales en el contexto
de las quejas a terceras partes.
En resumen, para la teoría del comportamiento del consumidor es importante
profundizar en el nexo entre las variables de marketing (ej., la gestión de las
quejas) y el resultado financiero. Específicamente, los resultados obtenidos en
este estudio pueden animar a las empresas a mejorar sus sistemas de suministro
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
230 • Resumen de la tesis doctoral
de servicios y también de recuperación (cuando se produce un fracaso en el
servicio), así como a implementar mejores sistemas de entrenamiento del
personal que inspiren a los empleados a reducir los efectos negativos de los
problemas que encuentran con los clientes. Esto llevaría a una reducción del
número de quejas hechas a terceras partes, un objetivo a alcanzar si se pretende
proteger el resultado financiero de los efectos negativos de este tipo de
comportamiento de queja.
Conclusión final
Finalmente, podemos dar respuesta a los objetivos generales de
investigación formulados en el apartado de introducción de esta tesis doctoral.
Los primeros dos objetivos se han examinado en las dos primeras
aplicaciones empíricas (Capítulos 4 y 5), mientras que el último objetivo se ha
estudiado en la tercera aplicación empírica (Capítulo 6).
Objetivo de investigación 1. ¿Cuáles son las principales variables que afectan
a las respuestas especificas tras un fracaso en el servicio (i.e., respuestas
privada y de voz) y a los juicios y/o comportamientos subsiguientes?
Los resultados muestran que en el contexto específico de servicio
considerado, las industrias aérea y bancaria, diferentes variables tienen un
impacto diferente en los modelos propuestos. En el Capítulo 4, el fracaso en el
servicio es un retraso en el vuelo. Hemos obtenido que las atribuciones de
control y la importancia de la puntualidad afectan al enfado con el fracaso del
servicio que, subsiguientemente, afecta a la satisfacción con el servicio y a las
intenciones de conducta (propensión a la queja e intención de recompra).
Adicionalmente, la importancia de la puntualidad tiene un efecto en el tiempo
percibido de espera.
Como un nuevo elemento en la literatura existente, hemos analizado
conjuntamente el enfado (reacción emocional) y la (in)satisfacción con el
fracaso del servicio (evaluación cognitiva y afectiva). Así, hemos analizado el
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Resumen de la tesis doctoral • 231
impacto de la emoción inicial negativa de enfado en el juicio inicial de
(in)satisfacción y en las intenciones de conducta posteriores.
En el Capítulo 5 el fracaso en el servicio analizado es un fracaso en los
servicios bancarios y examinamos escenarios de desviación doble (i.e.,
recuperaciones fallidas tras fracasos en el servicio). Los resultados muestran
que la magnitud del fracaso del servicio afecta la justicia distributiva y el enfado
con la recuperación del servicio directamente, y la satisfacción con la
recuperación indirectamente (a través de la justicia distributiva y el enfado con
la recuperación). Las atribuciones de control sobre el fracaso en el servicio
impactan en la satisfacción con la recuperación. Las disculpas y las
explicaciones (estrategias de recuperación) tienen un impacto directo en la
justicia distributiva y un impacto indirecto en la satisfacción con la recuperación
a través de esta última variable. Finalmente, la justicia distributiva tiene un
efecto directo y positivo sobre la satisfacción con la recuperación y el enfado un
efecto directo y negativo sobre esta última variable.
A nuestro entender, éste ha sido el primer intento en la literatura existente de
modelizar el efecto de emociones secundarias específicas en la (in)satisfacción
secundaria. También ha sido el primer intento de evaluar empíricamente un
modelo de (in)satisfacción con la recuperación del servicio en un escenario de
desviación doble.
En conjunto, estos resultados sugieren una importante implicación para la
teoría y para la práctica empresarial con respecto al desarrollo y uso de los
cuestionarios de satisfacción del cliente. Las puntuaciones de estos
cuestionarios de satisfacción suministran a las empresas un feedback o
retroalimentación formal que es normalmente usado para evaluar el resultado de
los empleados, para mejorar la gestión de las ventas y los programas de
entrenamiento, o para obtener percepciones acerca de los competidores, entre
otras utilidades. Sin embargo, Peterson y Wilson (1992) proponen que para ser
capaces de interpretar y utilizar de forma efectiva dichos cuestionarios de
satisfacción, las empresas deben comprender qué determina las puntuaciones
obtenidas así como las variables y/o factores que intervienen en su formación.
Los autores señalan que "se deberían llevar a cabo intentos para controlar de
forma explícita variables como las emociones, bien a través de los diseños de
investigación empleados bien a través de análisis estadísticos post-hoc (ej.,
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
232 • Resumen de la tesis doctoral
análisis de covarianza)" (p. 69). Así, nuestros resultados contribuyen a esta
rama de investigación a través del control explícito de la emoción negativa del
enfado en diferentes industrias de servicios y modelos conceptuales.
Objetivo de investigación 2. ¿Cuál es el papel que juegan las emociones
específicas (i.e., enfado) frente a las evaluaciones cognitivas en los juicios y/o
comportamientos de los clientes tras un fracaso en el servicio?
En el Capítulo 4, los datos apoyan el papel principal del enfado vs. la
satisfacción con el fracaso del servicio a la hora de afectar las intenciones de
conducta (i.e., intención de queja y de recompra futura). En el Capítulo 5, los
resultados muestran que el enfado con la recuperación del servicio tiene un
impacto mayor en la satisfacción con la recuperación del servicio que la justicia
distributiva. Adicionalmente, el enfado con la recuperación del servicio tiene un
mayor efecto mediador que la justicia distributiva en la relación entre la
magnitud del fracaso del servicio y la satisfacción con la recuperación del
servicio.
En conjunto, estos resultados evidencian el papel principal de las emociones
frente a las evaluaciones cognitivas en los contextos específicos examinados.
Por lo tanto, nuestros estudios se encuadran en la literatura existente de
marketing de servicios que propone que "las emociones deberían ser
conceptualmente incluidas [en los modelos de satisfacción con el servicio] y
combinadas con las evaluaciones cognitivas del servicio" (Liljander y
Strandvik, 1997, p. 168).
Objetivo de investigación 3. ¿Influyen las respuestas a terceras partes en el
resultado empresarial?
Finalmente, nuestros resultados muestran que las respuestas de los clientes a
terceras partes tienen un impacto en el resultado de la compañía a través de la
reacción de los inversores. Examinamos la reacción del mercado de valores
(i.e., los inversores) a la publicación del informe anual del Servicio de
Reclamaciones del Banco de España. Siguiendo la teoría de señales,
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Volver al Indice/Tornar a l'índex
Resumen de la tesis doctoral • 233
encontramos que los inversores responden a noticias negativas de la empresa (la
aparición de la empresa en el informe anual de quejas) vendiendo los activos de
la misma. Los resultados también muestran que el número de quejas por
oficinas tiene un impacto positivo en las rentabilidades anormales y que la
imagen corporativa de calidad modera negativamente la relación entre la
divulgación del informe de quejas y el número de quejas por oficinas.
Hasta ahora, la influencia de las respuestas de los clientes a terceras partes se
había analizado desde una perspectiva del cliente pero no sobre la base de su
impacto en el resultado empresarial. A nuestro entender éste ha sido el primer
intento de evaluar empíricamente la existencia de una relación entre las
respuestas de los clientes a terceras partes y el resultado empresarial.
UNIVER^O
C»:,i¡3i
a c o r a ó ot •
ir, p
/7
Üfl&fi
r
de
A D
DE
ALICANTE
-i d •• De
. ; do
'"""'-:j
ld
i-,
«ftctafere
r,
-
K ; L
-' -
' ciía de la f e c h a
- ' - ; - r - d l cié D o n / D ñ a .
id c s l i n c a c i ó n d
-,
^ r -
de
secretario,
\¿n . Casado
^c\a Uv
(i iC 1C4'-::
.../{...
ctaíejf— —.
ódaV>rf
J200f
Lx I ili"SiDi.',NTH
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
^'¿3
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
UNIVERSIDAD DE ALICANTE
Departamento de Economía Financiera,
Técnicas de Mercado y Publicidad
Encuestador:
Nº Cuestionario:
Good morning/afternoon. My name is (...) and I am an interviewer
from the University of Alicante. We are carrying out a study of consumer behavior in the Airport of
Alicante and we would like to have your help. We can assure you that your answers are confidential,
and we would be most grateful if you could spare a few minutes of your time.
Entrevistador: A continuación, marque con una X las celdas correspondientes a las respuestas de los
entrevistados.
Q1. Has your flight been delayed for more than 15 minutes?
Yes
No
Q2. How long have you been waiting since the expected
departure time?
horas
minutos
Q3-11. Which airline are you flying with?
Q3. Air 2000 Limited
Q4. Air Europa
Q5. Airtours Internacional Aviatio
Q6. Aviaco
Q7. Britannia Airways Ltd.
Q8. British Airways
Q9. Iberia
Q10. Monarch Airlines Ltd.
Q11. Another (.....................)
Q100. Don’t know
Q12-14. What sort of flight is it?
Q12. Regular
Q13. Charter
Q14. Transfer
Q15-27. Where are you flying to?
Q15. Ámsterdam
Q18. Dublín
Q21. Las Palmas
Q24. Mallorca
Q19. Dusseldorf
Q22. London
Q25.Paris
Q16. Barcelona
Q20. Frankfurt
Q23. Madrid
Q26.Tenerife
Q17. Berlin
Q27. Other (.....................................................................................................................)
Q28. How confident were you before you set off that your flight would be on time?(Mostrar tarjeta)
Not at all confident 1 2 3 4 5 Completely confident
Q29. How satisfied are you with the delay?(Mostrar tarjeta)
Not at all satisfied
1 2 3
4 5 Very satisfied
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Q30. Have the airline personnel told you the reason for the delay?
Yes
⇒
Pase a la pregunta siguiente Q31.
No
⇒
Pase a la pregunta Q32.
Q31. What have you been told?..................................................................................
Q32-37. What do you think is the main reason for the delay?
Q32.Mechanical problems
Q33. Airline personnel
Q34. Previous flight departure delayed
Q35. Other passengers
Q36. Weather
Q37.Others(...........................................)
Q100. Don’t know
Q38. Do you think that the reason for the delay is temporary and unusual, or is it something that
happens to this flight very frequently?(Mostrar tarjeta)
...............................................................
Very
unusual
1
Unusual
Normal
Frequent
2
3
4
Very
frequent
5
Q39. Do you feel the reason for the delay is something the airline had control over? (Mostrar tarjeta)
Totally uncontrollable
1 2 3 4 5 Totally controllable
Q40. To what extent do you think there are actions that the airline could take to shorten this delay but
has not taken?(Mostrar tarjeta)
It could have done anything 1 2 3 4 5 It could have acted
Q41-45. Look at the sentences below and decide to what extent you agree or disagree with them:
(Mostrar tarjeta)
Q41. It is important for me that my flight arrives
on time at its destination
Q42. I am angry with the airline for the delay
Q43. I am disappointed with the delay
Not at all
important
Not at all angry
Not at all
disappointed
Q44. I think I will complain to the airline Not at all inclined
personnel about the delay
to complain
Q45. I would not fly with this airline again if I Would not
had the choice
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
1
2
3
4
5
1
1
2
2
3
3
4
4
5
5
1
2
3
4
5
Very
important
Very angry
Very
disappointed
Very inclined
1
2
3
4
5
Would
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Q46. In case you are going to complain seriously, do you know which are the institutions or the
organisations where you must go?
⇒ Which are they?.....................................................................................
Yes
No
Q47. Are you travelling alone?
Yes
No
Q48-52. Which is the main reason for your trip?
Q48. Business
Q49. Holidays
Q50. Studies
Q51. Visiting friends/relatives
Q52. Another
Q53. Do you think that the airport offers enough things to do?(Mostrar tarjeta).
They are not enough
1 2 3 4 5 They are enough
Q54-57. How many times have you flown in the past year?
Q54. Between Zero and One
Q55. Between Two and Three
Q56. Between Four and Five
Q57. More than Six
Q100. Don’t know
DATOS CLASIFICACIÓN
Q58. Sex:
Male
Female
Q59-62. Age:
Q59. Between 18-35
Q60. Between 36-53
Q61. Between 54-71
Q62. Older than 71
Q63. Departure time:....................................................................................................
Q64. Where are you from?...........................................................................................
The interview has finished. Thank you very much for your collaboration and for the time you have
spent with us.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
FECHA:
HORA:
OBSERVACIONES:
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
UNIVERSIDAD DE ALICANTE
Departamento de Economía Financiera,
Técnicas de Mercado y Publicidad
Encuestador:
Nº Cuestionario:
Buenos días/tardes. Mi nombre es (...) y soy entrevistador de la Universidad de Alicante. Estamos
realizando un estudio sobre la conducta del consumidor en el Aeropuerto de Alicante y nos gustaría
contar con su colaboración. Le aseguramos que sus respuestas serán absolutamente confidenciales.
¿Querría dedicarnos unos minutos?
Entrevistador: A continuación, marque con una X las celdas correspondientes a las respuestas de los
entrevistados.
P1. ¿Se ha retrasado su vuelo más de 15 minutos sobre la hora prevista?
Sí
No
⇒
Pasar a cuestionario siguiente
P2. ¿Cuánto tiempo lleva esperando desde la hora prevista de salida?
horas
minutos
P3-11. ¿Qué compañía de transporte aéreo va a utilizar?
P3. Air 2000 Limited
P4. Air Europa
P5. Airtours Internacional Aviatio
P6. Aviaco
P7. Britannia Airways Ltd.
P8. British Airways
P9. Iberia
P10. Monarch Airlines Ltd.
P11.Otra(.......................................)
P100. NS/NC
P.12-14. ¿Con qué tipo de vuelo se dispone a viajar?
P12. Regular
P13. Charter
P14. Conexión
P15-27. ¿Cuál es el destino de su vuelo?
P15. Ámsterdam
P18. Dublín
P21. Las Palmas
P24. Mallorca
P19. Düsseldorf
P22. Londres
P25. París
P16. Barcelona
P20. Frankfurt
P23. Madrid
P26. Tenerife
P17. Berlín
P27. Otro(............................................................................................................................)
P28. Cuando usted llegó al aeropuerto, ¿esperaba que el avión saliera a su hora o por el contrario
estaba convencido/a de que sufriría un retraso? (Mostrar tarjeta)
Muy seguro de un retraso en la salida 1 2 3 4 5 Muy seguro de una salida puntual
P29. ¿Qué grado de satisfacción le produce este retraso en su vuelo? (Mostrar tarjeta)
Ninguna satisfacción 1 2 3 4 5 Mucha satisfacción
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
P30. El personal del aeropuerto, ¿le ha comunicado la/s causa/s del retraso?
Sí
⇒
Pase a la pregunta siguiente P31.
No
⇒
Pase a la pregunta P32.
P31. ¿Qué causa/s han sido comunicadas?..............................................................................
P32-37. ¿Cuál piensa usted que es la causa que ha motivado el retraso de su vuelo?
P32. Problemas mecánicos
P33. Prácticas de empleados/personal de vuelo
P34. Retraso del vuelo anterior
P35. Problemas generados por otros pasajeros
P36. Mal tiempo
P37. Otras..................................................................................
P100. NS/NC
P38. La causa que acaba de mencionar, ¿piensa que se produce habitualmente o por el contrario no es
frecuente que ocurra?(Mostrar tarjeta)
.........................................................................................
Nada
habitual
1
Poco
habitual
2
Normal
Habitual
3
4
Muy
habitual
5
P39. ¿Piensa que la causa señalada por usted con anterioridad estaba dentro del control de la compañía
aérea, es decir, que ésta podía haber hecho algo para evitarla? (Mostrar tarjeta)
No se podía controlar 1 2 3 4 5
Se podía controlar
P40. ¿Cree usted que la compañía aérea podía haber hecho algo para acortar el retraso y no lo ha
hecho? (Mostrar tarjeta)
No pudo hacer nada 1 2 3 4 5 Existían acciones a tomar
P41-45. A continuación le ofrecemos un conjunto de afirmaciones sobre el retraso de su vuelo.
Conteste por favor, cuál es su grado de acuerdo o desacuerdo con las mismas:
P41. Es importante para mí que mi vuelo
llegue a la hora prevista a su destino
P42. Estoy enfadado/a con la compañía por
el retraso en el vuelo
P43. Me siento decepcionado/a por este
retraso
P44. Creo que voy a quejarme a la
compañía por el retraso
P45. No volvería a volar con esta
compañía si se presentara la ocasión
Nada importante
1
2
3
4
5
Muy importante
Nada enfadado/a
1
2
3
4
5
Muy enfadado/a
Nada
decepcionado/a
Nada inclinado/a
1
2
3
4
5
Muy decepcionado/a
1
2
3
4
5
Muy inclinado/a
No volaría de
nuevo
1
2
3
4
5
Volaría de nuevo
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
P46. En caso de que tenga intención seria de quejarse, ¿conoce cuáles son los organismos o
instituciones a los que tiene que acudir para formalizar su queja?
Sí
No
⇒ ¿Cuáles son?...................................................................................................
P47. ¿Viaja solo/a?
Sí
No
P48-52. ¿Cuál es el motivo de su viaje?
P48. Negocios
P49. Vacaciones
P50. Estudios
P51. Visitar amigos/familiares
P52. Otros:....................................................
P53. ¿Considera que el aeropuerto ofrece suficientes entretenimientos cuando se tiene que esperar?
(Mostrar tarjeta)
No son suficientes
1 2 3 4 5
Sí son suficientes
P54-57. ¿Cuántas veces viajó en avión el año pasado?
P54. Entre 0 y 1
P55. Entre 2 y 3
P56. Entre 4 y 5
P57. Más de 5
P100. NS/NC
DATOS DE CLASIFICACIÓN
P58. Sexo:
Hombre
Mujer
P59-62. Edad:
P59. Entre 18-35
P60. Entre 36-53
P61. Entre 54-71
P62. Más de 71
P63. Hora inicial de salida del vuelo:
P64. ¿Cuál es su lugar de residencia habitual?........................................................................
La entrevista ha terminado. Muchas gracias por su colaboración y el tiempo que nos ha concedido.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
FECHA:
HORA:
OBSERVACIONES:
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Valencia, 31 de octubre de 2002
Estimado/a señor/a,
La Unión de Consumidores de la Comunidad Valenciana y la Universidad de Alicante están
realizando un estudio sobre el trato que dan las entidades bancarias a los clientes que sufren algún
tipo de problema con las mismas. El objetivo final del estudio es elaborar un informe sobre el estado
de la cuestión en nuestra Comunidad que sirva como guía para emprender acciones tendentes a
mejorar la calidad de la relación entre las entidades bancarias y los clientes, y que permita
defender mejor los derechos de los consumidores. Para ello, se ha utilizado la base de datos de
los individuos socios de la UCE con la intención de que participen en este importante proyecto.
Siendo usted socio/a de esta organización, le quedaríamos altamente agradecidos si
accediese a colaborar en este estudio y completase el cuestionario que se adjunta. La
información que le vamos a solicitar se refiere a sus experiencias como cliente de las entidades
bancarias, y a sus sentimientos y su comportamiento en caso de que haya tenido algún problema con
las mismas. Por supuesto, todos los datos que facilite serán tratados con absoluta confidencialidad,
de acuerdo con las leyes vigentes, y su identidad no será revelada en ningún informe que se derive
de los resultados de esta investigación.
De nuevo, nos gustaría animarle a participar en este proyecto, dado que de sus respuestas
depende el éxito de este trabajo. Para agradecerle su colaboración, sortearemos entre todas las
respuestas recibidas un práctico y moderno ordenador de mano PalmTM, serie m100, con el que
podrá organizar sus actividades diarias ayudándose de la tecnología más avanzada.
Muchas gracias por prestarnos su valioso tiempo para leer esta carta y completar el
cuestionario. Una vez completado, tan sólo tiene que introducirlo en el sobre con franqueo pagado
que se le ha facilitado y enviarlo por correo, sin ningún coste para usted, antes del 30 de
noviembre.
Sinceramente,
Ana Isabel Zarzuela Luna
Presidenta
UCE-Comunidad Valenciana
96 373 7161 / 96 373 7109
Ana Belén Casado Díaz
Profesora Titular de Escuela Universitaria
Universidad de Alicante
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
CUESTIONARIO
POR FAVOR, LEA ATENTAMENTE LAS SIGUIENTES INSTRUCCIONES PARA RESPONDER EL CUESTIONARIO Y, A
CONTINUACIÓN, COMPLETE EL MISMO SIGUIENDO EL ORDEN ESTABLECIDO
A) EN LAS PREGUNTAS QUE TENGAN UNA CASILLA PARA RESPONDER, POR FAVOR, INDIQUE CON UNA “X” SU
RESPUESTA. EN CASO DE QUE COMETA UN ERROR, TACHE LA RESPUESTA EQUIVOCADA Y PONGA UNA “X”
EN LA CASILLA CORRECTA.
B) EN LAS PREGUNTAS QUE DEJEN UN ESPACIO EN BLANCO PARA RESPONDER, ESCRIBA LA RESPUESTA QUE
USTED CREA ADECUADA EN LETRAS MAYÚSCULAS.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
1
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
1. ¿Ha tenido alguna vez una experiencia insatisfactoria o problema que recuerde especialmente con una
entidad u oficina bancaria?
 Sí
 No (Si no ha tenido NUNCA ningún problema, por favor, pase directamente a la última página del
cuestionario, rellene los datos de clasificación y envíelo igualmente).
2. En el momento en que ocurrió dicho problema, esa oficina bancaria ¿era su oficina principal, con la que
realizaba la mayoría de sus operaciones?
 Sí
 No
3. Por favor, escriba el nombre de la entidad bancaria a la que pertenece (pertenecía) esa oficina en el espacio
siguiente:
4. ¿Podría indicar los motivos que le han llevado (llevaron) a ser cliente de esa oficina?
5. En la siguiente cuestión le preguntamos sobre la frecuencia con la que usted tiene (tenía) contacto con su
oficina. Por contacto entendemos comunicarse personal o telefónicamente con algún empleado de la misma o
acceder a sus cuentas a través de Internet, pero no se incluye el uso de los cajeros automáticos. Así pues,
¿con qué frecuencia tiene (tenía) contacto con esa entidad/oficina a lo largo del mes?
 menos de una vez al mes
 una vez al mes
 dos veces al mes
 tres veces al mes
 una vez a la semana
 dos o más veces a la semana
6. Aproximadamente, ¿cuántos años es (fue) usted cliente de esa oficina?
año(s) (Si es menos de 1 año, escriba "< 1").
7.
A continuación, indique con una “X” su grado de acuerdo o desacuerdo con la siguiente afirmación (siendo
1=totalmente en desacuerdo, 2=bastante en desacuerdo, 3=ni de acuerdo ni en desacuerdo, 4=bastante de
acuerdo, y 5=totalmente de acuerdo):
“Antes de que ocurriera el problema, me consideraba un/a cliente leal a esa oficina bancaria”
Totalmente en desacuerdo :____: ____: ____: ____: ____: Totalmente de acuerdo
1
2
3
4
5
8. ¿Qué servicios/productos utiliza (utilizaba) usted en dicha oficina bancaria? (puede marcar más de una
opción).
 Domiciliación de recibos
 Domiciliación de nómina
 Tarjeta(s)
 Banca por Internet
 Préstamo hipotecario
 Préstamo personal
 Plan de pensiones/jubilación
 Compra/venta de acciones
 Fondos de inversión
 Depósitos a plazo
 Transferencias/Traspasos
 Otros (Especificar cuál(es):
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
)
2
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
9.
¿Trabaja (trabajaba) usted con otras entidades bancarias?.
 Sí
 No
En cualquiera de los dos casos, ¿podría indicar los motivos?
A CONTINUACIÓN, NOS VAMOS A CENTRAR EN EL PROBLEMA O EXPERIENCIA INSATISFACTORIA QUE SUFRIÓ.
POR FAVOR, CONTESTE EN EL ESPACIO QUE TIENE A CONTINUACIÓN:
10. ¿Cuándo ocurrió dicho problema? (si es posible, indique un año o fecha aproximada).
11. Exactamente, ¿qué ocurrió?.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
3
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
LAS SIGUIENTES CUESTIONES SE REFIEREN JUSTO AL MOMENTO EN QUE LE SUCEDIÓ EL PROBLEMA QUE
ACABA DE RELATAR. POR FAVOR, PONGA UNA “X” ENCIMA DE LA LÍNEA/NÚMERO QUE SE CORRESPONDA EN
MAYOR MEDIDA CON LOS SENTIMIENTOS QUE LE PROVOCÓ AQUEL PROBLEMA.
12. Basándose en toda su experiencia con las entidades bancarias, ¿cómo describiría usted el problema sufrido?
Nada importante :____: ____: ____: ____: ____: Muy importante
1
2
3
4
5
Nada serio :____: ____: ____: ____: ____: Muy serio
1
2
3
4
5
13. En ese momento, CENTRÁNDOSE SÓLO EN EL PROBLEMA, ¿cómo se sintió con respecto a la oficina?
Encantado/a :____: ____: ____: ____: ____: Disgustado/a
1
2
3
4
5
Satisfecho/a :____: ____: ____: ____: ____: Insatisfecho/a
1
2
3
4
5
Contento/a :____: ____: ____: ____: ____: Descontento/a
1
2
3
4
5
En una escala de 1 a 5, siendo 1=nada, 2=algo, 3=ni mucho ni poco, 4=bastante, y 5=mucho, responda a la
siguiente cuestión:
14. En ese momento, CENTRÁNDOSE SÓLO EN EL PROBLEMA, ¿hasta qué punto se sintió:
a. Enfadado/a?
b. Molesto/a?
c. Impotente?
d. Frustrado/a?
e. Irritado/a?
f. Engañado/a?
nada :____: ____: ____: ____: ____: mucho
1
2
3
4
5
nada :____: ____: ____: ____: ____: mucho
1
2
3
4
5
nada :____: ____: ____: ____: ____: mucho
1
2
3
4
5
nada :____: ____: ____: ____: ____: mucho
1
2
3
4
5
nada :____: ____: ____: ____: ____: mucho
1
2
3
4
5
nada :____: ____: ____: ____: ____: mucho
1
2
3
4
5
En una escala de 1 a 5, siendo 1=nada probable, 2=poco probable, 3=ni probable ni improbable, 4=bastante
probable, y 5=muy probable, responda a las siguientes cuestiones:
15. ¿Cuál es la probabilidad de que un problema similar ocurra de nuevo en esa oficina?
Nada probable :____: ____: ____: ____: ____: Muy probable
1
2
3
4
5
16. ¿Podría haber impedido la oficina que dicho problema ocurriera?
Nada probable :____: ____: ____: ____: ____: Muy probable
1
2
3
4
5
En una escala de 1 a 5, siendo 1=muy poco/a, 2=poco/a, 3=ni poco ni mucho, 4=bastante, y 5=mucho/a,
responda a las siguientes cuestiones:
17. ¿Cuánta culpa cree usted que tenía la oficina sobre el problema ocurrido?
Muy poca :____: ____: ____: ____: ____: Mucha
1
2
3
4
5
18. ¿Cuánto control cree usted que tenía la oficina sobre el problema ocurrido?
Muy poco :____: ____: ____: ____: ____: Mucho
1
2
3
4
5
19. ¿Había tenido usted algún problema con esa oficina antes del problema que acaba de describir?
 Sí
 No
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
4
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
LAS SIGUIENTES AFIRMACIONES SE CENTRAN EN EL TRATO QUE RECIBIÓ POR PARTE DE LA OFICINA DESPUÉS
DE HABER TENIDO EL PROBLEMA. POR FAVOR, PONGA UNA “X” ENCIMA DE LA LÍNEA/NÚMERO QUE SE
CORRESPONDA EN MAYOR MEDIDA CON SUS SENTIMIENTOS, SEGÚN LA ESCALA EMPLEADA:
En una escala de 1 a 5, siendo 1=mucho peor de lo esperado, 2=peor de lo esperado, 3=como esperaba,
4=mejor de lo esperado, y 5=mucho mejor de lo esperado:
20. La respuesta general a mi problema fue:
:______: ______: ______: ______: ______:
1
2
3
4
5
En una escala de 1 a 5, siendo 1=totalmente de acuerdo, 2=bastante de acuerdo, 3=ni de acuerdo ni en
desacuerdo, 4=bastante en desacuerdo, y 5=totalmente en desacuerdo:
21. El resultado que recibí fue justo
22. El tiempo empleado en resolver mi problema fue adecuado
23. Los empleados y/o el director estaban suficientemente
preocupados por mi problema
24. Obtuve lo que me merecía
25. Los empleados y/o el director me trataron con la cortesía
y respeto que merecía
26. Las explicaciones que recibí fueron adecuadas
27. Los empleados y/o el director me ofrecieron disculpas
28. Los empleados y/o el director se dieron cuenta de mi
problema sin que yo tuviese que quejarme
29. Los empleados y/o el director mostraron una flexibilidad
adecuada al tratar mi problema
:______: ______: ______: ______: ______:
1
2
3
4
5
:______: ______: ______: ______: ______:
1
2
3
4
5
:______: ______: ______: ______: ______:
1
2
3
4
5
:______: ______: ______: ______: ______:
1
2
3
4
5
:______: ______: ______: ______: ______:
1
2
3
4
5
:______: ______: ______: ______: ______:
1
2
3
4
5
:______: ______: ______: ______: ______:
1
2
3
4
5
:______: ______: ______: ______: ______:
1
2
3
4
5
:______: ______: ______: ______: ______:
1
2
3
4
5
30. En ese momento, CENTRÁNDOSE EN EL TRATO RECIBIDO EN LA OFICINA, ¿cómo se sintió con respecto
a la oficina?
Encantado/a :____: ____: ____: ____: ____: Disgustado/a
1
2
3
4
5
Satisfecho/a :____: ____: ____: ____: ____: Insatisfecho/a
1
2
3
4
5
Contento/a :____: ____: ____: ____: ____: Descontento/a
1
2
3
4
5
En una escala de 1 a 5, siendo 1=nada, 2=algo, 3=ni lo uno ni lo otro, 4=bastante, y 5=mucho, responda a la
siguiente cuestión:
31. En ese momento, CENTRÁNDOSE EN EL TRATO RECIBIDO EN LA OFICINA, ¿hasta qué punto se sintió:
a. Enfadado/a?
b. Molesto/a?
c. Impotente?
d. Frustrado/a?
e. Irritado/a?
f. Engañado/a?
nada :____: ____: ____: ____: ____: mucho
1
2
3
4
5
nada :____: ____: ____: ____: ____: mucho
1
2
3
4
5
nada :____: ____: ____: ____: ____: mucho
1
2
3
4
5
nada :____: ____: ____: ____: ____: mucho
1
2
3
4
5
nada :____: ____: ____: ____: ____: mucho
1
2
3
4
5
nada :____: ____: ____: ____: ____: mucho
1
2
3
4
5
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
5
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
ANTE EL PROBLEMA SUFRIDO Y EL TRATO QUE RECIBIÓ, ¿QUÉ CREE USTED QUE DEBERÍA HABER HECHO LA
OFICINA?.
ANTE EL PROBLEMA SUFRIDO Y EL TRATO QUE RECIBIÓ, ¿CÓMO ACTUÓ USTED?. POR FAVOR, PONGA UNA “X”
ENCIMA DE LA LÍNEA QUE SE CORRESPONDA CON LA(S) ACCION(ES) QUE EMPRENDIÓ. PUEDE MARCAR MÁS DE
UNA OPCIÓN:
32.
33.
34.
35.
36.
37.
38.
39.
40.
No hizo nada
Comunicó su insatisfacción al empleado
Se quejó al director de la oficina
Rellenó una hoja de reclamaciones
Acudió a la UCE para tramitar una reclamación
Se quejó a sus familiares y/o amigos
Decidió cambiar de oficina pero no de entidad
Decidió cambiar de oficina y de entidad bancaria
Pensó seriamente en cambiar de oficina y/o entidad
¿Por qué motivos no cambió de banco?:






 (EN CASO AFIRMATIVO PASE A LA PÁGINA SIGUIENTE)
 (EN CASO AFIRMATIVO PASE A LA PÁGINA SIGUIENTE)
 (EN CASO AFIRMATIVO, RESPONDA A CONTINUACIÓN:
ATENCIÓN: LAS SIGUIENTES DOS CUESTIONES DE ESTA PÁGINA SÓLO DEBE CONTESTARLAS SI NO HA
MARCADO LA(S) OPCION(ES) 38 y/o 39. EN CASO DE QUE HAYA MARCADO LA(S) OPCION(ES) 38 y/o 39 PASE A LA
PÁGINA SIGUIENTE.
41. EN LA ACTUALIDAD, tras todas sus experiencias con la oficina (incluidos el problema sufrido y el trato
recibido), ¿cómo se siente usted en general con respecto a la entidad/oficina?.
Encantado/a :____: ____: ____: ____: ____: Disgustado/a
1
2
3
4
5
Satisfecho/a :____: ____: ____: ____: ____: Insatisfecho/a
1
2
3
4
5
Contento/a :____: ____: ____: ____: ____: Descontento/a
1
2
3
4
5
En una escala de 1 a 5, siendo 1=muy probable, 2=bastante probable, 3=ni probable ni improbable, 4=poco
probable, y 5= nada probable, responda a la siguiente cuestión:
42. EN LA ACTUALIDAD, tras todas sus experiencias con la entidad/oficina (incluidos el problema sufrido y el
trato recibido),¿cuál es la probabilidad de que usted:
a. Comente cosas positivas sobre esa
oficina a otras personas?
muy
probable
:____: ____: ____: ____: ____: nada
1
2
3
4
5
probable
b. Recomiende esa oficina a otras personas?
muy
probable
:____: ____: ____: ____: ____: nada
1
2
3
4
5
probable
c. Anime a sus familiares y amigos a trabajar
con esa oficina?
muy
probable
:____: ____: ____: ____: ____: nada
1
2
3
4
5
probable
d. Siga trabajando con esa oficina en el futuro?
muy
probable
:____: ____: ____: ____: ____: nada
1
2
3
4
5
probable
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
6
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
DATOS DE CLASIFICACIÓN: PARA FINALIZAR, LE AGRADECERÍAMOS QUE NOS SUMINISTRASE INFORMACIÓN
SOBRE USTED, QUE SERÁ UTILIZADA EXCLUSIVAMENTE PARA CLASIFICAR LAS RESPUESTAS OBTENIDAS. POR
FAVOR, PONGA UNA “X” ENCIMA DE LA LÍNEA QUE SE CORRESPONDA EN MAYOR MEDIDA CON SU SITUACIÓN.
43. ¿Sexo?
 Hombre
 Mujer
44. ¿Edad?
________ años
45. ¿Estado civil?
 Casado/a
 Soltero/a
 Divorciado/a
 Separado/a
 Viudo/a
 Otro (Indicar cuál:
)
 Diplomatura
 Licenciatura
 Otro (Indicar cuál:
)
46. ¿Profesión?
47. ¿Nivel educativo?
 Sin estudios
 Estudios primarios
 Estudios secundarios (F.P., B.U.P., C.O.U.)
48. ¿Ingresos mensuales medios de su hogar, aproximadamente? (netos)
 Menos de 100.000 pesetas (601 euros)
 Entre 300.001 y 400.000 pesetas (1.804-2.404 euros)
 Entre 100.001 y 200.000 pesetas (601-1.202 euros)
 Entre 400.001 y 500.000 pesetas (2.405-3.005 euros)
 Entre 200.001 y 300.000 pesetas (1.203-1.803 euros)  Más de 500.000 pesetas (más de 3.005 euros)
49. ¿Número de individuos que conviven en su hogar, incluido/a usted?
 Uno
 Dos
 Tres
 Cuatro
 Cinco
 Más de cinco
50. ¿Cuánto tiempo hace, aproximadamente, que es usted socio/a de la Unión de Consumidores?
año(s) (Si es menos de 1 año, escriba "< 1").
51. Durante todo el tiempo que ha sido usted socio/a de la Unión de Consumidores, aproximadamente ¿cuántas
veces ha solicitado sus servicios para efectuar alguna RECLAMACIÓN?
 Nunca
 Una vez
 Dos veces
 Tres veces
 Cuatro veces
 Más de cuatro veces
52. Durante todo el tiempo que ha sido usted socio/a de la Unión de Consumidores, aproximadamente ¿cuántas
veces ha solicitado sus servicios para efectuar alguna CONSULTA?
 Nunca
 Entre 1 y 5 veces
 Entre 6 y 10 veces
 Entre 11 y 15 veces
 Entre 16 y 20 veces
 Más de 20 veces
53. ¿Ha trabajado usted alguna vez en un banco o caja de ahorros?
 No
 Sí (Indicar puesto desempeñado:
)
ESTO ES TODO. POR FAVOR, INTRODUZCA EL CUESTIONARIO EN EL SOBRE CON FRANQUEO PAGADO QUE LE
HEMOS SUMINISTRADO Y ENVÍELO LO ANTES POSIBLE. SI DESEA PARTICIPAR EN EL SORTEO DEL ORDENADOR
TM
DE MANO PALM SERIE m100 ESCRIBA SU NOMBRE COMPLETO O SU NÚMERO DE SOCIO EN EL ESPACIO
SIGUIENTE:
¡¡MUCHÍSIMAS GRACIAS POR SU TIEMPO Y AYUDA!!
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
7
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
QUESTIONNAIRE
PLEASE, READ THE QUESTIONS CAREFULLY AND FOLLOW THE SEQUENCE OF THIS BOOKLET TO ANSWER THEM
A) PLEASE, PLACE AN “X” FOR ANSWERING. IN CASE YOU MISTAKE, CROSS OUT THE WRONG ANWER AND WRITE
AN “X” IN THE RIGHT PLACE.
B) IN THE QUESTIONS WITH A BLANK SPACE FOR ANSWERING, WRITE DOWN THE ANSWER YOU CONSIDER
APPROPRIATE IN CAPITAL LETTERS.
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
1
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
1.
Have you experienced any problem that you specially remember with a bank or branch office?
___ Yes
___ No (If no, please skip to Question 44, complete classification data and send the questionnaire).
2.
In the moment that problem happened, was that branch office your main office, the one with which you did
most of your business?
___ Yes
___ No
3.
Please, write the name of the bank in the space below:
_________________________________________________________________________________________
4.
Could you indicate why do (did) you bank at this bank?
5.
The next question is about the frequency of your contact with the branch office. We understand contact as the
communication with an employee, personally or by telephone, and/or Internet banking, we do not refer to cash
dispensers. So, how often do you have contact with your bank in a month?
____ less than once a month
____ once a month
____ twice a month
6.
____ three times a month
____ once a week
____ two or more times a week
Approximately, how many years have (had) you been a customer of this branch office?
_________ year(s) (If less than one year, write "< 1").
7.
Now, place an “X” your degree of agreement or disagreement with the next sentence: “Before the problem
occurred, I considered myself to be a loyal customer of that bank office”
Strongly disagree :____: ____: ____: ____: ____: Strongly agree
1
8.
2
3
4
5
What services do (did) you use of this branch office? (prod1-prod12) (sorry, I need some help with the
translation, next week I will give it to you)
____ Domiciliación de recibos (bills)
____ Domiciliación de nómina (salary)
____ Credit card(s) (cards)
____ Internet banking (internet)
____ Préstamo hipotecario (mortgage)
____ Préstamo personal (loan)
____ Plan de pensiones/jubilación (pension)
____ Compra/venta de acciones (share)
____ Fondos de inversión (fund)
____ Depósitos a plazo (deposit)
____ Transferencias/Traspasos (transfer)
____ Otros (another)
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
2
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
9.
Do (did) you use several banks?.
_____ Yes
_____ No
Whatever your answer is, could you indicate the reasons?
NOW, THINK ABOUT THE PROBLEM OR DISSATISFACTORY EXPERIENCE YOU HAD. PLEASE, ANSWER IN THE
SPACE BELOW:
10. When did that problem happened? (if possible, indicate a year or concrete date)
11. Exactly, what happened?
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
3
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
THE FOLLOWING QUESTIONS REFER TO THE MOMENT YOU SUFFERED THE PROBLEM YOU HAVE JUST
DESCRIBED. PLEASE, PLACE AN “X” ON THE LINE/NUMBER THAT MOST CLOSELY CORRESPONDS TO HOW YOU
FELT ABOUT THE PROBLEM.
12. Based on all of your experiences with banks, how did you view this problem?
Not at all important :____: ____: ____: ____: ____: Very important
1
2
3
4
5
Not at all severe :____: ____: ____: ____: ____: Very severe
1
2
3
4
5
13. On that moment, thinking only about the problem, how did you feel about the bank/branch office?
Pleased :____: ____: ____: ____: ____: Displeased
1
2
3
4
5
Satisfied :____: ____: ____: ____: ____: Dissatisfied
1
2
3
4
5
Happy :____: ____: ____: ____: ____: Unhappy
1
2
3
4
5
On a five-point scale, being 1=not at all and 5=very, please answer the next questions:
14. On that moment, thinking only about the problem, to what extent did you feel yourself:
a. Angry?
Not at all :____: ____: ____: ____: ____: very
b. Annoyed?
Not at all :____: ____: ____: ____: ____: very
c. Powerless?
Not at all :____: ____: ____: ____: ____: very
d. Frustrated?
Not at all :____: ____: ____: ____: ____: very
e. Irritated?
Not at all :____: ____: ____: ____: ____: very
f. Deceived?
Not at all :____: ____: ____: ____: ____: very
1
2
1
3
2
1
3
2
1
3
2
1
3
2
1
3
2
3
4
4
4
4
4
4
5
5
5
5
5
5
15. From your point of view, how likely is it that a similar problem would occur again in that branch office?
Not at all likely :____: ____: ____: ____: ____: Very likely
1
2
3
4
5
16. From your point of view, could the branch office have prevented this problem from occurring?
Not at all likely :____: ____: ____: ____: ____: Very likely
1
2
3
4
5
17. From your point of view, how much blame had the branch office over this problem?
Very little :____: ____: ____: ____: ____: Very much
1
2
3
4
5
18. From your point of view, how much control did the branch office have over this problem?
Very little :____: ____: ____: ____: ____: Very much
1
2
3
4
5
19. Had you ever had any problem with this branch office before this problem?
___ Yes
___ No
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
4
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
THE NEXT SET OF QUESTIONS IS ABOUT YOUR EVALUATIONS OF THE BANK’S HANDLING OF THE PROBLEM
AFTER YOU HAD IT. PLEASE, PLACE AN “X” ON THE LINE/NUMBER THAT MOST CLOSELY CORRESPONDS TO HOW
YOU FELT, IN LINE WITH THE SCALE EMPLOYED.
On a five-point scale, being 1=much worse than expected and 5=much better than expected:
20. The bank’s overall response to my
problem was…………………………..………………:______: ______: ______: ______: ______:
1
2
3
4
5
On a five-point scale, being 1=strongly agree and 5=strongly disagree:
21. The outcome I received was fair………………….….:______: ______: ______: ______: ______:
1
2
3
4
5
22. The length of time taken to resolve my problem
was adequate……………… ..………………………...:______: ______: ______: ______: ______:
1
2
3
4
5
23. The employees and/or the director were
appropriately concerned about my problem…..….…:______: ______: ______: ______: ______:
1
2
3
4
5
24. I got what I deserved………………………………….:______: ______: ______: ______: ______:
1
2
3
4
5
25. The employees and/or the director gave
me the courtesy and respect I was due…………… :______: ______: ______: ______: ______:
1
2
3
4
5
26. The explanations I was given were adequate ……:______: ______: ______: ______: ______:
1
2
3
4
5
27. The employees and/or the director apologized
for my problem………………………….………………:______: ______: ______: ______: ______:
1
2
3
4
5
28. The employees and/or the director acknowledged
my problem without me having to complain………….:______: ______: ______: ______: ______:
1
2
3
4
5
29. The bank/branch office showed adequate flexibility
in dealing with my problem……………………………:______: ______: ______: ______: ______:
1
2
3
4
5
30. On that moment, thinking about the bank’s handling of the problem, how did you feel about the branch office?
Pleased :____: ____: ____: ____: ____: Displeased
1
2
3
4
5
Satisfied :____: ____: ____: ____: ____: Dissatisfied
1
2
3
4
5
Happy :____: ____: ____: ____: ____: Unhappy
1
2
3
4
5
On a five-point scale, being 1=not at all and 5=very, please answer the next question:
31. On that moment, thinking about the bank’s handling of the problem, to what extent did you feel yourself:
a. Angry?
Not at all :____: ____: ____: ____: ____: very
b. Annoyed?
Not at all :____: ____: ____: ____: ____: very
c. Powerless?
Not at all :____: ____: ____: ____: ____: very
d. Frustrated?
Not at all :____: ____: ____: ____: ____: very
e. Irritated?
Not at all :____: ____: ____: ____: ____: very
f. Deceived?
Not at all :____: ____: ____: ____: ____: very
1
1
1
1
1
1
2
2
2
2
2
2
3
3
3
3
3
3
4
4
4
4
4
4
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
5
5
5
5
5
5
5
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
THINKING ABOUT THE PROBLEM AND THE BANK’S HANDLING OF IT, HOW DO YOU THINK THE BRANCH OFFICE
SHOULD HAVE HANDLED THE PROBLEM?
THINKING ABOUT THE PROBLEM AND THE BANK’S HANDLING OF IT, HOW DID YOU REACT? PLEASE, PLACE AN
“X” ON THE LINE THAT MOST CLOSELY CORRESPONDS TO THE ACTION(S) YOU INITIATED:
32.
You did not do anything……………………………….:______:
33.
Communicated your dissatisfaction to the employee.:______:
34.
Complained to the director……….…………………....:______:
35.
Filled out a complaining card………
36.
Turned to UCE to process a complaint…………….…:______:
37.
Complained to relatives and/or friends .….…………..:______:
38.
Changed branch office but not bank…………….……:______: (If yes, skip to Question 44)
39.
Changed branch office and bank……………………...:______: (If yes, skip to Question 44)
40.
Thought seriously changing bank and/or branch office……. :______:
…………….:______:
(If yes, in the space below write down why did you not change bank?:
41. Nowadays, thinking about all your experiences with the branch office (including the problem suffered and the
bank’s handling of it), how do you feel overall about this branch office?.
Pleased :____: ____: ____: ____: ____: Displeased
1
2
3
4
5
Satisfied :____: ____: ____: ____: ____: Dissatisfied
1
2
3
4
5
Happy :____: ____: ____: ____: ____: Unhappy
1
2
3
4
5
On a five-point scale, being 1=extremely likely and 5=not at all likely, please answer the next question:
42. Nowadays, thinking about all your experiences with the branch office (including the problem suffered and the
bank’s handling of it), how likely is that you:
a. Say positive things about your
branch office to others?
extremely :____: ____: ____: ____: ____: Not at all
likely
1
2
3
4
5
likely
b. Recommend that branch
office to others?
extremely :____: ____: ____: ____: ____: Not at all
likely
1
2
3
4
5
likely
c. Encourage friends and relatives to do
business with that branch office?
extremely :____: ____: ____: ____: ____: Not at all
likely
1
2
3
4
5
likely
d. Do more business with that
branch office in the future?
extremely :____: ____: ____: ____: ____: Not at all
likely
1
2
3
4
5
likely
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
6
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
FINALLY, PLEASE PROVIDE THE FOLLOWING INFORMATION ABOUT YOURSELF. THIS INFORMATION IS USED FOR
CLASSIFICATION PURPOSES ONLY.
43. Gender?
____ Male
____ Female
44. Age?
________ years
45. Marital status?
____ Married
____ Single
____ Divorced
____ Separated
____ Widowed
____ Another (Say which: ____________________)
46. Job?
____________________________________________________________________________________
47. Educational level?
____ Without studies
____ Primary level
____ Secundary level (F.P., B.U.P., C.O.U.)
____ 3-years degree
____ 5 or 4-years degree
____ Another (Say which: ____________________)
48. Household monthly net income?
____
____
____
____
____
____
Less than 100.000 pesetas (601 euros)
Between 100.001 and 200.000 pesetas (601-1.202 euros)
Between 200.001 and 300.000 pesetas (1.203-1.803 euros)
Between 300.001 and 400.000 pesetas (1.804-2.404 euros)
Between 400.001 and 500.000 pesetas (2.405-3.005 euros)
More than 500.000 pesetas (3.005 euros)
49. Household size (individuals living at home)?
____
____
one
two
____
____
three
four
____ five
____ more than five
50. Approximately, how many years have you been UCE’s member?
_________ year(s) (If less than one year, write "< 1").
51. During this time, how many times have you use UCE’s services to make a complaint?
____
____
never
once
____
____
twice
three times
____
____
four times
more than four times
52. During this time, how many times have you use UCE’s services to ask for information?
____
____
never
once
____
____
twice
three times
____ four times
____ more than four times
53. Have you ever worked at a bank?
____
____
No
Yes (Say job position: ___________________________________________________)
THAT IS EVERYTHING. PLEASE, RETURN THE QUESTIONNAIRE IN THE PRE-ADDRESSED, POSTAGEPAID ENVELOPE PROVIDED AS SOON AS POSSIBLE. IF YOU WANT TO PARTICIPATE IN THE DRAW,
PLEASE WRITE DOWN YOUR FULL NAME OR NUMBER OF MEMBERSHIP IN THE FOLLOWING LINE:
¡¡THANK YOU VERY MUCH FOR YOUR TIME AND HELP!!
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005
7
Customers' responses to service failures. Empirical studies on private, voice and third-party responses. Ana Belén Casado Díaz.
Volver al Indice/Tornar a l'índex
Valencia, 15 de noviembre de 2002
Estimado/a señor/a,
Hace dos semanas le enviamos un cuestionario sobre el trato que dan las entidades
bancarias a los clientes que sufren algún tipo de problema con las mismas. Como ya le indicamos en
la carta de presentación, la Unión de Consumidores de la Comunidad Valenciana y la Universidad de
Alicante están realizando un estudio sobre este tema con el objetivo de elaborar un informe que sirva
como guía para mejorar la calidad de la relación entre las entidades bancarias y los clientes, y que
permita defender mejor los derechos de los consumidores.
Si ya nos ha enviado el cuestionario reciba nuestro más sincero agradecimiento. Si aún no
lo hecho, por favor hágalo cuanto antes. En caso de que hubiese extraviado el envío original, le
adjuntamos nuevamente el cuestionario y el sobre con franqueo pagado para que nos lo envíe sin
ningún coste para usted. Además, le recordamos que para agradecerle su colaboración,
sortearemos entre todas las respuestas recibidas un práctico y moderno ordenador de mano
PalmTM, serie m100, con el que podrá organizar sus actividades diarias ayudándose de la tecnología
más avanzada.
Muchas gracias por prestarnos su valioso tiempo para leer esta carta y completar el
cuestionario.
Sinceramente,
Ana Isabel Zarzuela Luna
Presidenta
UCE-Comunidad Valenciana
96 373 7161 / 96 373 7109
Ana Belén Casado Díaz
Profesora Titular de Escuela Universitaria
Universidad de Alicante
Tesis doctoral de la Universidad de Alicante. Tesi doctoral de la Universitat d'Alacant. 2005