MODELING CONSUMER RESPONSES TO NEGATIVE DISCONFIRMATION OF EXPECTATIONS: AN EMPIRICAL INVESTIGATION USING ITEM RESPONSE THEORY BASED MEASURES by JAGDIP SINGH, B. Tech. A DISSERTATION IN BUSINESS ADMINISTRATION Submitted to the Graduate Faculty of Texas Tech Univereity in Partial Fulfillment of the Requirements for the Degree of DOCTOR OF BUSINESS ADMINISTRATION ^ppro^^d Accepted August, 1965
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
MODELING CONSUMER RESPONSES TO NEGATIVE
DISCONFIRMATION OF EXPECTATIONS:
AN EMPIRICAL INVESTIGATION USING
ITEM RESPONSE THEORY
BASED MEASURES
by
JAGDIP SINGH, B. Tech.
A DISSERTATION
IN
BUSINESS ADMINISTRATION
Submitted to the Graduate Faculty of Texas Tech Univereity in
Partial Fulfillment of the Requirements for
the Degree of
DOCTOR OF BUSINESS ADMINISTRATION
^ppro^^d
Accepted
August, 1965
C.e>p, ^ - ACKNOWLEDGMENTS
A dissertation is a mosiac of contributions from
several people; without their cooperation and
unhesitating assistance the task would be impossible.
This acknowledgment commits to paper the writer's
gratitude and debt to all those who assisted in the
research.
The writer especially acknowledges the guidance and
encouragement of Dr. Roy D. Howell, Chairman of the
Committee. Dr. Howell remained a source of inspiration
not only during the dissertation process, but throughout
the doctoral program. To him the author owes a debt of
gratitude which can never be repaid.
The financial and moral support of Mr. Jan
Freiderich, Chief Executive Officer, Mr. Edward M.
Markham, Director of Management Information Systems, Axel
Hopp, and Kathy Komoll, all of Furrs Supermarkets
Incorporated, is sincerely appreciated. In particular,
the writer thanks Mr. Markham for his encouragement and
the many hours he patiently spent in providing valuable
managerial insights.
Special acknowledgments are also due to Dr. Danny
Bellenger, Dr. Robert Wilkes, Dr. James Wilcox and other
members of the marketing faculty. The author deeply
appreciates the valuable time that they unhesitatingly
ii
devoted in guiding this research.
A very special acknowledgment is due to the writer's
family, Neena and Tashi. Their love and affection made
this task much more meaningful and easier. It is to them
that this work is dedicated.
Finally, the author also wishes to recognize the
unfailing support and friendship of doctoral colleagues
especially, Phil Goodell and Gary Rhoads.
iii
TABLE OF CONTENTS
ACKNOWLEDGMENTS ii
LIST OF FIGURES vii
LIST OF TABLES ix
1. INTRODUCTION 1
Substantive Contribution 3
Methodological Contribution 6
The Choice of Service Industries 8
Outline of the Dissertation 9
2. A REVIEW OF THE CONCEPTUAL AND EMPIRICAL
RESEARCH 11
Conceptual Review 11
Empirical Review 30
3. THE CONCEPTUAL FOUNDATIONS OF A HOLISTIC
MODEL OF CCB 46
Conceptual Foundations of the Model. . . . 46
Partial Formalization of the Model . . . . 54
4. 0PERATI0NALI2ATI0N AND THE DEVELOPMENT OF
KEY HYPOTHESES 66
Operationalizations of Key Constructs. . . 68
Key Hypotheses 7d
iv
5. SURVEY PROCEDURES AND METHODS 92
The Two Phases of Research 92
Phase 1 93
Phase II 97
6. PHASE I RESULTS 105
Measurement Properties Using Traditional Methods 106
Structural Relationships Using Path Analysis 108
Measurement Properties Using IRT Based Procedures 110
Structural Relationships With
Reduced Scales 118
7. PHASE II RESULTS 137
Measurement Properties 137
Empirical Investigation of the Typology for the Predominant Predictor of CCB Intentions (Hypotheses H6-H8) . . . . 141
Empirical Investigation of the Framework for Predicting Specific CCB Responses (Hypotheses H14-H17). . . . 147
Process Model Versus Naive Model (Hypothesis H13) 155
Empirical Investigation of the Process Model (Hypotheses H4, H5 and H9-H12) 159
Empirical Investigation of Expectancy Value Judgments in the Four Industries (Hypothesis H18) 172
a. SUMMARY, IMPLICATIONS AND LIMITATIONS 197
An Overview of the Dissertation 197
Managerial Implications 207
Public Policy Implications 214
Theoretical Implications 216
Limitations 218
LIST OF REFERENCES 226
APPENDICES 237
vi
LIST OF FIGURES
2.1 Two Competing Conceptualizations of Consumer Complaint Behavior 43
2.2 A Process Model for Post-Purchase Phenomena . . 44
2.3 A Classification Schema for Consumer Complaints 45
3.1 A Holistic Model of Consumer Complaining Behavior 63
3.2 A Typology for the Nature of CCB Processes 64
3.3 A Model for Predicting Specific Consumer Complaint Actions 65
6.1 The Empirical Model Proposed to be Tested in Phase I 124
6.2 Eigenvalue Structure for the 35 Item Alienation Scale 125
6. 3 Eigenvalue Structure of the 82 Item Discontent Scale 126
6. 4 Path Analytical Diagram for Complete Scales 127
6. 5 Eigenvalue Structure for the Combined Pool of Discontent and Alienation Items . . . . 128
6.6 Scale Information Curves for Phase I Complete Scales 129
6.7 Scale Information Curves for Phase I
Shortened Scales 130
6.8 Plot of "b" Parameters from Two Studies . . . . 131
6.9 Scale Information Curves for Phase II Scales 132
6.10 Path Analytical Diagram Using Shortened Scales 133
vxi
6. 11 Path Analytical Diagram Based on Automotive Repair Data 134
6. 12 Path Analytical Diagram for the Discontent Group 135
6. 13 Path Analytical Diagram for the
Alienated Group 136
7. 1 A Naive Model of CCB Intentions 191
7. 2 A Process Model of CCB Intentions 192
7. 3 The Estimated Model of CCB Intentions for Grocery Data 193
7. 4 The Estimated Model of CCB Intentions for Automotive Repair Data 194
7. 5 The Estimated Model of CCB Intentions for Medical Care Data 195
7.6 The Estimated Model of CCB Intentions for Financial Data 196
viii
LIST OF TABLES
2.1 A Summary of Empirical Findings in the CCB Area 40
4.1 Operationalization of Key Constructs
in the Holistic Model 91
5. 1 Response Rates for Phase II Surveys 102
5.2 Chi-square Test for "No Problem" Respondents 103
6.1 Correlation Matrix for Discontent, Alienation and Attitude 122
6.2 "A" and "B" Parameters for the Discontent
and Alienation Scales 123
7. 1 Alpha Reliabilities of All Constructs 176
7.2 Partial Correlation Table for Grocery Data 177
7. 3 Partial Correlation Table for Automotive Repair Data 178
7.4 Partial Correlation Table for Medical Care Data 179
7.5 Partial Correlation Table for Financial
Data 180
7.6 Cell Means for "VOICE" Intentions 181
7.7 Cell Means for "W-O-M" Intentions 182
7.8 Cell Means for "FORMAL" Intentions 183 7.9 A Comparison of Naive and Process Models
for Grocery Data 184
7.10 A Comparison of the Naive and Process Models for Automotive Repair Data 185
7.11 A Comparison of Naive and Process Models for Medical Care Data 186
ix
7. 12 A Comparison of Naive and Process Models for Financial Data 187
7.13 Estimated Parameters for the Process Model: Maximum Likelihood Structural Parameters 188
7.14 Estimated Parameters for the Process Model: Standardized Measurement Parameters . . . . 189
7. 15 A Comparison of Expectancy Value Judgments Across the Four Industries 190
8.1 A Summary of the Various Hypotheses Tested in Phase 1 220
8.2 A Summary of the Various Hypothesis Tested in Phase II 221
8.3 Typical Verbatim Responses 223
CHAPTER 1
INTRODUCTION
Much of the research in consumer behavior is focused
on pre-purchase processes (e.g., information search,
beliefs, formation of attitudes, purchase preferences and
intentions). More recently, the importance of post-
purchase processes and their influence on consumer beha
vior in general and repurchase activity in particular has
been recognized. For example, the annual Conference on
Consumer Satisfaction and Dissatisfaction (CS/D) first
held in 1976, has now been broadened to include Consumer
Complaining Behavior (CCB) as well.
Post-purchase activity often involves a series of
steps in which consumers evaluate the perceived perfor
mance of a product against an expected level (or norm) of
performance and then act in a way influenced by the
resulting congruence or discrepancy (Gilly and Gelb 1982;
Woodruff et al. 1983; Bearden and Teel 1983; Oliver
1980). Several theoretical models have been proposed to
guide empirical investigation of the post-purchase
process (Andreasen 1977; Day 1984). In all models, the
overall feeling of satisfaction or dissatisfaction that
results from the post-purchase process is hypothesized to
affect repurchase beliefs, attitudes, intentions and
loyalty (Andreasen 1977; Engel and Blackwell 1982). From
a theoretical standpoint, the study of post-purchase
process provides increased understanding of the role of
prior experiences (previous satisfactions or dissatisfac
tions) in future purchase decisions. Additionally, it
affords the manager a richer understanding of the psycho
logical processes leading to brand loyalty over repeated
purchases.
The major objective of this dissertation is to
develop a theoretical framework for consumer dissatisfac
tion and complaint processes, and to empirically test a
portion of that framework. It is expected that a model
for the explanation and prediction of these post-purchase
processes which is supported by empirical observations
would provide a contribution to theoretical understanding
of consumer behavior as well as provide guidelines to
practicing managers for retaining the loyalty of their
customers. Further, empirical investigation is conducted
in four separate industries in order to test the validity
of the model in different situations and for a range of
products or services. Specifically, the proposed model
is tested for dissatisfactions involving grocery
shopping, automotive repair, medical care and financial
services. In order to delimit the scope and extent of
this dissertation, the specific substantive and methodo
logical contributions as well as the reasons behind the
choice of the four industries (for empirical investi
gation) are discussed.
Substantive Contribution
Research in the area of post-purchase processes has
followed two somewhat different directions. One is
widely referred to as the confirmation/disconfirmation of
expectations paradigm for explaining consumer satisfac
tion/dissatisfaction (CS/D). This paradigm asserts that:
Prior to purchase and use of a brand, the consumer forms expectations of its performance in a particular use situation. These expectations are predictions of the nature and level of performance the user will receive. After using the brand, the consumer compares perceived actual performance with expected performance. Confirmation results when the two performances match. A mis-match will cause a positive (perceived performance exceeds expectations) or a negative (perceived performance falls below expectations) disconfirmation. In turn, confirmation/disconfirmation leads to an emotional reaction called satisfaction/dissatisfaction (Woodruff et al. 1983).
Many theoretical frameworks have been suggested to
explain the process of confirmation/disconfirmation and
its relationship to CS/D. Early researchers in marketing
employed the cognitive dissonance theory suggesting that
consumers seek to reduce dissonance between expectations
and performance (Howard and Sheth 1969; Olson and Dover
1976). Lack of empirical support for this theory led
researchers to adopt the assimilation contrast theory as
a possible mechanism for the CS/D process (Anderson 1973;
Olshavsky and Miller 1972). Using the notion of latitude
of acceptance, this theory predicted an interaction
between the level of expectations and degree of discon-
firmation. These predicted effects, however, have had
little empirical support (Oliver 1980). Current research
tends to postulate independent effects of expectation
level and disconfirmation on the level of consumer satis
faction. Such a conceptualization is grounded in the
theories of comparison level (Latour and Peat 1979) and
adaptation level (Oliver 1980a). Each of the above
frameworks assumes that confirmation or positive discon-
firmation of the performance increases, while negative
disconfirmation of the performance decreases, the likeli
hood of brand repurchase.
The second stream of research, often referred to as
the Consumer Complaining Behavior (CCB), has sought to
understand, explain and predict actions following
unsatisfactory purchase or use experiences. Though much
of this work is descriptive, several studies show that
disconfirmation of expectations triggers a complaint
process (Bearden and Teel 1983; Richins 1982; Jacoby and
Jaccard 1981).
Several theoretical frameworks have been proposed to
explain the consumer complaint process--that is, when
would dissatisfied consumers seek redress, register
complaints, change future behavior (e.g., by withdrawing
patronage to the brand) or do nothing at all? Richins
(1979) suggests a psychological framework of
attitudes >intentions >behavior as the basis for
explaining the complaint process. Krishnan and Valle
(1979) propose an attributional theory perspective based
on attributions of dissatisfaction as the foundation for
understanding the CCB. Another framework based on
economic theory purports that consumers are rational
processors of information and specific complaint actions
are determined by expectancy-value Judgments that are
attached to each alternative course of complaint action
(Hirschman 1970). Althogh a clear understanding concern
ing the appropriate perspective in any given situation or
a framework that ties together the three theoretical
perspectives is lacking (Day et al. 1931), it seems clear
that retailer/manufacturer responses to specific consumer
complaint actions result in a final or overall feeling of
satisfaction/dissatisfaction with a purchase incident.
The CCB paradigm thus proposes that consumer's
satisfaction/dissatisfaction with the handling of his/her
complaint actions influences the likelihood of repur
chase. The disconfirmation of performance expectations
is, according to this paradigm, a necessary but not
sufficient condition for determining repurchase inten
tions (Andreasen 1977).
If these two dimensions of the post-purchase pheno
mena can be treated within a single framework, the
resulting synthesis would encourage and guide theoretical
as well as empirical research in the area. The consid
eration of how CS/D and CCB can be integrated into a
holistic model of post-purchase processes forms the major
objective of this dissertation. A research design is
presented which empirically tests a part of the proposed
holistic model of the post-purchase process in four key
service industries.
Methodological Contribution
From a methodological standpoint, this dissertation
research is one of the few attempts in the marketing
literature to use Item Response Theory (IRT) based
techniques as the basic tool for estimation of construct
reliability and validity. IRT is a measurement theory
(of. classical test theory), in that it sets down rules
for converting empirical observations into corresponding
theoretical constructs. IRT belongs to the latent trait
class of theories which explicitly specify the form of
the response curve for each item in the scale using
"item" as well as "person" parameters. IRT based tech
niques use this response specification in assessing
measurements of underlying constructs that are both
"sample-free" (not affected by the specific set of
respondents to whom the questionnaire is administered) as
well as "test free" (not affected by the specific set of
items used to measure the construct). This is in
contrast to classical test (or true score) theory which
guarantees only "sample-free" measurement and does not
account for "test" effects. IRT is well grounded in
statistical theory and has been rigorously developed by
researchers in the area of psychology and educational
psychology (Lord 1980; Brinbaum 1968; Hulin, Drasgow and
Miller 1983). IRT has not been embraced by marketing and
the classical test theory remains, despite its well docu
mented limitations, as the predominant paradigm of meas
urement in the marketing literature.
In this research, IRT is used to provide estimates
of construct reliability in terms of the standard error
of measurement defined for every level of the underlying
construct. This is a substantial improvement over the
global estimates of the lower bound of reliability, inde
pendent of the level of the construct being measured,
provided by the classical test theory (example: coeffi
cient alpha). Further, IRT provides item and test infor
mation curves as a function of the level of the under
lying construct. These information curves are proposed
to be used in the present research to achieve reliability
and parsimony in the various scales used as operational
measures of the constructs in the holistic model of post-
purchase process. Next, the specific service industries
selected for empirical investigation are discussed.
8
The Choice of Service Industries
This research investigates dissatisfaction and
complaint behavior in four industries: (a) auto repair,
(b) health services, (c) banks and financial services,
and (d) the grocery retail industry. Best and Andreasen
(1977) and Day and Bodur (1977) report that these indus
tries are characterized by high levels of consumer dis
satisfaction and complaints. For instance. Day and Bodur
(1977) report that the percentage of users dissatisfied
with auto-repair services is as high as 49.2X, with
health services in hospitals as much as 24.IX and with
banks and trust companies over 18X. Yet, only 61.IX of
these dissatisfied users, in the case of auto-repair,
take some action (e.g., complaint). This proportion
drops to 36.3X in health service industry, and is less
than 30X for grocery products (Best and Andreasen 1975).
This selection of industry groups, then, provides a wide
variation in post-purchase processes, level and nature of
dissatisfaction, and nature of complaints.
From a macro-marketing standpoint, Andreasen (1983)
predicts a rather gloomy picture of service industries
that are characterized by "loose monopolies." Hirschman
(1970) defines "loose monopolies" as industries, such as
health services where there are many suppliers but where
the consumer's freedom of "exit," that is, his/her abil
ity to switch suppliers, is restricted. Andreasen (1983)
predicts that when service industries function as loose
monopolies, they become increasingly inefficient and
unresponsive to consumer dissatisfaction. Ironically,
despite this rise in dissatisfaction, the nature of the
industry is such that consumers are discouraged from
seeking other suppliers. Such consequences entail great
cost, both for the individual consumer and the society as
a whole (Hirschman 1970).
This combination of increasing competition and
declining consumer satisfaction raises some serious
concerns for managers of consumer services. If a
systematic investigation of the processes, antecedents
and possible causes of dissatisfaction and complaints in
specific sex /ice industries can be carried out, it may
provide managers with guidelines for addressing
complaints satisfactorily and suggest programs for respo
nding to future dissatisfactions. Such a program could
lead to a higher level of consumer satisfaction which in
turn encourages loyalty over repeated purchases (Engel
and Blackwell 1982).
Outline of the Dissertation
The dissertation is organized as follows: (a) a
review of the theoretical and empirical work undertaken
in the CS/D and CCB area; (b) the development of a
holistic model that serves as the basis for the present
research; (c) the operationalization of the holistic
10
model and the development of key hypotheses; (d) the
survey procedures and methods adopted to collect data;
(e) the analysis of data using IRT and other traditional
methods to implement empirical investigation of these
hypotheses; and (f) a discussion of the implications,
contributions and limitations of the research.
CHAPTER 2
A REVIEW OF THE CONCEPTUAL AND EMPIRICAL RESEARCH
Conceptual Review
Terms and Concepts
Most of the research in CS/D and CCB concerns pro
ducts. Key terms in the area, such as perceived perfor
mance, expectation level, use experience, are defined in
terms of products rather than services. Since terms
merely represent theoretical concepts and concepts are
invariant across subjects (i.e., products or services),
these terms can be redefined for services. For instance,
the performance of a product is conceptually equivalent
to the benefits of a service, the use experience of a
product to the consumption experience of a service, and
so on. Rather than define a host of new terms, this
dissertation retains the usual terminology of the area
(performance, use, etc.) but implicitly assumes that
these terms represent concepts that are consistent with
the notion of services.
Theories of CS/D
The central concept in the various approaches to the
study of CS/D is the notion of discrepancy. It is implied
that satisfaction or dissatisfaction results from a
11
12
discrepancy, which in turn is a function of the conscious
comparison between "a cognitive state prior to the event
and a subsequent cognition state, usually realized after
the event is experienced" (Oliver 1980, p. 206). These
events could be product use experiences, the consumption
of services or other related consumption phenomena. The
various theories of CS/D are consistent in suggesting
that a feeling of satisfaction results when the subse
quent cognitive state "exceeds" the prior cognitive
state. Dissatisfaction results when the former "falls
short" of the latter (Woodruff et al. 1983). These
theories differ in the theoretical meaning and appro
priate operationalizations of these cognitive states, and
in the underlying mechanism of the comparison between
these cognitive states.
Cognitive Dissonance Theory
Early writers in marketing, including Engel, Koliat
and Blackwell (1968, pp. 512-15) and Howard and Sheth
(1969, pp. 145-50) proposed a cognitive dissonance theory
(Festinger 1957) as the underlying framework for CS/D.
They suggested that prior cognitions were expectations of
product or service performance and based on shopping
effort involved, and that subsequent cognitions pertained
to actual product performance. Thus satisfaction would
increase as the ratio of performance to expectations
increased (Cardozo 1965; Woodside 1972). The cognitive
13
dissonance theory predicted that consumers would be moti
vated to reduce the dissonance between expectations and
performance, yet empirical research in the laboratory and
the field showed little support for this framework and
suggested that CS/D may be a much more complex process
(Olson and Dover 1976, 1979; Oliver 1977; Swan 1977).
Assimilation-Contrast Theory
The Assimilation-Contrast theory (Sherif and Hovland
1961) framework was proposed by Anderson (1973) as a
possible mechanism for CS/D process. He posited that
consumers have a "Just noticeable difference" (Jnd) for
the magnitude of discrepancy. That is, there exists a
range of acceptable deviations around one's expectation
level which is not perceived as discrepant. This range
forms a "latitude of acceptance" which produces the
outside this latitude become psychologically magnified so
that the product is perceived as much better or worse
than it actually is: the contrast effect. Empirical work
based on this framework has yielded equivocal findings
(Olshavsky and Miller 1972; Olson and Dover 1979). Oliver
(1980) suggests that the failure of the assimilation-con
trast theory may be because:
emerging research across different products and contexts shows that expectations and disconfirmation are uncorrelated Crather than be interacting as suggested by assimilation-contrast theory] and
14
satisfaction is an additive function of the two (p. 207). ^
Comparison Level Theory
LaTour and Peat (1979) suggest Thibaut and Kelly's
(1959) comparison level theory as another promising
research direction. An extension of this theory to CS/D
phenomenon implies that a consumer has a desired expecta
tion level for each attribute of the product or the
service, which is essentially subjective and is based on
personal experience, significant other's experience and
the unique nature of the situation. This subjective
expectation level and its comparison with the net of
positive and/or negative disconfirmations of perceived
attribute levels for the brand (service) determine the
degree of satisfaction with the brand (service).
The major contribution of this framework is in its
definition of expectation levels that results from affec
tive evaluation of physical attributes as well as other
subjective considerations. This is in contrast to the
assimilation-contrast theory and the cognitive dissonance
theory, both of which conceptualize expectations as
dependent on objective features (attributes) of the
product or service. Though this framework appears inter
esting, empirical work in this area has suffered due to
the lack of rigorous conceptual development of the
comparison level theory as applied to consumer satisfac-
15
tion or dissatisfaction area (Oliver 1980, pp. 207-8).
Adaptation Level Theory
Recent work by Woodruff et al. (1983) and Oliver
(1980a) has attempted to model the additive character
istic of expectations and disconfirmations in determining
CS/D by using Helson's Adaptation Level theory (1948,
1959). The theoretical meaning of expectations in this
framework is very similar to that of its definition in
the comparison level theory (LaTour and Peat 1979). That
is, expectations are based on any number of factors
including prior experiences, word-of-mouth, manufac
turer's reputation, advertising, etc. Disconfirmation is
the perceived discrepancy obtained by comparing the
perceived performance against the expectation level. The
consumers' discrepancy ratings are hypothesized to be
normally distributed around their expectation level. The
consumer's net response of satisfaction or dissatisfac
tion is then given by:
Satisfaction = F (expectations, disconfirmation) (1)
The function "F" is an additive function of uncorre
lated factors (Oliver 1980a). Further, Oliver (1980a)
suggests that satisfaction experiences influence future
purchase intentions as well as post-purchase attitude,
say at time (t-^1). In other words, a dissatisfying
product or service use experience would decrease one's
16
inclination to re-purchase. Thus post-purchase attitude
at time (t+l) can be written as:
Attitude(t*l) = F (attitude(t), satisfaction) (2)
In all of the above, which Oliver (1980) calls as
the "over-theorizing" of CS/D phenomenon, one key
question has received comparatively much less attention:
What is the appropriate conceptualization of satisfac
tion, the dependent phenomenon? Is satisfaction a
cognition, an attitude, an emotion, a feeling or a
motivating force? Recent research in CS/D is now begin
ning to explore this issue seriously; this research focus
has served to bring together the CS/D and the CCB areas
of research on post-purchase process. The issue of the
conceptualization of satisfaction is now addressed.
Conceptualization of Satisfaction
Can satisfaction be conceptualized as an attitude
which results from disconfirmed expectation? In order to
qualify as an attitude the phenomena must be persistent
and fairly stable over time (Day 1984). Since satisfac
tion is the result of a particular consumption event, it
can neither exist before the event occurs nor necessarily
affect outcomes of future consumption events. This is
particularly true for frequently purchased services and
products. Thus, satisfaction does not seem to fit the
17
definition of an attitude (Day 1984).
How then should satisfaction be viewed? Recent
studies support the conceptualization of CS/D as an
emotional feeling (Westbrook 1983; Woodruff et al. 1983).
Emotion is defined as "a state of arousal which is mani
fested in conscious feelings, neurological processes and
observable expressions and behaviors" (Day 1984). Thus
emotions (satisfactions or dissatisfactions) may quickly
subside when the triggering stimulus (consumption event)
is removed or the situation changes (engaging in a new
consumption event). Westbrook (1983) presents the
results of an empirical study and Woodruff et al. (1983)
provide a theoretical framework incorporating an emo
tional conceptualization of CS/D.
Initial or Final Reaction
Andreasen (1977) further clarified the CS/D concept
by suggesting that there are two critical points in the
post-purchase process at which one can define CS/D. The
first or the "initial" reaction is S/D resulting from the
consumer's disconfirmation of expectations with product
or service performance. The second or "final" reaction
is the consumer's perceived S/D with the manner in which
complaints are handled. While the study of initial
reaction of S/D may predict that a "dissatisfying product
purchase should decrease one's inclination to repurchase"
(Oliver 1980) and support a satisfaction >post-attitude
18
link, the study of final S/D may suggest that this may
not necessarily be the sequence of events. In partic
ular, the final S/D reaction may predict that the
consumers who have the source of their initial dissatis
faction resolved by sellers' complaint handling
mechanisms, may end up with enhanced positive post
attitude and post-intentions of repurchase (Day 1981;
Andreasen 1977). Thus a CS/D can be conceived of as an
emotion that occurs both initially and again, after
sellers respond to reactions of dissatisfaction, with an
intervening complaint process.
Conceptualization of CCB
At least two competing conceptualizations of CCB
have been proposed in the literature (see Figure 2.1).
Bearden and Teel (1983) suggest that CCB is an action
resulting from emotions of dissatisfaction. Such a
conceptualization assumes no intervening variables
between CS/D and CCB. Much of the empirical work with
this conceptualization suggests a very weak relationship
between CS/D and CCB, with typically only 15% of the
variance explained (Day 1984). These findings strongly
suggest a misspecification of the relationship between
CS/D and CCB.
The alternative conceptualization posits that
complaining behavior is logically subsequent to dissatisfaction and is a distinct set of activities
19
which are influenced by a variety of situational and personal factors which appear to be unrelated to the intensity of dissatisfaction (Day 1984).
This implies that CS/D is an emotional state which will
under some circumstances motivate consumers to engage in
a complaining/non-complaining decision process, the out
come of which is the specific CCB. In other words,
dissatisfaction motivates the consumer to undergo a
subsequent decision making process which depends not so
much on how strong the emotions of dissatisfaction are
but on the consumer's perception of the attribution of
dissatisfaction, expectancy and value of outcomes, costs
involved, product importance, etc. This conceptual
ization is consistent with many empirical findings that
show that a large number of dissatisfied consumers do not
complain (Best and Andreasen 1975; Day and Ash 1979;
Warland et al. 1975).
What do these conceptualizations of CS/D and CCB
suggest for the relationship between post-purchase proc
esses and post-attitudes? If proper distinctions are made
between satisfaction from disconfirmation of expectations
and the satisfaction with resolution of subsequent
complaints, the following equations can be written:
the post-attitude at time t2 is a function of (a) atti
tudes at earlier time tl and (b) satisfaction with the
product performance. However, when CCB is triggered by
performance dissatisfaction, satisfaction with the reso
lution of these complaints significantiy influences atti
tudes at time t2. Post-purchase intentions at times
subsequent to t2 are a function of attitudes at time t2.
The process described by equations 3-7 is presented in
Figure 2.2.
What is the specific nature of this CCB process?
The theories of CCB are reviewed in an attempt to answer
this question.
Theories of CCB
It is suggested that a better understanding and
evaluation of theories can be achieved when the phenom
enon of interest is properly defined and its taxonomy (or
classification) satisfactorily developed (Hunt 1983).
Therefore, first the issue of a definition for CCB is
21
addressed and a comprehensive classification system for
CCB is proposed. Then the proposed classification system
is used as the framework for evaluating the various
theories of CCB.
Definition and Taxonomy
Landon (1980) states that complaint behavior has not
been thoroughly described and further work would be
helpful on a global definition and taxonomy of complaint
actions. Thus the development of a comprehensive
classification system is a desirable contribution to the
area.
What ought to be the properties of such a
classification system? Hunt (1983) suggests that classi
fication schema should: (a) adequately specify the
phenomena, (b) adequately specify the criteria of classi
fication, (c) provide categories that are mutually
exclusive, and (d) collectively exhaustive, and finally
(e) be useful.
In accordance with Hunt's first guideline, the
phenomenon must be adequately specified. A widely
accepted definition of CCB states (Day 1980; Fornell and
Didow 1980; Jacoby and Jaccard 1981):
The consumer complaint behavior or response is the
set of all non-behavioral and behavioral responses
which are triggered by dissatisfaction and involve
communicating something negative regarding a
22
purchase episode including the product or service.
Day and Landon (1977) proposed a two level hierar
chical classification of CCB. The first level distin
guishes between action and no action, while the second
distinguishes public actions from private actions.
Public actions include seeking redress or refund from the
seller as well as complaining to a consumer organization.
Private actions include word-of-mouth communication to
friends and relatives and the consumer's decision to stop
patronizing the product or service.
Day (1980) proposed another classification schema
for complaint behavior using the criterion of consumer
motive and the distinction between seeking redress and
the decision to change future behavior. He proposed
three responses based on this criterion:
(1) Seeking Redress: seeking a specific remedy.
(2) Complaining: communicating dissatisfaction for
reasons other than seeking remedy.
(3) Personal Boycott: discontinue purchase of
offending product or service.
Although this taxonomy addresses a deficiency in the
Day and Landon (1977) classification, it fails to
incorporate the distinction in private versus public
actions. However, Day (1980) suggests that his taxonomy
could be combined with that of Day and Landon (1977).
Research by Fornell and Westbrook (1983), Day and
23
Bodur (1977) and Day and Ash (1979) shows that there are
significant differences between consumers who complain to
sellers directly and those who take their complaints to
consumer organizations. Therefore, the distinction
between dyadic and third party complaint actions is,
probably, another useful criterion in the understanding
of CCB. A "dyadic" CCB response is limited to the
dissatisfied consumer and the seller whose product/-
service caused the dissatisfaction. "Third party" CCB
response involves consumer organizations, public
agencies, friends, relatives, etc.
Thus an exhaustive and complete classification of
CCB responses is achieved by incorporating within a
single taxonomy the distinctions between action/no
action, public action/private action, redress/future
behavior, and dyadic/third party. This taxonomy is in
part hierarchical since the last three distinctions are
meaningful only if some action is taken. For instances
of no action, further classification is meaningless.
A taxonomical framework using the above distinctions
is shown in Figure 2.3. Interpretation of the various
cells is fairly straightforward. For instance, voice
implies public actions for seeking redress that are
directed at the sellers. Similarly, exit results when
private actions for changing future patronage of the
specific seller/product/service are taken.
24
This classification schema provides a useful
grouping of the whole range of phenomena into mutually
exhaustive classes with well defined characteristics. An
interesting property of such a logical partitioning or
classification is that it reveals some "empty cells" that
are under-researched (cells 1 and 3 ) . Landon (1980)
previously noted that the CCB literature has completely
ignored informal actions taken occasionally for the
purpose of seeking redress from sellers.
This classification schema is the basic building
block in the theory building process and will be employed
in subsequent evaluation and empirical investigation of
CCB process.
Conceptual Frameworks of CCB
Phenomenological Model. Landon (1977) made one of the
first attempts in the marketing literature to develop a
theoretical model of CCB process. He proposed a phenom
enological model that restricts attention to the
consumer's perceptions of the relevant variables and
therefore does not directly incorporate the salient
characteristics of the environment. He suggested that
CCB process can be modeled by the equation:
Complaint behavior=F(dissatisfaction, importance, benefit from complaining, personality characteristics) (8)
25
Landon (1977) defined dissatisfaction as the func
tion of the discrepancy between performance and expecta
tion, importance of the product/service as a function of
cost, the search time involved in its purchase, the
possibility of physical harm that could result from
dissatisfaction and the ego involvement, benefit from
complaining as a function of the perceived payoffs and
costs of complaining, and personality as a sum of
concepts such as discontent, locus of control, attribu
tions, etc. Landon (1977) proposed personality to be
only a mediator in the process of CCB.
Landon's model integrated various research findings
in the CCB area and triggered a large number of studies
(Clabaugh et al. 1979; Day 1981; Bearden and Mason 1983;
Barnes and Kelloway 1980; Day and Ash 1979). However,
the model's lack of specificity allows neither a critical
assessment of empirical findings nor a clear under
standing of the process underlying the CCB responses.
Attribution Theory. Another approach suggested for
studying CCB is based on attribution theory (Krishnan and
Valle 1979; Richins 1983; Valle and Waliendorf 1977;
Foikes 1984). This theory suggests that people are
rational information processors whose actions are influ
enced by causal inferences (Foikes 1984). In other
words, when people are dissatisfied with a service, they
26
try to determine the cause of dissatisfaction and assign
responsibility.
Weiner (1980) suggested analyzing attributions in
terms of stability (temporary or permanent), locus (in
the consumer or in the seller), and controllability
(volitional or non-volitional). These attributions in
turn influence the type of action taken by the consumer
in response to a dissatisfying experience.
Empirical findings generally support the basic
tenets of the attributional theory. That is, the more
external, the more stable and the more controllable the
attribution, the greater the likelihood of engaging in
the VOICE responses. For instance, Foikes (1984) shows
that consumers who attribute the failure to the manu
facturer or store tend to engage more in seeking refunds.
However, Landon's model (1977) indicates that
attributions constitute but one of many determinants of
the CCB process. For instance, consumers may make exter
nal and volitional attributions of their dissatisfac
tions, but may decide not to complain because "it is not
worth the effort" or "the retailer wouldn't care anyway"
(Foikes 1984). Thus the explanation and prediction of
CCB based entirely on attributions of consumer dissatis
faction remain limited. Perhaps attributions are
antecedents to beliefs or expectancies regarding various
courses of action, which in turn affect the CCB process
27
(Landon 1977; Foikes 1984).
Economic Perspective. A far more promising theoretical
framework is proposed by Hirschman (1970) in the area of
economics. He suggested that the likelihood of CCB is
determined by two factors:
(1) an evaluation of the probability that taking a
particular action would have a desired result, e. g. ,
redress, refund, etc. This concept is similar to the
notion of expectancy of an outcome from a course of
action.
(2) a Judgment that the end result of the action is
worthwhile. This is the value Judgment of the
outcome.
Hirschman (1970) focused on macro-economic effects
of consumer responses to dissatisfaction. He proposed
that industries which are highly competitive would elicit
responses of EXIT from dissatisfied consumers. This is
so because many alternative products or services are
available and the cost of VOICE is high compared to the
easiness of EXIT. On the contrary, in monopolistic
industries, the consumers are "locked" in with the seller
and the avenue of EXIT is blocked. Hence, dissatisfied
consumers are more prone to VOICE their feelings of
dissatisfaction.
Fornell and Didow (1980) and Fornell and Robinson
(1983) present two empirical studies based on Hirschman's
28
theory. By modeling the effects of structural variables
(availability of alternative substitutes, expectancy of
outcomes, the number of competing firms in the industry
and distribution breadth), 63X of the variance in CCB
responses could be explained.
Andreasen (1983) has attempted to elaborate on the
economic framework of Hirschman (1970) as applied to the
CS/D and CCB processes in loose monopolies. He cites the
health industry as an example of a loose monopoly where
it is possible for the consumers (patients) to exit but
where exiting is unlikely for a host of reasons (restric
ted information about other doctors, ignorance of the
patient, psychological inhibitions against changing a
doctor). In such industries, the consumer's ability to
EXIT as well as the ability to VOICE is restricted. Thus
overall quality and welfare suffers. This particular
extension of the economic framework promises to be very
useful for the study of CCB, but as yet no empirical test
of this theory has been undertaken.
Psychological Perspective. The psychological approach to
the study of CCB process has gained considerable atten
tion from many recent researchers (Jacoby and Jaccard
1981; Richins 1979, 1982; Day 1980, 1984; Westbrook
1980). This framework suggests that the consumer's psyc
hological evaluations of the costs and benefits of
29
complaining actions and the resulting reaction of favor-
ableness or unfavorableness is a strong determinant of
CCB. In other words, attitude towards the act of
complaining is a causal antecedent of CCB and this
relationship can be modeled by the usual attitude >
intentions >behavior framework (Richins 1982; Day
1984). However, researchers in this tradition generally
allow that other variables (rational analysis of the
costs and benefits, attributions, etc.) probably act
along with attitudes to determine CCB. Though some
empirical studies have been undertaken using this frame
work, the actual form of the relationship is unknown.
Empirical research corroborates a strong relation
ship between attitudes and intentions of complaining,
though the relationship between attitudes and behavior is
generally poor (Richins 1982). The latter finding is a
reflection of the general lack of attitude-behavior
consistency found in the consumer behavior literature and
is probably due to many situational variables (e.g.,
frequency of patronizing, product cost, etc.) that
intervene in the attitude-behavior relationship (Richins
1982).
Other studies have investigated the effects of
generalized and stable affective influences, e.g.,
consumer discontent and alienation from the market place,
as a determinant of CCB (Westbrook 1980; Bearden and
30
Mason 1983). Generally, the explanatory power of these
generalized affective influences is weak (Bearden and
Mason 1983). This is perhaps due to the high level of
generality in consumer discontent and alienation from the
market place. CCB is a specific response to a specific
dissatisfaction.
The usefulness of a conceptual framework lies in the
extent of empirical research and support that it entails.
Therefore, a proper evaluation of these various concep
tual frameworks of CCB process should be in the light of
empirical findings in the area. Based on the review of
the empirical findings, an attempt will be made in the
next chapter to bring together the various conceptual
frameworks into a holistic model of CCB process.
Empirical Review
A recent review of the empirical findings in the CCB
literature was published by Robinson (1979) in the 1978
proceedings of the CS/D and CCB conference. This was one
of the first attempts in the CCB literature to summarize
previous research findings, identify shortcomings and
list areas of future inquiry. Since 1978, there has been
a steady growth in the body of knowledge about consumers
who complain when confronted by consumer problems: who
they are, what they complain about, to whom they
complain, how they are treated when they complain, and
how they differ from other consumers.
31
In the present dissertation, empirical findings are
discussed in light of Robinson's review of literature and
on empirical findings published since 1978. Table 2.1
shows a summary of key research findings including some
major empirical studies of 1977 (not covered in
Robinson's review). The criterion used for including
studies in table 2.1 are:
(a) adequate coverage of the large variety of
variables investigated as predictors of CCB,
and
(b) inclusion of studies which have generalizable
findings.
Validity of Robinson's Conclusions
Many of the conclusions stated by Robinson (1979)
regarding the state of research in CCB are still valid
for the research undertaken since then. Many empirical
investigations are based on recall information of a past
dissatisfaction and complaint behavior. The use of
scenarios is rare (see for exceptions Langmeyer and
Langmeyer 1979; Foikes 1983). Robinson (1979, p. 41) has
identified a specific problem with this method of re
search: "large samples sizes are required in most studies
(using recall data) in order to develop an acceptable
number of respondents who had experienced problems with a
product or service." Though this problem is reduced with
32
the use of scenarios (since each respondent is assumed to
face a given problem). The use of scenarios is not very
common in the CCB literature, perhaps due to the limited
generalizability of its research findings.
Robinson (1979) also reported that most studies
found low complaining rates for those consumers who did
experience a problem. Subsequent studies suggest a
similar conclusion. Day (1981) reported the number of
non-complainers to be between 22X to 46X for different
products, Richins (1983) found them to be as high as
32.2X and Bearden and Mason (1984) reported their number
to be as much as 61X to 76.6X. However, NO ACTION is a
legitimate complaint response (see classification of CCB)
and ought to be investigated as a part of the complete
range of complaint phenomena. Robinson (1979) points out
in his review that most studies focus on complainers,
ignoring non-complainers completely. Subsequent research
studies have attempted to address this deficiency. For
instance, Bearden and Mason (1984) and Richins (1982)
attempted to explain non-complainers based on their
attitudes towards complaining, Gronhaug and Zaltman
(1981) used "market place activity" as the explanatory
variable while Krishnan and Valle (1979) have attempted
to explain no-action consumers using attribution theory.
It was also pointed out in the previous review that
much of the research in the area has been limited to
33
analysis of demographic correlates. This emphasis has
changed considerably since the review was published
(1979). Many different theoretical streams have been
explored empirically. Perhaps one of the most active
research streams is the psychological perspective which
posits that attitudes toward the act of complaining are
predictors of CCB. Other theoretical frameworks
considered for empirical studies are attribution theory
(Foikes 1984) and Hirschman's economic model incorpo
rating availability of alternatives and structural
constraints.
Robinson (1979) had also concluded that much of CCB
research has oversimplified the concept of complaint
actions. Most researchers classify complaint actions
into a dichotomy of action/no action but the phenomena
has a range of possibilities, from legal action to word-
of -mouth communication. However, Day (1984) and Bearden
and Teel (1983) among others have attempted to concep
tualize complaint actions as several possible alterna
tives which can be ordered according to the extent of
effort involved.
In conclusion, Robinson's (1979) suggestion relative
to sample size remains valid. However, other reported
deficiencies, such as (a) the neglect of non-complainers,
(b) the over-simplification of complaint behavior, and
(c) limitation of the research to demographic correlates,
34
appear much less severe in current research studies. The
next section summarizes the empirical findings and the
key independent variables investigated as predictors of
CCB.
Key Predictors of CCB
Personality and Demographic Characteristics
Robinson (1979) reports six personality variables
that have been investigated as predictors of CCB;
dogmatism, internal-external locus of control, gener
alized self confidence, powerlessness, social isolation,
and political efficacy. Subsequent research studies have
in addition investigated aggressiveness and assertiveness
(Richins 1983) as predictors of CCB.
Among the demographic variables, Robinson (1979)
suggests age, income, education, and occupation as key
predictors found across many studies. Complainers are
usually younger, more educated, have higher incomes, and
have a greater tendency to hold managerial and profes
sional Jobs. In addition, Villareal-Camacho (1983) has
suggested that race and cultural differences could also
be related to complaint actions. However, Gronhaug and
Zaltman (1981) show that these demographic and person
ality correlates are, perhaps, artifacts of the extent of
market place activity. They show empirically that the
correlation between demographic and personality corre-
35
lates and CCB becomes insignificant when the effect of
market place activity is partialled out.
Attitudes and Affective Variables
There has been considerable investigation of the
effects of many attitudinal and affective variables on
complaint actions. While Robinson (1979) reports much
research on the concepts of consumer discontent
(Lundstrom and Lament 1976) and alienation from the
market place (Allison 1978), generally the explanatory
power of these generalized affective influences is weak
(Bearden and Mason 1983). This is probably due to the
high level of generality of these affective influences.
CCB is a specific response to a specific dissatisfaction.
Subsequent research has attempted to operationalize
the construct of attitude toward the act of complaining,
and to use it as predictor of intention to engage in
complaint actions (Richins 1980; Bearden, Teel and
Crockett 1980; Bearden and Mason 1984). These findings
corroborate a strong relationship between attitudes and
intentions of complaining though the relationship between
attitudes and actual behavior is generally poor (Richins
1982). The latter finding is a reflection of the general
lack of attitude-behavior consistency found in the
consumer behavior, and is probably due to many situa
tional variables (i.e., product cost, importance,
frequency of patronizing the store) that intervene the
36
attitude-behavior relationship (Richins 1982).
Evaluation of Alternatives
There is also a growing concern that the cost and
benefit evaluation of each alternative complaint action
should be incorporated within CCB model (Day 1984).
Richins (1980) has empirically shown the predictive
validity of this concept in explaining complaint actions.
Fornell and Didow (1980) show that consumer's perception
of possible alternatives available in a given situation
is a useful predictor in the case of complaint actions
with a range of products and services. Finally Richins
(1983) shows that consumers' evaluation of retailer
responsiveness when a specific complaint action is taken,
is a key determinant of the complaint behavior across two
product groups.
Attribution Variables
Other variables that have been investigated as
predictors of CCB include notions of external versus
internal attributions for the "cause" of dissatisfaction
(Foikes 1984), and the severity of the dissatisfaction or
the problem (Richins 1983). Each of these variables have
been examined empirically as antecedent to CCB.
Empirical findings generally support the basic tenets of
the attribution theory. That is, the more external, the
more stable, and the more controllable the attribution.
37
the greater the likelihood of engaging in VOICE
responses. For instance, Foikes (1984) shows that
consumers who attribute the product failure to the
manufacturer or store tend to engage more in seeking
refunds than to take NO ACTION. In addition, the
severity of the dissatisfaction seems to affect the
amount of effort that is expended in the CCB process
though not necessarily the specific complaint action
taken (Richins 1983).
Structural Variables
Fornell and Robinson (1983) and Fornell and Didow
(1980) present two empirical studies based on Hirschman's
theory. Their research indicates that structural
constraints such as the nature of industry (for example
competitive, monopolistic, etc.), and the distribution
breadth (for example widely distributed, etc.) seems to
affect the consumer's feelings of which complaint action
would be fruitful, in other words, the "expectancy" of
complaint actions. By modeling the effects of structural
variables (availability of alternative substitutes,
expectancy of outcomes, the number of competing firms in
the industry and distribution breadth), they explained
63X of the variance in the CCB responses.
Discussion
This chapter summarized and reviewed the literature
38
in CCB, discussed conceptual frameworks currently
employed and analyzed empirical findings in this area.
Two conclusions can be drawn. One, that the CCB area is
enriched by considerable theoretical contribution pur
porting to explain the process underlying the complaint
actions. Each of these theoretical frameworks has
received empirical corroboration for its hypotheses
(though limited in some cases) indicating that there may
be more than one route for explaining the CCB process.
This conclusion provides the basis for the second key
finding. The many different and sometimes contrasting
theoretical frameworks for explaining the CCB process
appear to give an impression of a relatively fragmented
structure of the research in the area. The empirical
corroboration of these perspectives suggests that each of
these frameworks may be valid under different conditions
and situations. Further research that investigates these
conditions and identifies the "valid zone" for the indi
vidual theories would be very useful in increasing our
understanding of the CCB process.
Further, a comprehensive conceptual model that
incorporates different streams of past research findings
and at the same time provides programmatic research
directions, would address a key deficiency in the area.
Richins (1979) states in the same vein:
Few if any studies (in the CCB area) have followed
39
or been embedded within a conceptual or theoretical framework. This deficit is an important one, and a conceptual model is needed to fully understand the nature of consumer complaining behavior.
Similarly, Foikes (1984) affirms that a theoretical
model is needed to "map out relationships between
specific thoughts about product failure and specific
complaining behavior."
Some recent attempts have been made to address these
issues. Day (1984) proposed a conceptual model that
incorporated the psychological perspective and the
rational decision making perspective of the economic
theory. However, the model is not formalized to allow
empirical testing of derived hypotheses. For instance,
the nature and operationalization of the construct of
"analysis of alternatives," which is a direct determinant
of CCB is not suggested. Thus the objective of the
following chapter is to develop and partially formalize a
holistic model of consumer complaining behavior--a model
that would incorporate the different theoretical streams
of thought in a single framework. The partial formaliza
tion of this model is undertaken to clearly define the
assumptions, axioms and law-like statements contained in
«o • a; • c to to o (/I t_ • . -O) Ol 4-> «- E o Q . 3 le 3 to I -to c OJ
O 4->
to o c
f- s »-• « - > • . - 0 * <o < . - .— C 1/1 r - . U • . - OJ OJ 4-> I/I .
4-> IQ I • >— to ( . <a to o)
— >>• «»- - D 3 O ^ .
C > > - ^ O)
4 ^ ^ •^ V) •»-> r r- 0»
OJ
u
« o c -.- •.-
<0 C O) — OJ 3 • • - "O o -lO C 01 > OJ t -
< 4 J *•-
c f— IQ
(o a. 1/1 g OJ o
t o T 3 C C O lO . 1 -
4 J C O
o <o • -4-> 4-> O C (O - i -
<*- <o </> 1 —
• • - Q .
IQ 5 t<0 O
.« t o c o •^ 4-) ( J l O
0) 4>> R>
> •»" (_ o .
. 10 •o <D c <a u
T3 C m
U I
=>
. lO
•o <o c le
o
4.J C
<o
Q .
§ U
— 'OJ o
o >
lO I/I OJ 4 J O ( J f -3 >
T3 »-O OJ L. to Q .
1—I HJ
o "D • •»- «J
o o • O Q . c lO Q£
o >- < "oJ » c o o a
to "W c «/l - ^ 01 O l 1-
.— C 0) • • - * J
a; •.- •.-4-> <T3 ^ i - ,— O)
O O . ( -
OJ § c «_ t j — t o to «- - D OJ (-
I S I/I o C 4-> o <J 1/1
OJ
"SI 4-) 4-> lO • . -C 4 ^ OJ 4-1
• . - lO
IT> OJ
>
O l to
5g.
i / i 4 J c
l O
I / I c o (J
OJ
I / I • O 4-> OJ c > •.-OJ r -O Q . 1- e OJ o a. u
Ol • o ^ O) — c
11 Ol
4-> I/ I OJ c
• c o
I / I .^J
c — 0> l_
.^^ 4-> IQ 4->
4- ) <0
O "O o — c •.-4.> C 4.1 u OJ u <0 4 J <o
4.) 01 to
f - >— 4 J <0 lO IQ
.— c to O . ( - I / I g OJ — O 4 J -O
u c %- o O Ol
J= OJ OJ 4-> I / I Q. 3 >> > . <o
t— ^ O
> l C • J 4>' OJ 0 )
• .-• .- .^ I/I — c_ C lO lO Ol E Q . - 0 O C Ol (. <« ^ Q . 4->
c
I/I <o 5 OJ — «-
• O Q . " ^
4-> 5 c •»- o o 4-> -r-*J O *-> < 4 J <0
. OJ u <o — Q .
1 . -
1 . -
•^ •o
•o c <o
t o c o •^ 4 - >
3
^ •^ t .
4 . ) 4->
<
1 e 5 u
• <*- t o O C
o t o • » -01 4->
o. u >. <«
4 J 4 J
•^ c
c •.-OJ IT> L. , _ OJ Q .
( - OJ
•>- u a. > OJ (_ (- a,
to O
4 J - D 3 C
< <o
c o to
H c -^ 01
"O • <o 0 0 0) O l
0 0 ^
^
u OJ Q . I / I
OJ
19
> •c C <o
. u o (. a .
oc c u <a < c
•^ &. ^
O l p ^ O l 1 - 4
\n vo
41
to Ol
C
T3 OJ 3 C
- H •P C 0
u
(N
U
OQ <
•o OJ
19 O l
I/I Ol >
0 0
o
OJ JZ
(-<o OJ to OJ oe.
OJ
^ <o
t o 4.> 1 — Q . IQ 0 » 3 U •o O • ^ <0 >
•'- >» •o •— C 1— t - l O
•^ OJ u > o • f - t o </l to <a OJ t . t o O l <o O l lO O l
c T3 • • -c c w •»"
l o OJ 1— > Q .
Z § (- u OJ I / I OJ t o > (O t -
OJ OJ u ;- (_
. u o • r -
> <o x: £ OJ OJ
^ ; ; = IQ C OJ
•.- o » c •.-
t o - r - O
OJ c <o > 4.< O — : <0 f - Q . > 4.> E C
•— 3 O <0 ( - ^ U ^ Q . - 1 - 4-»
f~ 01 OJ 4-) - o c ( - 4.1 3 OJ Q lO 4 ^ 4 ^
,— +3 o Ol IQ 4.> ^ C IQ OJ It} L. ( _
4.* 01 01 Q 4 J > E
^ C •>-(O • » - 4 J S
* - ' <0 4 J OJ O l —
O •>< OJ X 4.1 l « C OJ
E : • O OJ C ' - v t _ OJ 0 1 J O Q t o
4 J « — E 3
«/l • OJ I / I C I / I > C
• OJ r— O .
§ t/Y
• </)
. => OJ .c 4 . )
I .
o 1 . -
OJ 3
(-4.1
t o
• -OJ I / I
c OJ > OJ (_ OJ
l O C l O (O J C U O f u
• r - *^" ' r ->< 4 J — » X
ii ^^±
4 . )
"O c
o 4-» O 4.>
>» •— t o Ol c
^ IQ
; = ! t o "O «/l OJ t o
^ - <o
OJ t o
c c <o o
• . t o 4.> t - o OJ lO
E 3 4 . ) t o C C • # -O IQ t J . —
. c o
4.1
u 19
I t lO
•^ 4-» 19
Q . t o OJ E > O
•>- O 4 J «- O 01 *•> to l / l 4.> t l ( . 1 o
c to
52!
t o
•^ •o OJ
4 . ) lO
• r -
> OJ ^^ "iS
4. ) O c 11 (-<o t o
4. ) c <Q
a. § u
"S o
o >
"S 4.1
•^ u
• ^
^ o t o
c
• o OJ
> • ^
OJ o I . OJ a.
_ <9 3
..> t J IQ
<.-O
C O
4 J 19
4.> C . OJ t o t o C OJ o t _ - 1 -a . 4 . ) OJ U t . l a
<*-OJ to
4 J • » -(Q 4.1 L lO 3 l / l O (O O • • -HJ "O
1
Ol 4.) 4.) C o • • -lO IQ
c .— IQ Q .
x: E o o o >> 4J T3 •.- C > — < Q
<9 C (/I o o to -P-( . 4.) OJ l/>
a . •>-
. l/l c o
•^ 4 J
u <9
4.) c
•^ <o r * . .
a. c E 0) o .^ u le 4.>
<«-O l/l c
to O OJ •— a . 4 J > i o
1— <o
. t o
(_ 0) a. a. o ^ to
t /1
=> • o c TJ
"S
U I I .
» o to "^ c
^ PO U 0 0
•f- O l
"S
o 3
• o o
u OJ Q .
t o
u OJ o . to
o <o E <o o
<o
. u o l _
o.
f OJ c t c_ i n
— ^* •^
«« m 00 O l
I c_ Ol
4 J 4 J O C IQ "T" t- <o <Q I —
J Z Q . U E > i <J 4-> • I - - o I— c • IQ (Q t o C UI c o o o t o • » - ••— C- 4.> 4 . ) OJ l/l o
Q . - ^ IQ
OJ ( . o
4 . ) to
^ . OJ to
o S ( . 4 . *
o o
^ • 4 J O l/l o OJ (-3 a .
o .— < "oJ * C Ol
O <T>
I O C to OJ
••- OJ ^ • D C- «4- "Q
3 O 4J p— 4 J •Q IQ OJ C C C O . O — >,-.->«- - 4 ^ 4 ^
C O " O O " D <Q C • • - C IQ 4.) (Q 4 J
O C .— IQ E ••" IQ I t - Oj IQ
• I - UI •— .— 4 J •>- ^ Q . •— 4J o e C IQ < - O
•—' U I Q . O
to 4 J O 3
•o o • ( . UI a . OJ
o c •»-OJ > > I . OJ OJ l/l to
IQ
5
c < t OJ u UI "o •Q OJ • t- r -
• O !"«. C 0 ^
« « . - 4
^ I Q
•^ U o to
T3 C I Q
O l C •^ c •^ < Q
^ v
a. e o o to
• o (_ I Q
S o 4 ^
l/l OJ
"O 3
4 . )
• • 4.> 4-> «c
O l c o (-
4 - J
UI
01 (. ( Q
to 4 . )
c •^
I Q
—
• c
•^ I Q
— a. s 5 o o
4 J
10 c o
• p -
0 . 4 . ) B
o u O l c •^ . 1 ^
!S E O l c
•f—
c C-OJ u c o
c OJ
4 . >
c •^ 1 . -
o to
4 - >
c I Q
c •^ e 01
4 > J
OJ tJ "O
OJ • o 3
4 . ) • ^
4 . 1 • * » '
I Q
OJ >
4-->
•^ to O Q .
OJ (.
i I Q
OJ > I Q
. C
4 - J
l/l • ^
c-Ol E 3 UI c o u
. to (-OJ E 3 to C o u
t—
I Q
l -OJ c OJ O l
c " Q
X 4 - J
U I
t o OJ c
• ^
UI 3 ^ to • o (. I Q
X o *->
*J
to
•^ 4->
Q
OJ <.-OJ
T3
I Q
• o I Q
X
4 . 1
to I Q
C_ 4 . )
c o o c
•^ to C-OJ E 3 to c o o ^
I Q
(-OJ
c .$
. OJ •o 3
4 . )
•^ 4 . J
4 . 1
I Q
-C *-i
o ^ o
. - J
TJ OJ
4 . 1
I Q
^— u t .
> 1
— 01 >
•^ 4 - >
• -to O a. UI •^ 4 . )
to o u t . OJ E 3 to C o
t_)
1 •^ t .
4 . 1
! / •
• ^ •
C^
. c o • • 4 . 1
U I Q
<«-to •^ 4»>
I Q
U I
to •^ •o
> i 4 . )
•^ »— I Q
3 o-
T3 C I Q
OJ o •^ t .
OJ O
• ^
I -a. o *-> " a;
4 - 1
• Q
^— V L.
> p *
OJ >
•^ . - l
•#— UI o
o *-> " 1;
4 - 1
• Q
— OJ t -
> l f—
OJ >
•.— 4 . )
< Q
O l OJ
c Q.-C
to
• -X 4 . )
• o I Q
OJ c. ^ c o •.— 4 - >
3 Q . ^
c I Q
C o •^ 4 - >
u I Q
1 . -
to •f—
4 . )
I Q
to to
•r—
T3
. C o •^ 4 - >
l_) I Q
<.-l / l
- p -
4 . )
I Q
to to
• ^
TD
> l 4 . >
•^ — I Q
3
o-
I c o
l ^
•.- o 4-1 to
0 c — O) > 4 J IQ C •
x : •— Ol OJ c
XI T3 •>-C C
0 1 'Q '*— C (Q
• ^ to .— (- OJ a . O T3 E 4.> 3 O • • - 4 J O
UI - o
S IQ
UI ( - 01
• • - o IQ -f-Q . > OJ ( . ( - (U
l/l o
4 J - O 3 C
< : <Q
o *J c- c
IQ •O <— • O) Q . to > E 4->
•— O O OJ U OJ U •'-> «- -O ^ OJ C 3
a . IQ UI
•• O l l/l c OJ — u ^
• C 4J > , IQ O ( - U •— OJ — o o .— O Q . - 0 C Q . C - T IQ IQ
c o •^ *-> IQ t . 4 J
c OJ u c o o
St-
o 4 . )
u OJ
1 . -1 . -« t
c o
• F —
4.1 3
^ •^ 1 .
4 J t o
• r .
•o m
i/i
E •w
>*->*-o
(-Ol
E 3 l / l
c o u
•o c IQ
. X 4. )
•o <Q OJ c_
XI
c o •^ 4. ) U IQ f-t o
• -4-> IQ t o t o
• » • o
C
o 4.* l / l
o u
• > l
*J
•.— .— IQ 3 o-
• o
c IQ
0) o •^ C-Q .
J C 4-> • ^
X
T3 to C 4.) IQ U • 3 l/l O l
1 3 OJ C O > , . — — e — ^ ^ 0 . 4 . 1 IQ 4 J
UI ( . o VC O 3 «—
c o to
oc O l
• a ^ c IQ •
OJ c OJ
"O l~ o • IQ o a£ OJ I . CO
00 O T
o 0 0 O l
IQ > , O IQ O
UI X (-OJ O Q -c — IQ OJ <-> oc ^ <
o
u
'oJ • C PO I- 00 O Ol
CSJ
0 0 O l
42
tJ Ol 3 C •H
C 0
u
CN
U
OQ <
UI O l
c
u 3 "O o
(J
•.— l/l r— l/l . O OJ 3
r - Q .
OJ O )
> c IQ - 1 -• C ^
I Q
C 4.) O
• - OJ •*-> t o
o o IQ ^
4 . )
OJ 4-> C IQ IQ > JZ
• 1 - 4->
c a . UI
OJ O I T 3 C 3
• I - 4 J . M : •»-IQ 4. )
* J 4 J
UI
c o •^ 4 J
c 01
4>J
c •^ 4 - >
c •^ I Q
p w
01 c OJ
X 4 J
"O c I Q
to
c o •^ 4 J
U Q . IQ
o o
l « -
o o. •^ x: U I
c o •^ 4 J
I Q r —
OJ (Q Of
l/l ( . OJ OJ >
E • -3 4 J U I • ! -C U I
o o U Q .
. c o
4 J
u I Q
^ I Q
(-O " > I Q
. C OJ
X I
— I Q
3 4 J
U I Q
J C 4 J
•^ X
;-o o Q .
4 J
c •*— I Q
— a . s
o u OJ
^* O L
•^ 4 - >
— 3 E
c • r -
OJ O l I Q
O l
c OJ
o 4 . )
> l
u c OJ
•o c OJ
4 . 1
I Q
to • 1 —
• U I
( . o
> I Q
J C
OJ
^
> l
"c o OJ
JZ 4->
l/l
OJ
u c OJ
•^ t . OJ
a . X OJ
O
-5 U I
c_ OJ E 3 to
c o o
c OJ OJ X
* J
OJ
^ OJ
o c OJ
(-OJ
<«-O I < 4 -
c • f —
> l 3
^ to 3 O
> OJ
(-O L
• p .
• o
4 . )
c I Q
u •^ >»-•^ c
c OJ
-5 4 . )
o c
o • o
o • i
OJ to
o JZ
*•>
• o
c I Q
c o •^ 4 J
o I Q
OJ E o U I
OJ O l . ^
•^ to I Q
4 J
l / l
OJ
u c OJ
(. OJ
>*~ •*-•.— T3
u .c
OJ
> I Q
X
.
. o 4 . 1
OJ
» c o
0 . 4 . ) I Q
t . O l
o E OJ
T3
4 J
to
£ •
"S • p .
>•-to
•^ 4 J
I Q
U I
U I
•^ -o
I Q
U 3
"S * OJ
O l IQ
* OJ
c o u c •^ U I I Q
X
u 3 to
. to * J
u OJ
1 4 -
l « -
OJ
4 - ^
o OJ
(-• r—
•o c •^
> 1
c o
1
X OJ
* > l
4 . ) • r -
c OJ
> OJ to
1 — X
o C -
Q .
x: 4 - >
•^ »
"O OJ to I Q
OJ C-
u c
• p .
z 1
o 1
3
<•-o
• o o o
X
•^ ,— OJ . X
p —
t . OJ
o —
• o c I Q
. U I
c o
*r—
* J 3
X •^ C-4 J 4 ^ I Q
f—
I Q
C t . OJ
4 . J
. >»
• o OJ
1 (-I Q
OJ
•o •r—
> o
o.
o 4 J
to to OJ
c O l c •^ ^ ,— •^ s to
— I . OJ
^-• -I Q
4 . 1
OJ
(-
1
e Q U
o 4 . >
>> 4 . > • p -
t f l
c 01
a. o c Q .
C
•^ c o •^ 4 . > I Q
•*— (-I Q
> OJ
.c 4 . 1
<*-o
»« 0 0
^r
OJ • o 3
4 . )
•^ 4 . >
4 . > I Q
TJ C
I Q
• k
OJ O l I Q
> i X
•o OJ c I Q
^ Q . X OJ
OJ X
• o
3 O
u
c ' r -
<9 ^m.
a.
. O l c
• p -
c •^ I Q
r ^
O . E o o U I
•o C -
i Q
3 O
4 . >
1 •r—
to 3
X
OJ X 4 . J
.c *J • r -
J
*-> c OJ
4 ->
c o o l O
>t—
X>
Ol L .
I Q
O
-i to l -OJ E 3 to
c o u
O l
c — OJ ^ TS <Q 3 E 4->
•^ C 4.1 • • - 4-1
I Q
to 4_l OJ
c > IQ 4-> t - IQ
4 J O l l/l Ol OJ c
(- IQ UI l/l Ol OJ >
. — IQ X
OJ
> -o — c OJ IQ
u ( - to
^t •9—
U I IQ OJ — U I C L to g OJ O
c o
. to to OJ
c *^-Ul 3
X
to
•o 1 .
I Q
X o
4 . )
T3 OJ
4 - >
I Q
O l • r— 4 . )
to OJ
> c 1 . ^
0 0
o < _ }
OJ • 4.> t o IQ C > O
• ^ »r~ C- 4 J Q . O
I Q
• o C 4.) IQ C
•^ O IQ • ^ r— — Q .
-c s 3 O a. o
<•-o to O •^ 4->
to •f—
t -OJ
4.) U IQ
(-IQ
^ O
OJ . X IQ
4.)
o -i UI I . OJ E 3 to c o o
•a c IQ
to • C 4.J o o
•w- C 4.>
o o IQ "O
4J O
= -s 10
>— OJ a . l/l O 4J
c to 3 4J
E C O IQ U >—
Q . • c E 4-> O 3 O O
1 o < 4 -
o c 1 o
• o •«-C- 4 . ) O IQ 3 U
^ OJ 111
( Q
C • ' —
to • o
O X OJ to 3
5
• < !g t o
•a • OJ «c
to t /1
to 4.) c
•r— . IQ t o
UI -f-OJ IQ
4-» O l
u c 3 ••-< - ^
4.) IQ
UI OJ o
• p . UI > OJ ( .
I— OJ . X l/l IQ
«- -o a IQ
to
OJ X o t-IQ OJ UI OJ
OC
« 5
c o to
£ TJ c . IQ OC
u c <: OJ
T3 • u ^ IQ 0 0 OJ Ol
0 0 PH
1
o (-C L - O •
c to to IQ OJ 3 O O l/l •<-< - 4 J > t_ o t -IQ 3 01
> "O UI
. 1 - 4
00 O l
O l • <J 3 C O IQ IQ L .
X E Q-C 4.> o f— a£ ( . IQ U
o INI <
•g . 10 l/l
OJ O l o
c c • 1 - IQ J C •>-4 J f— O Q .
•— Q . U IQ
_ l
I Q
. C
U I C -
£ *^'l
to - o
c •f— •
X m U 00
— O l 0 £ 1-1
( - OJ • 1 - o IQ - 1 -
O . > OJ ( -
C <u U I
o 4J T3 3 C
« t IQ
• o C IQ
« t
» o . — CO OJ O l OJ •—1
» — • 4->
C 4.1 OJ OJ
TJ ^ C- U IQ O OJ C-
• o C-
a .
1 0 0 O < |
• "S •r—
< 4 -
•^ o 4-> OJ O Q .
Z to
• 0 0 > , IQ <_> .— .— <_) P — T —
OJ o i - o ^ IQ C t /1 Q . IQ
•5° 5 oioo t_ 1 - o
4-) O UI t /1 •
•o ^ C -o r*. 3 C 0>
_ l IQ 1—1
. u O (-Q .
tr> vo 00
(B£ARC£N AND TEEL, J^. FEB., 1983, PP. 21-8)
4 3
COGNITIONS AFFECTIVE :JNAriVE
A€GAT!VE / DlSCONFIRh*-! I -HON OF \E/PECrATIONS
DISSATISFAC-•riON P<KCESS
<
lOM'LAINT 3f .-lAv ; Jfi ,
CQC^ITIONS •OT. . ir iON AFFECTlVE/COGNl':aNS >:Ot*AT|,E
(WY, ACR PROCEEDINGS, 198*4, PP. 496-9)
FIGURE 2.1
TWO COMPETING CONCEPTUALIZATIONS OF CONSUMER COMPLAINT BEHAVIOR
44
CN
CN
U
D
<
z
u X CL b] cn < X u a: D a.
I
cn o a.
a: a u.
u Q a x:
en en u u a cc a.
45
TO S E E K
R E
D
R
E
S S
T O
C H A N G E
F U T U R E
B E H
A V [
!
D Y
A D I C
T H I R D P
A R T Y
D Y A D I
r. T H I R D
P
A R T
Y
PJBLIC AND
FORfV\L ACTIONS
" V O I C E " COrf>LAINTS
TO RETAILER OR
VVNUFACTURER
LEGAL ACTIONS
INTERVENTION BY
PUBLIC OR COfJSUr AGENCIES
SEEK BETTER SERVICE
FROf R E T A I L E R /
riANUFACTURER
COfPLAIN TO oilBLIC
OR C"i'i<;irER
AGENCIES
^ P [ ' . " T XUD
Vpy^'Ai ACTIONS
"irJOR IXIDE'JTS
IN"A:L.'E T COST
rp ccrr.pT
1
SEE" HELP OF
"^ 'ENDS AND/OR
" E X [ - ' = CHANGE
PAT^~-WGE
"iiORD-OF-MDirTVi"
COfTIUNICATION
FIGURE 2.3
A CLASSIFICATION SCHEMA FOR CONSUMER COMPLAINTS
CHAPTER 3
THE CONCEPTUAL FOUNDATIONS OF A HOLISTIC MODEL
OF CCB
The development of the holistic model of CCB Is
based an attempt to bring together the various conceptual
streams of CCB into a holistic framework. However, the
development of a general model presumes the existence of
some general laws that do not vanish from product to
product, service to service, or situation to situation.
Empirical support for the existence of such general
relationships is found in a recent study by Richins
(1983), who found that the relationship of problem
severity and retailer responsiveness to complaint actions
was not significantly different across product groups.
Conceptual Foundations of the Model
A holistic framework for the CCB process is
presented in figure 3.1. Consistent with the earlier
discussion on the conceptualization of CS/D and CCB, CCB
is postulated as a process triggered by CS/D. The
intensity or level of consumer dissatisfaction is assumed
to be the underlying motive which determines the amount
of effort expended in the CCB process (Richins 1983; Day
1984). However, since the level of dissatisfaction is
proposed to have no direct impact on the nature and kind
46
47
of complaint actions chosen by the consumer (Fornell and
Didow 1980), CS/D is not explicitly represented in Figure
3. 1.
Complaining intentions can be conceptualized as the
likelihood that any particular complaint action would be
chosen. Therefore, intentions to engage in specific
complaint actions are proposed to have a one-to-one
correspondence with actual complaint behavior. However,
situational constraints such as cost and frequency of
purchase may result in complaint behaviors that are
inconsistent with intentions (Richins 1982).
It is also proposed that there are two routes that
lead to complaining intentions. One route represents the
psychological perspective that a consumer's attitude
toward the act of complaining is related to his/her
intentions (Richins 1982; Day 1982; Bearden and Crockett
1981). The more positive the attitude, the greater the
likelihood of complaint actions.
The second route represents the economic framework
of Hirschman (1970) based on the concept of expectancy
and value for each alternative course of complaint
action. It is proposed that expectancy-value Judgments
directly affect the consumer's intentions to engage in
corresponding complaint actions. That is, the likelihood
of intentions is greater for an action that has a higher
expectancy-value attached to it. This conceptualization
48
assumes a multiplicative rule for combining expectancy
and value judgments. Other information processing rules
(lexicographic, etc.) are plausible and should be
investigated in future research.
Generalized affective feelings pertaining to
consumption activities are postulated to be antecedent to
the attitude construct and have no direct affect on the
intentions construct. Westbrook (1980) and Day (1980)
suggest that consumer discontent and alienation from the
market place fit the definition of the generalized
affective feelings construct. Research findings show
that whereas consumer discontent has a positive relation
ship with attitude towards the act of complaining,
alienation from the market place has a negative relation
ship (Bearden and Mason 1983; Lundstrom et al. 1979).
While it is hypothesized that consumer discontent is a
concept distinct from alienation and probably inversely
related to it, it is probable that the attitude toward
the act of complaining is related to discontent and
alienation in a nonlinear fashion. That is, attitudes
toward the act of complaining are negative for both, low
levels of discontent and high levels of alienation, with
positive values corresponding to region between these two
extremes. These generalized affective feelings of
discontent and alienation are in turn proposed to be a
function of personality characteristics, demographics,
49
and the socio-political environment (Jacoby and Jaccard
1981; Lambert 1981).
At least two determinants of expectancy of alterna
tive complaint actions are proposed: attributions of
dissatisfaction and prior experience in making
complaints. The attributions construct is conceptualized
based on attributional theory of CCB discussed in the
last chapter. That is, external attributions with a high
degree of stability and controllability would result in
higher expectancies of seller responsiveness attached to
these dissatisfactions. Similarly, consumers with
greater prior experience in making complaints would have
higher subjective probabilities that future complaint
actions would be successful. Neither the attributions
nor the the prior experience have a direct affect on
intentions but instead affect intentions indirectly
through their effect on the expectancy-value construct.
It is also proposed that "structural constraints"
(Fornell and Robinson 1983) which reflect the nature of
the industry may affect the expectancies of the various
complaint actions. In the model of figure 3.1 these
structural constraints are contained in the environmental
conditions prevailing for a given purchase episode.
Thus, when consumer dissatisfactions result from products
or services that are produced by an industry that is
known to discourage complaints, the expectancies of
50
complaint actions would be correspondingly affected.
This link is based on the propositions of Hirschman
(1970).
The value construct attached to each of the alterna
tive complaint actions is proposed to be a function of
prior knowledge, personality characteristics, and
demographics.
If the expectancy-value Judgments and attitudes are
two routes to the determination of CCB intentions, it is
legitimate to ask: How the expectancy-value Judgment and
attitude constructs interact to determine intentions?
Which route is the dominant mode in a specific situation
of dissatisfaction? Since a general model that incorpo
rates the psychological and the economic perspectives has
not been proposed before, no conceptual or empirical
answer is suggested by the CCB literature.
However, research on the role of involvement and
prior knowledge in multi-attribute models can provide
some theoretical guidelines to answer this question.
Bagozzi (1983) has proposed a typology that specifies
conditions when expectancy-value Judgments, attitudes, or
both together would be the predominant modes of the
underlying process that result in purchase intentions.
His typology uses the criteria of involvement and prior
knowledge (or learning). The concept of involvement as
used here implies the extent of effort expended in the
51
particular purchase activity. The corresponding concept
in the CCB process is the level of dissatisfaction since
it too represents the amount of effort that will be
invested in the complaint decision making process
(Richins 1984). The concept of prior knowledge is
similar in both areas, though it is more specific in the
case of CCB process, where it refers only to complaint
actions. Therefore, it can be hypothesized that a
typology similar to that proposed by Bagozzi (1983) may
be suitable for CCB process, when the level of dissatis
faction is substituted for involvement. This typology is
represented in Figure 3.2. It is reiterated that this
typology is only a hypothesis based on an extension of
general findings in the area of multi-attribute models,
and its applicability to CCB process is open to empirical
investigation.
The interpretation of Figure 3.2 is fairly
straightforward. For instance, when the level of dis
satisfaction is either medium or high and the consumer
has no previous complaint experience and knowledge (cells
2 and 3), the predominant mode of the CCB process is
hypothesized to be the expectancy-value Judgments.
Similarly, cells 5 and 6 suggest that with substantial
prior knowledge and learning about making complaints, the
tendency to adopt the attitude route for CCB actions is
high for moderate or high levels of dissatisfactions.
52
For low levels of dissatisfactions, the motivation to
expend effort in the CCB process is missing, and the
predominant mode is either impulse complaint behavior, or
a habitual response depending on the level of prior
knowledge and learning.
One reason for developing a theoretical model of the
CCB process is to facilitate predictions of specific
complaint actions (exit, voice, word-of-mouth, etc. ).
Day (1980) states that such an objective is desirable but
suggests no guidelines for making specific predictions.
Nevertheless, at least some predictions of complaint
actions at a level of specificity greater than general
complaint behavior can be made based on current research
findings. These hypothesized predictions result from
either high or low levels of expectancy value Judgments
and either positive or negative attitudes towards the act
of complaining. This represents a 2 X 2 table of
possible outcomes and is shown in Figure 3.3. The
specific prediction in each of the cells is based on the
work of researchers shown in the respective cells.
Specifically, Figure 3.3 hypothesizes that public actions
of VOICE as well as private actions of WORD-OF-MOUTH
communication would be used under conditions of high
level of expectancy-value Judgments and positive
attitudes toward the act of complaining (cell 1).
Similarly, private actions of WORD-OF-MOUTH communication
53
and EXIT are hypothesized to be expected when attitudes
are positive but expectancy-value Judgments are low. NO
ACTION is hypothesized to be the most probable response
under conditions of low expectancy-value Judgments
combined with negative attitudes towards the act of
complaining (cell 4). Finally, cell 2 indicates a pheno
mena that has not been investigated before, that is,
conditions of high expectancy value Judgments but
negative attitudes toward the act of complaining. It is
hypothesized that under the conditions of cell 2,
consumers would either VOICE their complaint (because of
high expectancy-value) or take NO ACTION (because of
negative attitudes). Again, it is important to note that
these are hypothesized predictions which will be tested
in the present research and that previous studies in the
area have not directly examined these relationships
within a single study.
A holistic model for the consumer complaint process
is proposed that brings together the economic perspec
tive, psychological perspective and the attribution
theory in a well specified framework. However, in order
for this to be useful, this model must be specified in
more detail to enable further theoretical development and
empirical testing. This task is referred to as theory
formalization in the philosophy of science literature and
is addressed in the next section.
54
Partial Formalization of the Model
Definitions
1. The purchase episode is the consumer's entire
experience in the purchase and consumption of a
particular product or service. Subsequent evalu
ations of additional experiences with the same
product/service are to be treated as new episodes
(Day, 1980, p. 211).
2. Dissatisfaction is a feeling or emotion triggered by
the consumer's comparison of the rewards and costs
of the purchase in relation to anticipated conse
quences within a purchase episode (Day, 1984, pp.
496-7).
3. Consumer Complaint Behavior (CCB) is the set of all
non-behavioral and behavioral responses which
involve communicating something negative regarding a
purchase episode including the product/service that
is triggered by dissatisfaction. The complaint
behavior is conceptually distinct from and not
necessarily related to the intensity of dissatisfac
tion with the purchase episode (Day, 1980, p. 211;
Day, 1984, p. 497; Jacoby and Jaccard, 1981, p. 61;
Fornell and Didow, 1980, p. 319). Further, the
process leading from the dissatisfaction stage to
the complaint response is called the consumer
complaining process.
55
4. The energizing components or the dynamic aspects of
human personality that activate and sustain a
process and account for its termination are the
motivating forces of the process (McGuire, 1976, p.
302).
5. Attitude is the enduring positive or negative
feeling about some person, object or issue (Petty
and Cacioppo, 1981, pp. 6-7).
6. Factual statements that a person perceives about
other people, objects or issues are termed as
cognitions. Cognitions are action neutral, i. e. ,
they are not charged with attractive and repulsive
characteristics of affect and may be proven to be
true or false in an objective sense. Common examples
of cognitions are beliefs, expectancies and
subjective probabilities (Bagozzi, 1980, pp. 40-2).
7. The subjective probability of an outcome or
expectancy is the cognition a consumer has that a
particular course of action or an event will have
some specific consequences (Landon, 1980, p. 335;
Fornell and Didow, 1980, p. 319).
8. The rational evaluation of the degree of desira
bility or undesirability of some specific conse
quence is termed as the value of that consequence
(Landon, 1980, p. 335; Bagozzi, 1980, pp. 41-2, 47).
56
Axioms
1. The cognitive consistency theories state that there
is a strong tendency for people to maintain
consonance (consistency) among the elements of
cognitive system, i.e., beliefs, subjective
probabilities, expectancies, affect and values.
These theories:
(a) attempt to describe the conditions for
equilibrium and disequilibrium among cognitive
elements,
(b) assert that disequilibrium motivates the person
to restore consistency among the elements, and
(c) describe procedures by which equilibrium might
be accomplished (Petty and Cacioppo, 1981, pp.
126-7).
2. The three major theories of cognitive consistency
are Balance Theory (Heider, 1958), Congruity Theory
(Osgood and Tannenbaum, 1968), and Cognitive
Dissonance Theory (Festinger, 1957).
3. Learning theories attempt to explain the conditions
and the processes by which relatively permanent
changes in the response tendencies including
attitudes, beliefs, evaluations, etc. result from
the effects of prior experiences. Common theories of
learning are Classical Conditioning (Staats and
Staats 1957; 1958), Operant Conditioning (Skinner
57
1938) and Cognitive Learning (Engel and Blackwell,
1982, pp. 236-63).
4. A consumer's dissatisfaction with a purchase
episode is the net response resulting from the
expectation level, plus or minus a magnitude of
disconfirmation. This disconfirmation of expec
tations as a phenomena is explained by many theories
such as Adaptation Level Theory which proposes that:
(a) Expectations are individual's perceived
Judgments of how a product/service should per
form based on a number of factors including
prior experiences, word-of-mouth, manufac
turer's reputation and advertising, etc.
(b) The perceived performance is the result of
experience with the current purchase episode.
(c) Disconfirmation is the distance or the discre
pancy between (a) and (b) above. Complaining
behavior only results when (a) is greater than
(b) (Oliver, 1980, pp. 206-9).
Law-Like Statements
1. If a consumer X is dissatisfied with a particular
purchase episode, then, a necessary but not
sufficient condition for engaging in complaint
response exists (Fornell and Didow, 1980, p. 319).
2. Consumer complaint responses result from specific
aspects of experiences with particular products or
58
services (as opposed to generalized feelings about
the market system) (Day, 1980, p. 211).
3. Consumer complaint response in specific situations
is a function of (a) intentions to engage in
complaining responses in order to obtain desirable
payoffs (Landon, 1977, p. 33), and (b) certain
situational factors such as the item's cost, and
frequency of patronizing (Richins, 1982, p. 503;
Landon, 1977, p. 33).
4. Consumer complaint intentions result from one or
both of the following conditions:
(a) psychological reactions favorable or unfavor
able to taking actions, i.e., attitudes toward
the act of complaining (Day, 1980, p. 214).
(b) a rational analysis of the benefits and useful
ness of taking any plausible courses of action
(i.e. expectancy-value) of each of the possible
actions (Day, 1980, p. 214).
5. The expectancy value of each of the plausible
courses of action (in a specific experience) is a
multiplicative function of both:
(a) the value gained from a successful action, and
(b) the probability or expectancy of achieving a
successful action (Hirschman, 1970;Landon,
1980, p. 335; Fornell and Didow, 1980, pp. 318-
19).
59
6. Previous experience in and knowledge about (a)
seeking redress in similar or other situations, (b)
buying and using the particular product/service and
(c) sellers policies, laws and consumerism are
determinants of a consumer's estimate of the proba
bility of achieving a successful complaint action
(Day et al., 1981, pp. 95-6). This generalization is
supported by theories of learning (e.g., classical
conditioning and cognitive learning) (Bagozzi, 1982,
p. 572).
7. Affective feelings or attitude toward the act of
complaining will directly affect complaining
intentions. This generalization is due to theories
of motivation, learning and purposeful behavior.
Consumers will be motivated to engage in those
actions that lead to satisfaction of needs and to
avoid those actions which are aversive (Bagozzi,
1982, p. 574).
8. The direct path from expectancy value Judgments to
complaining intentions results from a rational
consumer's choice based on preferences or values of
all alternatives and the availability of those
alternatives (probability of successful actions).
This is the economic theory of consumer choice
applied to post-purchase processes (Fornell and
Didow, 1980, p. 318-9).
60
9. The relationship between attitudes toward the act of
complaining and expectancy value Judgments of alter
native complaining actions is characterized by the
following process:
(a) They would tend to remain in balance driven by
cognitive-affective consistency requirements
(Bagozzi, 1982, p. 574).
(b) Affect can be expected to influence expectancy
value processes in contexts in which (i) the
behavior is impulsive or (ii) the affective
reaction originates from a relatively strong
arousing stimulus (Bagozzi, 1982, p. 574).
(c) On the other hand, expectanc/ value Judgments
of complaining actions can be expected to
influence attitudes toward the act of com
plaining in instances when the purchase episode
evokes beliefs and evaluations of actions/-
consequences that are cognitively arousing. The
mechanism of this influence is provided by
balance theory (Heider 1958) and the drive to
maintain cognitive-affective consistency
(McGuire, 1968).
10. Relatively stable influences pertaining to aspects
of the marketing system and the domain of consump
tion (e.g., the goods and services offered in the
market place, business practices, attitudes towards
61
the business, consumerism as well as sentiments of
pervasive consumer discontent), are distinct from
transient affective influences (i.e., temporary
favorable or unfavorable sentiments evoked by
specific purchase episodes). However, the former
have predictive relationship with the latter (Day,
1980, p. 211; Westbrook, 1980, p. 50).
11. The stable affective influences pertaining to
aspects of the marketing system and the domain of
consumption result from or are correlated with:
(a) Demographic characteristics including age,
income, social class, family life cycle, etc.
(Bearden and Mason, 1983, p. 6-7).
(b) Personality variables including assertiveness,
aggressiveness, etc. (Richins, 1983, p. 73-4;
Fornell and Westbrook, 1979, pp. 105-6).
(c) The cultural, economic and political environ
ment (Day et al., 1981, pp. 99-104).
12. The extent of dissatisfaction with the purchase
episode and the importance of the product/service
are the underlying motivating forces of the consumer
complaining behavior (Day, 1984, p. 497).
Discussion
This chapter first proposed a holistic model of
consumer complaint behavior that would incorporate within
62
a single model previous empirical findings and the
different conceptual frameworks for the study of CCB.
Second, the proposed model was partially formalized to
enable further theoretical development and empirical
testing. It is not claimed that the proposed holistic
model is "correct"--only empirical testing can Justify
the truth content of a model or theory. However, the
model is (a) consistent with past research, (b) helpful
in summarizing past research, (c) useful in providing
directions for future research, and (d) amenable to
empirical testing. This dissertation proposal presents a
framework for the empirical investigation of a part of
this model. The next chapter discusses the operationali
zation and development of hypotheses followed by research
methodology to be adopted.
63
—1/»
a:
D c/i ft: z o o u
n U] at: ID a
o
> < X UI OQ
- ] U ID Q Z a M E Z U M H cn M .J o X
<
a. o u
64
LEV
EL
OF
D
ISS
AT
ISF
AC
TIO
N
LOW
5
5 X
PRIOR EXPERIENCE OF Q>f>L;\iNiNG
LOW
IffVLSE BEHAVIOR
EXPECTANCY-VALUE
JUDGErtNTS
EXPECTANCY-VALUE
JlJDGE^€NTS
M I ^
HABITS
ATTITUEE -••--=13 THE
ACT OF C O M = ' _ - ; N I N G
ATTITUDE TOWARD THE
ACT OF COMPLAINING
FIGURE 3.2
A TYPOLOGY FOR THE NATURE OF CCB PROCESSES
65
z
1
.a J
J
E
PO
SIT
IVE
N
EG
ATI
VE
EXPECTANCY-VALUE ,IUDGEA€VTS
HIGH
'VOICE'
WORD-OF-MDUTH C0^nJNICATI0N
corf LA IN TO PUBLIC AGENCIES
7
'VOICE' OR 'EXIT'
LOW
'EXIT '
WORD-OF-MOUTH co^^UJICATION
NO ACTION
FIGURE 3.3
A MODEL FOR PREDICTING SPECIFIC CONSUMER COMPLAINT ACTIONS
CHAPTER 4
OPERATIONALIZATION AND THE DEVELOPMENT OF KEY HYPOTHESES
In chapter 3, a theoretical framework is developed
based on current literature in the CS/D and CCB area.
Then a formalization of this theory is attempted to
rigorously define the relationships among the various
"terms" and "concepts" in the theory. However, theories
that purport to explain and predict phenomena must
satisfy the "empirically testable" criterion (Hunt 1983,
pp. 243-8). That is, the relationships predicted by
theory ought to be empirically observable and either
supported or rejected based on these observations. Since
theories, being generalized conditionals, are not
directly testable, the task undertaken in this chapter
is to derive predictive type statements from the theory
that can then directly confront empirical data. These
predictive type statements derived from the theory are
referred to as Hypotheses. The confrontation of
hypotheses with empirical data would lead to either
rejection or acceptance of these hypotheses, which then
suggests an empirical evidence of lack of support or
corroboration of theoretical relationships.
Hunt (1983) proposes the philosophy of science
perspective on deriving hypotheses:
66
67
The requirement that research hypotheses be predictive type statements that are amenable to direct confrontation with data implies that all of the descriptive terms in the statements (or theory) must have rules of interpretation containing empirical referants (sometimes referred to as operational definitions, empirical indicators or epistemic correlations).
In other words, first the concepts or constructs in
the theory must be operationalized into empirical
referants or entities and then predictive type statements
called hypotheses can be stated in terms of these
empirical entities. Therefore, hypotheses are statements
of predicted relationships among empirical entities and
only indirectly refer to any concepts or constructs in
the theory.
What are these empirical entities? Bagozzi (1980)
suggests that empirical entities or concepts are observa
ble concepts that achieve their meaning through
operational definitions that specify procedures for
measuring observations in the world of experience. In
other words, operationalization of theoretical constructs
implies a definition of an empirical concept that has
correspondence with the theoretical construct as well as
the procedures for measuring the empirical concept.
Thus the purpose of the chapter is twofold: First,
to operationalize the theory of consumer complaining
behavior and the constructs therein; and second, to
develop specific hypotheses based on these
operationalizations.
66
Operationalization of Key Constructs
One desirable criterion for the operationalization
of theoretical constructs is to incorporate corres
pondence rules for individual concepts that have either
been employed or suggested in the literature. This would
assist in tying current research with previous research
findings and thus be useful in the accumulation of the
body of knowledge in the area. Therefore, the objective
in this section is to suggest operationalizations of the
concepts in the holistic model of consumer complaining
behavior that meet the above criterion and hence provide
a link with previous research.
The Dissatisfying Experience
The theory conceptualization defines the consumer's
experience of dissatisfaction with a purchase episode as
the trigger that activates the process of consumer
complaining. However, consumers experience different
dissatisfactions with different products and services.
Bad appliances can be repaired, whereas "bad" health
services can neither be "repaired" nor "refunded" in the
true sense. Therefore, the operationalization of dis
satisfying experience ought to involve a sampling across
products or services as well as a sampling across expe
riences.
Most of the research in this area operationalizes
69
the dissatisfying experience using recall information for
the "recent dissatisfying experience" (Richins 1983;
Bearden and Mason 1983; Day et al. 1981; Villarreal-
Camacho 1983; Fornell and Westbrook 1979; Andreasen 1977;
Best and Andreasen 1975). It is expected that dissatis
fying experiences are not often forgotten, irrespective
of action taken. Therefore, recall is used as the opera
tionalization for the construct of the dissatisfying
purchase episode. The expectancy value Judgments and
intended actions are then measured relative to a similar
but hypothetical incident occurring again in the future.
Four service groups will be investigated in the
present research: health services, auto-repairs, banking
services and grocery retailing. Auto-repairs are reported
to elicit a large number of voiced complaints, whereas
the response to dissatisfaction with grocery retailers is
generally to exit (Best and Andreasen 1975). Banking has
rarely been investigated while health services are sug
gested to restrict both the voice and the exit options of
dissatisfied consumers (Andreasen 1983). Therefore, this
selection of service groups seems to provide some varia
tion in the complaint responses and, hopefully, in the
underlying mechanism of the CCB.
Level of Dissatisfaction
The level of dissatisfaction is conceptualized in
the theory as the underlying motivating force in the CCB
70
process. It is not, however, directly modeled in the
process of complaining.
Westbrook (1980a) investigated the different
operationalizations of the satisfaction/dissatisfaction
construct. Based on this study, a percentage scale is
selected to measure the level of dissatisfaction. Since,
from a theoretical standpoint, initial feelings of
dissatisfaction are distinct from final reaction, the
percentage scale is used to measure the two levels of
dissatisfactions individually. The measurement property
of the percentage scale of dissatisfaction has been
discussed by Westbrook (1980a) and its reliability ranges
from 0.65 to 0.88 depending on the kind of product.
Prior Experience
The operationalization of prior experience as a
consumer and in complaining actions is suggested by Day
(1984). The operational measure is a four item scale
which taps the extent of previous experience respondents
have had in the various complaint actions. Specifically,
the scale attempts to measure the frequency with which
respondents have engaged in VOICEing their complaints
either directly to stores or manufacturers, in taking
FORMAL actions, such as to report to the Better Business
Bureau, in Word-of-Mouth communication to friends and
relatives and in taking LEGAL actions. The higher the
71
value of the index on this measure, the higher is the
prior experience of engaging in consumer activities and
complaining behavior. The measurement properties of this
scale will be investigated as a part of this research.
Stable Affective Influences
The construct of stable affective influences is
defined as those attitudinal structures that pertain to
the domain of consumption or market-place, are relatively
stable over time, and are not specific to an incident or
a situation. One operationalization of this concept sug
gested by Westbrook (1980) is the 82-item consumer
discontent scale developed in the marketing literature by
Lundstrom and Lament (1976). Consumer Discontent is
defined to be:
the collection of attitudes held by the consumer toward the product strategies of business, business communications and information, the impersonal nature of business and retail institutions, and the broader socio-economic forces which are linked with the business system (Lundstrom and Lament 1976, p. 374).
Thus the attitude of consumer discontent is fairly
generalized, pertains to the marketing system, and
appears to be generally stable over time (Westbrook
1980).
Another operationalization of the stable affective
influences that has often been used in the CCB literature
is the 35-item consumer alienation from the market-place
scale developed by Allison (1978). The operational
72
concept, consumer alienation is defined as:
the feelings of separation from the norms and values of the market-place. Such a state includes a lack of acceptance of or identification with market institutions, practices, and outputs as well as feelings of separation from the self when one is involved in the consumption role (Allison 1978, p. 570).
This conceptualization fits well the definition of stable
affective influences pertaining to the domain of
consumption.
These two operationalizations of stable affective
influences, namely discontent and alienation, are
frequently employed in the CCB research and appear to be
useful in explaining the CCB process (Bearden and Mason
1983, Day 1984). Thus both operationalizations will be
used in the present study. The alpha reliability of the
discontent and alienation scales is reported to be 0. 94
(Lundstrom and Lament 1976) and 0.8802 (Allison 1978),
respectively.
A particular operational problem with the use of
consumer discontent and alienation from the market-place
scales in survey research is the long length of the two
instruments. This is especially so for the 85-item
consumer discontent scale. It would be helpful if the
instrument could be reduced to manageable length without
losing much of the correspondence between the theoretical
construct and the empirical measure. This research will
also attempt to do so, as detailed later.
73
Attitude Toward the Act of Complaining
Richins (1982) has attempted to operationalize the
theoretical concept of psychological costs/benefits and
favorable or unfavorable reactions, that is, attitudes
toward the act of complaining. Richins's (1982) empirical
instrument has 15 items and purports to measure 3 dimen
sions of the attitude concept. The three dimensions are:
(a) generalized attitudes regarding the trouble involved
in complaining, (b) a personal norm concerning com
plaining, that is, appropriateness or inappropriateness
of complaining as a normative issue, and, (c) societal
benefits of complaining, that is, feelings regarding the
positive or negative impact on society of registering
complaints.
Richins (1982) has demonstrated external validity of
this operational measure by empirically supporting its
predictive relationship with propensity or intentions to
complain. Other researchers have also investigated the
measurement properties for similar measures of the
attitudes towards the act of complaining scale (Bearden
and Mason 1983; Bearden and Crockett 1981).
Day (1984) has proposed an operational measure of
this construct which shares some items with the Richins
(1982) operationalization. Day (1984), however, shows no
empirical support for the reliability and validity of his
74
operational measure.
It is proposed to operationalize the attitude toward
the act of complaining construct with an 18 item scale
composed of items from Richins (1982) and Day (1984).
This measure is composed of three facets similar to those
described by Richins (1982). The general facet has 10
items, and the personal norm and societal benefit facets
have four each. The possibility of an underlying general-
factor in these 18 items is unknown and is proposed to be
investigated as a part of this research.
Attributions of Blame
Krishnan and Valle (1979), Foikes (1984) and other
researchers have operationalized the construct of
attributions within the context of consumer complaining
behavior. The theoretical construct of attributions,
which implies imputing causal inferences to felt dis
satisfactions, is operationalized by Foikes (1984) to
measure its 3 dimensions. These dimensions are: (a)
locus, that is, cause of dissatisfaction located in the
consumer or external to him, (b) stability, that is, the
relatively temporary or fairly permanent nature of the
cause of dissatisfaction, and, (c) controllability, that
is, the extent of volitional or nonvolitional nature of
the cause. Krishnan and Valle (1979), however, operation
alized only the locus dimension of the attributions and
found its effects on CCB significant. Foikes (1984) in
75
her study found that the locus and stability dimensions
were most useful in explaining redress/future behavior
types of consumer complaining behavior.
Therefore, for the purpose of this study a five item
scale measuring the locus and stability dimensions of
attributions is defined as the operational referant of
the attributions construct. The scale is based on the two
studies cited above, but is adapted to suit each service
industry investigated.
Expectancy-Value Judgments
The construct of expectancy value Judgments has been
theoretically defined to be a multiplicative function of
value or benefit gained from a course of action and the
subjective probability that such an action would be
successful. Though the importance of this construct in
predicting and explaining consumer complaining behavior
was suggested in 1970 by Hirschman, few empirical
measures exist in the CCB literature. Therefore, it is
necessary that operationlizations of this construct be
developed for the purpose of this study.
Fortunately, guidelines for developing this opera
tional measure are available from the area of multi-
attribute models in consumer behavior. A large number of
empirical studies have operationalized the expectancy-
value construct for product choices (Bagozzi 1982).
76
Essentially, the operationalization ought to measure
the subjective probability of success and the importance
for each of the alternative courses of actions available.
A 15 item scale is thus developed for measuring the
expectancy and value Judgments separately. These items
are classified into three sets reflecting Judgments in
case of (a) word-of-mouth communication to friends and
relatives, (b) reporting to the seller or manufacturer,
and (c) complaining to a consumer organization or public
agency. The specific items in each of these sets reflect
the seeking redress and changing future behavior
dimensions of the various actions. However, expectancy-
value (E-V) Judgments for exit or no action are not
measured since they are expected to result from low
values of E-V Judgments for each of these three sets. The
measurement properties of the E-V scale are proposed to
be investigated as a part of this research.
Complaining Actions
The construct of complaining actions is directly
observable and thus requires no correspondence rules for
connecting it with some empirical concept. It is itself,
by definition, an empirical construct. However, as
defined in the classification schemata there are certain
actions such as word-of-mouth communication to friends or
relatives, or a personal decision to exit, that are
difficult, if not impossible, to observe directly. In
77
such cases, we are constrained to accept self-report data
on certain measures as valid representations of the
construct.
One such measure for complaining actions is proposed
by Day et al. (1981). This nine-item scale attempts to
measure responses to each of the nine possible courses of
complaining actions. Bearden and Teel (1983) operation
alized the complaining behavior construct based on the
measure of Day et al. (1981) using a five-item scale.
They also investigated the measurement properties of the
scale and demonstrated that it conformed to a Guttman
conceptualization. That is, increasing agreement with the
items reflects an increasing intensity of engaging in
complaining actions. The coefficient of reproducibility
is reported to be 0.98 with an index of scalability of
0.78 (Bearden and Teel 1983).
For the purposes of this research, the operation
alization of complaining behavior is a twelve-item scale
that is based on the five items of Bearden and Teel
(1983) and nine items of Day et al. (1981). The scale is
conceptualized to be Guttman type, and its measurement
properties are proposed to be investigated as a part of
this research.
Summary
The operationalizations of the various constructs in
78
the proposed holistic model of CCB are summarized in
Table 4.1. Other constructs in the theory but not
investigated in the present research are not discussed
here. Some operationalizations for these constructs are
available in the literature, such as for situational
variables (Richins 1982). In general, the research in
this area is still in its infancy and a focused study of
the key constructs in the theory would contribute to the
development of the body of knowledge.
Based on the above operationalizations of the
salient constructs, the next section states and discusses
the key hypotheses to be investigated in the present
research.
Key Hypotheses
The first part of the study investigates the
relationship among consumer discontent, alienation from
the market-place and attitude toward the act of
complaining. Lundstrom and Lament (1976) as well as
Lundstrom et al. (1979) show that consumer discontent is
conceptually distinct from alienation and inversely
related to it. In other words, consumers with high levels
of discontent are not necessarily alienated from the
market-place (Lundstrom et al. 1979, p. 154).
Yet Allison (1978) and Lambert (1980) among other
researchers suggest that consumer discontent and
alienation are similar concepts and are positively
79
related. An empirical study that incorporates both these
measures and directly addresses the question of the
relationship between discontent and alienation has not
yet been attempted.
Therefore, it is hypothesized that:
HI: The constructs of consumer discontent and alienation
from the market place are distinct concepts, possess
divergent validity, discriminant validity, and are
inversely related.
Research by Bearden and Mason (1983) and others
shows that alienated consumers express feelings of
helplessness, powerlessness, meaninglessness,
normlessness and cultural estrangement in their interac
tion with the market-place. Thus, these alienated
consumers feel that any actions of complaining against
businesses would be unrewarding and fruitless.
Other empirical findings, for instance by Lambert
(1980), suggest that alienated consumers have higher
feelings of dissatisfaction with the market-place and may
complain vociferously to acquaintances, friends,
relatives, consumer organizations, and public agencies.
They may even take legal action.
Thus it is hypothesized that:
H2: The greater the perceived alienation from the market
80
place, the less positive the attitudes towards the
act of complaining.
Many empirical studies have investigated the rela
tionship between consumer discontent and the complaining
behavior. Westbrook (1980) found that consumers with
lower feelings of discontent had higher levels of satis
faction with automobile products. Lundstrom and Lament
(1976) reported that a highly consumerist group with
higher propensity of engaging in complaining behavior
also had higher levels of discontent with the market
place.
However, theory suggests that stable affective
feelings such as consumer discontent influence com
plaining behavior through their effect on attitudes
toward the act of complaining. Little empirical evidence
is available regarding the relationship between the
affective feelings of consumer discontent and the
attitude concepts.
Therefore, it is hypothesized that:
H3: As the discontent with the businesses increases, the
attitudes toward the act of complaining tend to be
more positive.
Hirschman (1970) suggests that the value gained from
a successful complaint times the probability of that
successful outcome, that is, E-V Judgments, determine the
81
probability of voicing complaints. From a theoretical
standpoint, Fornell and Didow (1980) and Landon (1980)
support Hirschman's prediction. Day (1980) has also
suggested that consumer complaining behavior may be
partly predicted by a rational analysis of the benefits
and usefulness of each of the alternative courses of
action. Yet, no empirical investigation of this predic
tion is reported in the literature.
Thus it is proposed to empirically examine:
H4: The higher the expectancy-value Judgments of each
of the various courses of complaining actions, the
stronger the intentions to engage in complaining
behavior.
Richins (1982) empirically investigated the rela
tionship between attitudes toward the act of complaining
and the actual complaining behavior. She found that atti
tudes are significantly related to both the propensity to
complain and self reported complaint behavior. Other
researchers (Bearden and Mason 1983; Bearden and Crockett
1981) have also found empirical support for this rela
tionship.
From a theoretical standpoint. Day (1980) and
Richins (1979) suggest that attitudes toward the act of
complaining will directly affect complaining intentions
due to theories of motivation, learning, and purposeful
82
behavior. Thus for instance, if consumers have positive
feelings towards complaining, they will be motivated to
engage in such behavior.
Therefore, it is hypothesized that:
H5: The more positive the attitude towards the act of
complaining, the greater the tendency to engage in
complaining actions when faced with a dissatisfying
experience.
Day (1981, 1980) has theoretically explored the
issue of how the expectancy-value Judgments and attitudes
interact to determine intentions to complain. He posits,
"then it might be feasible to combine the two indexes
(the E-V Judgments and the attitudes) to predict whether
or not action would be taken" (Day 1980, p. 215). Yet he
provides no practical guidelines to combine these two
concepts. Other researchers have usually ignored the
issue.
However, this problem of combining E-V Judgments and
attitudes is addressed in the area of consumer behavior
in the case of product or brand choice. Bagozzi (1983)
has attempted to resolve the issue using the concepts of
involvement with the product or service and prior know
ledge or learning. His conceptualization has been
modified to be applicable to CCB using the concepts of
level of dissatisfaction and prior experience of com-
83
plaining. This modified conceptualization is discussed
in chapter 3 as a part of the development of the holistic
model of CCB.
Based on Figure 3.2, the hypothesized typology of
the predominant mode of CCB process, the following
hypotheses can be developed.
H6: For moderate to high levels of dissatisfaction, the
greater the prior knowledge and experience in
complaining, the stronger the relationship between
attitudes and intentions, and the weaker the
relationship between expectancy-value
Judgments and intentions.
H7: Similarly, for moderate or high levels of dissatis
faction, the lower the prior experience and knowle
dge in making complaints, the weaker the relation
between attitudes and intentions, and the stronger
the relationship between expectancy value Judgments
and intentions to complain.
H8: For lower levels of dissatisfaction, the intentions
to engage in complaining behavior are dependent
either only on prior experiences (i.e., habits) or
on impulse reactions.
The effect of attributions of product failure on
consumer complaining actions was empirically investigated
by Foikes (1983). She concluded that predictive ability
84
of attributional approach is limited, though attributions
of product failure may seem to be more useful as deter
minants of expectancies of various outcomes. Richins
(1979) also theoretically posed that expectancy-value
Judgments may result from such attributions. No empirical
investigation of this theoretical link has yet been
reported.
Thus it is hypothesized that:
H9: The greater the dissatisfaction is attributed to
the members of the distribution channel, rather than
to the consumer (external attributions), the higher
the expectancy-value Judgments perceived by
consumers. These attributions have no direct affect
on the intentions to complain but only affect
intentions indirectly through expectancy value
Judgments.
It is also hypothesized that feelings of consumer
discontent and alienation from market-place directly
affect attitudes but have no direct affect on intentions
to complain. Accordingly, Bearden and Mason (1983) found
alienation a poor predictor of consumer complaining
actions.
Therefore, it is hypothesized that:
H10: The feelings of alienation from the market-place
85
and consumer discontent have only indirect effects
on intentions to engage in complaining behavior.
This indirect effect is through attitudes towards
the act of complaining.
Much of the earlier research in CCB is often
criticized to be descriptive in nature (Robinson 1979).
Most of it described the demographic correlates, such as
age, sex, income, social class, etc., of complainers and
non-complainers. Thus it became a well-established
finding that complainers are usually younger, highly
educated, and belong to higher social classes.
In an insightful article, Gronhaug and Zaltman
(1981) suggested that the demographic correlates of
complainers, so often found in the CCB literature, may be
mere artifacts of market-place participation. That is,
consumers who are more involved in the market-place would
expect to have higher probabilities of being exposed to
buying problems. In the CCB model discussed in the
present research, the differences in these participation
levels are directly modeled by measuring expectancy-value
Judgments and attitudes. Therefore, it can be
hypothesized that:
Hll: The demographic variables of age, sex, income and
social class have no direct effects on intentions to
engage in complaining behavior but have indirect
86
effect through the constructs of expectancy-value
Judgments and the attitudes towards that act of
complaining.
Richins (1983) empirically investigated the rela
tionship between expectancy-value Judgments of retailer
responsiveness and complaining actions across two product
groups. She found the relationship between the two
concepts to be similar across the two product groups.
This finding supports the possibility of generalized
structural relationships that are not dependent on the
particular product or service. An empirical investigation
of the structural relationships across services is not
evident in the literature. Therefore, it can be hypothe
sized that:
H12: The strength and direction of structural relation
ships among expectancy-value Judgments, attitudes,
and intentions is similar across the four services
investigated in the present research.
Day (1984) among other researchers has suggested
that one key underlying rationale for proposing a process
conceptualization of CCB is that the level of dissatis
faction alone is found to have rather limited ability in
predicting CCB. For Instance, Bearden and Teel (1983)
could explain only 15% of the variation in CCB using the
level of dissatisfaction. Thus the process model of CCB
87
ought to perform better than the naive model of Bearden
and Teel (1983), An empirical investigation of the
explanatory power of the two models in a single study has
not been found in the literature. It is thus hypothe
sized that:
H13: The variance explained in the intentions to engage
in complaining behavior by the process model of
expectancy-value Judgments and attitudes is higher
than that explained by the degree of dissatisfaction
alone.
One of the key interests in developing a model of
CCB process is to be able to predict not only consumer
complaining behavior in general, but also specific
complaint actions, such as exit, word-of-mouth, voice,
etc. A framework for predicting specific complaint
actions is presented in chapter 3 as a part of the
holistic model of CCB (see Figure 3.3). Based on this
framework, the following hypotheses can be generated.
H14: For high levels of expectancy value Judgments of
seller responsiveness together with positive
attitudes towards complaining, the intentions to
take public actions would be high. The most pre
ferred action would be voice to seller and W-O-M to
friends and relatives (based on Hirschman 1970; Day
88
1980, pp. 214-5).
H15: Low levels of expectancy value Judgments of seller
responsiveness and positive attitudes towards
complaining would in general result in higher pro
pensity for W-O-M communication to friends and
relatives and/or exit actions (based on Richins
1983, pp. 74-6).
H16: In the case of low levels of expectancy value
Judgments of seller responsiveness and negative
attitudes towards complaining, the most preferred
complaint behavior would be to take no action (based
on Day 1981, pp. 93-6).
The nature of intentions to act in the case of high
levels of expectancy value Judgments of seller respon
siveness together with negative attitudes towards com
plaining are not previously recorded in the literature.
Since expectancy value Judgments are perceptions of
sellers responsiveness to VOICEd complaints combined with
value of outcomes, it may be expected that when E-V
Judgments are high, the propensity to take VOICE actions
would also be high. When high E-V Judgments occur along
with negative attitude toward the act of complaining, the
intentions for VOICE actions may be decreased somewhat,
yet VOICE may remain the most preferred action because of
the high probability of the seller responding positively
to the VOICEd complaints. Thus it is proposed that:
89
H17: If the level of expectancy value Judgments is high
but attitude toward the act of complaining is
negative, voice may still be the most preferred
action.
It is also desirable, from a managerial and public
policy standpoint to investigate the variation across the
different industries. Hirschman (1970) suggests that in
the most competitive industries, the channels for exit as
well as voice are open, whereas in monopolies, only voice
is the viable alternative. Andreasen (1983) has further
elaborated on "loose monopolies" such as health services,
where opportunities for both voice and exit are generally
blocked. It can, therefore, be hypothesized:
H18: When the level of dissatisfaction is controlled for,
the mean levels of expectancy Judgments are highest
for services that are perceived to be supplied by
most competitive industries, lowest for loose
monopolies, and somewhere in between for monopolies.
Discussion
This chapter has attempted to elaborate the various
hypotheses to be investigated in the present research.
The key objectives in their development are: (a) to
attempt to seek empirical testing of the structural
relationships suggested by the holistic model of CCB,
90
and (b) to examine those predictions that may have
significant theoretical, managerial and/or public policy
implications.
The next chapter provides the details of the survey
procedures and methods used to collect data for phase I
and phase II of this dissertation research. This is then
followed by data analysis and summary of results.
TABLE 4.1
OPERATIONALIZATION OF KEY CONSTRUCTS IN THE HOLISTIC MODEL
91
S. No.
1.
2.
3.
4.
5.
6.
7.
8.
CONSTRUCT
LEVEL OF DISSATISFACTION
PRIOR EXPERIENCE.
BASIC AFFECTIVE INFLUENCES.
ATTITUDE TOWARD THE ACT OF COMPLAINING.
ATTRIBUTIONS OF -BLAME'.
EXPECTANCY OF COMPLAINT ACTIONS.
VALUE OF COMPLAINT ACTIONS.
INTENTIONS TO COMPLAIN.
OPERATIONAL MEASURE
TWO SINGLE ITEM RATING SCALES.
BASED ON FIVE ITEM CONSUMER'S KNOWLEDGE AND EXPERIENCE SCALE
(a) 82 ITEM CONSUMER DISCONTENT SCALE.
(b) 35 ITEM ALIENATIO^ FROM THE MARKET--PLACE SCALE.
Based on (a) 15 ITEM ATTITUDES
SCALE, and
^b) 10 ir^M ATTiriioE TOWARD 'WE ACr OF COMPLAINING SCALE
PROPOSED FIVE ITEM SCALE.
PROPOSED 15 ITEM SCALE.
PROPOSED 15 ITEM SCALE.
Based on (a) 9 ITEM ACTIONS
SCALE. AND
(b) 5 ITEM COMPLAINT ACTIONS SCALE.
DEVELOPED 8r
WESTBBOOf.dHO JM. pp. 58-'2.
DAY, (19«4) ACR Proc.
LAMONT, ( H ; 6 ) .
ALLISON. (1978).
bOt*1 in 'MO .
RICHINS. i\^^2).
•^Ay. (\ '-: .
KRISHNAN AND vALLE (19791. and FOL<ES ( 19«4».
Bas°'J on BAGOZZI (1982).
•HAGn^ZI ( I98?i.
OA' ET «L.'1981.
ciFARDEN AND '•fC'
PHOPCSTIES
ALPHA=G.65 to 0.88
HOT
REPOR'tO.
ALPHA^O.94
ALPHA»0.88
NOT BEPURfED.
NOT REPORTED,
ro RE INVESTIGATED
ro a INVESTIGATED
'•O BE INVESTIGATED
'GUTTMAN' rrPE.
REPR0.'0.98 scat*.'0.78
CHAPTER 5
SURVEY PROCEDURES AND METHODS
This chapter focuses on the survey process itself--
addressing questions such as how the sample was selected,
how the survey was conducted, and the attempts made to
reduce nonresponse error. These aspects are important
since the quality of data is dependent largely on the
manner in which the data is collected.
The Two Phases of Research
The research was conducted in two phases, hereafter
referred to as phase I and phase II, In phase I investi
gation is limited to three constructs: discontent, alie
nation, and attitude towards the act of complaining. The
objective of phase I was twofold: (a) to test the rela
tionships among these constructs specified by hypotheses
HI, H2 and H3 (see chapter 4), and (b) to purify the
scales used for measuring these constructs so as to
reduce them to a manageable length while minimizing loss
of information. The specific questionnaire used in phase
I of this dissertation research is as per Appendix A.
Phase II was conducted subsequent to the completion
of phase I. The purpose of phase II is to test the
remaining hypotheses, H4 to HIS. The instrument for this
phase contains all constructs operationalized in chapter
92
93
4 including the shortened versions of the alienation,
discontent and attitudes scales obtained as a result of
phase I analysis. Further, each questionnaire for this
phase measures responses to dissatisfying experiences in
one of the four selected industries. That is, each
respondent would provide complete responses to only one
dissatisfying experience for one randomly assigned indus
try. The survey methods and procedures adopted for phase
I and II are now discussed in turn.
Phase I
Selection of Census Tracts
A two stage area sampling plan was employed for
phase I. The first stage involved selection of census
tracts with in the city of Lubbock. The second stage
involved a systematic selection of households from the
pre-selected tracts (e.g., every fifth member). The
critical step here is the selection of tracts to ensure a
representative sample. The sampling unit is a household,
and a sample size of 1000 households was selected to
ensure generalizability of findings.
Census tracts were divided into four groups based on
median household income as tabulated by the Census of
Population: (1) up to $10,000, (2) from $10,001 to
$15,000, (3) from $15,001 to $20,000, and (4) greater
than $20,000. Using random number tables, two tracts
94
were selected from group 1 (12.02 and 13), another two
from group 2 (23 and 3) and one each from group 3 (20)
and group 4 (17.02). Within each of these tracts, the
interviewers were instructed to select every fifth house
hold along each block in the tract to be a part of the
sample.
Method of Data Collection
Undergraduate business students were recruited to
drop off questionnaires to the selected sample of house
holds. An effort was made while selecting the students
to match their characteristics with those of the
potential respondents. The selected group of students
consisted of 4 white, 1 black and 1 Hispanic. They were
trained to interact with and request participation from
the potential respondents in an appropriate manner. A
text of the standard introduction to be used by each
student in seeking cooperation of the household is
enclosed in Appendix B. Each student was also instructed
to wear, at all times, a label clearly showing his name
and the sponsoring party (Texas Tech). The dress code
selected was informal since it was believed that a formal
attire may give an impression of a "door-to-door"
salesman.
Two methods were adopted for pick-up of completed
surveys in phase I. Approximately one-half of the total
sample (that is 500) was provided with a stamped return
95
envelope to mail back completed responses. For the
remaining, the students were instructed to go back after
a week and pick up the completed responses. In the
latter method the students were instructed to inform the
respondent of the time and day of the pick up at the time
they delivered the survey. The two methods were employed
to study the differences in response rates when different
methods for collecting the responses were adopted.
However, this particular study did not constitute a part
of this dissertation research. Further, irrespective of
the method, the students were trained to keep a record of
the addresses of the households who agreed to participate
in the survey. This was done to ensure proper follow up
and callbacks.
The phase I survey was conducted on all days of the
week excluding Sunday (for religious reasons) and Friday
(end of week syndrome). The survey was conducted between
4.30 p.m. and 8 p.m. on all days of the week, and between
9 a.m. and 1 p.m. on Saturday. It was believed that the
probability of contacting the respondents would be high
during these time slots. An effort was made to contact
Not-at-Homes during different days of the week than that
on which the original attempt was made.
Nonresponse and Callbacks
When student pick up was used, up to three callbacks
96
were made to collect the responses. For the second
method, telephone callbacks (up to 3 times) were used to
contact the respondents to ensure their participation.
Telephone numbers were obtained from the city directory
based on the addresses collected earlier (Lubbock City
1983).
A total of 512 responses were received out of 1000
surveys given out--a 51.2% response rate. However, only
460 responses could be used because of incomplete
responses. An analysis of the 460 completed responses
shows that 85.5% are white, 4.8% are black, 7.7% are
Hispanic and the balance 1.9% are of other ethnic groups.
These percentages compare well with those based on
Lubbock city census data--81% white, 8X Hispanic and 7%
black household population (Census Information 1983).
The median household income of the sample is in the
$20,001 to $30,000 range and the mode is in the $10,001
to $20,000 bracket. This represents fairly well the
city's 1979 median income of $15,735 and a mode in the
$5,000 to $9,999 bracket (Census Information 1983). The
latter figures will correspond better once they are
corrected for the inflation. The sample, it appears, is
generally representative of the parent population and the
non-response has, perhaps, not eroded the quality of the
sample--at least based on demographic characteristics.
97
Phase II
Selection of Tracts
The census tracts were selected in phase II using a
procedure similar to that in phase I. Income was used to
stratify the census tracts and then a random sample of
tracts was selected from each stratum. Since phase II
required sending out four surveys (corresponding to the
four industries), each with a sample size of 1000, this
procedure was repeated four times. It was desirable not
to send more than one questionnaire to any given house
hold. Therefore, an effort was made not to select the
same census tract in more than one survey. Appendix C
provides the diagrams of the census tracts selected in
each of the four surveys using the above procedure.
The second stage, which involved selecting the
specific households, was also similar to phase I. A
systematic sample of households was selected from the
already selected census tracts. Using a city directory,
every 10th household from each census tract selected
became part of the sample till the total sample size
desired was obtained (City Directory 1983).
Method of Data Collection
Considering the large sample size for phase II (4000
in all) a mail survey was selected as the appropriate
method for collecting the data. This method was also
98
better suited to the phase II surveys, since the ques
tionnaires were more involved and lengthy. Using the
city directory, a packet was mailed to each of the selec
ted households. The packet contained a covering letter
explaining the purpose and sponsorship of the survey, the
questionnaire, and a pre-paid return envelope. Copies of
the questionnaires and the covering letters for the four
industries are enclosed in Appendix D.
Callbacks
The city directory provides the addresses as well as
the telephone numbers of all households. These telephone
numbers were used to contact the selected households 7-10
days after the mailing of the questionnaire packet. On
contact, a request for their participation was made, in
case they had not done so, and the importance of their
responses was reiterated. This procedure was adopted for
three out of the four industry surveys. For the last
survey, a reminder card was mailed to the sample, 5 days
after the packet of questionnaire was sent out. A dif
ferent method for reminding the respondent was used for
financial services survey in order to study the effects
of different follow up methods on response rates in mail
surveys. However, this particular study did not consti
tute a part of the present research. The four surveys
were conducted one after the other over a period of three
months (February-April 1985). Response rates for the
99
four surveys are shown in Table 5.1.
Nonresponse
Nonresponse is an important concern here since the
response rates in phase II were far below that obtained
in phase I (51.2%). Since low response rates do imply a
source of bias in the obtained data and the corresponding
analysis, the response rates obtained in phase II of this
research do impose a limitation on the validity of the
results. However, the following discussion attempts to
show that, because of the particular nature of this
survey, the severity of the nonresponse bias may be
somewhat mitigated.
Several reasons can be attributed to the lower
response rate--the questionnaire was longer and more
involved. Most important is the nature of the question
naire. The whole questionnaire revolved around a
particular "recent problem" or "dissatisfaction" that the
respondent had with the given industry. What if the
respondent felt he/she had had no. "recent problem" of any
significance, with grocery shopping for example? It is
suggested that such a situation resulted in nonresponse.
This conclusion is based on two evidences.
When respondents were contacted by phone, two
typical responses were observed. People who felt they
had been satisfied with their grocer, for example, had a
100
higher probability of having discarded the questionnaire,
and showed no interest in participation. On the other
hand, people who had bad experiences were eager to parti
cipate and "let some one know" about the problem(s) they
had faced. Motivation to complete and mail the question
naire was high in the latter situation.
The second evidence comes from the wave analysis of
returned responses. As the responses were received, they
were classified into waves by industry. There was a gap
of at least 1 to 2 days between the two consecutive
waves. The percentage of respondents indicating "no
problem" was calculated for each wave. Later waves had a
much higher percentage of "no problem" than did earlier
waves. For example, in the grocery industry, the "no
problem" respondents were 17% in the first wave, 29% in
the second wave and over 72% in the third wave. A chi-
square test for the difference in cells by chance was
significant at alpha=0.05 level of significance for all
the four surveys. This shows that respondents who had
less interest in the survey (later waves) had a much
higher probability of experiencing "no problem." Table
5.2 gives the chi-square statistic for each of the four
industry surveys.
Further, respondents who had faced a recent problem
or dissatisfaction were analyzed to determine if the
latter (wave) respondents were significantly different
101
from the earlier (wave) respondents. Specifically, dif
ferences were examined for several key constructs, such
as attitude toward the act of complaining, VOICE, W-O-M,
and FORMAL intentions as well as expectancy value
judgments. Respondents were divided into three groups
based on waves (the third and fourth wave had to be
collapsed because of small number of responses in the
fourth wave). An analysis of variance was done with the
three groups as the treatment and the individual
constructs as the dependent variable. The null hypothe
sis of no difference in the construct means among the
three waves was not rejected at 0.05 level of signi
ficance for each of the four industry data. Infact, the
F value for significant difference in the means was less
the 2.0 for 26 out of 28 construct means investigated.
This implies that respondents who had experienced a
recent problem but had responded in the latter waves
were, perhaps, no different from respondents who had also
faced a recent dissatisfying experience but responded to
the questionnaire early.
Since the objective of this research is to only look
at those people who are dissatisfied, and investigate the
kind of complaint behavior they engage in and why, non-
response from people who do not have a high probability
of recently experiencing a dissatisfaction would,
perhaps, not adversely affect the quality of data.
102
TABLE 5.1
RESPONSE RATES FOR PHASE II SURVEYS
Industry
Grocery
Auto Repair
Medical Care
Financial Services
#Mailed
1000
1000
1000
1000
#Received
176
155
166
172
Response Rate
17.6 %
15.5 %
16.6 %
17.2 %
Usable Responses
124
116
125
104
Usable Rate
12.4 %
11.6 %
12.5 %
10.4 %
TABLE 5.2
CHI-SQUARE TEST FOR "NO PROBLEM" RESPONDENTS
1 0 3
INDUSTRY CHI-SQUARE STATISTIC DP P-VALUE
Grocery
Auto Repair
Medical Care
Financial Services
59.5
36.4
7.83
10.65
3
3
3
3
0.0000
0.0000
0.049
0.0049
CHAPTER 6
PHASE I RESULTS
The objective of phase I is to investigate the
nature of the relationship among the three affective
constructs--alienation from market-place, discontent, and
attitude towards the act of complaining. Specifically,
it is proposed to empirically examine hypotheses H1-H3
(Figure 6.1). Hypotheses HI and H2 attempt to examine
the measurement properties of the alienation and discon
tent constructs. This is of particular interest, since
researchers debate the validity of treating these two
constructs as distinct and different from each other.
For instance, Allison (1978) who developed the scale of
alienation from the market place suggested that alie
nation and discontent are similar and, perhaps identical
constructs. On the other hand, Lundstrom et al. (1979)
argue, from both a methodological and theoretical stand
point, that the alienation and discontent are distinct
concepts, and are in fact inversely related to each
other. Since this dissertation proposes to use these two
constructs, phase I attempts to resolve this dilemma
before proceeding to phase II.
Further, empirical examination of hypothesis H3
should also provide insights in to the measurement
104
105
properties of alienation and discontent scales. H3
examines the relationship of these two constructs with an
external criterion variable--attitude toward the act of
complaining. If these two constructs are truly distinct
and possess discriminant validity, then the relationship
between each of these two scales and attitude toward the
act of complaining should neither be identical nor
similar. The empirical investigation of H3, therefore,
will help clarify the nature of alienation and discontent
scales.
Several methodologies are used to test the hypoth
eses H1-H3. In particular, factor analysis is used to
examine the measurement properties, and path analysis is
used to investigate the structural relationships in
hypothesis H3. In addition. Item Response Theory (IRT)
is used to analyze the measurement properties of the
alienation and discontent scales. IRT is a measurement
theory which affords stronger conclusions about scales
and what they are measuring than the traditional
classical test theory (CTT) (Hulin, Drasgow and Miller
1983). Further, IRT can be very useful for scale devel
opment and modification since it explicitly accounts for
the information provided by each item on the scale (Lord
1980). Perhaps, the most attractive consideration for
the adoption of IRT for phase I analysis, is that it
provides measurement and information properties of items
106
on a scale that is independent of the sample selected,
except for the effect of sampling variations (Lord 1980).
Although marketing has been slow in the adoption of this
powerful technique to address measurement problems, this
technique is widely in use in educational psychology, and
to some extent in psychology.
In the following discussion, hypotheses H1-H3 are
examined with both, the traditional methods of factor
analysis as well as Item Response Theory. In that sense,
the objective also to show the two methods can be used in
a complimentary manner to enrich the understanding of
what is being measured by a pool of items and with what
efficiency.
Measurement Properties Using Traditional Methods
The measurement properties of the three constructs
are first investigated before examining the structural
relationships among them. The alpha reliability of the
18-item attitudes towards the act of complaining
construct is 0.626, the 35-item alienation scale has a
reliability of 0.902 and the 82-item discontent scale has
an alpha of 0.944. This compares well with the
reliability values reported in the literature for these
three constructs. For instance, Allison (1978) reported
an alpha of 0.88 for the alienation scale; Lundstrom and
Lament (1976) found an index of 0.96 for the reliability
107
of the discontent scale. A proper development of the
attitudes towards the act of complaining construct has
not been undertaken in the marketing literature, and its
measurement properties can not be compared. However, the
scale used in phase I is based on the efforts of various
researchers in the area, specifically Richins (1982).
Items that compose a scale are expected to relate
positively to each other since they are purportedly
tapping the same trait. Items that have a negative (or
zero) correlation with the total score on a given scale
do not meet this criterion and should be deleted to
improve the measurement properties of the scale. Using
this procedure, four items were dropped from the attitude
scale and the revised reliability of the scale improved
to 0. 685.
The next step in the measurement analysis is to
investigate the factor structure for each of the three
constructs. A common factor analysis of the 35-item
alienation scale shows the presence of a single dominant
factor. This is depicted in the scree plot shown in
Figure 6.2. The first factor is associated with an
eigenvalue of 8.23 which represents 65% of the shared
variance. The second eigenvalue in contrast is only
1. 44. This eigenvalue structure is representative of
unidimensional scales (large first eigenvalue and the
remaining values about the same range but small). This
108
finding supports the unidimensional structure for the
construct of alienation from the market place and is
consistent with the results of Allison (1978) (see
Bearden, Lichtenstein and Teel 1983 for a different
conclusion).
Similarly, the attitudes construct on factor
analysis of the correlation matrix with squared multiple
correlations on the diagonals shows that the first factor
represents 70% of the shared variance (eigenvalue 2.40).
The second eigenvalue in this case is less than half of
the first eigenvalue. This supports a unidimensional
conceptualization of the attitudes towards the act of
complaining scale. Finally, the discontent scale also
indicates a similar conclusion. The scree plot of the
eigenvalues is given in Figure 6.3. The first eigenvalue
is 16.58 representing 45% of the total shared variance.
The next eigenvalue is only 3. 18. The factor analysis,
therefore, suggests that each of the three constructs
have good measurement properties and the items that
compose them tap a distinctive and well defined trait.
Structural Relationships Using Path Analysis
It has been suggested previously that considerable
controversy surrounds the nature and measurement of
discontent and alienation constructs. Hypothesis H1-H3
directly addresses this issue by examining the divergent
109
validity of these two constructs by examining their
relationship with a third, external variable, attitude
toward the act of complaining. A correlation matrix
using summated indexes for the three scales is given in
Table 6.1.
Discontent and alienation are positively related
(r=0.896, p<0.0001), which suggests that these two
constructs are not distinct concepts and do not possess
discriminant validity. When attenuated for reliability
of the two constructs, the correlation increases to
0.971. Consequently, hypothesis HI is rejected. A
somewhat different conclusion results when the corre
lations with the attitudes construct are examined.
Alienation from the market place does not have a
significant relationship with attitude towards the act of
complaining (p=0.1186), whereas discontent bears a sig
nificant and positive effect on the attitudes construct
(p<=0.0001). The unattenuated correlation for the latter
is 0. 22. It is recommended that path analysis be used in
situations where relationships among three or more
constructs are being examined in the presence of corre
lated constructs (Johnson and Wichern 1982). This is
relevant here since discontent and alienation are corre
lated while at the same time their effect on attitude
toward the act of complaining is under investigation.
Figure 6.4 gives the path analytic diagram for phase I
110
data. This figure presents a striking result: the path
coefficient between discontent and attitudes is 0.555
(unattenuated 0.690) and the path coefficient between
alienation and attitudes is -0.426 (unattenuated -0.542).
This implies that discontent has a positive effect on the
attitudes construct, whereas alienation has an inverse
relationship. Therefore, hypothesis H2, which proposed
that alienation is negatively related and hypothesis H3
which indicated that discontent is positively related to
the attitudes construct, are both supported by this data.
And yet, it is also observed that discontent and alie
nation are positively and significantly correlated so as
to appear to lack discriminant validity.
This finding clearly supports hypotheses H2 and H3,
and casts a doubt on the earlier conclusion about
rejecting HI. In other words, discontent and alienation
are almost perfectly correlated while simultaneously
possessing strikingly different path coefficients with
respect to the construct of attitudes towards the act of
complaining. What then is the real nature of these
constructs and what do they measure? How can these two
opposite pieces of evidence be explained? Item Response
Theory is an appropriate tool for this investigation.
Measurement Properties Using IRT Based Procedures
The assumption of unidimensionality of a given pool
Ill
of items is a necessary precondition before employing IRT
based procedures. Do the data meet this requirement? It
has been shown earlier that each of the three
constructs--alienation, discontent and attitudes--
evidence a unidimensional eigenvalue structure. As the
concern here is the distinction between alienation and
discontent, it is important to investigate if the items
pooled from the discontent and alienation scales also
show a unidimensional factor structure. To investigate
this, the 35 alienation items and the 82 discontent items
are combined into a single pool of 117 items. This
combined pool is then factor analyzed. The eigenvalue
plot for this correlation matrix shows that the first
eigenvalue is approximately 6 times the second value and
explains about 40% of the shared variance (see Figure
6.5). This is evidence of the unidimensionality of the
trait underlying this pool of 117 items. Thus, IRT
procedures can be safely employed for this data.
A two parameter logistic function was fitted to the
response to each item in the pool using the Legist
program (Wingersky, Barton and Lord 1982). The output
from this procedure includes the approximate maximum
likelihood estimates of "a" and "b" parameters, their
standard errors and the "thetas," that is, estimates of
the level of the trait for each respondent. Based on
these parameters and the IRT procedures it is possible to
112
calculate the amount of information contained in a group
of items for different levels of the underlying trait,
theta. The theta in the present investigation is the
affective feelings of discontent and alienation. Each
item on the scale provides some information about this
theta, called the item information. Information provided
by an item is not necessarily constant throughout the
range of theta. For instance, if the item is very
"easy," it may provide very little information at higher
levels of the underlying trait. Nevertheless, the same
item may provide a substantial amount of information
around the low levels of theta. Further, the total
information or test information provided by a scale of
items is simply the sum of the informations provided by
individual items in the scale (Lord 1980). The test
information also provides a measure of reliability. When
information is high, uncertainty about the measurement of
the underlying trait is low and therefore reliability is
high. Since the test information function varies for
different levels of theta, the reliability of a scale is
also a variable, possessing different values at different
levels of the underlying trait (Hulin, Drasgow and Miller
1983).
These information curves can provide useful insight
into the measurement properties of a pool of items. In
the present case, two information curves are estimated.
113
one for the a2-item discontent scale and another for the
35-item alienation scale. Figure 6.6 shows that the
discontent items provide their maximum information
(highest reliability) at approximately -0.7 trait level,
while the alienation items have their maximum information
at approximately -0.55. The trait level is standardized
with mean zero and standard deviation one. The two infor
mation curves suggest that the given set of items provide
much of their information in the middle range of the
underlying trait and relatively poor information at the
two extremes. The closeness of the two information
curves also explains the high correlation observed
between discontent and alienation constructs.
Some key implications of the above analysis can now
be stated. The pool of items purportedly measuring the
discontent affective feelings and the other pool suppose
dly tapping the feelings of alienation from the market
place are in fact measuring the same underlying trait--
with one difference. The alienation items provide the
bulk of their information at a trait level slightly
higher than that for the discontent items. Since,
alienation and discontent are somewhat different
(evidenced by the support of hypotheses H2 and H3), it is
possible that higher levels of discontent lead to
alienation. In other words, as people become more and
more discontent they also tend to become alienated.
114
Therefore, an appropriate measurement of discontent can
be achieved if its information curve peaks at lower
levels (less than the mean) of the underlying trait.
Similarly, the alienation construct can be better tapped
if its information curve peaks at higher levels (greater
than the mean) of the common underlying trait. An exami
nation of Figure 6.6 shows that current scales for the
above constructs do not meet this criterion. How, then,
should these scales be modified to provide a better
measurement of the alienation and discontent constructs?
The item difficulty parameter, "b, " obtained from
IRT procedures provides necessary guidance in the
selection of items for the modified discontent and
alienation scales that meet the above condition. For any
given item, the "b" parameter represents the value of the
trait at which the given item would have a peak for its
information curve. Since the total or the test informa
tion for a complete scale is simply the sum of the
information from each of its items, it would be desirable
that all discontent (alienation) items have their "b"
parameter less (greater) than zero. Thus, the following
criteria for the selection of items are proposed:
Assume: The underlying trait being measured--the
generalized affective feelings of discontent
and alienation--is standardized with mean 0 and
standard deviation 1.
115
Criterion #1: The "b" parameter values for all discon
tent items should be negative (less than
mean).
Criterion #2: The "b" parameter values for all aliena
tion items should be positive (greater
than mean).
Criterion #3: No two items should have the same "b"
parameter value. In fact, the items
selected should have "b" values that span
the entire range from •»-3 to -3.
Criterion #4: The standard errors of the "b" parameters
should be small compared to the parameter
value.
Criterion #5: Each selected item must have a strong "a"
parameter value (for instance > 0.4) with
small standard error compared to the
parameter value.
Criterion #6: The item should satisfy face validity
requirements for inclusion as a discontent
or an alienation item.
Criterion #7: The number of items selected in each of
the two scales should be between 8-15
items.
This procedure will yield shortened versions of the
discontent and alienation scales that (a) tap a substan
tial range of the underlying affective trait, (b) provide
116
adequate measurement (information), and (c) can differen
tiate better between the feelings of alienation and
discontent. It may be noted that no external criterion
variable is being used in the selection of items for the
shortened scales (Nunnally 1978).
Using this procedure, thirteen items were selected
out of 35 original items to compose the alienation scale
and 12 items were selected out of 82 items for the
discontent scale (Appendix E). The "a" and "b" parame
ters for the selected items are tabulated in Table 6.2.
Figure 6.7 presents the revised information curves for
the shortened scales.
A comparison of Figure 6.7 with figure 6.6 shows
some striking differences attributable to the above
procedure. The information curves are now distinguish
able, the peak of the discontent items occurring around
-1.25 trait value and that of the alienation scale at
around 0.75. Thus the revised scales do seem to measure
distinctively the two dimensions of the underlying trait.
The discontent dimension is measured with greater relia
bility than the alienation dimension because of its
higher information peak. In addition, the alienation
scale does not provide much information about the more
alienated persons (trait value>2.0), though there is a
definite improvement over the original scale. It is
suggested that newer items may have to be constructed to
117
measure effectively the higher levels of the alienation
dimension.
As discussed earlier, IRT procedures provide
estimates that are in general invariant across different
samples. Therefore, the "b" parameters and consequently
the information curves should correlate highly across two
independent studies. To investigate this, IRT parameters
of the shortened scales were calculated based on the data
collected in phase II for the grocery and automotive
repair industries. The data consisted of 240 completed
responses. Some of the items had to be dropped since
their standard errors could not be estimated accurately.
The reduced scale consisted of 12 discontent items and 9
alienation items.
Figure 6.8 is the plot of "b" parameter values
calculated independently from the two different samples.
The slope of the line (correlation) is 0.864 (p<=0.0000)
indicating close correspondence. Similarly, the infor
mation curve derived from phase II data ought to be
similar to Figure 6.7 if IRT procedures are valid.
Figure 6.9 shows the information curves for the dis
content and alienation dimension for the grocery and
automotive repair data. Once again, a close correspon
dence is evidenced between the information curves derived
from the two different studies, suggestive of the
stability and validity of the measurement procedure. In
118
addition, if the measurement of a construct is thus
improved, then its relationship with other constructs
should also be better defined. The next section investi
gates the structural relationship of the reduced
discontent and alienation scales with the external
construct of attitude toward the act of complaining.
Structural Relationships With Reduced Scales
Figure 6.10 shows the path analytical diagram
(similar to Figure 6.4) using summated indexes from the
shortened versions of the discontent and alienation
scales but based on phase I data. The correlation
between discontent and alienation is now 0.592, substan
tially lower than the earlier value of 0.93, indicating
the existence of two different concepts: discontent and
alienation. This reduced correlation also helps in part
to stabilize the two path coefficients and reduce their
standard errors. The path coefficients support the
earlier conclusion, that is, discontent has a positive
effect, while alienation has a negative effect on the
attitudes construct. In other words, the revised scales
provide partial support for hypothesis HI, in that
alienation and discontent are distinct facets, though of
a common underlying trait. Support for H2 and H3 is
provided as well. This is an improvement in the understa
nding of these constructs, since the significant differe-
119
nces in the two path coefficients can be attributed to
the two dimensions of the underlying generalized
affective feelings.
A contrary argument that can be made is that since
the items selection was based on the analysis of phase I
data, and the structural relationships were tested using
the same data, it is plausible that the conclusion is
merely an artifact. To examine this argument, a path
analysis is undertaken using the automotive repair data
in phase II (Figure 6.11). The magnitude and direction
of path coefficients is very similar to that observed for
phase I data (Figure 6.10). Discontent and alienation
are positively correlated but the correlation is relati
vely smaller than 0.93 (Figure 6.4). This suggests that
for the automotive repair data, discontent and alienation
also appear to be distinct dimensions. Discontent has a
positive path while alienation a negative path leading to
the attitudes construct, implying support for hypothesis
H2 and H3. A comparison of the path coefficients in
Figure 6.10 and 6.11 shows that they are of comparable
magnitude. The automotive repair data, therefore,
provides an independent support of the conclusions
obtained for phase I data. Thus the procedures adopted
for the phase I of this dissertation appear to be valid
and reliable.
The question remains as to whether alienation and
120
discontent are positively or negatively related, or
positively related in one range and negatively in the
remaining range? To explore these questions, the data
obtained from the grocery and automotive repair surveys
(240 responses in all) were divided into two groups,
based upon thetas (value of the underlying trait) being
either below or above the mean. Below zero (average)
theta represent predominantly discontent and theta above
zero (average) denote mainly alienated respondents. As
shown in Figures 6.12 and 6.13 the correlation between
discontent and alienation drops from 0. 51 in the
discontent sample to 0.45 in the alienated sample. This
suggests that the distinction between these two concepts
improves at higher levels of the underlying trait. In
other words, it is hypothesized that alienation may not
be identifiable at lower levels of the underlying trait.
Figures 6.12 and 6.13 show that while the
discontent >attitudes path remains of almost the same
magnitude across the two groups, the alienation >atti-
tudes path relatively changes substantially from 0.027
(no relationship) to -0.10 (negative relationship). This
provides an additional evidence in support of the sug
gested conceptualization of these two constructs.
Summary
Based on the data obtained in phase I, the analysis
suggests that: (1) discontent and alienation are two
121
dimensions or facets of an underlying generalized
affective feelings concept; (2) discontent and alienation
are directly and positively related; (3) alienation
becomes better defined and distinct at higher levels of
the underlying trait; (4) alienation is negatively
related to the attitudes construct; and (5) discontent
has a positive relationship with the attitude towards the
act of complaining. This implies that hypothesis HI is
supported in part while hypotheses H2 and H3 are both
supported by the data.
Further, shortened versions- of the three scales were
also constructed using IRT based procedures. The reduced
scales now consist of 13 items in the alienation scale,
12 items in the attitude scale and 12 items on the
discontent scale. These shortened versions are used in
phase II to test the remaining hypotheses.
122
TABLE 6.1
CORRELATION MATRIX FOR DISCONTENT, ALIENATION AND ATTITUDES
Discontent
Alienation
Attitudes
Discontent
1.00
0.896* (0.0000)
0.176* (0.0001)
Alienation
1.00
0.073 (0.1186)
Attitudes
1.00
* Significant at 0.01 level.
123
TABLE 6.2
AND "B" PARAMETERS FOR THE DISCONTENT AND ALIENATION SCALES
1. Grocery "have more trained and efficient cashiers"
"out of date dairy products and packaged meats"
"check for outdated stock more often"
"see that sale items are put into the computer"
"give their sackers lessons on how to fill the grocery sacks"
"include proper cleaning of store as a regular part of an employees duties"
"have adequate supply or not advertise item at bargain prices"
"take bad products off the shelves"
"make sure all their checkers know the specials and have the right price"
"better employee training programs"
2. Auto Repair "salesmen needs to be more honest-when a person asks for steel radials they give you polyester radials for the cost of steel radials"
"the repair part cost $0.35. . they want $300 labor fee"
"stop overcharging and be honest in repairing"
"try to be more conscientious instead of "half fixing" something and hoping the customer would be too tired or busy to bring it back"
224
TABLE 8.3 (Continued)
Industry Typical Responses
"better train their workmen"
"be fair to women"
"close shop 1"
"the problem was fraud-what could be done?"
"not to be afraid of talking to a dissatisfied customer face the problem"
"get approval before doing any additional work"
"(try) to please them (customers) as they will return if satisfied, otherwise one customer can really hurt business"
3. Medical "recognize that my time is as important as theirs"
"have greater "sincere" concern for the patients problems"
"I was robbed of both my teeth and money"
"reduce cost"
"there is not much that can be done with a greedy person .... except to try to inform others about the bad experiences"
"I think it is real simple--they are just too greedy"
"they need to answer to someone .. most patients are too frightened or feel too poorly to take up for their rights"
"stop trying to fight alligators .. and figure ways to clean the swamp"
225
TABLE 8.3 (Continued)
Industry Typical Responses
"run the AMA out of the country"
"take time--maybe talk a little more (other than medical talk!)"
"they need to be more caring"
4. Financial "if they would show the slightest bit of concern, I would feel better, but most could care less"
"train employees to give courteous responsible service"
"full banking service on at least one week night on Saturday morning"
"upon closing my account after 30 years plus, no concern or question as to why I moved my account"
"set fire to their computers"
"I know banks do make errors but it sure helps to have .. people who are sincere in wanting to help you. That makes all the difference"
REFERENCES
Allison, Neil K. (1978), A Psychometric Development of a Test for Consumer Alienation from the Marketplace, Journal of Marketing Research. Vol. 15, Nov., pp. 565-75.
Anderson, Ralph E. (1973), Consumer Satisfaction: The Effects of Disconfirmed Expectancy on Perceived Product Importance, Journal of Marketing Research. Vol. 10, Feb. , pp. 38-44.
Andreasen, Alan R. (1977), A Taxonomy of Consumer Satisfaction/-Dissatisfaction Measures, Journal of Consumer Affairs. Vol. 11, No. 2, pp. 11-24.
(1983), Consumer Satisfaction in Loose Monopolies: The Case of Medical Care, Journal of Marketing and Public Policy. Vol. 2, pp. 122-35.
Andrich, David (1978), A Rating Formulation for Ordered Response Categories, Psychometrica. Vol. 43, No. 4, pp. 561-73.
Ash and Quelch (1980), Consumer Satisfaction, Dissatisfaction and Complaining Behavior: A Comprehensive Study of Rentals, Public Transportation and Utilities, Refining Concepts and Measures of Consumer Satisfaction and Complaining Behavior. Hunt and Day Eds., Indiana Univ.
Bagozzi, Richard P. (1980), Causal Modelling in Marketing. John Wiley and Sons, N. Y.
(1982), A Field Investigation of Causal Relations Among Cognitions, Affect, Intentions and Behavior, Journal of Marketing Research. Vol. 29, Nov., pp. 562-84.
(1983), A Holistic Methodology for Modelling Consumer Response to Innovation, Operations Research. Vol. 31, No. 1, pp. 128-76.
(1984), A Prospectus for Theory Construction in Marketing, Journal of Marketing. Vol. 48, No. 1, pp. 11-29.
226
227
(1984), A Prospectus for Theory Construction in Marketing, Journal of Marketing. Vol. 48, No. 1, pp. 11-29.
Barnes, James G. and Kelloway, Karen R. (1980), Consumerists: Complaining Behavior and Attitudes Toward Social and Consu-mer Issues, Association of Consumer Research Proceedings. Jerry Olson Ed., Vol. 7, pp. 329-34.
Bateson, John E. G. (1979), Why We Need Service Marketing, Conceptual and Theoretical Developments in Marketing. Ferrell, Brown and Lamb Eds., AMA.
Bearden, William O. and Crockett, Melissa (1981), Self Monitoring, Norms and Attitudes as Influences on Consumer Complaining, Journal of Business Research. Vol. 9, pp. 255-66.
Bearden, W. 0., Lichtenstein, Donald R. and Teel, Jesse E. (1983), Reassesment of the Dimensionality, Internal Consistency and Validity of the Consumer Alienation Scale, AMA Educators Conference Proceedings, pp. 35-40.
Bearden, W. 0., and Mason, J. B. (1983), Empirical Evidence on Marketplace Alienation, Journal of MacroMarketing. Fall, pp. 6-20.
, and (1984), An Investigation of Influences on Consumer Complaint Reports, ACR Proceedings. Kinnear Ed.
Bearden, W. 0. and Teel, Jesse E. (1983), Selected Determinants of Consumer Satisfaction and Complaint Reports, Journal of Marketing Research. Vol. 20, Feb. , pp. 21-8.
Bearden, W. O., Teel, J. E. and Crockett, M. (1980), A Path Model of Consumer Complaint Behavior, Marketing in the 80's, 1980 AMA Educators Conference Proceedings. Bagozzi et al. Eds.
Bernhardt, Kenneth L. , Robinson, Larry M. and Semans, Debra (1983), Consumer Complaint Handling: Self Regulation Vs. Government Regulation, AMA Educators Conference Proceedings, pp. 369-73.
Berry, Leonard L. (1980), Service Marketing is Different, Business Week. May-June, pp. 24-29.
228
Best, Arthur and Andreasen, Alan R. (1977), Consumer Responses to Unsatisfactory Purchases: A Survey of Perceiving Defects, Voicing Complaints and Obtaining Redress, Law and Society Review. Vol. 11, Spring, pp. 701-42.
Biehal, Gabriel J. (1983), Consumer Prior Experiences and Perceptions in Auto Repair Choice, Journal of Marketing. Vol. 47, Summer, pp. 82-91.
Blalock, H. M. (1968), The Measurement Problem, in H. M. Blalock and A. Blalock (Eds.,) Methodology in Social Research, pp. 5-27, McGraw Hill, N. Y.
Brinbaum. A (1968), Test Scores, Sufficient Statistics, and the Information Structures of Tests, in F. M. Lord and M. R. Novick, Statistical Theories of Mental Test Scores. Reading, Addison-Wesley.
Census Information (1983), 1980 Census Information for City and County, Planning Department. City of Lubbock. Aug., 1983.
City Directory (1985), Lubbock City Directory. R. L. Polk and Company, Chicago, 111.
Clabaugh, Jr., Maurice G., Mason, J. B. and Bearden, W. 0. (1979), Consumer Alienation and Causal Attribution as Moderators of Consumer Satisfaction-/Dissatisfaction and Compalint Behavior, New Dimensions of Consumer Satisfaction and Complaining Behavior, Day and Hunt Eds., Vol. 3, Indiana University, pp. 2-9.
Cliff, Norman (1979), Test Theory Without True Scores?, Psychometrica. Vol. 44, No. 4, pp. 373-93.
Cudek, R., McCormick, D. J., and Cliff, N. (1979), Monte Carlo Evaluation of Implied Orders as a Basis of Tailored Testing, Applied Psychological Measurement. Vol. 3, pp. 65-74.
Day, Ralph L. (1980), Research Perspectives on Consumer Complaining Behavior, Theoretical Developments in Marketing, AMA Educators Conference Proceedings. Lamb and Dunne Eds., pp. 211-215.
(1984), Modelling Choices Among Alternate Responses to Dissatisfaction, Association of Consumer Research Proceedings, Vol. 11, Kinnear Ed., pp. 496-9.
229
Day, Ralph L., and Ash, Stephen B. (1979), Consumer Responses to Dissatisfaction with Durable Products, Association of Consumer Research Proceedings. Vol. 6, Ed. Wilkie, pp. 438-44.
Day, R. L., and Bodur, M. (1977), A Comprehensive Study of Satisfaction with Consumer Services, Consumer Satisfaction. Dissatisfaction and Complaining Behavior. Day Ed., pp. 64-70.
Day, R. L. , Grabricke, K., Schaetzle, T. , and Staubach, F. (1981), The Hidden Agenda of Consumer Complaining, Journal of Retailing. Vol. 57, Fall, No. 3, pp. 86-106.
Day, R. L. and Landon, Laird E., Jr. (1977), Towards a Theory of Consumer Complaining Behavior, Consumer and Industrial Buying Behavior. Arch G. Woodside, Jagdish N. Sheth, and Peter D. Bennett Eds., North-Holland, NY, pp. 425-437.
Drasgow, F. and Parsons, C. K. (1983), Application of Uni-dimensional Item Response Theory Models to Multidimensional Data, Applied Psychological Measurement. Vol. 7, No. 2, Spring, pp. 189-99.
Engel and Blackwell (1982), Consumer Behavior. 4th edition, Dryden Press.
Engel, J. F., Koilat, D. T. and Blackwell, R. D. (1968), Consumer Behavior. 2nd Ed., Prentice Hall.
Enis, Ben M. and Peering, Kenneth (1981), Services Marketing: Different Products, Similar Strategies, Marketing of Services, Donnelly and George Eds., AMA
Festinger, L. (1957), A Theory of Cognitive Dissonance. Stanford University Press, Stanford.
Fishbein, Martin and Ajzen, Icek (1975), Belief Attitude Intentions and Behavior: An Introduction to Theory and Research. Addison Wesley.
Foikes, Valerie S. (1984), Consumer Reactions to Product Failure: An Attributional Approach, Journal of Consumer Research. Vol. 10, March, pp. 398-409.
Fornell, Claes and Didow, Nicholas M. (1980), Economic Constraints on Consumer Complaining Behavior, Association of Consumer Research Proceedings, Vol. 7, Olson Ed., pp. 318-23.
230
Fornell, Claes and Larcker, D. V. (1981), Evaluating Structural Equations Models with Unobservable Variables and Measurement Error, Journal of Marketing Research. Vol.18, Feb., 1981, pp. 39-50.
Fornell, C. and Robinson, W. (1983), Industrial Organization and Consumer Satisfaction/Dissatisfaction, Journal of Consumer Research. March, Vol. 9, pp. 403-12.
Fornell, C. and Westbrook, Robert A. (1979), An Exploratory Study of Assertiveness, Aggressiveness and Consumer Complaining Behavior, Association of Consumer Research Proceedings. Vol. 6, Wilkie Ed., pp. 105-110.
Gilly, Mary C. and Gelb, Betsy (1982), Post-Purchase Consumer Process and the Complaining Consumer, Journal of Consumer Research. Vol. 9, Dec., pp. 323-8.
Gronhaug, K. and Zaltman, G. (1981), Complainers and Noncomplainers Revisited: Another Look at the Data, ACR Proceedings. Vol. 8, Monroe Ed.
Guttman, L. (1950), Chapters 2, 3, 6, 8, and 9, in S. A. Stouffer et al. (Eds.,), Measurement and Prediction. Princeton University Press, Princeton, N. J.
Hambelton, R. K. and Van der Linden, W. J. (1982), Advances in Item Response Theory and Applications: An Introduction, Applied Psychological Measurement. Vol. 6, No. 4, Fall, pp. 373-8.
Helson, H. (1964), Adaptation Level Theory: An Experimental and Systematic Approach to Behavior. Harper and Row, N. Y.
(1959), Adaptation Level Theory, in S. Koch (Ed. ), Psychology: A Study of Science. Vol. 1, McGraw Hill, N. Y.
Hirschman, Albert O. (1970), Responses to Decline in Firms. Organizations and States. Harvard University Press, Cambridge, Mass.,
Howard, John A. and Sheth, Jagdish (1969), The Theory of Buyer Behavior. New York, Wiley.
Hulin, Charles, Drasgow, F. and Parsons, C. (1983), Item Response Theory; Applications to Psychological Measurement. Dow-Jones-Irwin.
231
Hulin, C., Lissak, R. I. and Drasgow, F. (1982), Recovery of Two- and Three-Parameter Logistic Item Characteristic Curves: A Monte Carlo Study, Applied Psychological Measurement. Vol. 6, No. 3, Summer, pp. 249-60.
Johnson, Richard A. and Wichern, Dean W. (1982), Applied Multivariate Statistical Analysis. Prentice Hall, Engelwood Cliffs, N.J.
Jacoby, Jacob and Jaccard, James J. (1981), The Sources, Meaning and Validity of Consumer Complaining Behavior: A Psychological Review, Journal of Retailing. Vol. 57. No. 3, Fall, pp. 4-24.
Joreskog, Karl and Sorborm, Dag (1981), LISREL VI: Analysis of Linear Structural Relationship by Maximum Likelihood and Least Sguares Method. lES, Chicago.
Kangun, Norman, Cox, Keith K., Higginbotham, J. and Burton, J. (1975), Consumerism and Marketing Management, Journal of Marketing. Vol. 39, April, pp. 3-10.
Kendall, C. L. and Russ, F. A. (1975), Warranty and Complaint Policies: An Opportunity for Marketing Management, Journal of Marketing. Vol. 39, April, pp. 36-43.
Krishnan, S and Valle, Valerie A. (1979), Dissatisfaction Attributions and Consumer Complaining Behavior, Association of Consumer Research Proceedings, Wilkie Ed. , Vol. 6, pp. 445-9.
Lambert, Zarrel V. (1980), Consumer Alienation, General Dissatisfaction and Consumerism Issues, Journal of Retailing. Vol. 56, No. 2, Summer, pp. 3-24.
Landon, Jr., E. Laird (1980), The Direction of Consumer Complaint Research, Association of Consumer Research Proceedings, Vol. 7, Olson Ed., pp. 335-8.
(1977), A Model of Consumer Complaint Behavior, Consumer Satisfaction, Dissatisfaction and Complaining Behavior. Ralph Day Ed., pp. 31-5, Indiana University Press.
232
Langmeyer, Mellnda and Langmeyer, Daniel (1979), A Behavioral Assesment of Complaining and Its Implications for Further Research, New Dimensions of Consumer Satisfaction and Complaining Behavior. Day and Hunt Eds., Vol. 3, Indiana Univ., pp. 26-9.
Latour, Stephen and Peat, Nancy (1979), Conceptual and Methodological Issues in Satisfaction Research, ACR Proceedinqs. Vol. 6, Wilkie Ed.
Lawley, D. N. (1944) The Factorial Analysis of Multiple Item Tests, Royal Society of Edinburgh Proceedings. Vol. 62, pp. 74-82.
(1943) On Problems Connected With Item Selection and Test Construction, Royal Society of Edinburgh Proceedings. Vol. 61, pp. 273-87.
Lazarsfeld, P. F. (1950), Chapters 10 and 11 in Measurement and Prediction. S. A. Stouffer et al. Eds., Princeton University Press, N. J.
Linn, R.1., Levine, M, V., Hastings, C. N. and Wardrop, J. L. (1981), Item Bias in a Test of Reading Comprehension, Applied Psychological Measurement. Vol. 5, pp. 159-73.
Lord, Fredrick (1980), Applications of Item Response Theory to Practical Testing Problems. Lawrence Erlbaum, N. J.
(1952), A Theory of Test Scores. Psychometrica Monograph, No. 7.
Lord, F. M. and Novick, M. R. (1968) Statistical Theories of Mental Test Scores. Reading, Mass., Addison-Wesley Publishing Co. **
Lovelock, Christopher (1983), Classifying Services to Gain Strategic Marketing Insights, Journal of Marketing. Vol. 47, No. 3, pp. 9-20.
Lundstrom, William J. and Lament, Lawrence M. (1976), The Development of a Scale to Measure Consumer Discontent, Journal of Marketing Research. Vol. 13, Nov., pp. 373-81.
Lundstrom, W. J., Gerald Skelly and Donald Sciglimpagila (1979), How Deep are the Roots of Consumer Discontent: A Study of Rural Consumers, New Dimensions of Consumer Satisfaction and Complaining Behavior. Day and Hunt Eds., Indiana Univ.
233
MacDonald, Roderick P. (1982) Linear Versus Linear Models in Item Response Theory, Applied Psychological Measurementr Vol. 6, No. 4, pp. 379-96.
McGuire, W. J, (1976), Some Internal Psychological Factors Influencing Consumer Choice, Journal of Consumer Research. Vol. 2, No. 4, pp. 302-19.
McGuire, W. J, (1968), Personality And Susceptibility To Social Influence in Handbook Of Personality Theory And Research. E.F. Borgatta and W.W. Lambert Eds., Chicago, Rand McNally, pp. 1130-87.
Miniard, Paul W., Obermiller, Carl and Page, Thomas J. Jr., (1983), A Further Assesment of Measurement Influences on the Intention-Behavior Relationship, Journal of Marketing Research. Vol. 20, May, pp. 206-12.
Nunally, J. C. (1978), Psychometric Theory. McGraw Hill, N. Y.
Oliver, Richard L. (1980), A Cognitive Model of the Antecedents and Consequences of Satisfaction Decisions, Journal of Marketing Research. Vol. 17, Nov., pp. 460-9.
(1980a), Theoretical Bases of Consumer Satisfaction Research: Review, Critique and Future Directions in Theoretical Developments in Marketing, AMA Educators Conference Proceedings. Lamb and Dunne Eds. , pp. 206-10, AMA, Chicago.
(1977), Effect of Expectation and Disconfirmation on Post-exposure Product Evaluations: An Alternative Interpretation, Journal of Applied Psychology. Vol. 62, Aug., pp. 480-86.
Olshavsky, Richard W. and Miller, John (1972), Consumer Expectations, Product Performance, and Perceived Product Quality, Journal Of Marketing Research. Vol. 9, pp. 19-21.
Olson Jerry C. and Dover, Philip (1976), Effects Of Expectation Creation And Disconfirmation On Belief Elements Of Cognitive Structure, 1975 Annual ACR Conference Proceedings. Beverlee B. Anderson Ed. , pp. 168-175.
Ott, Lyman (1977), An Introduction to Statistical Methods and Data Analysis. Duxbury Press, Mass.
234
Petty, Richard E and Cacioppo, John T. (1981), Attitudes And Persuation: Classical And Contemporary Approachegr Wm. C. Brown Company Publishers.
Rasch, G. (1960), Probabilistic Models for Some Intelligence and Attainment Tests. Danish Institute for Educational Research, Copenhagen.
Richins, Marsha L. (1979), Consumer Complaining Processes: A Comprehensive Model, New Dimensions of Consumer Satisfaction and Complaining Behavior. Day and Hunt Eds., Vol. 3, Indiana Univ., pp. 30-4.
(1980), Consumer Perceptions Of Costs And Benefits Associated With Complaining In Refining Concepts And Measures Of Consumer Satisfaction And Complaining. Hunt and Day Eds., pp. 105-8, Indiana University.
(1982), An Investigation of Consumer's Attitudes Toward Complaining, Association of Consumer Research Proceedings. Vol. 9, Mitchell Ed. pp. 502-6.
t
(1983), An Analysis of Consumer Interaction Styles in the Marketplace, Journal of Consumer Research. Vol. 10, June, pp. 73-82.
Ritchie, J. R., Brent, M. and Claxton, John (1979), In Search of Actions to Reduce Consumer Shopping Problems, Association of Consumer Research Proceedings, Vol. 6, Wilkie Ed., pp. 466-71.
Robinson, Larry M. (1979), Consumer Complaint Behavior: A Review With Implications for Further Research, New Dimensions of Consumer Satisfaction and Complaining Behavior. Vol. 3, Day and Hunt Eds., pp. 41-50, Indiana University.
Russo, Edward J. (1979), Consumer Satisfaction-/Dissatisfaction: An Outsiders View, Association of Consumer Research Proceedings. Vol. 6, Wilkie Ed. , pp. 453-55.
Samejima, F. (1977) Weakly Parallel Tests In Latent Trait Theory With Some Criticisms Of Classical Test Theory. Psychometrica, Vol. 42, No. 2, pp. 193-199.
Samejima, F. (1972) A General Model For Free-Response Data. Psychometric Monograph No. 18.
235
Sherif, M. and Hovland, C. I. (1961), Social Judgement: Assimilation and Contrast Effects in Communication and Attitude Change. Yale University Press, New Haven.
Shostack, G Lynn (1977), Breaking Free from Product Marketing, Journal of Marketing. Vol. 41, April, 73-80.
Swaminathan, H. and Gifford, J. A. (1979), Estimation of Parameters in the Three Parameter Latent-Trait Model. Laboratory of Psychometric and Evaluation Research, Report No. 90, University of Massachusetts, Amherst, Mass.
Swan, J. E. (1977), Consumer Satisfaction With a Retail Store Related To The Fulfillment of Expectations on an Initial Shopping Trip, in Consumer Satisfaction Dissatisfaction and Complaining Behavior Proceedings, Day Ed., Indiana University.
Thibaut, J. W. and Kelly, H. H. (1959), The Social Psychology of Groups. New York: John Wiley and Sons.
Thrustone, L. L. (1928), Attitudes Can Be Measured, American Journal of Sociology. Vol. 33, pp. 529-54.
Torgerson, W.S. (1958) Theory And Methods Of Scaling, New York: John Wiley and Sons.
Tucker, L. R. (1946) Maximum Validity Of A Test With Equivalent Items, Psychometrica. Vol. 11, pp. 1-13.
Valle, V. A. and Waliendorf, M. (1977), Consumer's Attributions of the Cause of Their Product Satisfaction and Dissatisfaction, in Consumer Satisfaction. Dissatisfaction and Complaining Behavior Proceedings, Day Ed. , Indiana Univ.
Villareal-Camacho, Angelina (1983), Consumer Complaining Behavior: A Cross-Cultural Comparison, AMA Educators Proceedings, pp. 68-73, AMA Chicago.
Warland, Rex H., Herrmann, Robert 0. and Willits, Jane (1975), Dissatisfied Consumers: Who Gets Upset and Who Takes Action, Journal of Consumer Affairs. Winter, pp. 198-163.
Weiss, D. J. and Davison, M. L. (1981), Test Theory and Methods, Annual Review of Psychology, Vol. 32, pp. 629-58.
236
Westbrook, Robert A. (1980), Intra-Personal Affective Influences on Consumer Satisfaction with Products, Journal of Consumer Research. Vol. 7, June, pp. 49-54.
(1980a), A Rating Scale for Measuring Product/Service Satisfaction, Journal of Marketing. Vol. 44, pp. 68-72.
(1983), Consumer Satisfaction and the Phenomenology of Emotions During Automobile Ownership Experiences, in International Fare in Consumer Satisfaction and Complaining Proceedings. Day and Hunt Eds, Indiana Univ.
Westbrook, Robert A. and Joseph W. Newman (1978), An Analysis of Shopping Dissatisfaction for Major Household Appliances, Journal of Marketing Research. Vol. 15, Aug., pp. 456-66.
Wingersky, M. S., Barton, M. A. and Lord, F. M. (1982), LOGIST User Guide. ETS, Feb., 1982.
Wood, R. L. and Lord, F. M. (1976), A Users Guide to LOGIST, Research Memorandum 76-4. Educational Testing Service. Princeton, N. J.
Woodruff, Robert B., Cadotte, Ernest and Jenkins, Roger (1983), Modelling Consumer Satisfaction Processes Using Experience Based Norms, Journal of Marketing Research, Vol. 20, Aug., pp. 296-304.
Woodside, A. G. (1972), Positive Disconfirmation of Expectations and Effect of Effort on Evaluation, 80th Annual APA Convention Proceedings. APA, pp. 743-744.
Wright, B. (1980), Foreward in Probabilistic Models for Some Intelligence and Attainment Tests. University of Chicago Press.
Wright, B. and Mead, R. J. (1977), BICAL: Calibrating Items and Scales with the Rasch Model. Research Memorandum No. 23, Univeristy of Chicago press.
Zeller, Richard A. and Carmines, Edward G. (1980), Measurement in The Social Sciences, Cambridge University Press.
APPENDICES
A. PHASE I QUESTIONNAIRE B. A TEXT OF THE INTRODUCTION USED BY THE INTERVIEWERS C. PHASE II CENSUS TRACTS D. PHASE II QUESTIONNAIRES E. SELECTED DISCONTENT AND ALIENATION ITEMS
237
238
APPENDIX A: PHASE I QUESTIONNAIRE
Con^umer Sat 's^iCtipn Sijrve/
This questionnai ro is designed to determine consumers' sa t i s fac t ion , bel iefs and at:it j<les " " i ' - : : ; ^jsi '^'ss. - ^ * ) ' adver t is ing , e t c . There are no r ight or wrong answers to the questions f^at *^orow. Howpvar, /^ur :.>^sor4' : 3 ' - - : important. Therefore, please answer aj_l questions as best as you can.
" . t ) o r ' . la' Please express the extent to which you agree or disagree with each of the foPowng stjtemeitj. C--: • best represents your opinion.
The questionnaire is rather long and has six sections. Some of the questions -wy appear sii-'ir ").' jctja'. "^ic question is different and designed to measure a unique aspect of consumer satisfaction.
1.
2.
3.
4,
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
Section I
5trongly Dlsaqroo
Most Companies are responsive to the demands of the consumer I
The business community has been a large influence in raising our country's standard of living i
Business profits are too high
It seems wasteful for so many companies to produce the same basic products.
Styles change so rapidly a person can't afford to keep up
People who sell things over the telephone -ire always trying to gip you
Unethical practices are widespread throughout business
Advertising is a good source of information
Credit manes things too easy to buy •
Stores do not care why people buy their products just as long as they make a profit
Many times I need assistance in a store and I'm just not able to get
Warranties would not be necessary if the manufacturer made the produc" right in the first place
Shopping is usually a pleasant experience.
Salesmen really take an interest in the consumer and make sure he finds what he wants
Products that last a long time are a thing of the past
People are unable to help determine what products will be sold in the store.
Business takes a real interest in the environment and is trying to improve it
Food which is not nutritious Is another example of business trying to make a buck and not caring about the consumer
Advertising and promotional costs unnecessarily raise the price consumer has to pay for a product
People rate other people by the value of their possessions
Business firms usually stand behind their products i guarantees
What a product claims to do and what it actually does are two different things
When a product is advertised as "new" or "improved" it is the same old thing only in a different package
Industry has an obligation to clean up the waste they have been dumping but they aren't doing it
23.
24.
25. Mass production has done away with unique products
2
2
2
2
2
2
7
2
2
2
2
2
2
2
2
2
2
2
2
2
•^
•>
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
•3
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
239
Section I I
Chain stores ire getting so big that they --oally don't treat the customer personally
,*'0'-
Disac 5 y f OO
2,
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23,
24.
25.
Permanent price controls are the only way to end inflation
Misrepresentation of product features is just something we have to live with.
The quality of goods has consistently improved over the years
Many times the salesman says one thing to the shopper but he knows it's just the opposite
Harmful characteristics of a product are often kept from the consumer.
Many times it's easier to buy a new product rather than trying to fix the old one
The only person who cares about the consumer Is the consumer himself.
It is embarrassing to bring a purchase back to the store
The actual product I buy is usually the same as advertised
It is hard to make a buying decision because of all the products to choose from
I tend to spend more than I should just to Impress my friends with how much I have
The small businessman has to do what big business says or else!,
Most companies have a complaint department which backs up their products and handles consumer problems
Even with so much advertising it is difficult to know what brand is the best
Business is the one using up our natural resources (oil, gas, trees etc ) but it does nothing to replace what has been taken.
Many companies listen to consumer complaints but they don't do anything about them
A sale is not really a bargain but a way to draw people into the store....
Generaly speaking, products work as good as they look
Products fall apart before they have had much use
It is difficult to identify with business practices today
Products are only as safe as required by goverment standards, but no more.
Stores advertise "special deals" just to get the shopper into the store to buy something else
It is difficult to Identify with current trends 4 fads in fashion.
Companies are helping minorities and under privileged by providing them with jobs
4 q r a a
5 6
5 "
5 6
5 6
Section 111
1. The information on most packages 1$ enough to make a good decision.
2. I often feel guilty for buying so many unnecessary products
3. Most salesmen who call at home try to force the consumer Into buying something
Strongly Disagree
Sfongly Agree
240
:)is: •o-g ,/
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24,
25.
All business really wants to do is to maxe the most icrey it can
Most brands are the same -ith just different names and labels
The business community is actively involved in solving social jroolems
Most people know that advertising lies a "little."
A product will usually break down as soon as the warranty is up
Companies encourage the consumer to buy more than he really needs
The goverment should enforce ethical business practices
Business is responsible for unnecessarily depleting our natural resourc»s.
The consumer knows exactly what he is buying with food products
because tne ingredients are on the package
Companies aren't willing to listen or do anything about consumer gripes..
Companies try to influence goverment just to better themselves
Recycling o* products is one way business is cleaning up the environment.
Business does not help local residents because it's not profitable
One must be willing to tolerate poor service from most stores
When the consumer is unsure of how good a product is, he can get the correct information from a salesman
The consumer is usually the least important consideration to most compan'->s.
It is difficult to know what store has the best buy
Salesmen are "pushy" just so they can make a sale
If all advertising stopped, the consumer would be better off
Business's orime objective is to make money rather than satisfy the consumer
Sales clerks in stores just don't care about the consumer anymore.
Most products are safe when they are used right
2
2
?
2
2
2
2
Z
2
2
7
2
2
2
2
2
2
2
2
2
2
Section IV
1. ; often feel •'rustrated when I fall to find what I want in the store.
2. Advertised "specials" aren't usually in the store when the shopper goes there
3. Service departments "pad" the bill by charging for unneeded work,
4. After making a purchase, I often find nyself wondering "why." ...
5. The price I pay Is about the same as the quality I receive
6. Companies try to take a personal interest in each consumer rather than treating him as a number
7. It is hard to understand why some brands are twice as expensive as others.
8. As soon as they make the sale, most businesses forget about the buyer
9. Comiercials make a person unhappy with himself because he can't have everything he sees
Strongly Oisi gree
10. It is not unusual to find out that business has lied to public.
2
2
2
2
2
2
2
2
2
Strongly Agree
6
6
6
6
6
6
6
6
6
241
11.
12.
13.
14.
L5.
16.
17.
18.
19,
20.
21.
22.
23.
24.
25.
26.
27,
28.
29.
30.
Health and safety wamifigs on packages are not adequate enough to inform the consumer of possible danger ,
Service manuals aren't provided for products because the company wants to make money servicing products as well as selling them...
Buying beyond one's means is justifiable through the use of :redit.
What is seen on the outside of a package is many times not what you get on the inside
There are too many of the same types of products which is a waste of money..
It is often dif f icult to understand the real meaning of most advertisements.
In general, companies are honest in their dealings with the consumer
Prices of products are going up faster than the incomes of ordi nary consumer
Products are designed to wear out long before they should
Advertising tempts people to spend their money foolishly
Most claims of product quality are true
Business profits are high yet they keep on raising their prices.
I am often dissatisfied with a recent purchase
Companies generally offer what the consumer wants
The wide variety of competing products makes intelligent buying decisions more di f f icul t
Business has commercialized many meaningful holidays, such as Christmas.
Advertisements usually present a true picture of the product
The main reason a company does things for the society is to make more sales
I often don't like to return or exchange products I am dissatisfied wi*:-.
An attractive package many times influences a purchase that isn't necessary
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
•>
5
5
5
5
5
5
5
5
5
5
5
6
0
S
6
6
6
6
6
S
6
Section V
1. Store employees are often quite unpleasant to customers who return unsatisfactory products
2. A large variety of products allow the consumer to choose the one that he really wants
3. Making a complaint about a defective product usually leads to frustration.
4. Self-service stores leave the consumer at the mercy of how the products looks
5. I envy people who take the effort and courage to complain about unsatisfactory products
6. Conpanles "jazz up" a product with no real improvement, just to get a higher price or sell more
7. Complaining to business is usually done by people with l i t t l e else to do.
8. Most of the things I buy are over-priced
9. It bothers me quite as bit i f I don't complain about an unsatisfactory product
Strongly D'sagree
2
2
2
2
2
2
2
Strongly Agree
5
5
5
5
5
5
5
242
s
10. Prices are reasonable given the h'gh cost of doing business
11. People have a responsibility to tell stores when a product 'hey purchase 1s defective
12. Promotional or "junk" mail is just a waste
13. It sometimes feels good to get ny dissatisfaction and frustration
with the product off my chest by complaining
14. Repairs take too long because the right part is often not in stock
15. People are bound to end up with unsatisfactory products once in a while, so they should not complain
16. Advertising tells the shopper about things he would not ordinarily hear about
17. Complaining isn't much fun. but it's got to be done to keep business from becoming irresponsible ,
18. A warranty or guarantee may be a good one but the service
department is often unable to do the work correctly
19. Making a complaint about a defective product usually takes a lot of time ,
20. Repair work is usually done right the first time ,
21. I often complain when I an dissatisfied with business or products because I feel it Is my duty to do so ,
22. Business takes advantage of poor people or minorities by charging higher than normal prices
23. By making complaints about unsatisfactory products, in the long
run the quality of products will improve ,
24. The stock market is controlled by big financial institutions
25. "iost stores are willing to adjust reasonable complaints ,
26. Consumer activists, like Ralph Nader, do more harm than good to busines
27. I feel a sense of accomplishment when I manage to get a
complaint to a store taken care of
29. It is difficult to identify with business practices today
29. By complaining about defective products. I may prevent other consumers from experiencing the same problem ,
30. Many business say they want their customers satisfied but they ar» not willing to stand behind their word
31. I don't like people who make complaints to stores, because usually their complaints are unreasonable
3
Sectlon VI
IN THE NEXT SECTION WOULD YOU PLEASE GIVE US SOME BACKGROUND INFORMATION?
1. Are you: ^Male ^Female
2. Please check the category that represents your age.
15 to 20 years 36 to 40 years "21 to 25 years 41 to 45 years 26 to 30 years 46 to 50 years ~Ti tn -x^ w«»r« 51 to 55 years
56 to 60 years "61 years or more
_ 26 to 30 years _^31 to 35 years
243
3. Are you: ^Married ^Divorced Separated Wiiowed Sing--
4. How many persons, including yourself, presently live in your household? ^ _ _ _ _
5. What is your specific occupation? [job title]
6. What is the last year of formal education you completed? (Circle one number)
7. Please check the category that represents your total household Income ( jo in t i f married) "i :^83'
less than $10,000 $50,001 to $60,000 S100.00'. to $:20.000 "$10,001 to $20,000 $60,001 to $70,000 S'.JO.OO'. to SUC.QOO "$20,001 to $30,000 $70,001 to $80,000 $140,001 to $160,000 $30,001 to $40,000 $80,001 to $90,000 $160,001 to $180,000 $40,001 to $50,000 $90,000 to $100,000 $130,001 o^ ^
8. With which ethnic or racia l group do you ident i fy yourself? [Mark one only]
White ^Black ^Hispanic ^Ot"e' ioec i 'y )
THANK YOU FOR YOUR COOPERATION ( I f you desire further information on this survey, please attach your f u l l address)
244
APPENDIX B: A TEXT OF THE INTRODUCTION USED
BY THE INTERVIEWERS
Good Morning (Evening, Afternoon):
My name is
I am a student at the Texas Tech University and am
helping with a survey being conducted by the marketing
department.
Many people have problems or dissatisfaction with
products or services at department stores, etc. In this
study we are interested in finding out about your
feelings on such problems.
We would like your help in completing the survey
which will take only a short time and which can be
completed at your convenience. Your responses will be
kept confidential and will be used only to help us do our
survey. This is a university survey and does not involve
any businesses. Therefore, no salesman will ever call
you based on the information you provide.
I would like to return in one week and pick up the
survey between 6 p.m. and 8 p.m. If you find that you'll
be out at this time, please leave the survey in the
mailbox.
Thank you for your assistance.
APPENDIX C: PHASE I I CENSUS TRACTS
245
SOURCE; US. CENSUS - 1980
J C » > t I • 5 ' " ^ t l
246
\ r..^^v. I > [ ^
SOURCE: US CENSUS - 1960
247
( V 1<-'*^
1 .566
SOURCE; US. CENSUS - 1960
248
SO( ICE;U.S CENSUS - 1980
» L C >' • i ' M l L l t
~>7
2 4 9
APPENDIX D: PHASE I I QUESTIONNAIRES
This qu«»t1onniir» is designed to iJ«ten»in» th« s p e c ' ' c conoUmtj '"lit -ran^jMr?'. ..c"^ IS <-j -i»« t a ' C f SttOQpInq ) t t grjcgry store. There trt no right or xroflg insii«rs. ^'omti'r your : • ' • . : - ! ' :D "':"S i'» •"oar-? 1 e « t re«d the instruction* c j re fu l ly «nd 4ns«*r i l l questions.
Stctton I F i rs t I ' l l 90lnq to Kk you 50»e gefltril qu«$t1ont ibout /our :o<n*ons ' . i i r r - q ous -'s-. 's " •<• I oould Uko you to re»d eKh t t t t ea tn t «t i t wpetrs . rh«n mdlci'.e ".ni! • i t * " : :* / c - i : - - - ^ * - ' by c i rc l ing tha nupbc^ tb«t best d«jcrib«» your reKtIof l to the stitcMent. > « ei'.*'?:'-'»s r »
(6) Stronqly Aqree («) Mree Sowewhit 2; : S K " » (5) Aqreo (3) OiJiqree 5aiie«h*t 1 i ' s i ; - ' ^ '>:'-.^:
Mean 4 .51 2.56
4.55
4.30 2.38
4.17
5.35
2.94 4.67
3.95
3.26 4.12
4.85
1.74 4.28
4 .60
2.85 4.36
3.94
2.76 3 .52
4.59
1.71 4 .36
2.18
2.47
( s td) 1.16) 1.29)
1.32)
1.28) 1.12)
1.30)
0 .80)
1.23) 1.09)
1.33)
1.25) 1.29)
1.13)
1.08) 1.14)
1.13)
1.37) 1.46)
l^j.
1. CoMpU'iing to business <s usually done by oeoo'e vir.r\ l i c e
2. Most coi"oanies c*re nothing at all about the consjner
1 0 . .
3. Industry has an obligation to clean jo tne «aste they ia»* been l-^o they aren't doing it
«. I t bothers me quite a bit if I do not comolain about in ^nsat'S'act :)ry j ' - ja^ct. i
5. The business community has been a Urge '"fluence in r j i smq :ur : ; j r ' t ' y ' ; standard of l iv ing *
6. Chain stores ire getting so big that they don't treat the consjr^er j e ' s o n i H / - - 5
7. People have a resoonstbiHty to te l l stores inen a product t^ey j ' . rcase is defective *
8. Shopping is usually an unpleasant e«perience
9. Stores advertise "special deals" Just to get the shoooer nto t"e store to buy something else
10. I t sometimes 'eels good to get my dissatisfaction and ' rus t ra fon «ith f e product off my chest by complaining
U . People ire unable to detennine «hat products wi l l be sold n tie store.
12. All business real ly xants to do is to maHe the most money i t can
13. People are bound to end up . i t h unsatisfactory products once "i i . h i i e , so they should not complain
14. Misrepresentation of product features is just something «• - ive to I've « > t i . . . i
15. C(W)*ii«s encourage the consumer to buy more than he ' - 1 / needs 6
16. Canplaining isn' t much fun. but i t ' s got to be done to .??o Businesses 'ram becoming irresponsible
17. The small businessman has to do <<hat big business says ;-• sUe! i
13. I t is hard to understand why some brands are twice as e.pensive is K i e r s a
1 . 3 9 ) 19. I often complain when I n dissatisf ied with business or products because : 'ee' i t is my duty to do so
1 . 2 9 ) 20. The consumer is usually the least important consideration to most companies.... 6
1 ' 3 9 ) 21 . As soon as they make a sale, nost businesses forget about the buyer 6
1 . 0 0 ) 22. By making conplilnts about unsatisfactory products, in tne 'ong run -ne
qual i ty of products wi l l improve
1 . 1 2 ) 23. One must b« wi l l ing to tolerate poor servce 'rom most stores 6
1 . 3 6 ) 2A. P r icM of products are going up faster than the incomes of o rd ' i j ry co"s.«ie' . . . S
0 . 6 6 ) 25. Most Stores *re w i l l ing to adjust reasonable complaints *
0 . 8 5 ) 2». C9»p«nies general ' / of 'er what tne consigner wants *
5
5
5
5
5
5
3
5
S
4
a
)
4
4
4
4
4
4
4
4
4
4
4
•?1 S - ) - » s .
250
Mean ( s t d ) ;•''
• ( 1 - 1 ' ^ ) 27. 3us.ness p r o f t ; are - , , « , e , »ney .eep on r j .s ing -n j . r p r : , , ,
4 . 6 5 ( 1 . 1 1 ) 2 8 . I feel i sense of accompl ->nn,eot * . n ; , , „ , , , to ^et a comp. n-. -o , ;• - . taken care o * . , , , , , ,
i
2 . 1 8 ( 1 . 0 5 ) 2 9 . In general companies are pU.n dishonest 'n tne'r l e j i n s , , ; , . . . c.^s^r.' . •,
'' 30. 4n attract ive package sometimes in'ljences i ourcmse ••^jt isn t - f c e s i ) ' / o 4 . 8 1 ( 0 . 9 1 ) 31. By complaining about defective products. I nay or. ,go: otner - - , ^ » - , ' rv -
eipenencing the same problem ^
2 . 8 1 ( 1 . 1 1 ) 3 2 . I w often dissatisf ied with a recent purchase ^
4 . 3 1 ( 0 . 9 7 ) „ , ^ . ,. JJ. Lomganies jazz up" a product with no real norovement, i.st -^ a»r , - .n.r
price or sell more . . , . . . . ' . . . ' . . . . . . i
•' 34. I don't l ike people who complam to stores, because ..%^i^',y t"*"- c jnp i i ' - t s i re unreasonable ' . . . . . 6
Z . i o ^ i . U J ; 35_ 4 ij^^g variety of o'cducts allows consumers to choose 'ne -"• 'ne/ ' > i ' ' / »*nt 6
4 . l y ( 1 . Z U ; 36. Companies try to influence government just to better tnemse'«»s i
2 . 7 0 ( 0 . 8 5 ) 37. Business firms stand behind tnoir products and guaranties i
Section I I In this section, I vould l ike to ask you about your experience in handling problems md co»»' i ' - t5 wntie shooo"« at a grocery store.
1 . Have you, in tne last six ( 6 ) , months contacted any store/manufKturer regaro'ng j n , snoso''; orob'«n. -,-;•< as overcharge, bad product, refund, replacement, etc.?
Yes 10
I f yes, how often? about I or 2 times
about 3 to 6 times about 7 to 12 t nes
•nee tnan 12 t-mes
3 .12 (1 .40 )
2. How often do you talk to your friends and relatives regarding problems and complai'^ts / C J ni»» . n ' « siooo'-'g for grocery products (c i rc le appropriate number)?
6 5 1 3 ? ; Often le.er
3. -lave you in the last six (6) months contacted a consumer :r OJOI IC igenc/, sucn is t-e 3 e : t i ' 3ui"ess ^ji-eau regarding any complaint?
'es '*Q
I f yes, roughly how many times m the last six months? times aoprox'njte' /
4. -lave /ou ever taken formal action against a store or •na-^-acturer regarding any : ' /o^r :omp'i"ts?
res
I f yes, roughly now many times? 10
t 'mes loprox imate 'y
Section I I I Next, I would l ike for you to think of * recefit prpblea that you reacmber most c'early stth shopping at a grocery store. For exa4>lt . you may have had probl««« with the store i t s e l f , its cNployces or with products/produce purchased at th« store. Keeping in mtnd your particular p r o b l « , please ansaer the following questions.
1 . Please describe br ie f ly your problem or dissatisfaction:
2. NaM and location of tne jrocery store involved:
231
Meafl ( s t d ) *• " * "•""*' ••* '"»ol»ed, wnat was it: Product (food, o i l . produce, otnen ^ •rand name of tne uea ( i f aodlicao'ei "
t * r m ??^"* ' "^"^^ ' '^* *" " " ""' • ' • " ""•"'ons about the pr,bl«> and f o you 'ee' s respors.b't 'or t L U f l i l i l K* T f? ''••^ • • ' ' ' " ' *•* 'o"<*'"9 «4t€iM«tj and then indicate the extent of /our igre^enc :r j i ; . aqrevent by clrcHno the appropriate nurter. ' ' '
2 . 1 5 ( 0 . 7 8 ) - • . The store policies arc to bl«ie for the above probioi 3 »
Z . J 3 ( 0 . 7 2 ) 5. The itorp personnel are simply careleis ;
1 . 5 8 ( 0 . 8 0 ) 6. The store didn't mean for the problem to xcur, but mistakes haooen 3 :
' . I exptcted too much of the store ;
2 . 0 9 ( 0 . 5 2 ) 8. It Is probably my fault, since I Should be more careful in Shopping : ;
1 . 9 0 ( 0 . 9 1 ) 9. The manufacturer Is responsible for the bad product ( if a product 's '".o'»eoi
10. Overall how dissatisfied were you before you did anything about the prob'jm s^c r. :: oic< 'o t-e -.t--*, etc. ) . Please circle the number that best represents your feeimg.
lOOS 90t 80S 70S 60S SOS 40S 30S 20S lOS tv Completely Oissatisfied lot Ji.sat'sf-ed at «i'
U . Which of the following action (s) did you take after you expenencad tne ioo«e oroe>"' -•:>'•• t"! ' '•» msaer Is ok).
After I experienced the above problem I:
_ ^ ^ _ ^ Forgot about the incident and did nothing Complained to the store on my next trip
Went back or called the store immediately and asked 'nem to take care : ' iy o'p»i«» Decided never to shop again at that store
Told my friends and relatives about my bad experience _^____ Complained to a consumer agency, such as the Better Business 3ur*iu
Took some formal action against the store/manufactjrer Other, please specify
12. ow please tell us how you felt about the whole Incident after you nad taken tne above Kt-on (s). Please circle the number that best represents your feeling.
lOOS 90S SOS 70S 60S SOS 40S 30S 20S lOS OS Completely Sat-sfied Not Satisfied at A''
13. What do you think grocery stores ought to do to solve the type of problem you nad?
Section I Next, imaglna that soMttmc in the future you were shopping at your most frequpntly v't'tP^ grocery store and a* Incidtnt sniilar to the one you Indicated above Kcurtil agam. Hem read each or tli« foiioving statannts aM indicate tht extent to which you thinli they are likely to happen. The categories an:
1. Assume you reported the incident to the store, how likely is it that the store would:
4.16(1.96) / 7 0 / 1 & M * ' 4Pb'09i'* liut ^ nothing 6 S 4 3 2 I < » . / o v i * 0 1 ^ b. take appropriate action to take care of your prob'em (refund, ' t c . ) . . . S S 4 3 2 1 3 . 8 6 ( 1 . 6 4 ) c. solve your problem and give better service to you m the future 6 5 4 3 2 1 _ * _ - ) , *^-% d. be more careful in future and everyone would benefit 6 3 4 3 2 1 3 . 7 J \ 1 . 6 7 )
2. Assi«« that you mentioned the problem to your friends and relatives who shop . at the same store, how likely is it that they would:
A -17/1 A 7 \ a. go on buyin? as usual 6 S 4 3 2 I n.^i^i-.tij b. be more careful when buying froei that grocery store 6 5 4 3 2 1 1 . 8 5 ( 1 . 1 7 ) c. stop buying froa that grocery store altogether 6 $ 1 3 2 1 2 7 0 ( 1 6 0 ) * ' " • ' * '"*' ***'** '""^ orobleei 8 5 4 3 2 1
252
» 3 4 !
"1
Mean ( s t d )
3. Assui* tnat you '-oort^o tie inco .n t to a consider iqe'c/ s . c is •»« " ' ^ 3 1 6 n 7 S 1 * * " • • • Business Bureau, now ' <e / s s tnit they .OJ :
3 . 1 2 ( 1 7 6 1 *• """ ' ' • ' ' 9 ' ' * /ou j n t i l many o f e ' s nave s m ' a r -jno i " - - . i - : J - > Q / i * t \ b. take no acton '" ^
3 . 2 9 ( 1 . 6 7 ) c. make the store t « e care of yo^r p rob lem! ! ! " ! ! m ] ] j ^ ] [ [ j [ ] ' ' ] j ] . " | . .; 2 . 9 4 ( 1 6 3 V "• ^o ' * * /•""• problem and ensure that tne store '% : i r . ^ / " i n " - - »
^ future 5
how I would l ike to ask you about what you might K t u a l l y do in the case tne above you were shopping at your most frequently visited grxery store.
J . 0 5 ( 1 . 5 3 ) 4. I f a similar problem occured again, how l ike ly is t tnat you .-.. i: '—^ ' 4 . 6 2 ( 1 . 7 5 ) *• 'orget the incident and do nothing .; i • : 2 : -1 n o / 1 7 Q \ ''• ' lefinately complain to the store naniqe' ;n /c^r -»<t " o •; i '• I Z '. J ' U O v I . /y) c. decide not to shop at tnat r^cery store iq j '^ ^ 5 : ; ; 4 . 7 0 ( 1 7 4 ) '^- 9" ' " ' ' ' "• <^*" *''• *tore imeO' i te ly and ask them 'o •Vi*'-i'-' z' 1 an/A •ic\ your problem 5 • . ' • } : : J .oj\L. /O) e. speak to your friends and relatives about your Bad experience 5 5 • ] 2 '. 2 . 2 0 ( 1 3 6 ) ^' ^o"''""^* y>^'' friends and relatives not to snop at tnat stor? i ^ : 3 : ;
" ^ * ' 9- complain to a consumer sgency and ask them to nake tne store ti«e 2 . 1 8 ( 1 . 5 6 ) care of your problem 5 5 4 3 2 . 1 4 7 ^ 0 9 7 1 ''• • ' ' ' ' • * let ter to the local newspaper about your bad ><oerience 5 ' . ' • } ' . .
^ ^ ' i . report to a consumer agency so tnat they cm .ar- ^r^.r cons j ^ o ' s . . . . 6 5 : 3 : 1 1 . 9 6 ( 1 . 4 8 ) J- take some formal action against the store/manufacturer s 5 1 3 2 1
1 . 5 0 ( 1 . 0 3 ) As the last part of this section, I m going to ask you to think about how I k e ' / is t tnat you . . . ' j - n e «ny action (s) I f you were pretty sure about the response you »rt going to get. For example, •'tn, consjue's .c j :d complain only I f they were confident that someone would take care of tneir concer". whi'e tnere are many otners who would coaplain regardless of the response they get. how read each of the foi'owing statements and then 'no*. cate the extent to which you tre l ike ly to take that action.
Jmry
- <e / 5. How likely is it that you would report the incident to the store, if you
were pretty sure that the store would: 2 .26 (1 .50 ) c ^/.(o Jfl\ ' • *Po' ' '9 ' 'e but do nothing 6 J .J'^K'J ' to) (,_ tj|,g j j^g gf ygy^ problem to /our satisfaction 6 5 . 5 6 ( 0 . 7 6 ) c. solve your problem and give you better service in the 'u t j re 6 c 'iTfn Q1N d. be more careful in future and everyone would benefit 6
6. How l ike ly is i t that you would mention the incident to your friends ind T n n / i £.n\ re lat ives i f you were pretty sure that they would: i . y\j( i . oU j 4 . 8 7 ( 1 . 0 9 ) *• 9° ' " ^"y'"9 ** usual 5 - ' - , \ , ' c.e.\ *• ' * ' ' '°' '* ^^ref'i\ when shopping from that grocery st: '? 5 3 . 3 6 ( 1 . 6 6 ) c. stop buying from that grocery store altogether 6 4 U \ ( \ 3 9 1 ''• *^*'' ^"^ solve your problem 6
5 5 5 0
4 4 4 4
3 3 3 3
2 2 2 2
>'.'J - 1 « * ' /
I I 1 I
5 0
5 5
4 4 4 4
3 3 3 3
2 5
) 2
1 1 \ 1
2 .27 (1 .63 )
7. How l ike ly is i t that you would report the incident to i rrnsjner agency, such as the Better Business Bureau, i f you were pretty sure •.•'at tney would:
a. not believe you unti l many others have simi l i r j j n p ' i m t s 6 1 . 7 0 ( 1 . 1 7 ) b. take no action 6 (. an (\ \ \ \ c. make the store take care of your problem 6 ' • . 7 / V . l . i i ^ d. solve your problem and ensure tnat the store is c irefu' in tne
5 . 2 7 ( 0 . 9 3 ) future 5
5 5 0
4 4 4
3 3 3
2 2 2
1 1 ;
Section V In this last section, I would l ike to ask you a few bKkground questions:
1 . At which grocery store do you buy most of your groceries (cheek one)?
F j r r ' s
^ _ _ united Alber tsons
3ive n Gam Ot-er, o'ease n»Be
2. How many times do you v is i t this store in a (1) month?
3. How long have you been shopping at this g rxery store?
t imes
( i n - i ; - t n s )
4. At wnat other store (s) do you snop for g rxery products (more tnjn ;re s o«i ' Furr's
res Oon't f'-: No
253
j n i t e d AlOertSOnS
Save n 3 l ' n Otn»r. p l e i i e n » »
5. How frequently do you shop at these other stores eacn month? Furr's
United Albertsons
Save 'n j l ' n other, o'ease - «••!
S. Oo any of tht grocery stores you v is i t have computer 'automat'ici'ly 'eads ir-n .-^:<- ,•
"es to
I f yes, which ones?
7. Oo you think computer (automatically 'eads price) check-outs in grocery stor.s nis -ide ."^o: - j 'a:.;'''
3. Oo you think the computer check-out procedure on the whole, is more accurate f i n t ' l o t c n a : ; - » : • - : . method?
More accurate -ess icr . r i - . ho about the yft
9. All things considered, how do you feel about the use of computer cneck-outs m grocery stores?
10. How often do you buy grocery products each month (other than for f i i u i n items)?
times approximately
u.
12.
13.
14.
Are you: Male
Female
Please check the category that represents
15 to 20 years 21 to 25 years
26 to 30 years
Are you: Single
What is your occupation (Job t i t l e ) :
your age:
31 to 35 /ears 36 •• •) years
to 45 years
Married Divorced
16 to 50 years 51 to 55 /ears
56 to 60 /ears Over W yrs
Separated w1dowed
15. What is the last year of formal education that you completed? (check one)
High school or less Trade school College
16. With which ethnic or racial group do you identify yourself? White
Black Hispanic
Other, please specify
Graduate scnool
17. Please check tht category that represents your total household income ( jo in t if marr*ed) m 1984?
lest than {10,000 SIO.OOO to SZO.OOO
"~ $20,001 to J30.000
S30,001 to t50,000 S50.001 to t70,000
~ $70,001 to $90,000
$90,001 to $110,300 Over $110,001
Thank you fpr your cooperation
254
Cowscer Survey
This questionnaire Is designed to determine tne specif': tomeli'nts tn«t toi^j^rs'. \jcn is -ou n««e to-'*-- -g auto-reoair. There tre no r'gnt or wrong answers, •'owever your pe'sona' o o " :ns t't m o o r ' . f . ' t m --i: the instructions carefully and answer all questions.
Section I first I'm going to ask you some general questions about your opinions regarding Ous"esses •• •"• ."'tei ^'ites. I would like you to read each statement as it appears, ''hen indicate tne eitent ;' OJ-- i-jr.-«•"•. ;- :isiqr.«aent by elrelint the nueper that best describes your reaction to the statement, 'ne tite^o'es i'?
(6) Strongly Agree (5) Agree
' * ) Agree Somewnat (3) Disagree Somewnat
Mean (std) 4.54(1.34)
2.68(1.31)
4.32(1.12) 4.49(1.18)
2.26(1.07)
4.63(1.30) 5.46(0.80)
3.28(1.44) 4.88(1.00)
3.96(1.44)
3.16(1.26)
4.41(1.17) 5.07(1.02)
1.72(1.12)
4.33(1.19)
4.58(1.25)
2.74(1.42)
3.69(1.54)
3.74(1.44)
2.87(1.33)
3.64(1.35)
4.47(1.09)
2.08(1.40) 4.35(1.41)
2.27(0.79)
•2' O'Sl^r^.
ill :isi;-'«
1. Comolaining to business is usually done o/ people with iitt'e e'se to do 5
2. Most companies care nothing at all about tne consumer 5
3. Industry is not cleaning up the waste tney nave oeen dumping 5
4. It bothers me quite a bit if I do not comoliin about an jnsatisfactor/ orodjct. 5
5. The business community has oeen a large influence in raising our country s standard of living i
6. Cham stores are getting so oig that they don't treat tne consjmer oe'so"!''/.. 5
7. Peoole »ive a responsibility to tell stores ineri a product tney ourcnase is defective '
8. Shopping is usually an unpleasant experience 6
6 9. Stores advertize "special deals" just to get the shopper into the store to
buy something else
10. It sometimes feels good to get my dissatisfaction and frustration «itn tne product off my chest Oy comolaining
11. Consumers ire unable to determine what products will Se sold m tne stores 6
12. All business really wants to do is to make the most money it :in 5
13. People ire bound to end up with unsatisfactory products once n a while. so tney should not complain *
6
17. The small Businessman has to do what big Business says :<• else! 6
18. It is hard to understand why some Brands ire twice as expenswe as otners 6
19. I often complain when I'm dissatisfied with Business or products Because t 'eel it is my duty to do SO
20. '*ie consider is usually the least important consioerition to most tomom'es.
21. As soon as they maxe a sale, most businesses forget about tne ouye'
22. By making complaints about unsatisfactory products, m tie long run tne quality of products will improve ... 6
23. .One must be willing to tolerate poor service from most stores 6
24. Prices of products tr* going up faster than the incomes of oramary consuners.. 6
25. Most stores tre willing to adjust reasonable complaints 6
14. Misrepresentation of product features is Just something <e -i/e to I've wuh... S 5
15. Companies encourage the consumer to Buy more than ne/s-^ '?il!/ needs 6 5
16. Complaining isn't much 'un, out t's got to se done to <?eo Businesses '^om Becoming irresponsible 5
5
5
5
5
5
5
5
5
5
• : - • ) ' /
no'fe
255
Mean (std) 2.66(0.92)^ -
'26. -ompan'es jeneriM/ o"»r «nit the consjuer .^-^s i 3 . 8 3 ( 1 . 4 0 ) 2 7 . Business profits »re nigh yet tne/ (eep on -usi-g -ne- jr-.s i 4.52(1.03),. , , ,
28. I reel a sense of accomplishment when [ manage to get i cono i nt -o t r o " taken care of ' ^
2.28(1.1I)„ ,
29. m general companies tre plain dishonest in tneir oeilmgs .ith tne :o^s--»' .. 4
. 5 / ( 1 . 0 9 ) 3 0 . An attractive package sometimes influences a purchase tnat M n ' oer.ssi-/ 5 4.82(0.97),, .
31. By complaining about defective prpducts. I may or.yoot o'ler --"s -«-s -r-n experiencing the same problem 5
2.82(1.18),, , 3Z. I am often dissatisfied «itn a recent purchase S
. £ . J \ , l . ] m } j 2 . Companies "jajz up" a product with no real mprovement. jjst -o ge' i -jr.r price or sell more i
t* .OJ(. 1 . U y ) 34. I don't like peoole who complain to stores. Because jsually tne <• como'i'its ire unreasonaole 6
2.25(0.96) 35. A large variety of products a i : : «s consumers to choose tne one tne/ ' e i '/
'-±1
want.
4 . 4 0 ( 1 . 2 1 ) 3 6 . Companies try to influence government just to Better themsel/es 5 5 4
2 . 8 4 ( 0 . 9 3 ) ^ ' ' *"*'"•" firms stand Behind their products and guarantees 5 5 4
Section II In this section, I would like to ask you about your experience in handling problems and complaints tf-e'-n^q automobile repair.
1. nave you, in the last six (6), months contacted any repair shop/manufactjrer -egardTng jn, jr^j>,_ ;,;, ,, poor quality work, bad product, etc.?
Yes No
If yes, how often? aoout •. or 2 times
iDout 3 to 6 t"»es about ' to 12 ties
more than 12 t'mes 2. How often do you talk to your friends and relatives regarding oroo'ems and complai-^ts .ou nue :once'"'"g
automobile repair (circle appropriate number)?
3.33(1.28) Often s« »r
3. Have you in the last six (6) months contacted a consume' ;'• ojolic agency, sjcn as tne Better Business B^'eiu regarding any complaint?
'es 10
If yes, roughly how many times in tne last six months? times iporo«'iiate'/
4. Have /ou ever taken legal action against a repair shop or nanuf acturer regardi'g iny of your lompamts'
'es
If yes, roughly now many times? 10
;imes iporo«imateiy
Section U I Next, I would like for you to think of a problea that you remmbtr most clearly concerning your experience with autoaoOilt repair. For exampit, you may have been unhappy with the quality of work done on your car. tne behavior of shop aaployeti or with the repair shop policies. Keeping in mind your particular problem, please answer the following questions.
1. Please describe briefly your problem or dissatisfaction:
256
M e « n ( s t d ) 2. i<me and location of tne rett^r snop - . o ' . e d :
3. '.' a product was mvol/ed, wnat was t 'roduct Brand '^ute of tne item ti' appliciP'e
At this point I would'like to ask you a few questions aeout tne oroo -m a";! wnom og '•• s -.sj-'s 3'- '-- •• .- ! ! ? _ - ' ! ! • I^" '? '••*'* •*•* '^ "•• 'bllowing statements and then noicate tne extent :' /ou' i:ree~»-t ;r jisl *9'"ee*ent by circling the appropriate ni*oer.
2.28(0.74), rs . - "" " '^" — -». ihe repair shop's policies ire to blime for the above 3roo>n j
^ . 3 3 ( 0 . 8 1 ) 5. The repair shop personnel are simply careless !
1.88(0.78) 6. The repair shoo didn't mean for tne problem to occjr, out mstj.es "apoen 1 .'
. U J V U . 5 1 ; 7 I expected too much of the repair shop 3 ; ;
2 . 0 6 ( 0 . 4 9 ) 8. It is proBaOly my *ault, since I Should be more careful 'n -,J ---q •— ' . p j -shop 3 , .
2 . 0 1 ( 0 . 8 8 ) 9. The manufacturer of the product is responsio!* for tne sad oroduct ( f t o':3jt IS involved j j
10. Overall how dissatisfied were you Before you did anything aoout tne I'-z'o'^ s.:n as ;o ;i:» to t-e 'epa -Shop etc.). Please circle the numoer that Best represents /our 'eeiiig.
lOOS 90S 30S 70S 60S SOS 40S 30S 20S i:x )X Completely Oissatisfied lot : ssitisfi«3 it 4' '
11. Which of the following action (s) did you take after you eoer?nced f e iBo<e orooiem' i c * -i- -ne answer IS OK).
After I experienced tne above problem I:
Forgot about the incident and Oid notni-ig Complained to the repair shop 'imediately or on iy -if. t-o
Went Back or called the repair snop immediately ana isteo tnem to tj«« cir? ;f :>•( oroo Decided never to use that repair snop
Told -ny friends and relatives about my Bad experience Complained to a consumer agency, sucn is tne Better Sjsmess B--?ij
Took some legal action against the '•eoair shop •na'iji'i:t.,-»r Other, please specify
12. Now please tell us how you felt aoouC tne whole incident after you nad tuen ;"e iso-e ic: o" s). 'lease circle the number that best represents your feeling.
lOOS 90S aOS 70S SOS SOS 40S 30S 20S \'X :t Completely Satisfied 'lot Sifsr'^d at Ai:
13. What do you think auto repair snops ought to do to solve ~'> tyoe of orooiem you nad?
Section IV Next, Imagine that sometime in the future you had gone to your most frequently visited repair shop and tn incident similar to the one you Indicated above occured again. Now read each of the 'ollowing statements and -ndi. cate the extent to which you think they are likely to happen. The categories are:
(6) Very Likely (4) Somewhat Likely (2)'Jnliket/ (S) Likely (3) Somewhat Unlikely (1) Very jniuely
1. Assjue that you reported the incident to tne repair shoo, now 'sely is it that tne repair shop would:
Very .fy '-'<e'/ .m >eiy
4.08(1.92) a. apologitt but do nothing 6 5 * 3 4.36(1.72) b. take appropriate action to take care of your problem ('e'j-d. etc.)... 6 5 4 3 1 1L(\ (sl\ ' '"'*• ><""• problem and give better service to you in tne '..tjre 6 5 4 3 i.l'^Kl.ol) J oe more careful in future and everyone would benefit s 5 4 3 3.48(1.69)
Mean(std)
257
2. Assume tnat /ou -nent'oned tne orooiem to /our '-• tne same --epair snop. now Uxely is it that tie/ 3s ana 'e
go on using the i-epair shop as jsual ^ Be more careful »ntn jsing that ^eoir shoo i stop using that repair shop altogetner ^ help you solve your problem '!!!!!!!!!!!!!!!!...! 5 that you recforted the incident to a consumer agency, sucn is t-» Business Bureau, how likely is is tnat tney wouM:
not Believe you until many others nave similir :o<"o'i"'ts 5 take no action 5 make the reoair shoo take tare of your problem 5 solve your problem and ensure that the 'epair shop is care'-' in future..., 5
Now I would like to ask you about what you might actually do in case the iBo you had gone to your most frequently visited repair shop.
4. If a similar problem occurred again, how lilcely is it that you would:
ve inc- :ei'. o::.'*<3 i-ji'
.,ar y
'. < e < '±J.
t 4 4
4 4 4
i 4 4 4
3 ) 3 3 3
3 3 3 1
2.55(1, 5.44(0, 5.52(0, 5.38(0,
64) 98) 78) 84)
a. forget the incident and do nothing 5 • b. definitely complain to the store manager on your next trip 5 : c. decide not to use that repair shop again 5 : d. 30 oacK or call tne repair shop immediately and ask tnem to ti<e
care of your problem S : e. speak to your friends and relatives aoout your bad experience 5 ! f. convince your friends and relatives not to use that repair snop 6 : g. complain to a consumer agency ind ask them to naxe the repair snop
taxe care of your problem S : n. write a letter to the local newspaper about your oad experience 5 i. report to a consumer agency so tnat they can warn otner considers 6 : J. take some legal action against the repair shop/manufacturer 6
As the last part of this section, I am going to ask you to think about how likely is 't tnat y:j aou d take any actlon(s) if you were pretty sure about the response you tre going to get. ^or example, many co^sjuers •ou'd complain only if they were confident that someone would take care of their concern, w m l e tn«r. are iiany others who would complain regardless of the response they get. Now read each of the following statements and tnen indicate the extent to which you tre likely to take that action.
Very _. ,ery
-'»ei/ 'J"''»?'/ 5. How likely is it that you would report the incident to the repair shop,
if you were pretty Sure that the repair shop would:
a. apologize but do nothing 5 0. taxe care of your proolem to your satisfaction 6 c. solve your proolem and give you Better service • -e future 6 d. Be more careful in future and everyone would be'-^-'t 6 '.
3.26(1.73) 4.85(1.18) 4.32(1.63) 4.40(1.64)
2.22(1.55) 1.83(1.28) 5.22(0.85) 5.33(0.86)
4 1 4 4
3 3 3 3
-•ow litely IS it that you would mention the incident to relatives if you were pretty sure that they would:
Our friends and
1. go on usiig that repair shop as usual 6 0. Be more careful onen using that repair shop 5 c. stop jsmg that repair shop altogetner 5 d. nelp you solve your prpblem 6
7. How likely is it that you would report the incident to a consumer agency, as tne Better Business Bureau, if you were pretty sure that they would:
Such
not Believe you until many others have similar compliints 6 take no action S make tne repair shop take care of your oroblem 6 solve the problem and ensure that the repair shop is careful m tne future *
5 5 5 5
4 4 4 4
3 3 3 3
5
J
> J
1
5 3 0
4 4 4
3 3 3
5
2 2
3 2
Section V
In this last section, I would like to ask you a few oackground questions:
1. Which repair shop do /ou use iwst frequently (please 3i<e "am* ind :3cit ;<•
2. How many timet did you use this repair snop in tne last six o<ie /ear?
times ipproxmats /
3. How long have you been using this repair shop?
less than 6 months ' ess than 2 /ears But no'e tnan 5 nontns
4. What other repair shops do you use for auto repair?
5. How frequently do you use these other repair snops. say m tne last one /ear'
258
6. Are you:
7.
3.
Please check
Are you:
the
15
Male i-emaie
category that repi
to 21
20 years to 25 years 26 to 30 years
Single
31 to 35 years 36 to 40 years
— 41 to 45 years
Warned Divorced
9. What is your occupation (job title):
46 to 50 years 51 to 55 years
56 to 60 /ears Over 60 yrs.
Separated widowed
10. What is tne last year of formal education that you comp eted? (check one)
High school or less I'rade school College Graduate scnool
11. With wnlch ethnic or racial group do you identify yourself?
White Black
Hispanic ~ Other, please specify
12. PI east check the CJtegory that represents your total nousenold mcome Ho 'n t • ' married) in 1984?
less than $10,000 $10,000 to $20,000
— $20,001 to $30,300
$30,001 to $50,000 $50,301 to $70,300
— $70,001 to $90,000
$90,001 to $110,000 Over $110,001
Thank you for your cooperation
259
Medical Care Survey
This questionnaire Is designed to determine the spec"': complaints tnat peoo'e -)>• medical tare 'rom a physician and/or hospital. Tnere are no -'gnt or wrong i-s«»-s nions are important. Please read the instructions carefully and answer i ' g^estions.
'O.r oersonj ;pi.
Section 1 First I'm going to ask you some general questions about your opinions regara'-g o^rst i' United States. I would like you to read each statement as if appears, ''len -rtite tn< or disagreeaent by circling the n\aib*r that Best describes your react'on to the state**-'
4 . 6 5 ( 1 . 3 0 ) I. Complaining to hospitals/physicians is usually done by people "itn little •ise
to do 5
2 . 6 6 ( 1 . 4 1 ) 2. Most hospitals care nothing at all aoout the patient 5
• ^^ ^ 3. Most physicians care nothing at all about the patient 5
3 . 7 3 ( 1 . 5 5 ) 4. It Bothers me quite a bit if I do not complain about poor medical t-eit-ent 5
2 22(1 25) ^ ' ' 5. The medical industry has Been a large influence in raising our count'/ s
standard of nealth 5 t*. ID(1. £.0) g Hospitals ire getting so big that they don't treat the patient oersonal'/ 5 5 . 5 2 ( 0 . 7 8 ) 7. Peoole have a responsibility to tell hospital/pny$ic1ans wnen the treatment
or service they receive is poor 6
3 . 7 3 ( 1 . 4 9 ) 8. Getting satisfactory medical care is a real problem 6
3 . 2 9 ( 1 . 4 8 ; q [J sometimes feels good to get my oissatisfactton and frustration with medical care off my chest oy comolaining 5
3 . 5 6 ( 1 . 5 9 ) Q_ Patients do not have any influence on the medicines that are prescrioed to them *
3 . 7 0 ( 1 . 5 1 ) 11. All hospitals really want to do is to make the most money they can 6
3 . 6 5 ( 1 . 3 7 ) 12. All physicians really want to do is to make the most money tney can 5
4 . 7 6 ( 1 . 4 4 ) jj_ People are bound to end up with unsatisfactory medical tr»it-,?nt once in a while, so tney should not complain S
3 . 9 4 ( 1 . 2 2 ) 14 Physicians prescribe more medicines than the patient rei / -eeds S 3 . 7 2 ( 1 . 3 5 ) 15. Complaining has to be done to xeep hospitals from becomi-g "esoonsible 6
3 . 6 7 ( 1 . 4 2 ) j5_ Complaining has to Be done to keep physicians from oecon-g irrespons'Ole 6
i 4 l 7 5 ( 1 . 2 3 ) 17. It is hard to jnderstand why some physicians are twice as expensive as otners.. i
3 . 3 9 ( 1 . 5 2 ) ,j J fjg„ comolain when I'm dissatisfied with medical care Because ! feel it IS my duty to do so
2 . 9 2 ( 1 . 3 9 ) j5 T g patient is usually the least important consideration to most nospitals 6
2 . 6 2 ( 1 . 2 8 ) 2 0 . The patient is usually the least important consideration to most pnysicuns.... 6
4 . 3 5 ( 1 . 4 9 ) ^ 1 AJ „,„ J, they discharge a patient most hospitals forget aoout tne patent.... 6
4 . 3 7 ( 1 . 3 1 ) 2 2 . By making complaints about unsatisfactory medical care, m the long rjn tne quality of health service will improve *
1 . 7 6 ( 1 .^'48) 23. One must be willing to tolerate poor medical service *
5 . 3 7 ( 1 . 0 7 ) 2*- f^^ces of medical treatment is going jp faster than the incomes of ordinary ^ consiiners
2 . 6 4 ( 1 . 0 9 ) ^ 5 ,|jjj hospitals are willing to accept reasonable complaints S
2 . 9 5 ( 1 . 0 9 ) 2 7 . Hospitals generally offer the services that the patent re«|!, ,,,..5 ,
4 . 2 8 ( 1 , 4 3 ) 2 8 . •hospital profits tre mgn yet they xeep on raismq their prices 5
4 ' 1 6 n ' A 9 1 '' ''''*'''*" "'•''"" *••• '"9»» yet they keep on raising tneir jnces
30. I feel a sense of accomplishment ^ntn I manage to get r-onolimt '0 1
hospital taken care of !.......!..! •
2 j A 0 ( l . 3 8 ) 3 1 . In general hospitals ire plain dishonest m tneir Jeilmgs .un t-e pifen:.... i
32. By complaining about poor medical service. I may prevent T^^r ^.ool* ''om
experiencing the same prpBlon .\_..' ^
. H I . 5 0 ) 33. I j^ gf5g„ dissatisfied with the medical care I receive i
4 . 5 5 ( 1 . 2 2 ) 3 4 . I don't like people who complain to physicians, because JS,.I11/ -neir complaints ire unreasonable ! 5
2 . 8 0 ( 1 . 3 1 ) 3 5 . A large variety of physicians allows people to choose the one tne/ 'ei'ly want
4 . 5 6 ( 1 . 2 1 ) 3 6 . The medical industry trys to influence government for tneir own oenefit 5
; 1
» J
» I
3
: )
i I
4 ]
1 3
Section 11 In this section, I would like to ask you about your experience in handling problems and cowlamts toncermna medical care. '
1. Have you, in the last year contacted any physician/hospital regarding any oroo em. such as mco-o.••'•:• waiting time, charges, etc.?
res 10
If yes, how often? about 1 or 2 times
aoojt 3 to 6 times about 7 to 12 t -nes
Tiore than 12 times
3.38(1.49)
2. How often do you talk to your fnends and relatives regarding proolems and compii -ts you -i»e toncemiq medical care (circle appropriate numOer)?
6 5 4 3 2 1 Often Never
3. -lave you m the last year contacted a third party, such is --e "Medical Association or tne Better Business Bureau regarding an^ medical complaint?
'es No
If yes, roughly how many times in the last yetrl times approximately
4. Have you ever taken legal action against a physician ana or nospital regarding any of /our comp'amts?
'es
If yes, roughly now many times? I0
times approximately
Section III Next, I would like for you to think of a problem that you remaaber most clearly concerning your experience with medical cart. For exaaplt, you may have been unhappy with the competence of the doctor ana/or nospital staff, tht waiting tiat at tht doctor's offlct/hospltal or tht manner in which you were treated. Keeping m mind your particular problaa, please answer the following questions.
I. Please describe briefly your problem or dissatisfaction:
261
Mean(8td) , 2. When did tms proplem occur:
•*""" ""• '«»' 8 •«>"'« iK»re than 6 »onths ago «jr. -,r , ,, ,,. but within the last I year
?*JlIIl!l ??l!* ' •*•'* ""• *• "* '^ • '*• "»•«''»"« *"»«»«« w e xoPla* and whom you feel s 'esoonsBie 'or -.-!?^..! .* *?• " •"••* ••«* »' "•• follo«ln, statements and then indicate the extent of your igreeme-t :r lil •VetMnt by elrelint tht appropriaU ni»Ptr. '^ ' i- - . »
3. Tht phytlcian i$ to blaaw for tht above problem (If applicable) 3
5. rht physician/hospital didn't mean for the prpblem to occur, out mistaxes nappe"... 3 : ^ . 1 7 ( 0 . 5 8 ) g, I expected too much of the physician/hospital 1 ;
2 . 0 9 ( 0 . 5 4 ) 7. It is probably my fault, since I should be leore careful m selecting -ne physician/hospital j ; ;
2 . 2 9 ( 0 . 8 4 ) 8. The hospital is responsible for tht problem (If a hospital is involved) 3 2
9. Overall how dissatisfied were you before you did anything about tne prpoiem. "ease ;•-:*« tne ---loer -..m best represents your feeling.
lOOS gOS SOS 70S 60S SOS 40S 30S 20S lOS 3S Completely Dissatisfied Not Dissafsf'ed it All
10. Which of the following actlon(s) did you take after you experienced tne apove problem? more tnan on* mswer is OK).
After I experienced the above problem I:
Forgot about the incident and did nothing Complained to the hospital and/or physician inwiediately
Went back or called the hospital and/or physician aoout the problem Otcldtd never to go to that hospital and/or physician
Told my friends and relatives about my bad experience Complained to a third party, sucn as the Medical Association or tne tetter Business Bureau
Took some legal action against tne hospital and/or pnysicin Other, please specify
11. Now please tell us now satisfied you were with the whole incident after you had taxe" :nt aao<e action (si. Please circle tht nwotr that otst represents your feeling.
lOOS 90S BOS 70S 60S SOS 401 30S 20S lOS 7S Completely Satisfied tot Satisfied at A M
12. What do you think physicians and/or nospitals ought to do " solve the type pf problem you had?
Section IV Dtit, imagint that ^owtimt in tht futurt you had gont to yotr t t t frtquftly visittd phyticien/hospital and an incidtnt similar to tht ont you indleattd aPovt occurrtd again. No* rted tacn of tht fpllaving stateatnts and indlcatt tht eittnt ta which you thinfc thty are Hktlr to happtt. Tht eattgpritt art:
1. AssMt that you reported tht incidtnt to tht hospital and/or physician, how llktiy is it that tnt physician and/or hospital would:
3.06(1.99) a. apologllt but do nothing 6 5 4 3 2 1 3.45(1.88) b. takt appropriate action to takt cart of your prooleai 6 5 4 3 2 1 •> 1 1 / 1 a n \ c. solvt your proPlea and givt bttttr strvlct to you in tnt futurt 6 5 4 3 2 1 J.ll^l.oUj d. ot mort cartful in future and everyone would btntfit 6 S 4 3 2 1 3.07(1.76)
262
Hean(std)
3.07(1.72) 4.77(1.32) 2.75(1.74) 2.79(1.63)
2.61(1.56) 2.13(1.45) 2:47(1.53) 2.14(1.40)
- ' / Assume that you mentioned the problem to your friends and r.ij'.w.s ">-the same nospital and/or pnysidan. how likely is it that they wou'd:
a. continue to go to the same hospital md/or physican 5 0. be more careful when they go to that physician and/pr lospiti' 5 c. stop going to that hospital and/or physician altogether 6 d. help you solve your problem ^
AssiMt that you reported m e incidtnt to a third party, such as tne Med'cai Association or the Setter Business Bureau, how Hkely is is that they •;. ::
i. not believe you until many others have similar coiwliints 5 b. take no action 5 c. make the physician/hospital take care of your proolem 6 d. solve your problem and ensure that the pnysidan and/or nospital
is careful in future 4
Now I would like to ask you about what you might actually do in case the above incident occ' you had gont to your most frequently visited hospital and/or pnysidan.
4. If a similar problem occurred again, how likely is It that you would:
2.69(1 5.32(1 5.39(1 5:44(1
a. forget the incident and do nothing 5 5 « 3 b. definitely complain to the hospital and/or physician on your next
trip 6 decide not to go to that hospital and/or physician again 6 go back or call the nospital and/or pnysician i:mmediately and as< them to take care of your prpblem 6 speak to your friends and relatives about your pad experience 6 convince your friends and relatives not to go to that nospital and/or physician.. 5
g. comolain to a consumer agency and ask them to make the hospital and/or physician take care of your problem 6
h. write a letter to the local newspaper about your bad experience 6 i. report to a third party so that they can warn other consumers 6 j. take some legal action against the hospital and/or physician 6 5 4 3 2 1
.69)At tht last part of this section, I am going to ask you to think about how likely Is It that you would take any action(t) if you were pretty sure about the response you art going to gtt. For exawle. aany people would complain only If they wtrt confidtnt that soewont would takt cart of thtir conctrn, wnilt there ire «any others who would complain regardless of the response they gtt. Now rtad each of the following statements and then indicate tht extent to which you are likely to take that action.
Very lery L'xely Jn' <e^t
5. How likely is it that you would report the incident to the '-spital and/or physician if you were pretty Sure that tne hospital i-d/or physician would:
c. d.
e. f.
.69)
.09)
.12)
.07)
3.73(1.80) 5.08(1.18) 4.06(1.80) 4.37(1.60)
2.60(1.68) 1.86(1.35) 4.62(1.55) 5.00(1.36)
a. apologize but do nothing 6 b. take care of your proolem to your satisfaction 6 c. solve your problem and give you better service •' tne future 6 d. be more careful in future and everyone would be^^'it 6
How likely Is it that you would mention the incident to /our friends and relatives if you were pretty sure that they would:
a. continue to go to that hospital and/or physician 6 b. be more careful when they go to tnat hospital and/or physician 6 c. stop going to that hospital and/or physician altogether 6 d. help you solve your problem 6
HOW likely if it that you would report the incident to a third oarty, sucn as the Medical Association or tht Better Business Bureau, if you were pretty sure that thty would:
a. not Dtlitvt you until many othtrs nave similar complaints < 6 b. takt no action 6 c. makt tht hospital and/or physician taxe care of your problem 6 d. solve the problem and ensure that the nospital and/or pnysidan
is cartful in tht future 8
5 5 5 5
4 4 4 4
3 3 3 3
t
1
2 2
I 1 I I
5 5 5 5
4 4 4 4
3 3 3 3
2 > 2 I
i
1 1 i
5 5 5
4 4 4
3 3 3
2 2 2
1 I i
263
Stctlon V
In this last stctlon, I would likt to ask you a few background questions:
. I. How frequently have you visited any physician for medical care m the last /eir'
less than once more than once but less than 5 times
fi re •"!
'•iii t-a
•no'e t"l" .J t
2. How frequently have you visited any hospital for iiedlcal care m the last /ear?
less than once more than once but less than 5 times
lore ••'1" 0 • --i 0
'ess tnan ;: t ••a
more than 13 fm«s
3. What do you think physician and/or hospitals can do to improve service to patients?
4.
S.
«.
7.
Are you: Male
Female
Please check the category that represents
15 to 20 years 21 to 25 years
26 to 30 years
Are you: Single
What is your occupation (job title)?
your age:
31 to 35 years 36 to 40 years
41 to 45 years
Married Divorced
46 to 53 years 51 to 55 years
36 to CO years Over 60 yrs
Separated Widowed
8. What Is the last year of formal education that you cafflo:'?tj:? (check one)
High school or less Trade school College Graduate school
9. With which ethnic or racial group do you identify yourself? -
White Black
Hispanic Other, please specify
10. Please check the category that represents your total household income (joint if married) in 1984.
less than $10,000 $10,000 to $20,000
~~ $20,001 to $30,000
$30,301 to $50,000 $50,001 to S70,000
"~ $70,001 to $90,000
$90,001 to $110,000 Over $110,301
Thank you for your cooperation
264
financial Services Sjr»«/
This questionnaire Is designed to determine t»e specific to-oiai^ts i q or-oems tnat o-^o a variety of financial institutions; Including banks, savings and 'oars, ced't jn^ois, et people tre not happy with the service charges banks put on cnecxmg accounts, tne oa'i-c-manner In which loan applications are handled by many Banxs. 'nere tre no rijn; or .r-,-g personal opinions tre important. Please read the instructons care'uily and answer a ;.
Section I First I'm going to ask you some general questions about your opinions reqaro'ig ajnis (.n savings and loans, and credit unions) in the United States. I would l'«« /ou to reaj eic appears. Then Indicate the extent of your agreement or disagreement oy Q-':'."-q tn« ^jy^t your reaction to the statement. The categories tre:
(6) Strongly Agree (5) Agree
«) Agree Somewhat (3) Disagree Somewhat
Mean(std)
4.96(1.07) 2.93(1.26)
3.77(1.40) 3.06(1.37)
4.40(1.34)
5.45(0.80)
3.25(1.33) 3.37(1.40)
3.24(1.63)
4.40(1.31) 4.54(1.22)
2.91(1.28)
3.64(1.44) 4.88(1.21)
3.60(1.39)
2.82(1.32)
3.20(1.38)
4.01(1.32)
2.11(l.'>4e)
4.38(1.26)
2,61(1.06) 2.71(1.01)
4.25(1.22)
2) 5'sagr (I) Disagr
iqr.e
1. Complaining to Banks is usually done By people with little else to do 5
2. Most banks really care about their customers 5
3. It bothers me quite a bit if I do not complain about poor service from a oam.. 6
4. The banking Industry has Been a large influence in improving our country's financial position S
5. Banks are getting so Big that they don't treat the customer perspnally 6
6. People have a responsoility to tell banks when a service they recewe IS not satisfactory 6
7. Getting satisfactory service from Banks is a real problem 6
8. It sometimes feels good to get my dissatisfaction and frustration in dealing with banks off my chest by complaining 6
9. People do not have any influence on the manner m «hlch the Banks conduct
their Business 6
0. All Banks really want to do is to make the most money they can 6
1. People are bound to end up witn unsatisfactory service once >n a wmle.
so they Should not complain 6
2. Banks encourage people to spend more than they really shoJd 6
3. Complaining nas to be done to keep Banks from Becoming "-esoonsiBle 6
4. It is hard to understand why financial services (such as :-ecxing accounts) at some banks are twice as expensive as others 6
5. I often complain when I'm dissatisfied with bank services Because I 'eel it is my duty to do so 5
6. The customer is usually the least important consideration to most banks 6
7. As soon as they open your account, most banks forget about the customer S
8. By making complaints apout unsatisfactory service, m tne long run tne quality of banks will improve 5
9. One must be willing to tolerate poor service from most oanxs 6
0. Costs of financial services are going up faster than the -"omes of many people *
1. Most banks tre willing to adjust reasonable complaints 6
2. Banks generally offer what people want 6
3. Banks profits are high, yet they keep on raising the cost of their services.... 6
• " t . « " : t < -q • ' t n • : - •ttitot. - i ' ,
3' s t a t e " « " t s tn<] ••<# "Sw«rs. - - . » , e r , : . , r S t i p - S .
s 1 1 : , ? • • -• - I t : e $
4
4
*
*
4
4
4
4
4
O l - i S .
IS t .",-.' oes
D ' i i r ' t
2
2
2
2
2
2
2
2
1
I
•
1
1
I
I
265
»qree '•'•"•'i /.
M8an(3td)
4 : 4 3 ( 1 . 1 6 ) 2 4 . I feel a sense of accompl ishme'-t when ! nanage to je' i compli'^' -o a r>, taken care of ,
2.01(1.21) ^ , ' 25. In general banxs tre plain dishonest in tneir dealings .itn -re jeop e 5
4 . 0 9 ( 1 . 3 2 ) 2 6 . By complaining about unsatisfactory service. I may pre«»nr o'-*' o-oo'e --om
experiencing the same problem ] , , ' ^
2 . 5 7 ( 1 . 3 0 ) 2 7 . I am often dissatisfied with the service I receive n most oanns ^
4 : 6 3 ( 1 . 2 3 ) 2 8 . I don't like people who complain to oanxs. Because usually t-e- :omo i "ts
are unreasonable ,;
2 . 2 8 ( 1 . 1 2 ) 2 9 . A large number of Banks allows people to Choose the one tney -ea'l/ .i-t 5
4 . 0 7 ( 1 . 2 5 ) " 8*"'" ''•y to influence government Just to Better tnemseUes s 2 . 0 7 ( 1 . 2 2 ) 31. Banks firmly stand Behind their deposits m d guarantees 5
Section H In this section, I would like to ask you about your experience in handling problems and complaiits m ;oti! various financial services from a bank (such as cheeking accounts, ioans. etc.).
1. Have you, in the last six (6), months contacted any Bank regarding m y oroO'e^. sjcn as errors m you-account, poor service, etc.?
->
1
1
J
1
3
3
-l_i
•r,
If yes, how often? No
about 1 or 2 times iDout 3 to 6 •. mes
iBout 7 to 12 f-nes more tnan 12 t "nes
2. How often do you talk to your friends m d relatives regarding oroolems m d c»iplai"ts /OJ ^i.» :o'':e''"'-g financial service (circle appropriate numoer)?
6 5 4 3 2 1 2.39(1.29) 21-SH v.er
3. Have you in the last six (6) months contacted a consumer or puolic agency, sucn as tne Bctt*- 3jsiness Bureau regarding any complaint?
Tes No
If yes, roughly how many times in the last six months? times approximately
4. Have you ever taken legal action against i bank regarding i-y of your comoiaints?
'es
If yes, roughly how many times? •<o
times approximately
Section ::i Next, I would like for you to think of a problem that you remember most clearly concerning your experience with banks. For exa«ple. you may have oeen unhappy with the way your account was handled or balanced, the jnfair treatment of your loan application, the behavior of bank employees or the various service charges put on your account. Keeping in mind your particular problem, please answer the following questions.
1. Please describe Briefly your problem or dissatisfaction:
2. When did this problem occur:
within the last 6 months more than 6 montns But less than 1 year
more tnan a yetr ago
3. N M O and location of the bank involved:
266
M«an (std)
2.24(0.80)
1.88(0.75) 1.67(0.79)
2.20(0.58) 1.76(0.55)
At this point I would like to ask you a few questions about the proB'tm irj ^^ ^^ - „ , -taon^ •> . I would like you to read each of the following statements and tn-- ind-ite :ne ..-ent of ,->.r ,a ,-• agreement By circling the appropriate number. ^
^:'^f : i;rje
4. The bank policies'are to Blame for tne aoove proolem 3
5. The bank personnel tre simply careless ]
6. The bank didn't mean for the problem to occur, out mistakes happen ;
7. I expected too much of the Bank j
8. It is prooably my fault, since I should Be more careful m select ig -e oi''« 1
9. Overall how dissatisfied were you Before you did anything loct tie or-o^-i - eis,? ; -: •-•? -.-:•>-best represents your feeling.
lOOS 90S 30S 70S 60S SOS 40S 30S ITt ITt OS Completely Dissatisfied Not Tssat'sf'^J it J''
10. Which of the following action (s) did you take after you experienced -• ipove o-oo -'-'' n-:-» f i " is Ok).
After I experienced the above problem I:
Forgot about the incident and did nothing Comolained to tne Bank immediately
Went Back or called tne Bank aoout the proolem Decided never to use that Bank
Told •ny friends and reJati/es about my oad e'oerie^ce Complained to a consjner agency, such as tne 3ett»r 3jsi^ess 3.'-?i-
Took some legal action against tne amx Other, please specify
11. Now please tell us how satisfied you were about the wnole incident atte-- /ou •iio tnen --e )Oo<e act Please circle tne numoer that Best represents your feeling.
lOOS 90S 30S 70S SOS SOS 40S 30S 20S U t » Completely Satisfied 'Jot Sat 'sf'ed at 4' i
12. What do you think banks should do to solve the type of problem you nad?
•o' t. -"• 3 1 s -
on '$).
Section !V Next, imagine that sometime in the future you had gone to yo.' ••ost f ^ ' ^ n y H / f!*^^*^ **"* *"* *" '"eident similar to the one you indicated above occurred again. Now -jad each of the following statements and 'noic the extent to which you think they are likely to happen. The categories are:
cate
(6) Ury Likely (5) Likely
(4) Somewhat .lely (3) Somewhat .nlikaly
( 2 ) J n l ' i i e l y f l ) very Unluely
3,70(2.10) 4.30(1.88) 3.85(1.90) 3.46(1.86)
1.90(1.22) 4.23(1.59) 1.77(1.22) 2.92(1.82)
Very
Assume tnat you reported the incident to the Bank, now lUely is it that the otnK would:
a. apologlte but do nothing S b. take appropriate action to take care of your proolem 6 c. solve your proolem and give Better service to you in tne futjr* 6 d. oe more careful in future and everyone would Benefit 6
Very
Assune that you mentioned the problem to your friends and relatives «no jse the same bank, how likely is u that they would:
a. go on using that bank as usual 6 b. be more careful when using tnat bank 6 c. stop using that bank altogether 6 d. help you solve your proolem 6
3. Assume that you reported the incident to a consuner agency, such as tne Setter Business Bureau, how likely is it that thty would:
4. not btlitve you until many pthtrs havt similar complaints b. take no action .-c. make the bank takt care of your problem d. solvt your problem and ensure that the bank is careful m future.
tmry . ' i ' y
How I would likt to ask you about what you might actually dp in case the above incident occ-'-e; yot had gont to your most frpqutntly visittd bank.
ierf
4. If a similar problem occurred again, how likely is it that you would:
4. forget tht incident and do nothing g b. definitely complain to the manager $ c. dtcidt not to ust that bank again g d. go back or call tht bank and ask them to take care of your proolem... 5 e. speak to your friends and relatives about your bad experience 6 f. convince your friends and relatives not to use that bank 6 g. complain to a consuner agency and ask them to make the bank taxe
care of your proolem $ h. write a letter to the local newspaper aoout your bad exoenence 6 I. report to a consumer agency so that they can warn other cons.4Mrs 6 j. take some legal action against the oank 6
'Jit
igam wnile
As tht last part of this stctlon. I am going to ask you to think about how likely is it that you would take any action(s) if you were pretty sure about the response you are going to gtt. For example, many people would complain only If thty wtrt confidtnt that someone would take cart of thtir conctrn, while there are many others who would complain regardless of the response they get. Now read each of the following statamtnts and tnen Indicate tht extent to which you are likely to take that action.
Very Very Lixely 'Jniuelf
5. How likely is It that you would report the incident to the bank, if you were pretty sure that the bank would:
a. apologize but do nothing 6 b. take care of your problem to your satisfaction 6 c. solve your oroblem and give you better service in the future 6 d. be iwre careful in the future and everyone would benefit 6
6. How likely is it that you would mention the incident to your fritnds i'-relatlves If you were pretty sure that thty would:
a. go on using that bank as usual 6 b. bt more careful when using that bank 5 c. stop using that Bank altogether i d. help you solve your problem 6
7. How likely is It that you would report the incident to a consumer agency, as the Better Business Bureau, if you were pretty sure tnat thty would:
Such
a. not believe you until many others have similar complaints 6 0. take no action 6 c. make the bank take care of your problem 6 d. solve tne problem and ensure that the bank is careful in the future... 6
5 5 5 5
4 4 4 4
3 3 3 3
2 2 2 2
I 1 I I
Stctlon V
In this last stctlon. I would likt to ask you a few background qutstions:
1. Which is your primary bank? (pleast give nme and location)?
268
2. How long nave you Been using this bank?
'•»« '141 5 months less tnan 2 years o.t -nore -:-• :-)o > ,,ars than 6 months ^
3. What other banks do you use for financial services?
4. What do you think Banks can do to improve service to customers?
5. Oo you use automatic teller machines (ATM) for depositing or witndri«ing money'
yes No
6. Oo you think the automatic te l le r machines (that jse Bank cards) ntie -iioe aanic-g fasier^
agree don't know disagree
7. All things considered, what is your opinion about the automatic teller macm-es^
8. Are you:
Male Female
9. Please check the category that represents your age:
1: to 20 years 31 to 35 years H to :•? /ears 21 to 25 years 36 to 40 years 51 to 55 /ears
26 ta 30 years 41 to 45 years S6 to SO .ears Over 60 <'S.
10. Are you: Single Married Separated
Divorced wiiowed
11. What is your occupation? (joo title):
12. What 1$ the last year of formal education that you cono'-^ted? (check one)
Hign school or less ''ride school College graduate sfooi
13. With which ethnic or racial group do you identify yourself?
rfhite Black
Misaanic Otner. please specify
14. Please check the category that represents your total household income (Joint if narr.o) in '.984.
less than $10,000 $30,001 to $50,000 590.001 to $110,000 $10,000 to $20,000 $50,301 to $70,000 Over S'tlO.OOl
$20,001 to $30,000 $70,001 to J90.300
Thank you for your cooperafon
269
APPENDIX E: SELECTED DISCONTENT AND ALIENATION ITEMS
ITEM hfUMBER iTEM DESCRIPTION
DiKOBtoot #1
#2
#3
#4
#6
#«
#7
#8
#0
#10
#11
#12
Industry hM an obligation to cleui ap the wMt« th«y have b««a damping bat they w«i't doing It.
Chain ptorap are gvtting PO big that th«y do not treat the consomer peraonally.
Stores advertM 'special d«ais* jast to get the alMppnr into the store to boy something else.
All business really wants to do is to maks ths most money it can.
Companies encourage the consnmer to b«y more than he/she really nseds.
It is hard to onderstan-: why sooie brands are twice as expensive as others.
As soon as they make a sale, most forget abont the bvyer.
PriceB of prodncts are going up (aster than the incomes of ordinary consumers.
Business profits are high yet they keep on raising their prices.
An attractive package sometimes influences a purchase that isn't necessary
Companies 'jaii up* a product with oo real improvement, just to get a higher prxe or sell more.
Companies try to influence the govemment jnst to better themselves.
2 7 0
ITEM NUMBER ITEM DESCRIPTION
ABenntioo #1 Most companies care nothing at all about the
#2
# 3
#4
#5
#«
#7
#8
# 9
#10
Shopping is usually an nnpleasaat experience.
People are unable to determine what products wiU be sold ia the store.
The small businessman has to do what the big business says or else!
The consumer is usually the least important ronsifiersiion to most companies.
One must be willing to tolerate poor service from most stores.
Companies generally offer what the consumer wants (R).
In geasral companies are plain dishonest in their dealing with the consumer.
I am often dissatisfied with a recent purchase.
Business firms stand behind their products and guarantees (R).