Close
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Immoral by association: Customers' reactions to information about an employee's morality
(USC Thesis Other)
Immoral by association: Customers' reactions to information about an employee's morality
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
IMMORAL BY ASSOCIATION: CUSTOMERS’ REACTIONS TO
INFORMATION ABOUT AN EMPLOYEE’S MORALITY
by
Allison Rachelle Johnson
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BUSINESS ADMINISTRATION)
August 2005
Copyright 2005 Allison Rachelle Johnson
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3196825
Copyright 2005 by
Johnson, Allison Rachelle
All rights reserved.
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy
submitted. Broken or indistinct print, colored or poor quality illustrations and
photographs, print bleed-through, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
®
UMI
UMI Microform 3196825
Copyright 2006 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEDICATION
I dedicate this work to the people who supported me through the process of
obtaining my degree. I dedicate this work to my husband, who is my best friend and
my biggest fan, with all my love. I dedicate this work to my friend Gustavo, whose
support and friendship were invaluable to me. You are gone too soon, my friend,
and I will always miss you. I dedicate this work to my mother, who has always had
a sympathetic ear and a warm hug at the ready. I dedicate this work to my advisor,
for her guidance and commitment. I dedicate this work to the Cohens, who helped
me keep my sanity through the final year. And finally, I dedicate this work to all my
family, whose love and support I’ve never had to do without. I thank you all very
much, and it wouldn’t have been the same without you.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iii
TABLE OF CONTENTS
Dedication ii
List of Tables iv
List of Figures V
Abstract vi
Chapter I: Introduction 1
Chapter II: Study 1 27
Chapter III: Study 2 45
Chapter IY: Study 3 66
Chapter V: Study 4 77
Chapter VI: Conclusion 87
References
105
Appendix A
111
Appendix B
113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IV
LIST OF TABLES
Table 1: Dependent variable means by
Table 2: Dependent variable means by
Table 3: Dependent variable means by
Table 4: Dependent variable means by
condition in Study 1 40
condition in Study 2 56
condition in Study 3 72
condition in Study 4 82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V
LIST OF FIGURES
Figure 1: Conceptual model 23
Figure 2: Company morality judgments and purchase intentions in Study 1 41
Figure 3: Company morality judgments and purchase intentions in Study 2 59
Figure 4: Company morality judgments and purchase intentions in Study 3 73
Figure 5: Company morality judgments and purchase intentions in Study 4 84
Figure 6: Summary of the factors affecting judgments of company morality 91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
My dissertation tests the proposition that judgments of company morality
can be made based on association with an employee’s immoral (or moral) behavior.
If moral judgments are made based on association, rather than on consumers’
deliberation of the relevance of the employee’s action to judgments of the company,
the employee’s morality should be generalized to judgments of the company’s
morality. Four experiments examine associative judgments of company morality
based on information about employee morality, providing evidence of judgments
based on associative processing by contrast with judgments more likely to be formed
based on deliberative processing. The findings indicate that, consistent with an
associative processing interpretation, judgments of company morality can lack
objective justification and can vary based on a consumer’s perception of the strength
of the employee’s association with the company. The action of an employee who is
highly representative of the company, such as the company president, was found to
have a greater effect on company morality judgments than the action of a less
representative employee, such as an assembly line worker. However, an immoral
action by a less representative employee can have a significant, though lesser, effect
on judgments of company morality. Motivating deliberative judgments of company
morality was also found to decrease the effect of an employee’s morality. Such
motivation to process information deliberatively was created based on accountability
for the judgment, as well as by priming the rules by which judgments of morality are
deliberatively formed. In sum, this research contributes to the literature on corporate
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
social responsibility by demonstrating that the mere association of a company with
an employee’s morality can influence judgments of company morality and purchase
intentions.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
CHAPTER I: INTRODUCTION
Judgments involving moral issues pervade the daily lives of consumers. For
example, consumers may consider whether they should purchase shoes that might
have been manufactured in a “sweatshop” overseas, whether they should switch their
telephone service to a company that donates 10% of its profits to charity, whether
they should buy gas from a company that created an environmental disaster with an
oil spill, or whether they should choose groceries that are packaged with recycled or
recyclable material. All these decisions are influenced by judgments of the morality
of the companies’ behavior. Research on corporate social responsibility has found
that consumers’ judgments and decision-making are often affected by the perceived
morality of the company’s behavior in fulfilling societal obligations toward
employees, customers, the community and the environment (e.g., Brown and Dacin
1997). Corporate charity and other moral behavior can lead to more positive
consumer attitudes toward the company and can build brand equity and corporate
image (e.g., Dawar and Pillutla 2000; Hoeffler and Keller 2002). Practices that are
seen as immoral can lead to more negative attitudes toward the company and its
products (e.g., Folkes and Kamins 1999).
Research findings on corporate social responsibility and morality in
marketing seem to suggest that consumers are processing information about
company morality in a logical, deliberative fashion. Flowever, this may not always
be the case. Research in psychology suggests that moral judgments can be made
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
based on associative reasoning, and can be associative, intuitive judgments rather
than deliberated judgments formed through rule-based reasoning (Haidt 2001;
Tetlock et al. 2000). The distinction between associative, automatic, heuristic, or
intuitive processing and deliberative, controlled, logical, or systematic processing
suggests that there are important differences between judgments made by the two
systems (e.g., Sloman 1996; 2002). Associative judgments can be made based on
mere association. For example, one might assume that an employee of a non-profit
organization that is considered very moral would also be a moral person. In
contrast, deliberative judgments would be made based on logical justification. For
example, that employee might have chosen to work for the non-profit because the
job paid well or because it was convenient to where he or she lives rather than based
on any moral inclinations. Flowever, if there was no motivation to examine one’s
judgment of the employee, these alternatives might never be considered. If moral
judgments of companies can be made associatively, such judgments may at times
lack logical justification, may be difficult to change, and are likely to be influenced
by the type and strength of associations in the judgment context.
My dissertation examines consumers’ moral judgments in a context that
allows associative and deliberative judgments of company morality to be
distinguished. When an employee commits moral or immoral behavior that occurs
in his or her private life and that has nothing to do with the company or its products,
there would seem to be no logical justification to hold the company morally
responsible for the employee’s actions. If consumers’ reactions to the company and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
its products are affected by an employee’s morality, this would suggest that
consumers’ moral judgments are formed associatively rather than deliberatively.
As well as having theoretical import, researching moral judgments in this
context has important marketing implications by providing a greater understanding
of the effect that an employee’s morality can have on customers and their reactions
to the company. Recent situations, such as the decline in sales suffered by Martha
Stewart’s company when she was charged with obstruction of justice, suggest that it
is important to understand the potential effects of employees’ actions in their private
lives on consumers’ reactions. Although the publicity surrounding Stewart’s
infraction was rather extreme, it is not unusual for a person’s place of employment
to be included when exceptionally moral or immoral actions are being reported in
the media. Hence, it is important to understand when and why such information
could influence consumers’ reactions to the company.
In sum, my dissertation advances theory on morality in consumer behavior to
predict the effects of moral judgments on consumers’ reactions to information about
employee morality. First, I review prior research on corporate social responsibility,
moral judgments, and associative processing to provide background for these
predictions. I present the results of four experiments, conducted to test the
predictions and to rule out alternative explanations. Finally, I discuss the
implications of these findings for future marketing research and practice.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
MORAL JUDGMENT OF COMPANIES
The literature on corporate social responsibility describes the effect of
information that provides a logical justification for judgments of company morality
(e.g., Brown and Dacin 1997; Handelman and Arnold 1999; Sen and Bhattacharya
2001). However, that research does not necessarily provide evidence that moral
judgments are always made through logical, rule-based processing of the relevance
of moral information. Intuitive, associative judgments of company morality have
implications for the factors that may influence such judgments as well as for the
effects of those judgments. A dual-process theory that distinguishes between
associative and deliberative processes of forming moral judgments would provide a
framework for predicting consumers’ morality judgments as well as their decisions
based on moral information. The literatures on associative processing and on moral
judgments suggest that such a theory might be usefully applied to judgments of
company morality.
Associative Judgments and Intuitive Processing
Many scholars have described two separate processes or systems of forming
judgments and making decisions, where one system is more associative and intuitive
and the other system is more deliberative and rule-based (e.g., Eagly and Chaiken
1993; Kahneman 2003; Sloman 1996, 2002). Sloman (1996, 2002) has described
evidence to support a dual-process theory of two independent systems of reasoning
that encompass many of the distinctions made in other popular dual-process
conceptualizations (e.g., Chaiken, Liberman, and Eagly 1989; Petty and Cacioppo
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
1986). The first is an associative system that depends on associations among
objects, relying on intuition and images, and the second is a rule-based system that is
more deliberative, depending on causal and logical relations. The associative system
draws inferences based on similarity and contiguity, or other types of perceived links
between entities and concepts, which form the set of associations for any given
situation or judgment target. The rule-based system draws inferences based on a
process of applying rules of reasoning, which are abstract representations of the
logical relations between a set of defined variables. The rule-based system of
reasoning results in a judgment that is explained by logical justification, whereas the
associative system results in a judgment that may not have a compelling
justification.
However, this does not imply that associative judgments are inherently non-
cognitive or irrational. As Sloman (1996, 2002) argues, associative judgments are
based on learned patterns of association that allow judgments to be made quickly,
and associative judgments can be made in situations where the deliberative judgment
may be unclear. The associative system has a greater tendency to occur
automatically than the rule-based system, which often requires sufficient motivation
and opportunity to devote cognitive resources to the judgment task. Associative
processing is a valuable strategy when a situation has not been encountered
previously and when there are no defined rules that the individual can apply to the
reasoning situation. Though associative judgments can seem irrational in situations
that are often explicitly designed by researchers to create seemingly illogical
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
judgments, under most circumstances the judgments arrived at through associative
processing are adaptive in the sense that they are relatively accurate and allow the
individual to solve problems and devote cognitive resources to other aspects of the
situation.
Convergence o f deliberative and associative judgments. Unless a situation
results in conflicting judgments based on associative versus deliberative processing,
it is difficult to distinguish the type of processing that has arrived at the judgment.
In most situations, judgments based on associative processing may seem rational and
justified, because they are based on the learned pattern of associations that has
predicted correct judgments in the past. Associative judgments may often converge
with deliberative judgments, because the rules of reasoning point to the same
conclusion as the associations in the context.
For example, judging a company immoral because they used sweatshop labor
could be arrived at through rule-based deliberative processing or through associative
processing. One might search out or process information about negligence or
intention to cause harm, malevolent disregard for the fairness of wages, and
manipulative practices in recruiting labor, to justify the judgment based on the logic
of responsibility for moral behavior. The rule-based process of assessing
blameworthiness involves judging causality, foreseeability, and intention
(Mascarenhas 1995; Mikula 2003). In contrast, though an associative judgment of a
company using sweatshop labor might also conclude that the company is immoral,
the judgment would likely be based simply on the association of the company with
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
the concept of sweatshop labor. Thus, in a situation like this, the judgment reached
by the two systems of processing might be the same.
However, deliberative and associative judgments can also diverge. Rule-
based processing could arrive at a judgment that the company was not morally
responsible for the decision (e.g., the decision was made by a contractor), whereas
the association of the company with the immoral contractor and with sweatshop
labor could lead to a judgment that the company is immoral. Thus, the two systems
could arrive at different judgments of the same situation.
Differences between associative and deliberative judgments. Given
knowledge of the rules of logic that are likely to be applied to a deliberative
judgment and of the associations that might influence a judgment, it is possible to
predict the judgments that are likely to result from the two types of processing. In
addition, the two types of judgments can be differentiated by the conditions under
which they are made. Deliberative processing requires cognitive effort and
motivation. In comparison, associative processing often requires less effort and may
occur automatically, even when there is motivation to process information
deliberatively (Sloman 1996, 2002). When judgments of the same situation conflict
between individuals or between occasions, it may be possible to distinguish the two
types of processes based on whether there was motivation or opportunity to
deliberatively process information.
Associative and deliberative judgments can also be simultaneously present.
When the two systems result in different judgments, the rule-based judgment may
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
suppress or be used to correct the associative judgment, given sufficient awareness,
motivation, and ability to do so (Sloman 1996; Wilson and Brekke 1994; Wilson,
Centerbar, and Brekke 2002). However, the more automatic associative judgment
may frame or anchor the ultimate judgment or decision. For example, upon
encountering a member of a stereotyped group, stereotypical associations may
spring to mind to create an impression of that individual that is very different from
the deliberative impression that one arrives at by analyzing the person’s actual
characteristics and behavior. The deliberative or rule-based impression may be used
to correct and suppress the stereotypical associative impression, but the activation of
the stereotyped impression may result in the deliberative impression being
contrasted away from or anchored to the stereotype (Sloman 2002; Wilson et al.
2002).
Associative judgments are likely to be difficult to change or ignore
completely, even when corrected by rule-based reasoning. Because associative
processing can occur automatically, the process by which the judgment is formed
may not be available to conscious introspection (Bargh 1997; Wilson et al. 2002).
This implies that associative judgments may be difficult to change because their
basis cannot be challenged. Automatic processes are difficult to change once they
are internalized, though they can be affected by repeatedly challenging the results of
an associative processing judgment until the new associative process is internalized
(Bargh and Ferguson 2000). In contrast, because deliberative rules are consciously
manipulated and accessed, changing the process would be more easily
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
accomplished, and rule-based judgments are likely to be more easily changed
because their justification can be directly challenged.
In sum, associative judgments and rule-based deliberative judgments differ in
the extent to which they are justified, the motivation and cognitive effort required to
form the judgment, and the ease of altering and intervening in the process of making
the judgment. These differences have important implications in the context of
judgments of company morality, if such judgments can be made associatively.
Moral Judgment
Morality judgments are one of the most common types of judgments made in
everyday life, and moral considerations are often important in decision-making (De
Bruin and Van Lange 2000; Folkes and Whang 2003; Reed and Aquino 2003;
Tangney et al. 1998). The ability to make moral judgments is part of the process of
development and socialization that is required to be a productive and valuable
member of society. Parents teach their children to judge right from wrong, and
individuals learn what is acceptable behavior in their own judgment and in the view
of important others. The process of internalizing moral standards and applying them
to one’s own and others’ actions in everyday life results in moral judgments that are
often automatic or intuitive once an individual has reached maturity (Haidt 2001).
This does not imply that moral judgments cannot be rationally considered or that
moral standards cannot be changed, but it does imply that moral judgments can and
often do take place with very little effort and without conscious deliberation of the
reasoning behind them.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0
Moral judgments are based on the fit of a perceived action or situation with
internalized moral standards that define what is acceptable behavior and treatment of
others. For example, the extent to which an action causes harm or injustice is the
most consensual standard of morality across cultures (Haidt, Koller, and Dias 1993;
Tetlock et al. 2000). Actions that are seen as harmful or unjustified are judged to be
immoral, and conversely, actions that prevent harm or benefit the deserving are seen
as moral. There are other standards of what individuals judge to be moral or
immoral behavior, such as moral prescriptions based on religious beliefs. Moral
judgments depend on the moral standards to which an individual subscribes or has
internalized from his or her culture. The more fully internalized the moral standard,
the more likely it is that moral judgment based on that standard will occur
automatically when a situation is perceived as relevant to it (Haidt 2001).
The process o f moral judgments. It has been proposed that moral judgments
are most often an intuitive process involving an automatic comparison of the
situation at hand to internalized moral standards, or a process involving a gestalt
sense of “right” versus “wrong,” or “good” versus “bad” (Haidt 2001). People’s
reactions to information about moral behavior, and moral infractions in particular,
might often be more intuitive than deliberate. Some argue that justifications for
moral judgments may often not be consciously formed unless the person is called on
to defend their judgment or reasoning (see Haidt 2001 for a review).
Investigations in situations where moral judgments seem to lack logical
justification provide some evidence to support the conceptualization of moral
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
judgments as intuitive rather than deliberate. Tetlock et al. (2000) investigated
reactions to scenarios describing decision-makers considering taboo tradeoffs.
Taboo tradeoffs are situations in which there is a conflict between secular and sacred
values, such that one has to be compromised to the other. For example, a tradeoff of
human life for profit is perceived as taboo. A person who even considers making
this tradeoff is seen as tainted and immoral, even if the taboo tradeoff is only
considered briefly and quickly rejected. Haidt et al. (1993) found similar reactions
to taboo scenarios in which the characters’ actions created an intuitive judgment of
morality that could not be justified by reference to definable moral standards. For
example, a woman who cut up an old decommissioned national flag and used it to
clean her bathroom was seen as immoral. In such situations, where no harm was
caused and no injustice or public disrespect was committed, it seems that there is
little justification for a judgment of immorality. In fact, though people considered
taboo actions or tradeoffs to be extremely immoral, they often could not provide
justifications for that judgment, other than visceral or emotional reactions that such a
thing is “just wrong.” This type of evidence supports the conceptualization of moral
judgments as intuitive and based on “gut reactions” as opposed to being formed
based on deliberate reasoning and logic.
Though situations can be identified in which there seems no justification for
a strongly held moral judgment, more often moral judgments can be justified by a
definable, logical moral standard. However, it seems that moral judgments are not
necessarily made in a deliberative fashion, even when they can be deliberately
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
justified. The evidence that moral judgments can be formed without reference to a
consciously accessible moral standard suggests that the cognitive processing
involved in moral judgments can be intuitive, automatic, and associative rather than
logical, controlled, and deliberate. However, both associative and deliberative
processing can occur in moral judgments.
Moral judgments o f companies. One might expect that moral judgments in a
consumer realm would be more likely than moral judgments in a purely social
situation to be based on logical justification because consumer decisions seem to
rely on relatively objective or tangible evidence - for example, evidence of product
quality - more than on subjective judgments. However, an abundance of evidence
has shown that consumers’ judgments and decisions are often based on heuristics
and associations, and that they do not necessarily involve extensive processing of
available information (e.g., Aaker 2001; Bettman, Luce, and Payne 1998; Hahn,
Park, and Maclnnis 1992; Hutchinson and Alba 1991; Maclnnis and Jaworski 1989).
Therefore, it is reasonable to predict that moral judgments of companies would not
require rational justification or deliberative processing. Rather, I predict that
judgments of company morality can be more intuitive and associative than
deliberate. This distinction is important because there are important differences
between judgments made based on deliberative, rule-based processing and those
made based on associative, intuitive processing.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
Corporate Social Responsibility
Though the distinction between associative and deliberative processing of
moral judgments has not been examined in marketing, the literature on consumers’
reactions to corporate social responsibility is consistent with a deliberative
processing perspective. Research on corporate social responsibility has tended to
imply that consumers make a deliberative decision about the implications of social
responsibility information for judgments of the company. For example, it has been
suggested that corporate social responsibility information may affect evaluations of
the company and its products because it signals the company’s trustworthiness
(Brown and Dacin 1997). Corporate social responsibility information is thought to
be used by consumers to judge a company’s “character” (Sen and Bhattacharya
2001). Though these assertions do not rule out the possibility of associative
processing, they do suggest that consumers are actively seeking out and deliberating
on the implications of social responsibility information.
Most research on corporate social responsibility measures general
evaluations of the company or purchase intentions rather than moral judgments, per
se (e.g., Dawar and Pillutla 2000; Sen and Bhattacharya 2001). However, evidence
of consumer decisions based on information relevant to a company’s morality
suggests that consumers use rational considerations and objective justifications to
form evaluations and purchase decisions. For example, research on boycotting
decisions has found that consumers are affected by the perceived likelihood of the
success of the boycott and by the acceptability of available substitutes for the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
company’s products (Sen, Guerhan-Canli, and Morwitz 2001). In addition, evidence
that the company meets at least a minimum standard of social responsibility seems
to be required for consumers to consider purchasing from a company (Handelman
and Arnold 1999). The effects of these logical considerations and criteria might
suggest that purchase decisions are deliberated based on moral information about the
company.
Though most studies focus on the outcomes of company morality, one study
that measures judgments of company ethics could be interpreted as evidence that
moral judgments are based on logical considerations as well. That study shows that
marketing a harmful product to a vulnerable population affects consumers’
perceptions of the company’s ethics, in addition to intentions to purchase and
recommend the company’s products (Smith and Cooper-Martin 1997). Judgments
of the company’s ethics were more negative depending on how harmful the product
was and how vulnerable the population. Those judgments may be based on
deliberated considerations of potential harm and responsible behavior, and on the
implications of those considerations for company morality. However, research on
companies’ socially responsible behavior focuses on situations such as this in which
there is obvious justification for consumers’ judgments of the company and for
purchase decisions.
When there is such obvious justification for judgments of the company’s
morality, it is difficult to discern whether such judgments are formed based on
deliberative consideration of the justification for such a decision, or rather are
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
formed based on an associative or intuitive judgment of moral or immoral behavior.
Though it has been suggested that consumers use corporate social responsibility
information to judge the company’s trustworthiness or “character,” there is no direct
evidence that such judgments are based on rational deliberation of the implications
of social responsibility information. In at least some situations, judgments of
company morality could be formed based on association and intuitive processing.
Hypotheses
Evidence in psychology suggests that moral judgments can be formed
through associative rather than rule-based processing (Haidt et al. 1993; Tetlock et
al. 2000). I predict that judgments of company morality can also be intuitive and
based on associative processing. The theoretical basis of associative processing, and
prior evidence supporting its existence, implies that judgments lacking logical
justification, and judgments that vary based on mere association, are being
influenced by associative processing.
Implications o f associative moral judgments. The distinction between
associative processing and deliberative processing has been tested in situations
where the two systems of reasoning are likely to arrive at different judgments (e.g.,
Haidt et al. 1993; Kahneman 2003; Sloman 1996, 2002). In the context of moral
judgments of companies, one such situation is when consumers get information
about the morality of an employee. Information about an employee’s actions in his
or her private life - actions that have no direct relevance to the company or its
products - should not affect perceptions of the morality of the company as a whole.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
For example, if an employee is arrested for drunk driving while coming home from
drinking in a bar with friends, there is little reason to believe that the company as a
whole is immoral. Judgments of responsibility and blame, which should influence
deliberative judgments of morality, are based on the extent to which the individual
or group caused or had control over the event, whether the event was foreseeable, or
whether there was intent to cause harm (Mascarenhas 1995; Mikula 2003; Morse
1998). Thus, for an employee’s action in his or her private life, there is no reason to
expect that the company would meet any of these conditions for moral
responsibility. However, if moral judgments of companies are made associatively,
the link between an employee and the employing company could result in the
generalization of the moral judgment onto the company based solely on association.
Thus, based on the expectation that moral judgments in this context will be formed
based on associative processing, I predict that information about employee morality,
even when irrelevant to the company itself, will affect judgments of company
morality.
HI: Judgments of company morality are more positive when consumers have
information that an employee’s action is moral than when the
employee’s action is immoral.
Evidence that moral judgments of a company are affected by information
about an employee’s morality would provide some support for the proposition that
judgments of company morality can be based on associative processing. However,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
more compelling evidence would be provided by a comparison of deliberative
judgments with associative judgments. The dual-process theory of reasoning
implies that different judgments may result if the customer is motivated to process
information deliberately versus associatively (Sloman 1996, 2002). In a situation
where the logic that is likely to guide rule-based processing contradicts the
implications of association, judgments arrived at through the two types of processing
should differ. In this context, the rule that customers are most likely use to judge the
implications of the situation for the company’s morality are whether the company is
directly or indirectly responsible in any way for the employee’s actions (Lickel,
Schmader, and Hamilton 2003; Mikula 2003). Because the company is unlikely to
be seen as responsible, customers who are motivated to process the information
deliberatively are likely to conclude that the employee’s actions are irrelevant to the
company’s morality. In contrast, customers who process information about the
employee’s actions associatively may generalize the employee’s morality to
judgments of the company’s morality to a greater extent.
Thus, in the context of an employee’s morality in his or her private life,
associative processing is likely to lead to the judgment that the company’s morality
is similar to that of the employee, whereas rule-based processing is more likely to
result in a judgment that the employee’s morality is relatively unrelated to the
company’s morality. Though the associative judgment may automatically occur,
consumers who are motivated to apply rule-based reasoning should adjust their
judgment of the company’s morality based on the extent to which they can logically
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
justify that judgment. Consumers are unlikely to be motivated to process
information about company morality deliberatively, unless the company is highly
personally relevant to them or the judgment has implications for an important
consumption decision.
Motivation to process judgments deliberatively can also derive from the
knowledge that one will be held accountable for the judgment in some way, such as
knowing that the judgment will have to be justified to some audience.
Accountability has been found to lead to more deliberative judgments that can
correct for the potential bias of an associative judgment (see Lemer and Tetlock
1999 for a review of the effects of accountability). Accountability can motivate
deliberate information processing when participants are informed prior to making
their judgments that they will be asked to provide their reasons for those judgments
(Lemer and Tetlock 1999; Tetlock 2002). Accountability should result in different
morality judgments than when there is no accountability, if judgments of company
morality in this context are processed associatively.
H2: Accountability moderates the effect of employee morality on judgments
of company morality, attenuating the effect when accountability is high.
Strength o f association. The strength of association between the employee
and the company may also affect moral judgments made through associative
processing. If associative processing occurs, generalizations to the company’s
morality should be more likely to occur to the extent that the employee is seen as
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
representative of the company. Perceived employee representativeness indicates the
perceived strength of the association between the company and the employee in
consumers’ minds and, therefore, should predict the tendency to form moral
judgments based on that association.
The link between an employee and a company is similar to the connection of
any individual to a social group of which he or she is a member. Social identity
theory describes the tendency to define one’s own and others’ identity in part based
on their membership in social groups (e.g., Tajfel and Turner 1985). Such social
groups can be based on shared characteristics, such as racial, ethnic, or gender
groups, or groups based on personal relationships or formal membership, such as
groups of friends or professional groups (Brewer and Gardner 1996). Group
membership affects perceptions of an individual’s identity based on inferences that
the person is similar to other group members and shares the characteristics that
define the group. These types of inferences are based on association and can lead to
the generalization of perceptions of the group onto an individual member as well as
the generalization of perceptions of an individual onto the group as a whole
(Greenwald et al. 2002; Heider 1958; Lambert et al. 1998).
Investigations of social identity in the context of employees’ relationship
with companies have indicated that membership in a company as a social group is
similar to other types of group memberships (e.g., Bergami and Bagozzi 2000; Hogg
and Terry 2000). In addition, it has been theorized that consumers use information
about a company’s employees to make inferences about the company itself
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
(Bhattacharya and Sen 2003). An employee is likely to be seen as at least somewhat
representative of the company as a whole.
However, some employees may be seen as more representative of the
company’s identity than others. As with social groups and other types of categories,
some members are seen as more prototypical of the group and more representative
of the characteristics of the group as a whole (Lickel, Hamilton, and Sherman 2001).
The concept of representativeness implies that more prototypical employees will
have a stronger perceived association with the company. If moral judgments of
companies are based on associative processing, then the representativeness of the
employee should affect the extent to which moral judgments are generalized from
the employee to the company. For example, though the actions of any employee
have the potential to affect perceptions of the company, the actions of the company
president may have a greater effect than the actions of someone who works on the
assembly line. Thus, the extent to which the employee is perceived as representative
of the company is expected to moderate the effect of the employee’s actions on
judgments of the company’s morality.
H3: The actions of an employee who is perceived as more representative of
the company will affect judgments of the company’s morality to a
greater extent than the actions of a less representative employee.
Purchase intentions. Evidence in the literature on corporate social
responsibility suggests that consumers’ decisions regarding the company and its
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
products are affected by judgments of company morality (e.g., Smith and Cooper-
Martin 1997). Though there are many influences on purchase decisions, when
judgments of company morality are salient, they are likely to influence purchase
intentions. Consumers are likely to be more willing to purchase from a moral
company than an immoral company. Because judgments of company morality are
predicted to be formed through associative processing, and because purchase
intentions are likely to be based on those judgments, purchase intentions are likely to
vary with the factors that affect the outcome of the judgment process. Thus,
representativeness and accountability, as well as other factors that vary the extent to
which consumers rely on associative moral judgments, are expected to influence
purchase intentions. In addition, judgments of company morality should be
responsible for the effect of employee morality on purchase intentions, as opposed to
employee morality having a direct effect on purchase intentions. Thus, for
consumers who would consider purchasing the company’s products in the absence
of any information about the employee’s action, judgments of company morality
should account for the relationship between employee morality and purchase
intentions.
H4: Judgments of company morality mediate the effect of employee morality
on purchase intentions, for consumers who are inclined to purchase the
company’s products.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
In sum, the context of employee morality will be used to examine associative
judgments of company morality and the resulting effect on purchase intentions. In
this context, employee representativeness and the motivation to process created by
accountability should moderate the effects of associative judgments of company
morality (see Figure 1). Employee morality should be generalized to judgments of
company morality based on mere association, and this effect should be attenuated
when employee representativeness is low or when the judgment must be justified.
These results would provide evidence that moral judgments of companies can be
formed associatively and are not necessarily based on deliberative justification.
OVERVIEW OF EXPERIMENTS
Four experiments investigated judgments of company morality in the context
of an employee’s actions in his or her private life. An employee’s private actions
were used as the stimuli in these experiments to allow comparison between
associative and rule-based, deliberative judgments. Because in this context rule-
based and associative processing should result in different judgments of company
morality, it should be possible to distinguish between the two types of processes. In
addition, this context reduces the likelihood that the employee’s actions would affect
the quality of the company’s products or threaten consumers directly. An
employee’s actions as a private citizen should also be less likely to be attributed to
the influence of the company. For example, a stockbroker who recommended
unwise investments might have been influenced by the company’s unofficial policy
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
encouraging that type of immoral behavior. However, an employee who is
convicted of child abuse would be unlikely to have been acting under the influence
of the company, and it would be unlikely for that action to affect product quality.
Thus, the context helps reduce the potential effects of factors that might obscure the
effects of associative moral judgments in other contexts.
FIGURE 1: CONCEPTUAL MODEL
Employee
morality
Purchase
intentions
Motivation
to process
information
Employee
representative
ness
Judgments of
company
morality
Study 1 was designed to test H2, to demonstrate that employee morality can
be generalized to judgments of company morality, and that rule-based deliberative
processing attenuates the effect. Accountability was manipulated, to vary the
motivation to process information about the employee’s action deliberatively. The
effect of motivating consumers to process information deliberatively provides
evidence that company morality judgments are made based on associative
processing in this context. In addition, Study 1 manipulated the level of consumer
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
identification with the company to examine the effect of employee morality on
consumers who have a prior connection with the company.
The second experiment manipulated the representativeness of the employee
and manipulated the valence of moral judgment by varying the action of the
employee. This experiment was designed to test HI, H3, and H4, and to provide
additional support for the contention that moral judgments of companies are made
associatively. To test HI, both moral and immoral actions of employees varying in
representativeness were presented to participants to observe the effect of valence on
the extent of generalization to company morality. Though negative information
about immoral actions is likely to have a greater effect, as has been observed in
many contexts (e.g., Ahluwalia 2002), positive information about moral actions also
could affect customers’ perceptions of the company. In addition, potential
alternative explanations for associative processing were examined, such as the
perceived responsibility of the company for the employee’s actions, and the extent to
which the employee’s actions are seen as diagnostic of the morality of the
company’s other employees.
The third experiment examined the effect of priming the concept of
responsibility and mitigating circumstances for the employee’s action on the extent
to which customers generalize the moral judgment to the company. Information
suggesting a potential cause of the employee’s immoral action was provided, and its
effect was compared to irrelevant information and a no information control
condition. Information relevant to the cause of the employee’s action may motivate
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
consumers to process information more deliberatively about the responsibility of the
company for the employee’s actions, and therefore about the extent to which moral
judgments should be generalized to the company. In addition, this study further
examined the effect of moral versus immoral actions to provide additional tests of
HI and H4.
The fourth experiment investigated additional factors that may prime
consumers to process moral judgments deliberatively versus associatively. This
study examined the effect of information about actions taken by the company on the
extent of generalization of the employee’s morality. Actions that alter the
employee’s association with the company should affect moral judgments based on
associative processing, whereas actions that suggest that the company does not
endorse the employee’s action should inspire more deliberative processing of
company morality judgments.
These experiments examine moral judgments of companies based on
employee morality, manipulating the context and content of information about the
employee’s morality to observe effects based on associative versus deliberative
processing. Because scenarios involving a fictitious company and events are used,
the context is devoid of many potentially confounding features, such as prior
associations with the company. Though scenarios are a commonly used
methodology, they also lack many of the features that make real-life situations
compelling to consumers. However, this rather restricted context should provide a
rigorous test of the hypotheses, and the effect of moral judgments could be even
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26
stronger in real-life instances of conspicuous employee morality that might be
connected to the company.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
CHAPTER II: STUDY 1
Study 1 was designed to investigate the effect of employee morality on
judgments of company morality, and the effect of accountability on the extent of this
generalization. Customer-company identification was also varied to observe its
effect on judgments of company morality. Both of these variables have the potential
to affect consumers’ motivation to form deliberative judgments of the company’s
morality.
Accountability
To test H2, a method of distinguishing the effects of deliberative and
associative processing was used. Participants were informed that they would have to
provide reasons for their responses before they were exposed to information about
the employee’s actions. This method of accountability has been used by dual
process researchers in investigations to distinguish between relatively automatic,
intuitive, associative processing as compared to more deliberate, reasoned, rule-
based processing in forming judgments (Lemer and Tetlock 1999; Sloman 2002).
Associative processing results in a judgment based on association or similarity to
other entities or concepts for which an attitude is readily accessible. In contrast,
deliberative processing involves a more effortful assessment of the evidence and its
relevance in forming a judgment that can be justified with an explanation. Thus,
when an explanation of the reasons for a judgment is required, deliberative
processing is motivated to form justifiable judgment. I predict that the moral
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 8
judgment of the company, and the effects on purchase intentions that result from that
judgment, will be based on the mere association of the employee with the company
in the absence of explicit motivation to deliberate on these judgments. Thus,
motivating participants to process information deliberatively should attenuate the
effects of information about an immoral action by an employee.
Identification
This study also manipulated participants’ identification with the fictitious
company described to observe its effects on moral judgments. Identification may
reduce the effects of employee morality on judgments of company morality, if
identification motivates deliberative information processing. The internalization of a
relationship, connection, or association with a social group is defined in social
identity theory as identification (Doosje et al. 1998; Tajfel and Turner 1985).
Research in marketing suggests that consumers can identify with companies, and
identification may motivate deliberative processing because information about the
company is personally relevant and involving (Arnett, German, and Hunt 2003;
Bhattacharya, Rao, and Glynn 1995; Bhattacharya and Sen 2003). There is some
evidence to suggest that customers who are identified with a company would be
more forgiving and more likely to process information deliberatively when they are
exposed to negative information about the company or its products than consumers
who don’t identify with the company (Ahluwalia, Bumkrant, and Unnava 2000).
For example, a consumer may purchase from and identify with The Body Shop
based on their environment- and animal-friendly products and policies. That
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
customer may be motivated to view the company in a favorable light that preserves
the company as a source of positive identity, and therefore might be positively
biased and motivated to process information deliberatively in interpreting potentially
negative information about the company.
However, the treatment of identification in marketing has not considered the
effects of the strength of customers’ identification. Strength of identification
depends on the importance of the company to the customer’s self-concept (Arnett et
al. 2003; Bhattacharya et al. 1995; Bhattacharya and Sen 2003; Doosje et al. 1998).
Research in psychology has found that strong identification with social groups has a
positive effect on interpretations of negative information (e.g., Doosje et al. 1998).
However, lower levels of identification are not as likely as high levels of
identification to provide sufficient motivation for customers to bias their perceptions
or form deliberative judgments of information related to the company.
It is important to note that deliberative processing does not imply unbiased
processing - rather, it simply implies that deliberative judgments are justified in the
mind of the individual (Jain and Maheswaran 2000). Such justification may or may
not be objectively unbiased. Similarly, associative processing is not necessarily
biased, because inferences made based on association may or may not be objectively
true or impartial. High levels of identification are likely to lead to more biased and
more deliberative processing of information about a company’s morality. However,
low levels of identification are more likely to lead to associative processing that is
not biased by a desire to maintain a positive perception of the company because
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
information about the company is not sufficiently personally relevant or involving.
Though generalizing perceptions of an employee’s morality onto perceptions of the
company’s morality does imply a biased judgment when it contradicts more
objective justification, this is not likely to come about because consumers with a low
level of identification are engaging in biased reasoning.
In the context of this investigation of the process of judgments of company
morality, the effect of low levels of identification is an important factor to consider.
Low levels of identification based on a minimal connection to a company may be
relatively common and are undoubtedly more likely to occur than the intense
relationship with a company implied by a high level of identification. Though less
likely than high levels to result in biased processing of moral information about the
company (Doosje et al. 1998), low levels of identification may result in more
positive judgments of company morality than the judgments of consumers who lack
any identification with the company.
Simply purchasing from a company or publicly endorsing a company could
create a low level of customer-company identification. Research in psychology has
shown that a minimal connection is capable of creating a low level of identification.
Using a methodology called the minimal group paradigm, researchers have shown
that simply assigning a meaningless group label or creating temporary and
meaningless groups of individuals can create a discernible level of identification
with the group (e.g., Cadinu and Rothbart 1996). Thus, a company’s customers,
even if they do not have a strong or intense relationship with the company, may
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
identify with the company to some extent just based on the connection created by
having purchased from the company in the past.
In fact, the majority of a company’s customers are likely to have a relatively
low level of identification with the company. For most companies, the negative
effects of decreased purchase intentions in existing customers are likely to have the
greatest potential to affect the company’s sales, compared to effects on willingness
to purchase for consumers who were not among the company’s existing customers.
Thus, it is important to study the effects of moral information on customers with a
low level of identification with the company. Study 1 manipulated identification to
examine the potential for low levels of identification to positively bias moral
judgments and to examine the effect of employee morality on customers’ purchase
intentions.
Method
Ninety-eight students participated in the study for partial course credit in an
introductory marketing class. The sample included 46% females (mean age = 20.7
years). The study used a 2 x 2 full factorial between-subjects design. The first
factor manipulated identification (low vs. none), and the second factor manipulated
accountability (high vs. none). High accountability was created by informing
participants that they would be asked to provide reasons for their judgments and
decisions, before they received information about the employee’s action. The no
accountability condition did not require judgments to be justified and did not
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
32
intervene in the process by which participants formed judgments of the company’s
morality.
Identification pretest. A pretest was conducted to verify that low levels of
identification could be induced by eliciting from participants an endorsement of the
company in the form of expressing a willingness to buy the company’s products.
Sixty-four university students participated in the experiment for partial course credit
in an introductory marketing class. The sample consisted of 53% females (mean age
= 20.3 years). A scenario methodology was used, describing a fictitious clothing
company, identified only as Company K. A clothing company was selected because
customer-company identification may be more likely, and perceptions of company
identity more important, in situations where the company’s products are symbolic,
hedonic, or publicly consumed. Though an investigation of the situations in which
social identity has the greatest effect is beyond the scope of the current investigation,
it is important to recognize that the context of the company and its products are
likely to have an effect on the extent of the effects of identification.
Participants were first presented with a description of the high quality of the
company’s clothes. The questionnaire stated that a consumer magazine had reported
ratings of the company’s clothes, which were described as among the top brands of
clothing, in terms of style, design, and durability. The potential for customer-
company identification was manipulated by describing the company’s customers as
either similar or dissimilar to the college student participants. The company’s
customers were described either as “college students (aged 18-24)” in the low
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
identification condition or as “senior citizens (age 65 and over)” in the no
identification condition. Because other customers of the company are an indicator
of the company’s image or identity (Bhattacharya and Sen 2003), this manipulation
was expected to affect participants’ tendency to identify with the company, as well
as their willingness to express an intention to purchase the company’s clothes for
themselves.
Participants were then asked to rate their willingness to purchase and
recommend the company’s clothes and to rate the clothes’ likely quality. The
process of declaring a willingness to purchase the company’s products should induce
a degree of identification with the company, because it is tantamount to making a
public commitment to become a customer of the company. Such a commitment
should increase the potential for identification, in addition to the potential created by
the similarity of the company’s customers to the participant (Bergami and Bagozzi
2000). Participants were significantly more likely in the low identification than the
no identification condition to say they would purchase from the company [M= 6.76
vs. M = 2.11; /* '( 1, 62) = 226.00,p < .01] and to say they would recommend the
company’s clothes to friends and people they know [M= 6.05 vs. M = 2.02; F( 1, 62)
= 118.17,/? < .01]. However, as intended, quality ratings were not significantly
affected [M= 7.16 vs. M — 6.12; F(1, 62) = 2.30,p > .10]. Because participants who
thought the company’s customers were senior citizens were unlikely to say that they
would purchase from the company, they were unlikely to identify with the company.
On the other hand, participants who saw the description of the company as a high
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
quality producer of clothing aimed at college students were likely to say that they
would purchase the company’s clothes, which made it likely that they would identify
with the company. After rating their willingness to purchase and recommend the
company’s products, participants rated their identification with the company.
Identification was measured on a three-item scale adapted for this context
based on measures of identification developed by Tropp and Wright (2001) and
Reed and Aquino (2003). The measure indicates the extent to which participants felt
that they were 1) associated with, and 2) connected to, the company, as well as the
extent to which being a customer of the company 3) reflects “who I am.” Each of
these items was rated on a nine-point scale, from “definitely disagree” to “definitely
agree.” The three-item measure loads on a single factor accounting for 78% of the
variance in a principal components analysis and shows good reliability (a = .86).
The results indicate that the identification manipulation was successful. The
manipulation significantly predicted identification with the company, with a
moderate level of identification in the low identification condition (M= 5.20) and
very low ratings of identification in the no identification condition [M = 1.96; F( 1,
62) = 67.01 ,p < .01]. Importantly, the low identification condition did not result in
very high levels of identification with the company. The identification scores
induced by this procedure were not significantly different from the midpoint of the
scale [i.e., midpoint = 5.00, t{33) = 0.59,p > .10], whereas scores in the no
identification condition were significantly lower than the midpoint of the scale [/(29)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
= -16.17, p < .01]. Thus, the manipulation of identification was successful in
creating a relatively low or moderate level of identification with the company.
Main study procedure. The identification induction used in the pretest was
presented at the start of the questionnaire. Participants were given quality
information about the company’s clothes, rated either by college students or senior
citizens depending on the randomly assigned condition, and participants rated their
purchase and recommendation intentions. The motivation to process instruction was
placed at the bottom of the initial page that presented the identification induction
manipulation. Participants in the high accountability conditions were told that they
were about to receive additional information, and that they should be aware that they
would be asked to provide reasons for the judgments and decisions they made based
on that information. For the other half of participants in the no accountability
conditions, no justification of judgments was required. On the second page of the
questionnaire, all participants were presented with information that the head
designer of the company’s clothes had been convicted of child abuse. No details of
the offense were given beyond that description, and participants were told that the
head designer was given probation, because it was a first offense, and continued to
work for the company. Following this information, participants were asked to rate
their purchase intentions, the morality of the employee and company, and the other
dependent variables in this study.
Measures. Each of the constructs was measured with multiple items that
were rated on nine-point scales (see Appendix A for a list of items used in all four
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
studies). Purchase intentions and quality were measured with two items each.
Additional items were deemed unnecessary and likely to induce participant fatigue.
Purchase intentions were indicated on semantic differential scales from “not at all
likely” to “very likely” and from “definitely would not” to “definitely would”
purchase. The likely quality of the clothes was rated on two semantic differential
scales from “poor” to “excellent” and from “very bad” to “very good.” Each of
these measures showed high correlations within the two-item scales (r > •94, p <
.01), suggesting that they are reliable indicators of the respective constructs. The
order of measuring quality and purchase intentions was randomized to control for
order effects, with half of participants receiving the quality measure first, and the
other half receiving the purchase intention measure first. However, order did not
significantly affect responses on these variables.
Participants’ judgment of the company’s morality was then measured with
three semantic differential scales. The three scales are used to indicate the extent to
which the company was considered 1) “highly immoral” versus “highly moral,” 2)
“not at all harmful” versus “very harmful” (reverse coded), and 3) “very cruel”
versus “very compassionate.” This measure exhibited high reliability (a = .88). The
measure is derived from a measure of judgments of company ethics (Smith and
Cooper-Martin 1997), and it was adapted to focus specifically on the definition of
moral judgment used here. The employee’s morality was measured using the same
three items, which also exhibited high reliability (a = .82).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
Open-ended items were included only in the high accountability conditions.
These items prompted participants to provide their reasons for responding the way
they did to the items regarding the company’s morality and their purchase intentions.
Following these measures, participants’ identification with the company was
measured, as a manipulation check.
Results
The results support the contention that moral judgments are formed
associatively. As expected, employee morality was generalized to judgments of
company morality when they were not motivated to form deliberative judgments.
As predicted by H2, the effects of information about an employee’s immoral action
were attenuated by the motivation to process deliberatively that was created by
accountability. The data were analyzed using MANOVA to assess the effects of
identification and accountability.
Manipulation check. Ratings of identification indicate that the manipulation
was successful, based on a significant main effect of the identification manipulation
[M= 3.69 for low identification vs. M = 2.24 for no identification; F(l, 93) = 16.86,
p < .01]. There is no significant main effect of accountability on identification
ratings [F(l, 93) = 2.82,p > .10], and no interaction effect [F(l, 93) = 0.05,p > .10].
Also, as intended, moral judgments of the employee’s actions do not vary with the
manipulations of accountability, identification, or their interaction [F(l, 93) = 1.99;
F (l, 93) = 0.03; F (l, 93) = 0.12; p > .10], See Table 1 for mean ratings of all
variables in this study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
Judgments o f company morality. Judgments of company morality show the
expected attenuation effects of the accountability manipulation, suggesting that
judgments are formed associatively when there is no induced motivation to process
the information deliberatively. There was a significant main effect of accountability
[F(l, 93) = 13.66,p < .01], but no significant main effect of identification [F( 1, 93)
= 0.11 ,p > .10] and no significant interaction effect [/-’ (1, 93) = 0.00,/? > .10]. The
company was rated as significantly more moral when participants were motivated by
accountability to process the information about the employee’s immoral action
deliberatively (M = 5.38) than when they did not have to justify their judgments (M
= 4.07). This main effect suggests that associative processing occured when there
was no accountability to provide reasons for judgments of company morality.
Informing participants that they would have to provide justification for their
company morality judgments resulted in significantly more positive judgments than
when justification was not required (see Figure 2A). This suggests that the more
negative judgments in the no accountability conditions were based on associative
processing for which justifications were not formed spontaneously in arriving at the
judgment.
Purchase intentions. Purchase intentions reflect the effects of the
accountability and identification manipulations. There is a significant main effect of
accountability [F(l, 93) = 12.58,/? < .01], and a significant main effect of
identification [F(l, 93) = 93.56,/? < .01]. These effects are qualified by a significant
interaction [F(l, 93) = 6.76,/? < .01]. Low identification participants in the high
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
accountability condition were significantly more likely to say they would purchase
from the company than low identification participants in the no accountability
condition [see Table 1 and Figure 2B; t{53) = 4.08, p < .01]. As expected, purchase
intentions in the low identification, no accountability condition were still
significantly higher than in either of the two no identification conditions [1(45) =
5.38 for no accountability; t(48) = 4.38 for high accountability; p < .01]. This is due
to the effect of the identification manipulation on baseline purchase intentions (i.e.,
college student participants were never going to consider purchasing clothing made
for senior citizens). Overall, the results for company morality and purchase
intentions support H2 and suggest that associative processing is responsible for the
effect of an employee’s immoral actions on customers’ negative reactions.
Quality. The results for quality ratings suggest that perceptions of quality are
not responsible for these effects on purchase intentions. Quality ratings do not
significantly vary with the manipulation of identification [F (l, 93) = 0.27, p > .10],
accountability [F(l, 93) — 0.60,p > .10], or their interaction [F(l, 93) = 0.16, p >
.10]. In addition, though quality is a significant covariate of purchase intentions
[F(l, 92) = 4.98,p < .05] and a marginally significant covariate of judgments of
company morality [.F(l, 92) = 3.09,p < .10], controlling for the variation accounted
for by quality does not change the results for these variables. Accountability still
significantly predicts judgments of company morality, and identification and
accountability significantly predict purchase intentions, when the effect of quality
perceptions is accounted for. Thus, the effect of information about an employee’s
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
immoral actions is not due to changes in the perceived quality of the company’s
products.
TABLE 1: DEPENDENT VARIABLE MEANS BY CONDITION IN STUDY 1
Low identification No identification
Dependent
variablea
High
account
ability
No
account
ability
High
account
ability
No
account
ability
Identification
3.89
(2.34)
3.41
(1.53)
2.55
(1.34)
1.93
(1.31)
Employee 1.68 1.47 1.78 1.43
morality (1.06) (0.72) (1.23) (0.79)
Company 5.31 4.01 5.43 4.13
morality (1.68) (1.58) (1.79) (1.85)
Purchase 5.86 3.94 2.04 1.75
intention (1.80) (1.67) (1.35) (0.85)
Quality 6.93 6.83 7.28 6.88
perception (1.73) (1.80) (1.44) (1.75)
n 28 27 23 20
a Higher scores on each of these measures indicate more of that construct (i.e. higher numbers mean
more moral, more identified, greater purchase intention, higher quality).
Open-ended responses. Examination of the reasons given to justify
participants’ moral judgments and purchase intentions in the high accountability
conditions indicates that they were likely to come to the conclusion that the
employee’s actions were irrelevant to judgments of the company and to purchase
intentions. The reasoning provided by the majority of participants indicated that
they felt that one employee was too small a sample on which to base conclusions
about the company or that the employee’s actions were irrelevant to the quality of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
FIGURE 2: COMPANY MORALITY JUDGMENTS AND PURCHASE
INTENTIONS IN STUDY 1
A)
Company
morality
9
B High accountability
■ No accountability
Low No identification
identification
B)
Purchase
intention
I High accountability
I No accountability
Low identification No identification
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
the company’s products. For example, participants said things like “the company
had nothing to do with the designer’s actions,” or “I don’t care what one employee
does as long as the quality of the clothes is good.” See Appendix B for a complete
list of the open-ended responses.
Responses for both company morality and purchase intentions indicate that
when motivated to form deliberative judgments, participants were more likely to
conclude that the employee’s action should not be taken into account. The reasons
provided for judgments of the company’s morality generally indicated that the
company either had no control over the head designer’s actions or that the designer’s
actions were irrelevant to judgments of the company. Thirty-five out of fifty-one
responses (69%) indicated that the employee’s action should not be generalized to
judgments of the company’s morality. In addition, the reasons provided for
purchase intentions indicated that the designer’s action was seen as unlikely to affect
product quality and would not affect purchase intentions. Thirty-seven out of fifty-
one responses (73%) indicated that the employee’s actions were irrelevant to their
purchase intentions.
Discussion
The results of this study suggest that information about an employee’s
immoral action affects customers’ judgments of company morality and purchase
intentions. This supports the prediction that an employee’s morality can be
generalized to the company, and indicates that an employee’s actions can affect
purchase intentions. The results also support H2 and suggest that consumers are
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
predisposed to process information about company morality associatively rather than
deliberatively. When motivated to process information deliberatively, participants
were less likely to generalize the employee’s morality to the company, and this
occurred regardless of their level of identification with the company. Accountability
resulted in more positive company morality judgments, and in higher purchase
intentions for low identification participants (see Table 1 and Figure 2).
As indicated by the open-ended responses, participants could come up with
reasons to discount the information about the employee’s actions when they were
motivated to do so. However, when they were not explicitly motivated by the need
to provide reasons for their judgments, participants generalized the employee’s
morality onto the company. It seems that associative processing may be responsible
for an intuitive effect of the association between the company and the employee’s
action on judgments of company morality.
Accountability seems to motivate deliberative processing, resulting in
different judgments of company morality than when accountability was not
manipulated. In contrast, low levels of identification do not seem to motivate
deliberative processing and positively biased perceptions of the company, which
seems to contradict predictions that have been made for customer-company
identification. However, because strength of identification has not been considered
in the marketing literature on identification, these findings are consistent with a view
that high levels of identification are required to motivate biased and deliberative
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
processing, whereas low levels of identification do not provide sufficient motivation
to bias perceptions of the company or to process information deliberatively.
In sum, Study 1 provides initial evidence to support the prediction that
judgments of company morality in the context of an employee’s immoral action are
based on associative processing. Evidence supporting H2 suggests that
accountability motivates deliberative processing, resulting in more positive morality
judgments and purchase intentions than when participants were not explicitly
motivated. The more negative judgments and purchase intentions when there was
no accountability suggests that associative processing influences judgments and
decisions in this context. However, one cannot directly test for the presence of
associative processing, because, as demonstrated in the accountability conditions in
this study, it is difficult to observe the process of reasoning without causing
reactivity in that process and the resulting judgment. Therefore, it is necessary to
seek additional evidence that is consistent with an associative processing explanation
of these results.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
CHAPTER III: STUDY 2
Study 2 was designed to provide further evidence that associative processing
is responsible for judgments of company morality, and to investigate the effects of
the valence of the employee’s action. Associative judgments should be affected by
employee representativeness, whereas deliberative judgments should not. Because
representativeness is an indicator of the strength of the employee’s association with
the company, moral judgments of the company as well as purchase intentions should
be moderated by employee representativeness. Hypothesis 3 predicts that the
actions of a more representative employee will have a greater effect on customers’
reactions than a less representative employee. The actions of a former employee
were included as a comparison condition to provide a baseline for the effects of
information about a person who should not be perceived as representative of the
company.
Two alternative explanations for the effects attributed to association on
customers’ reactions were also investigated. Though accountability seems to
motivate deliberative, rule-based processing resulting in more positive judgments of
company morality, it is possible that consumers are using a decision rule in the no
accountability conditions in Study 1 that they were unwilling to declare as a
justification for their decisions. They might have changed their responses to
conform to a justification that they found more defensible and parsimonious - i.e.,
that the employee’s morality was irrelevant to judgments of the company.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
Two potential justifications that would imply rule-based processing are
responsibility and diagnosticity. The extent to which the company is seen as
responsible for the employee’s action could explain the effect of employee morality
on perceptions of company morality. The employee’s action could also be seen as
diagnostic of the likely behavior of other company employees. Each of these
alternative explanations would suggest that participants in the no accountability
conditions in Study 1 might have been expressing deliberative rather than
associative judgments, but that when the participants in the high accountability
conditions were asked to provide reasons they selected more defensible reasons and
adjusted their judgments accordingly. These potential alternative explanations need
to be ruled out to lend greater confidence to the assertion that associative judgments
can occur in this context.
Collective Responsibility
Social groups are sometimes held collectively responsible for the actions of a
group member, even when the group did not directly sanction or actively contribute
to the individual’s actions (Lickel et al. 2003). Research has shown that social
groups that have a higher level of interaction and shared characteristics with an
individual, as well as those that have authority over the individual, are more likely to
be held responsible for the individual’s actions. A company could fulfill both of
these roles in relation to an employee. Employees are likely to have a relatively high
level of interaction with the company and with other employees, and the company is
likely to have at least a certain amount of authority over the person, in that the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
person depends on the company for his or her livelihood. Based on this interaction
and authority, the company as a whole or the company’s other employees may be
seen as responsible to some degree for the moral or immoral action of the employee.
The responsibility of the company for the employee’s immoral action may
have been difficult for participants in Study 1 to put into words, or may have seemed
an indefensible or invalid justification for negative judgments of the company’s
morality. However, if participants are presented with questions about the
responsibility of the company’s employees to prevent or anticipate the employee’s
immoral action, they may see it as a legitimized reason for their relatively negative
judgments of the company’s morality. In addition, perceptions of collective
responsibility could account for the effects of the employee’s representativeness, if
the company is seen as more responsible for the actions of a more representative
employee. Thus, collective responsibility needs to be ruled out as a potential
alternative explanation for the effects attributed to associative judgments.
Diagnosticity
Another potential justification for the negative judgments of company
morality found in Study 1 is the perceived diagnosticity of the employee’s action in
predicting the behavior of other company employees. Diagnosticity is the degree to
which information is perceived as relevant to forming impressions, evaluations, and
attitudes (e.g., Skowronski and Carlston 1987). A cue or piece of information that is
seen as diagnostic will have a greater effect on attitudes than information that is less
diagnostic. Diagnosticity has been shown to influence person perception, as well as
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
attitudes toward products and services (Folkes and Patrick 2003; Herr, Kardes, and
Kim 1991; Skowronski and Carlston 1989). The perceived diagnosticity of
information about one employee in predicting the behavior of other employees has
been found to influence inferences about other employees and the company as a
whole (Folkes and Patrick 2003). Diagnosticity of the employee’s morality for the
likely behavior of other employees could explain the generalization of employee
morality onto the company as a whole. If other employees were seen as more likely
than otherwise to behave immorally as well, judging the company to be immoral
could be based on this justification.
Diagnosticity may also covary with the manipulation of employee
representativeness in Study 2. A more representative employee may be seen as more
diagnostic of the company, leading to greater generalization of the employee’s
morality. This would imply that participants are forming a deliberative rather than
associative judgment of company morality. However, if both diagnosticity and
collective responsibility can be ruled out as alternative explanations, the assertion
that associative processing is responsible for company morality judgments in this
context can be made with greater confidence.
Valence
The valence of the employee’s actions was varied in Study 2 to observe the
effects of employee representativeness on judgments of company morality for both
moral and immoral employee actions, as predicted by HI. Hypothesis 1 predicts
that judgments of company morality will be more negative given information about
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
an immoral employee action than when given information about a moral employee
action. This comparison provides additional evidence that employee morality is
affecting judgments of company morality by association. However, the valence of
the employee’s action was also varied to observe a possible negativity effect on
customers’ reactions.
Numerous studies have found that negative information is more likely than
positive information to affect attitudes when the perceiver has no motivation to bias
information processing (e.g., Ahluwalia 2002). Theoretically, a positive association
with an employee’s moral action should positively affect judgments of company
morality, in the same way that immoral actions negatively affected judgments in
Study 1. However, more extreme and more consistent positive information about
employees’ actions maybe necessary to create a significant increase in judgments of
company morality and purchase intentions.
Diagnosticity is also relevant to the potential effect of the valence of the
employee’s action. Research by Folkes and Patrick (2003) demonstrated that a
positive service experience was perceived to be more diagnostic of the likely quality
of service that one could expect from other members of the company than was
negative information. Thus, those researchers found a positivity effect for
inferences from service employees. However, it is more likely that there will be a
negativity effect in the context of morality than in the context of ability and quality
(Martijn et al. 1992). To distinguish the potential for negativity versus positivity
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
effects, diagnosticity for the likelihood of both positive and negative behavior by
other employees of the company will be measured.
Collective responsibility and diagnositicity as potential justifications for
judgments of company morality are not predicted to account for the effects of
employee morality. However, these are plausible alternative explanations for the
effects that have been attributed here to associative processing. These explanations
need to be ruled out in order to provide further support for the contention that
customers are forming moral judgments based on mere association, rather than
forming judgments based on a reasoned conception of why an employee’s actions in
his or her private life implicate the morality of the company as a whole.
Method
One hundred and ninety-one university students participated for partial
course credit in an introductory marketing course. The sample consisted of 61%
females (mean age 20.6 years). The study used a 2 x 3 full-factorial between-
subjects design. The first factor manipulated the valence of the moral implications
of the employee’s actions (moral vs. immoral). In the immoral action conditions, the
action described was child abuse, using the same scenario as in Study 1. In the
moral action conditions, the action described was saving a child from abuse, and
participants were told that the local police department had praised the employee’s
bravery.
The second factor varied the representativeness of the company employee.
Three positions with the company were used to operationalize representativeness
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
(high vs. low vs. control). Attributing the actions to the head designer of the
company’s clothes was used for the highest level of the employee’s
representativeness. The actions of an assembly-line worker were used for relatively
low employee representativeness.
Control group participants read about the actions of a former assembly-line
worker for the company, which serves as a baseline comparison for reactions based
on an individual who has no influence in the company and should not be seen as
representative of the company. In contrast to the high and low representativeness
conditions, the control group was told that the person had not worked for the
company at the time of the action described. To decrease representativeness further,
participants were told that the former employee had worked for Company K for only
a brief time and that the person had not worked for the company in years.
Procedure. The procedure of this study was similar to that used in the low
identification condition of Study 1. The identification induction was presented first,
describing the high quality of the company’s clothes for college students (aged 18-
24). On the second page of the questionnaire, information about the employee’s
action was presented. Depending on the condition to which the participant was
assigned, the employee was described as the head designer, an assembly-line
worker, or a former employee of the company. The action was described as either
saving a child from abuse or as child abuse. Following the description of the
employee and the action, participants were asked to rate their willingness to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
purchase the company’s clothes. These measures were followed by measures of the
other variables in the study.
Measures. Purchase intentions were measured with the same two items as in
Study 1. The measure again showed high correlation within the two-item scale,
suggesting high reliability (r = .80,p < .01). The participants’ judgment of the
employee’s morality, as a manipulation check, and their judgment of the morality of
the company as a whole were measured with the same items used in Study 1. Both
of these measures exhibited high reliability (a > .86).
Participants then rated the extent to which the employee was perceived as
representative of the company’s identity. This measure was derived from measures
used by Lickel et al. (2003) and from those authors’ analysis of factors that influence
perceptions of entitativity. The employee’s representativeness was measured with
five items assessing the extent to which the employee was perceived as having 1)
power and 2) prestige in the company, as well as the extent to which the employee
was likely to 3) be similar to other employees, 4) have a lot of interaction with other
employees, and 5) to have a lot of influence on other employees. Each of these
items was measured on a nine-point scale, from “strongly disagree” to “strongly
agree.” The measure exhibited high reliability (a = .82).
Participants also rated their perceptions that all the employees of the
company share some responsibility for the representative’s actions. The
responsibility of other employees of the company was measured by gauging the
extent to which participants agreed with statements, from “strongly disagree” to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
“strongly agree.” The items used to measure this construct is listed in Appendix A.
The measure is based on that used by Lickel et al. (2003) to measure the collective
responsibility of social groups for an individual member’s actions. The measure
exhibited high reliability (a = .80).
Perceptions that the actions of the representative were diagnostic of the likely
actions of other people associated with the company were also measured.
Diagnosticity for the likelihood of positive behavior as well as for the likelihood of
negative behavior was assessed. The likelihood of actions differing in their valence
were measured separately based on the possibility for a negativity effect resulting in
greater diagnosticity of the employee’s action for negative behavior than for positive
behavior, or conversely the possibility for a positivity effect resulting in greater
diagnosticity for positive than negative behavior. The two scales measured the
extent to which participants agreed with statements about the likely actions of people
associated with the company, including employees and customers. Participants rated
the extent to which they agreed with six statements (three for each of the two scales),
from “strongly disagree” to “strongly agree.” The items used are listed in Appendix
A. These measures also showed high reliability (a > .85).
Results
The results indicate that employee representativeness moderates the extent of
the effect of employee morality on company morality judgments. The results thus
support H3, as well as HI and H4. The evidence suggests that moral employee
actions can lead to more positive judgments of company morality and immoral
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
54
actions to more negative morality judgments. In addition, judgments of company
morality mediate the effect of employee morality on purchase intentions. The data
were analyzed using MANOVA, with the employee’s representativeness and action
valence as independent variables, and with judgments of the employee’s morality
and of the company’s morality as well as purchase intentions entered as dependent
variables.
Manipulation check. Analyses were conducted to confirm that the
manipulations of valence and representativeness were successful. Ratings of the
morality of the employee’s actions exhibit a significant main effect of valence [M =
8.30 for moral action vs. M = 1.58 for immoral action; F( 1, 181) = 1966.94,p < .01].
As intended, employee representativeness does not have a significant main effect, or
an interaction with valence, on judgments of the employee’s morality [F( 2, 181) =
2.01, F(2, 181)= 1.24,^>.10],
The manipulation check for representativeness shows a significant main
effect of the employee described as committing the behavior [F(2, 181) = 73.02,p <
.01], as well as a main effect of valence [F(l, 181) = 14.46, p < .01], The interaction
between the two factors was not significant [F(2, 181) = 1.54,p > .10]. As
expected, the head designer was seen as more representative of the company (M =
6.11) than the assembly-line worker [M= 4.16; /(143) = 8.1 \,p < .01], who in turn
was seen as more representative than the former employee [M = 3.01; t(\ 15) = 4.31,
p < .01]. The main effect of valence indicates that employees were seen as more
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
55
representative when the behavior was positive than negative (M = 4.83 vs. M —
4.03).
Though unexpected, the main effect of valence is interesting because it is
consistent with an identification-induced bias. Even customers with a low level of
identification may be motivated to bias their perception of an employee’s
representativeness when it gives the potential for a more positive identification.
This can be accomplished by downplaying the representativeness of an immoral
employee or by emphasizing the representativeness of a moral employee. Though
these two possibilities cannot be distinguished here, the overall result provides
additional support for the importance of identification in this context.
Judgments o f company morality. Judgments of the company’s morality
exhibit the predicted effects of valence and representativeness, consistent with HI
and H3. Representativeness had a marginally significant main effect on judgments
of company morality [F(2, 181) = 2.44 ,p < .10], and the valence of the employee’s
actions had a significant main effect [F(l, 181) = 57.76,/? < .01]. These main effect
results were qualified by a significant interaction between representativeness and
valence [F(2,181) = 21.54,/? < .01].
Multiple comparisons indicate that the interaction reflects the greater effect
of information about the highly representative employee’s actions and a greater
effect of negative than positive information (see Table 2 for the means in each
condition). Judgments of the company’s morality were significantly more negative
when the head designer committed the negative action than when the assembly-line
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
worker did [7(71) = -2.39, p < .05]. In addition, negative actions by the assembly-
line worker resulted in significantly more negative perceptions of the company as
compared to negative behavior by a former employee [t(57) = -3.78, p < .01].
TABLE 2: DEPENDENT VARIABLE MEANS BY CONDITION IN STUDY 2
Dependent
variablea
High
representativeness
Low
representativeness
No
representativeness
Moral Immoral Moral Immoral Moral Immoral
Employee 8.54 1.71 7.98 1.58 8.37 1.45
morality (0.70) ( 1.46) ( 1.29) ( 1.08) (0.91) (0 .76)
Representative 6.27 5.89 4.56 3.76 3.64 2.38
ness (1.41) (1.51) (1.28) (1.41) (1.53) (1.21)
Company 7.20 3.88 6.73 4.76 5.97 6.29
morality (1.52) (1.52) (1.41) (1.62) (1.39) (1.34)
Purchase 7.03 4.32 6.65 5.58 7.02 7.43
intention (1.54) (2.10) (1.35) (1.95) (1.39) (1.30)
Company 3.98 3.34 3.64 3.15 3.29 3.09
responsibility (1.78) (1.57) (1.50) (1.30) (1.47) (1.32)
Negative 2.26 2.32 2.47 2.26 2.94 2.33
diagnosticity (1.46) (1.91) (1.34) (1.42) (1.57) (1.46)
Positive 4.89 5.33 4.28 5.28 4.87 5.70
diagnosticity (1.69) (2.32) (1.83) (1.10) (2.21) (1.59)
n 37 37 37 36 21 23
a Higher scores on each of these measures indicate more o f that construct (i.e. higher numbers mean
more moral, more representative, greater purchase intention, etc.).
When the employee’s actions were positive, high representativeness led to
more positive perceptions of the company than when the employee was not
representative [t(56) = 3.06,p < .01]. The low representativeness employee’s effect
on company morality fell between these two extremes and was significantly higher
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
57
than the non-representative employee [t(55) = 1.98,p < .05], though not
significantly different from judgments based on the highly representative employee
[#(71) =-1.36,p > . 10].
In support of HI, judgments of company morality were more negative in
response to the immoral employee action than the moral action (see Figure 3A).
When the head designer committed child abuse, judgments of company morality
were significantly more negative than when the head designer saved a child from
being abused [7(72) = -9.40,p < .01]. When the assembly-line worker’s action was
immoral, judgments of company morality were again significantly more negative
than when the action was moral [t(70) = -5.50, p < .01]. However, judgments of
company morality did not significantly differ based on the morality of the former
employee’s action [t(42) = 0.78,p > .10].
Purchase intentions. The results for purchase intentions also show
significant effects, especially when the action is negative. Ratings of purchase
intentions follow the same pattern as judgments of the company’s morality (see
Figure 3B). Representativeness has a significant main effect [F(2, 181) = 11.61 ,P<
.01], and the valence of the representative’s action significantly predicts purchase
intentions [F( 1, 181) = 19.20, /? < .01]. These main effects are qualified by a
significant interaction between representativeness and valence [F(2,181)= 11.74, p
< .01]. Multiple comparisons indicate that purchase intentions are significantly
lower when an immoral action is attributed to a high than a low representativeness
employee [see Table 2; t{l\) = -2.66,p < .01]. In addition, purchase intentions are
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
significantly lower when an immoral action is attributed to the low- than to the non
representative employee [7(57) = -4.02,p < .01]. Purchase intentions are not
significantly different between representativeness conditions when the employee’s
action is moral [7(72) = 1.12, t(56) = -0.01, t(56) = 1.00,/) > .10], However,
purchase intentions are significantly higher in the moral than immoral action
conditions for both the head designer |7(72) = 6.31, p < .01] and assembly-line
worker [7(71) = 2.72, p < .01] but not for the former employee [7(42) = 1.01 ,p >
.10],
Regression analysis was conducted to test H4, with ratings of employee
morality and company morality entered as independent variables, and purchase
intentions entered as the dependent variable. Employee morality is a significant
predictor of purchase intentions when judgments of company morality are not
controlled ([3 = .34, p < .01). When judgments of company morality are included,
employee morality is no longer a significant predictor of purchase intentions ((3 =
.03,p > .10). However, judgments of company morality significantly predict
purchase intentions, controlling for employee morality ([3 = .51, p < .01). Thus,
judgments of company morality fully mediate the relationship between employee
morality and purchase intentions, supporting H4 (Sobel’s test for mediation: z =
5.93,p < .01).
Evidence o f alternative explanations. Collective responsibility and
diagnosticity could potentially account for the effects found here and in Study 1.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
FIGURE 3: COMPANY MORALITY JUDGMENTS AND PURCHASE
INTENTIONS IN STUDY 2
A) Company
morality
9 Y
''HSSSsI,. Il.lliil k
Immoral action Moral action
□ High
representativeness
M Low
representativeness
■ No
representativeness
B) Purchase
intention
9
8
7
6
5
4
3
2
1
Moral action Immoral action
HHigh
representativeness
91 Low
representativeness
□ No
representativeness
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
60
Each of these alternative explanations could provide a justification for customers’
judgments of company morality based on the employee’s action in Study 1 and
might be expected to covary with the manipulation of representativeness here. If
these explanations were found to account for the relationships observed, it would
suggest that moral judgments are made deliberatively in this context rather than
associatively as has been predicted. Using ANOVA and ANCOVA, diagnosticity
and collective responsibility were examined as covariates of employee
representativeness and the valence of the employee’s actions in predicting
perceptions of company morality.
Though the measure of the collective responsibility of the other employees
does vary with the manipulation of valence, it does not explain the effects of
representativeness and valence on perceptions of company morality. Ratings of
collective responsibility for the employee’s actions indicate a willingness to give the
company some credit for employees’ actions, but a hesitance to assign blame. There
was a significant main effect of valence [F(l, 184) = 3.78, p < .05], with greater
perceived responsibility for positive actions (M= 3.63) than for negative actions (M
= 3.19). There was no significant main effect of representativeness on ratings of
collective responsibility [F(2, 184) = 1.44, p > .10] and no significant interaction
[F(2,184) = 0.28,p > .10], The measure of the company’s responsibility was not a
significant covariate of the relationship between employee representativeness and
valence in predicting company morality [F(l, 183) = 0.67, p > .10].
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
Perceptions that the employee’s actions are diagnostic of the likely positive
or negative behavior of other employees also cannot explain the effects of
representativeness and valence on judgments of company morality. The perceived
diagnosticity of the employee’s actions in predicting the likelihood that other
employees of the company would behave immorally (i.e. negative diagnosticity) was
not significantly affected by representativeness [F(2, 182) = 0.70, p > .10], valence
[F(l, 182) = \ .\ l ,p > .10], or their interaction [F(2, 182) = 0.63, p > .10], Negative
diagnosticity was also not a significant covariate of the effects of representativeness
and valence on perceptions of company morality [F (l, 180) = 1.27, p > .10],
Positive diagnosticity also cannot explain the effects found for judgments of
company morality. Positive diagnosticity, which is the extent to which the
employee’s actions suggested that other employees would behave morally, was not
significantly affected by representativeness [F(2, 182) = 1.19, p > .10], or by the
interaction of representativeness and valence [F(2, 182) = 0.44, p > .10]. However,
positive diagnosticity showed a significant main effect of the valence of the
employee’s actions [F(l, 182) = 7.59,p < .01]. Positive behavior by other
employees of the company was seen as more likely when the employee’s actions in
his or her private life were immoral (M = 5.44) than when they were moral (M =
4.68). However, the direction of this difference is opposite to the observed changes
in judgments of the company’s morality, which would imply a perception that
employees are less rather than more likely to behave positively in the immoral than
moral action conditions. Positive diagnosticity was a significant covariate of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
62
representativeness and valence in predicting perceptions of company morality [F(l,
180) = 7.70, p < .01], but including positive diagnosticity as a covariate did not
change the extent to which representativeness and valence predicted perceptions of
company morality or purchase intentions. The pattern of the results remained the
same and the relationships remained significant.
Overall, these results provide evidence to rule out alternative explanations
for associative judgments of company morality. These results suggest that the
actions of an employee affect perceptions of company morality by association, rather
than based on a justification or theory customers have about what the employee’s
actions mean about the company. Judgments of the company’s morality seem to be
based on the association between the company and the employee in a manner
consistent with associative processing of moral judgments.
Discussion
The results of this study provide further evidence that associative processing
occurs in judgments of company morality. Study 2 replicated the results for the
immoral action examined in Study 1. The employee’s morality was generalized to
the company for the immoral action, and to a slightly lesser extent, for the moral
action. Employee representativeness moderated the degree of generalization of the
employee’s morality onto the company and the effect of this judgment on purchase
intentions (see Table 2 and Figure 3). Associative processing is clearly implicated
by these results, because the strength of the employee’s association with the
company influenced judgments of company morality, as predicted by H3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
The results for the manipulation of valence indicate that there is also likely to
be a negativity effect in the influence of employee’s actions on customers.
Information about an immoral action by an employee had a greater effect on
customers’ reactions than did information about a moral action. These reactions are
consistent with the abundance of evidence that has shown a greater effect of
negative than positive information on attitudes (e.g. Ahluwalia 2002; Folkes and
Kamins 1999; Martijn et al. 1992; Ofir and Simonson 2001; Sirdeshmukh, Singh,
and Sabol 2002). However, positive information about an employee’s action did
have significant effects on judgments of company morality. The company was
perceived as more moral when the head designer or assembly-line worker behaved
morally than when the former employee did. These results suggest that a
representative employee’s moral actions can have a positive effect on judgments of
the company. However, purchase intentions did not reflect this effect, showing no
significant differences based on an employee’s moral action. This result, as well as
the greater magnitude of the effect of the immoral action on judgments of company
morality, indicates that there is likely to be a negativity effect for information about
employee morality.
It is also important to note that, though the highly representative employee’s
actions had the greatest effect, an immoral action by an employee perceived as
relatively low in representativeness did also have a significant effect on customers’
reactions to the company. The assembly-line worker’s immoral action resulted in
more negative perceptions of the company’s morality and lower purchase intentions
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
than when the action was committed by a former employee. Though
representativeness increased the degree to which information about the employee’s
action affected customers’ reactions, it seems that even actions by a relatively low-
level employee can have an effect. Associative moral judgment effects may depend
on the degree of the association between the employee and the company, but any
current association seems to be sufficient to create an effect based on information
about an immoral action.
In sum, the effect of an employee’s actions in his or her private life on
customers’ reactions to the company in this context seems to be based on
association. The results of this study provide additional evidence to support this
interpretation and to rule out alternative explanations. If associative processing were
not responsible for customers’ reactions to information about an employee’s action,
the representativeness of the employee should not moderate the extent to which the
employee’s actions are generalized to the company’s morality. If customers are
simply denigrating the company in order to reduce cognitive dissonance, for
example, then their reactions should not vary with the representativeness of the
employee. Cognitive dissonance could occur because a previously positive view of
the company conflicts with a negative view about one of the company’s employees,
and this could occur regardless of the employee’s position in the company. Thus,
the results for employee representativeness rule out some possible alternative
explanations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
The variables measured as potential reasons for customers’ generalization of
the employee’s morality onto the company as a whole also provide support for the
associative processing interpretation by ruling out alternative explanations.
Diagnosticity and the company’s responsibility cannot account for the effects of
representativeness on customers’ reactions. Indeed, these measures do not vary
based on representativeness, as one would expect if they were confounded with the
manipulation. Instead, they vary, if at all, with the manipulation of valence.
These results suggest that customers form their judgments of the company’s
morality based on the mere association of the employee with the company. It seems
the resulting effects on purchase intentions derive from this association as well,
supporting H4. The evidence that participants do not seem to form their judgments
based on any explicit justification suggests that moral judgments of the company are
formed based on associative processing rather than deliberative processing.
Participants in Study 2 seemed to form their judgment based merely on the degree of
association between the employee and the company.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
66
CHAPTER IV: STUDY 3
The purpose of Study 3 was to provide further evidence of associative
processing and of the effect of the valence of the employee’s action. The relevance
of information provided about the employee was manipulated to observe the effect
of priming judgments of the employee’s responsibility. Any information that is
relevant to the cause of or motives for the employee’s action may inspire more
deliberative processing of the information. This deliberation on the cause of the
employee’s action may in turn affect the extent to which moral judgments of the
company are processed deliberatively rather than associatively. Information about a
potential cause of the employee’s action may cue consumers to process information
more deliberatively based on the notion of responsibility, mitigation, intent, and the
extent to which punishment or praise is justified (Haidt 2001; Mascarenhas 1995;
Mikula 2003).
Cueing a rule-based judgment of the employee’s behavior should serve as a
priming manipulation that focuses participants on the extent to which the
employee’s action is relevant to judgments of the company. Priming can affect
information processing and the type of information that is used to make judgments,
such as the moral judgments being investigated here (Bargh and Chartrand 2000). In
contrast, information that is irrelevant to the cause of the employee’s action should
not prime the concept of responsibility and should not cue deliberative information
processing. Thus, when irrelevant information is provided, associative processing
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67
should be more likely to occur, leading to greater generalization of employee
morality onto the company.
The secondary purpose of this study was to further explore the effect of the
valence of the employee’s action on customers’ reactions. The results of Study 2
suggest that there is a possible negativity effect, in that an immoral employee action
has a larger and more consistent effect on judgments of company morality and
purchase intentions. Study 3 examines valence in the context of information that is
relevant versus irrelevant to the cause of the employee’s action. Priming the concept
of responsibility may decrease the effect of moral as well as immoral employee
actions on judgments of company morality, because deliberative processing could
lead to the same conclusion that the employee’s action is irrelevant to judgments of
the company’s morality.
Method
Two hundred and thirty-two students participated in exchange for partial
course credit. The sample included 54% females (mean age 20.7 years).
Participants were randomly assigned to one of six conditions in a full-factorial 2 x 3
between-subjects design. The first factor manipulated the valence of the employee’s
actions (moral vs. immoral), using the same actions as in Study 2. In the immoral
action conditions, the action described was child abuse. In the moral action
conditions, the action described was saving a child from abuse.
The second factor manipulated the causal relevance of the information
provided to participants about the employee (relevant vs. irrelevant vs. no
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
68
information control). In the relevant information condition, participants were told
that the employee - the head designer of the company’s clothes, as in Study 1 - had
been abused himself as a child. This information was chosen for the relevant
information condition because it is relevant to both the moral and immoral actions,
in terms of providing a possible cause or motivation for the behavior. In the
irrelevant information conditions, participants were told that the employee was a
collector of unusual clocks and “has over 100 clocks of different shapes and sizes in
his home.” This information was chosen because it is obviously irrelevant to the
employee’s moral or immoral action. In both the relevant and irrelevant information
conditions, the employee was described as a man from a city in the Northwest, and
participants were told that the news report in which they learned of the employee’s
behavior carried a picture of the employee. Both these types of information are
likely to be included in media reports. In the no information control conditions,
participants only got information about the employee’s action, replicating the
conditions in Study 1 and Study 2 for comparison.
Procedure. The questionnaire used in this study was similar to that used in
Study 1 and Study 2. Though the questionnaire was administered by computer
rather than as a paper and pencil task, the content was much the same. The
advantage of the computer-based questionnaire was that participants received one
item or set of information at a time and they could not go ahead or back in the
survey out of turn. Participants first completed the same identification induction
used in the low identification conditions of study 1. First, they read positive quality
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
information about the company’s clothes as rated by other college students, and then
rated their initial purchase intentions and quality perceptions, as well as their
identification with the company. They then saw the manipulated information about
the employee’s action and the causal information, which varied depending on
randomly assigned condition. For example, a participant in the moral action,
irrelevant information condition was told that the head designer of the company’s
clothes had saved a child from being abused and that he was a collector of unusual
clocks.
After reading this information (participants were prevented from moving on
to the next portion of the survey until they had viewed that page for at least 60
seconds), participants proceeded to rate their purchase intentions, judgments of the
company’s morality and of the employee’s morality, on the same items used in the
first two studies. Reliability for these measures was high (a > .84; r = .87,p < .01).
Results
The results support HI and H4, replicating the findings of Study 1 and Study
2 that employee morality is generalized to judgments of company morality and
affects purchase intentions. Priming rule-based processing by providing information
relevant to the cause of the employee’s action was successful in attenuating the
effect of immoral information on judgments of company morality and purchase
intentions. In comparison, irrelevant information about an immoral action did not
affect company morality judgments or purchase intentions. The data were analyzed
using MANOVA, with the causal information manipulation and action valence as
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
independent variables, and with judgments of the employee’s morality, judgments of
company morality, and purchase intentions entered as dependent variables.
Manipulation check. Results indicate that the manipulation of the valence of
the employee’s actions was successful (see Table 3). There was a significant main
effect of valence on ratings of the employee’s morality [F(l, 226) = 1085.39, /? <
.01]. The employee was seen as extremely immoral when he committed child abuse
(M= 2.19) and as extremely moral when he saved a child from being abused (M =
7.94). There was also a marginally significant main effect of causal information on
judgments of the employee’s morality [F(2, 226) = 2.58,/? < .10]. The interaction of
the two factors was not significant [F(2, 226) = 0.57,/? > .10]. Averaged over the
moral and immoral action, participants in the irrelevant information conditions
considered the employee slightly more immoral (M= 4.76) than participants in the
relevant information (M = 5.44), though this comparison was not significant [r( 152)
= -1.36,/? > .10]. Judgments of the employee’s morality were not significantly
different in the relevant or irrelevant information conditions from the control
condition [M= 5.06; *(155) = 0.75, *(151) = -0.57,/? > .10].
Judgments o f company morality. Judgments of the company’s morality
showed that the relevant causal information did mitigate the effects of an employee’s
immoral action on consumers’ reaction to the company (see Table 3 and Figure 4A),
though the results for the moral action are less clear. Action valence and the
manipulation of causal information had significant main effects on judgments of
company morality [F(l, 226) = 188.09,/? < .01; F(2, 226) = 3.24,/? < .05]. These
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
main effects were qualified by a significant interaction between valence and causal
information [F(2, 226) = 5.50, p < .01]. In the immoral action conditions, relevant
information led to significantly more positive judgments of company morality than
in the control condition [see Table 3; t(74) = 3.14,/? < .01] and in comparison to the
irrelevant information condition [t(75) = 2.96,p < .01]. Irrelevant information did
not lead to significantly more positive judgments of the company than in the control
condition [t(75) = 0.06, p> .10]. However, in the moral action conditions,
participants who received irrelevant information had significantly more positive
judgments of the company’s morality than in the control condition [t(74) = 2.24, p <
.05], A moral action and relevant information did not significantly affect ratings of
company morality in comparison to the control condition [t(79) = 0.55, p > .10] or in
comparison to the irrelevant information condition [t(75) = -1.58, p > .10],
Thus, relevant information resulted in more positive judgments of company
morality when the employee’s action was immoral but did not affect judgments
when the action was moral. In contrast, irrelevant information led to more positive
judgments of the company when the employee’s action was moral but did not affect
judgments when the action was immoral. However, neither relevant nor irrelevant
information affected judgments of company morality to the extent that there were no
effects of the valence of the employee’s morality. The immoral employee action led
to more negative judgments of company morality than the moral action, regardless
of the manipulation of causal information [t(76) = -8.55, t{73) = -9.48, t{ll) = -5.61,
P < -01].
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
TABLE 3: DEPENDENT VARIABLE MEANS BY CONDITION IN STUDY 3
Dependent
variablea
Relevant
information
Irrelevant
information
No information
Moral Immoral Moral Immoral Moral Immoral
Employee 8.11 2.56 7.73 2.02 7.98 1.98
morality (1.18) (1.49) (1.42) (1.19) (1.14) (1.52)
Company 6.77 5.06 7.27 4.10 6.61 4.12
morality (1.42) (1.28) (1.33) (1.55) (1.24) (1.32)
Purchase 6.37 5.87 6.47 4.49 6.53 4.70
intention (1.64) (2.09) (1.70) (1.98) (1.66) (1.94)
n 41 38 36 39 40 38
a Higher scores on each o f these measures indicate more of that construct (i.e. higher numbers mean
more moral, greater purchase intention).
Purchase intentions. Purchase intentions were also affected by the priming
manipulation (see Figure 4B). There was a significant main effect of action valence
[F( 1, 226) = 35.27, p < .01] and a marginally significant main effect of the causal
information manipulation [F(2, 226) = 2.60, p < .10]. These effects were qualified
by a significant interaction of valence and causal information [F(2, 226) = 3.84, p <
.05]. Given an immoral employee action, purchase intentions were significantly
higher in the relevant information condition than in the control condition [see Table
3; til A) = 2.53,/? < .05] or in the irrelevant information condition [/(75) = 2.98,/? <
.01]. Similar to the effect on judgments of company morality, purchase intentions
were not significantly affected by the information irrelevant to the immoral action as
compared to the control condition [t(75) = 0.47,/? > .10], Purchase intentions were
not significantly affected by causal information in the moral action conditions [t(74)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73
FIGURE 4: COMPANY MORALITY JUDGMENTS AND PURCHASE
INTENTIONS IN STUDY 3
A)
Company
morality
9
Immoral action Moral action
H Relevant
information
H Irrelevant
information
□ No information
B)
Purchase
intention
i
Immoral action Moral action
B Relevant
information
IB Irrelevant
information
□ No information
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
74
= 0.14, t{19) = 0.43, t(75) = 0.28,p > .10]. However, when relevant information
was provided about the immoral employee action, purchase intentions were
improved to the point where they were not significantly different from purchase
intentions when relevant information was given for the moral action [7(77) = -1.18, p
< .10]. In comparison, purchase intentions remained significantly higher in the
moral than immoral action conditions when irrelevant or no causal information was
provided [t(76) = 4.47, t(73) = 4.64,p < .01],
Regression analysis was conducted to test H4, with ratings of employee
morality and company morality entered as independent variables, and purchase
intentions entered as the dependent variable. Employee morality is a significant
predictor of purchase intentions when judgments of company morality were not
controlled ((3 = .45, p < .01). When judgments of company morality were included,
employee morality is no longer a significant predictor of purchase intentions (P =
.07, p > .10). Judgments of company morality significantly predict purchase
intentions, controlling for employee morality (P = .54, p < .01). Thus, judgments of
company morality fully mediate the relationship between employee morality and
purchase intentions (Sobel’s test for mediation: z = 6.52,/? < .01), providing further
evidence to support H4 and replicating the effect found in Study 2.
Discussion
The results of Study 3 support HI and H4, providing further evidence that
employee morality can be generalized to company morality judgments and
replicating the effect of valence found in Study 2. These results indicate that the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
priming effect of causal information can mitigate the effect of an employee’s
immoral actions on customers’ reactions to the company. Information relevant to
the cause of the employee’s immoral action resulted in less generalization of the
moral judgment to the company, even though such information did not significantly
affect judgments of the employee’s morality (see Table 3 and Figure 4). It seems
that information relevant to the immoral action inspired more deliberative processing
of the implications of the employee’s actions for judgments of the company’s
morality. In addition, purchase intentions were significantly improved by
information relevant to the immoral action, supporting H4.
The results for moral actions were less clear. Irrelevant information
improved participants’ judgments of the company’s morality for the moral employee
action, though for the immoral action there was no significant effect. It was
predicted that relevant information might decrease generalizations from a moral
employee action, but the significant result for irrelevant information is surprising.
This might indicate that different processes affect judgments of moral versus
immoral actions. Though it seems that immoral employee behavior can result in
associative moral judgments in this context, situations involving moral employee
actions might be processed somewhat more deliberatively or be more susceptible to
bias at even low levels of customer identification.
Overall, the results of Study 3 suggest that relevant causal information can
prime deliberative processing and affect moral judgments. These results suggest that
associative judgments might be altered or prevented when consumers are given
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
information that is relevant to the cause of an employee’s immoral actions. The
effect of relevant information does suggest that it is not inevitable that information
about an employee’s immoral action will be generalized to company morality. Such
information may prime consumers to focus on the company’s responsibility for the
employee’s immoral action and to form more deliberative judgments of the
company’s morality.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77
CHAPTER V: STUDY 4
The purpose of Study 4 was to investigate an alternate mechanism of priming
deliberative judgments of company morality based on an immoral employee action.
The first three studies provide evidence that associative judgments can occur in
response to information about an employee’s morality. However, Study 1 and Study
3 demonstrate that consumers can be motivated or primed to form more deliberative
judgments of company morality. This study investigates two potential approaches
that a company might take to such a situation that have implications for the
likelihood that consumers will be primed to focus on associations versus primed to
deliberate on the reasons for their moral judgments. One approach is to try to sever
the negative association by distancing from the employee who has committed the
immoral action. For example, one way to distance from the employee is to
announce the person’s resignation. This approach may be unlikely to motivate
consumers to form a deliberative moral judgment of the company, but it is likely to
affect moral judgments because the association on which judgments are based has
been altered. Because the association has been severed, judgments of company
morality should be more positive than when the company takes no action in response
to the offense.
An alternative approach to countering the effects of an employee’s immoral
action is to communicate that the company has taken some positive, moral action, in
the hope of inspiring customers to moderate their judgments of the company. This
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
benevolent response could involve making some attempt at trying to prevent similar
situations from occurring in the future. For example, the company could make a
contribution to a charity that works to prevent child abuse and provides aid to abused
children. This type of response might be more likely to motivate consumers to form
a deliberative judgment of the company’s morality, and thus may also result in more
positive judgments of company morality than when the company takes no action in
response to the offense. Consumers might be more likely to deliberate on their
judgment of the company based on the implications of the company’s moral
response in contrast to the employee’s immoral action. In comparison to severing
the company’s association with the employee, communicating a benevolent response
might have greater relevance to the company’s moral character. The effects of
information about these two types of responses will be investigated to provide
further evidence of the effects of associative moral judgments.
Method
One hundred and eighty students participated in exchange for partial course
credit. The sample included 56% females (mean age = 20.3 years). Participants
were randomly assigned to one of six conditions in a 2 x 3 full factorial between-
subjects design. The first factor manipulated the presence or absence of an immoral
action by a company employee (immoral action vs. control). The action described
was child abuse by the president of the company. To create control conditions, half
of participants received no information about the immoral action.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
The second factor manipulated the response of the company (benevolent
response vs. distancing response vs. no response). In conditions where the immoral
action was described, these responses were described after the offense. When no
offense was described, the company’s actions alone were described. One of three
descriptions of the company’s actions was included. In the benevolent response
condition, participants were told that the company announced that they were
donating a percentage of their profits to charities that prevent child abuse by
providing parenting classes and give aid to abused children. In the distancing
condition, participants were told that the company had forced the president to resign
his position. In the no response control condition, participants were told that the
company and its PR department had issued a “no comment” statement.
The procedure of this study was otherwise the same as that used in the first
three studies. The same procedure was used to induce low levels of identification
with the company before presenting information about the employee and company’s
actions. The same measures of company and employee morality and purchase
intentions were used, and all the measures again showed high reliability (a > .80; r =
.97,p < .01).
Results
The results provide further support for the proposition that associative
processing is responsible for the effect of employee morality on judgments of
company morality. The data were analyzed using MANOVA to examine the effects
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
80
of the company’s response and the presence or absence of an immoral employee
action on the dependent variables.
Manipulation check. The results indicate that the independent variables
affected judgments of the employee’s morality. The main effect of information
about the offense indicates that the manipulation was successful [F(l, 172) = 380.71,
p < .01]. The president was perceived as significantly less moral when he was
described as having committed child abuse (M — 1.93) than when no employee
action was described (M = 5.78).
In addition, the main effect of the company’s response [F(2, 172) = 16.78,/?
< .01] and the interaction effect [F(2, 172) = 13.04,/? < .01] were significant
predictors of judgments of the president’s morality. These effects seem to be due to
differing employee morality judgments based on information about the company’s
action in the conditions when no offense was described. There were no significant
differences in judgments of employee morality in the conditions where child abuse
was described [see Table 4; £(58) = 0.97, £(58) = 0.39, £(58) = 0.56,/? > .10].
However, there were significant differences in judgments of the president’s morality
in the control conditions. When the company was described as giving to charity, the
president was judged more positively than in the neutral control condition [i.e., no
immoral action and no response; £(56) = 2.58,/? < .01]. Judgments of the president
were more negative in the distancing control condition than in the neutral control
condition [£(58) = -5.67, p < .01]. In other words, the actions of the company affect
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
perceptions of the president either positively or negatively when there is no
information about the president’s own actions.
Judgments o f company morality. The company’s response also affects
judgments of the company’s morality. The presence or absence of an offense [F( 1,
172) = 38.68,p < .01] and the company’s response [F(2, 172) = 8.33,p < .01]
significantly affect judgments of company morality. These effects are qualified by a
significant interaction [F(2, 172) = 22.71, p < .01]. Perceptions of the company’s
morality are most positive when there is no offense and the company contributes to
charity, significantly higher than in the neutral control condition [see Table 4 and
Figure 5A; £ ( 56) = 3.41 ,p < .01]. Perceptions of the company’s morality were most
negative when the company took no action in response to the president’s offense,
significantly lower than the neutral control condition [£(59) = -7.35,/? < .01].
Judgments of the company’s morality were significantly improved by a distancing
response, as compared to no response to the offense [£(59) = 4.04,/? < .01]. The
benevolent action marginally improved judgments of company morality over no
response [£(59) = 1.88,/? < .10]. Distancing improved judgments of company
morality directionally more than did the benevolent response, though the difference
is not significant [£(58) = 1.52,/? > .10].
Purchase intentions. Participants’ purchase intentions are also affected by
the company’s response and by employee morality, which affects purchase
intentions through its effect on judgments of company morality as predicted by H4.
The main effect of the immoral action is a significant predictor of purchase
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82
intentions [F(l, 172) = 21.24, p < .01]. Though the company’s response does not
have a significant main effect on purchase intentions [F(2,172) = 2.16, p > .10],
these main effect results are qualified by a significant interaction of the employee’s
action and company response [F(2, 172) = 6.06, p < .01].
TABLE 4: DEPENDENT VARIABLE MEANS BY CONDITION IN STUDY 4
Dependent
variablea
No response
Benevolent
response
Distancing response
Immoral Control Immoral Control Immoral Control
Employee 1.79 6.01 2.11 6.95 1.90 4.37
morality (0.83) (1.01) (1.62) (1.70) (1.30) (1.23)
Company 4.05 6.34 4.92 7.46 5.66 4.92
morality (1.48) (0.87) (2.08) (1.56) (1.62) (1.08)
Purchase 4.08 5.88 4.53 6.53 5.57 5.48
intention (2.06) (1.83) (2.24) (1.67) (1.63) (1.80)
n 30 31 30 29 30 30
a Higher scores on each of these measures indicate more of that construct (i.e. higher numbers mean
more moral, greater purchase intention).
Multiple comparisons show that purchase intentions are significantly lower
when the company takes no action in response to the offense than in the neutral
control condition [t(59) = -3.61 ,P< .01], indicating that information about the
offense does significantly affect purchase intentions, as in the first three studies (see
Table 4 and Figure 5B). Distancing from the president significantly improves
participants’ purchase intentions compared to taking no action in response to the
offense [7(59) = 3.12, p < .01] and compared to the benevolent response [7(58) =
2.04, p < .05]. Donating to charity does not result in significantly higher purchase
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
intentions than the company taking no action at all in response to the offense [t(59) =
0.82 ,/? >.10]. However, distancing results in purchase intentions that are not
significantly different from the neutral control condition where no offense occurred
[7(58) = -0.71,/? > .10]. When there has been no offense, contributing to charity
results in higher purchase intentions, though not significantly higher than in the
neutral control condition [7(57) = 1.42,/? >.10]. In sum, both the immoral action and
the company’s response affect purchase intentions, and the distancing response
seems to be more effective in improving customers’ reactions to the company than
the benevolent response that was examined.
Regression analysis was conducted to test H4, with ratings of employee
morality and company morality entered as independent variables, and purchase
intentions entered as the dependent variable. Employee morality is a significant
predictor of purchase intentions when judgments of company morality are not
controlled ((3 = .37,/? < .01). When judgments of company morality are included,
employee morality is no longer a significant predictor of purchase intentions ([3 =
.08,/? > .10). However, judgments of company morality significantly predict
purchase intentions, controlling for employee morality ((3 = .55, p < .01). Thus,
judgments of company morality fully mediate the relationship between employee
morality and purchase intentions, providing evidence to support H4 as in Study 2
and Study 3 (Sobel’s test for mediation: z = 5.63, p < .01).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
84
FIGURE 5: COMPANY MORALITY JUDGMENTS AND PURCHASE
INTENTIONS IN STUDY 4
A) Company
morality
5 S
« ^ d s s s s a l.illl. i „ g
Immoral action Control
S Distancing
response
E Q 3 Benevolent
response
0 No response
B)
Purchase
intention
I
Immoral action Control
■ Distancing
response
B B Benevolent
response
■ No response
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
Discussion
The results of Study 4 provide additional evidence of the effects of employee
morality on customers’ reactions to the company. The company’s response to an
immoral employee action can affect the extent to which employee morality is
generalized to the company, as evidenced by the dramatic improvement in
judgments of company morality and purchase intentions when the company
distanced itself from the employee (see Table 4 and Figure 5). This result is
consistent with the interpretation that judgments of company morality in this context
can be based on association, because severing that association eliminated
generalizations from the employee’s morality.
However, it cannot be stated with confidence that judgments of company
morality in the distancing condition were based on associative processing, because
deliberative judgments would be likely to result in the same conclusion. Distancing
from the employee provides justification for a relatively positive judgment of the
company’s morality. Though this action operates on associations, this evidence
cannot speak to the likelihood that the judgment is based on associative rather than
deliberative processing.
The benevolent action did not improve company morality judgments or
purchase intentions to the point where they were not significantly different from the
baseline, as the distancing action did. This may be because, instead of priming
participants to focus on responsibility for the employee’s action as in Study 3, the
benevolent response focused participants on the motivation of the company.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
86
Participants may have concluded that the company’s contribution to the charity was
motivated only by the desire to look good rather than out of any real sense of moral
imperative, which has been found to decrease any potential positive effects of
corporate charity (Dean 2004).
These results provide some initial guidance for managers in addressing the
effects of an employee’s immoral actions. Distancing seems to be the most effective
option for attenuating negative customer reactions. The benevolent action examined
here does not seem to improve customers’ reactions. However, information that the
company is contributing to a children’s charity after receiving information that the
company president committed child abuse may be perceived as a cynical attempt to
curry the favor of consumers rather than a genuinely moral act. Other types of
benevolent actions might be more effective, especially those that are perceived as
more genuine or that preceded the immoral employee action. In addition,
benevolent actions that do not bring further attention to the transgression should
more effectively improve judgments of company morality and purchase intentions.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
87
CHAPTER VI: CONCLUSION
Corporate social responsibility is an increasingly important issue in an ever-
smaller world of internet communications, global commerce and manufacturing, and
corporate scandals. Information about company activities around the world is
available over the internet, and over the internet, consumers can organize boycotts
and spread positive or negative word-of-mouth well beyond their circle of personal
acquaintances (Stauss 1997). In addition, corporate scandals such as WorldCom and
Enron have made consumers more skeptical about corporate morality. The publicity
surrounding the trial and incarceration of Martha Stewart and the accompanying
decline in her company’s sales have illustrated the potential consequences of a
company’s association with an employee’s immoral behavior. Though her example
might be seen as exceptional, there are other examples of employee morality that
have the potential to tamish judgments of the company they work for. For example,
when the then current CEO of Smith & Wesson was found to have committed armed
robbery in his youth, the company quickly distanced from him for fear of the
immoral association.
The research presented here suggests that mere association with an
employee’s morality can indeed change consumers’ judgments of a company’s
morality. The results of these four studies provide evidence that judgments of
company morality can be based on associative processing in the context of
information about an employee’s morality. These four experiments provide
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
converging evidence that judgments of company morality can be made associatively,
by examining judgments made in a context that can distinguish between associative
and deliberative judgments. Because generalizing onto the company an employee’s
morality in his or her private life is not easily justified, when such generalizations
are made, they can reasonably be interpreted as morality based on mere association.
In addition, manipulations that require justification or inspire deliberative processing
result in different company morality judgments than those made based simply on
information about the employee’s action.
Summary of the Evidence
The first experiment demonstrated that motivating consumers to process
information deliberatively reduced the extent to which judgments of company
morality were affected by an employee’s immoral action. In support of H2,
accountability attenuated the negative judgments of company morality that were
observed when there was no accountability for the judgments’ justification. In
addition, Study 1 examined the effect of low levels of customer-company
identification on judgments of company morality. There was no protective effect of
low levels of identification on company morality judgments, resulting in judgments
as negative as those who had no identification with the company. This result
suggests the potential for negative reactions by customers of the company who do
not identify with the company strongly enough to process information deliberatively
or to bias their interpretation of the company’s association with the employee. This,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
89
as well as the negative effect on purchase intentions found, indicates the potential for
a company’s sales to be affected by information about an employee’s morality.
The second experiment demonstrated that judgments of company morality
were moderated by the strength of the employee’s association with the company.
Consistent with H3, the effect of a more representative employee’s morality on
company morality judgments was greater than the effect of a less representative
employee’s morality. Though a less representative employee’s action did have a
significant effect on reactions to the company, reactions were affected to a greater
extent by the actions of a highly representative employee. In addition, the most
likely potential justifications for generalizing the employee’s morality onto the
company could not account for the effects found, which provides additional support
for the associative process interpretation of these results.
The effect of the valence of the employee’s action was also investigated in
Study 2. Consistent with HI, judgments of company morality were more positive
when the employee’s action was moral than when it was immoral. In addition, when
compared to baseline ratings of company morality and purchase intentions, both
moral and immoral employee actions had significant effects. Though the immoral
employee action had a more pronounced negative effect on judgments of company
morality and purchase intentions, a moral employee action did have the potential to
positively affect judgments of company morality. Though there did seem to be a
negativity effect of employee morality on reactions to the company, positive
reactions did occur for a moral action by a highly representative employee. Thus,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
90
association with both moral and immoral behavior affected customers’ reactions to
the company.
The third experiment demonstrated that priming the basis on which moral
judgments are typically justified reduced the effect of an employee’s immoral action
on company morality judgments. Information relevant to the cause of the
employee’s immoral action resulted in more positive judgments of company
morality, although it did not result in significantly more positive judgments of the
employee’s morality. Priming the concept of justification seems to have motivated
more deliberative processing of judgments of the company, whereas irrelevant
information or no information resulted in relatively negative judgments of the
company.
The fourth experiment examined the effect of the company’s actions in
response to an employee’s immoral action as a means of distinguishing deliberative
versus associative judgments. A company response that alters the company’s
association with the employee was found to be more effective at improving
customers’ reactions to the company than an action that might have focused
customers on the justification of their judgment of the company. It seems that the
deliberative priming strategy may have backfired, focusing consumers on the
authenticity of the company’s motivation to behave morally rather than inspiring
deliberative processing of the justification for the moral judgment.
In sum, these four experiments provide evidence that judgments of company
morality can be made associatively. Figure 6 summarizes the factors explored here
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
91
FIGURE 6: SUMMARY OF THE FACTORS AFFECTING JUDGMENTS OF
COMPANY MORALITY
Factors
affecting the
process of
forming
judgments
Identification
I I
None Low
i
Motivation to process information
Low High
(no accountability, (high accountability,
no priming) relevant priming)
1
High
Associative
judgment
▲ ▲
Deliberative
judgment
Factors
affecting the
outcome of
judgments
Employee
representativeness
Company response or
prior judgments of
company morality
Company
responsibility
(causality,
intention,
foreseeability)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
92
that can affect both the process and outcomes of judgments of company morality in
this context. When customers aren’t motivated to process information deliberatively
through accountability, high levels of identification, or priming, associative
processing may result in morality judgments that are not necessarily logically
justified.
Purchase intentions also varied with the factors that affected the likelihood of
associative judgments of company morality, and judgments of company morality
mediated the effect of employee morality on purchase intentions, as predicted by H4.
In addition, the manipulation that severed the employee’s association with the
company in Study 4 did result in purchase intentions that were not significantly
different from the baseline measure in the neutral control condition.
Associative judgments of company morality may also be difficult to change.
Though purchase intentions did significantly increase when participants were
explicitly motivated to form deliberative judgments, purchase intentions after being
given information about an immoral employee reaction were often less positive than
initial purchase intentions based only on information that the clothes were high
quality. For example, in Study 1, purchase intentions after being given information
about the employee’s immoral action were significantly lower than initial purchase
intentions, even when participants were motivated to process information
deliberatively [M= 4.83 pre-exposure vs. M = 4.14 post-exposure; t(50) = 3.36, p <
.01]. Thus, associative morality judgments may affect customers’ reactions to the
company even when they are motivated to process information deliberatively.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
93
Overall, these results suggest that it is important to consider the implications of
associative processing of company morality judgments.
Theoretical Implications
This research contributes to theory on corporate social responsibility by
suggesting that researchers should consider the process of forming company
morality judgments. The research presented here offers evidence that morality
judgments can be made based on associative rather than deliberative processing,
which has not been considered in the literature on corporate social responsibility.
Associative judgments of company morality have important theoretical implications.
The literature on corporate social responsibility has considered the ways in which
social responsibility information can affect customers’ responses but does not often
address how consumers process such information.
The literature on corporate social responsibility has used a number of
theoretical perspectives to predict the effects of social responsibility information.
For example, institutional theory has been used to place corporate social
responsibility in the context of cultural norms and expectations of a company as part
of a community (Flandelman and Arnold 1999). Institutional theory helps to explain
the generally positive relationship between social responsibility and consumers’
reactions to the company’s products based on consumers’ expectations. An
expectancy-disconfirmation paradigm has also been used to predict the effect of
prior evaluations of the company on consumers’ responses to moral or immoral
company behavior (e.g., Dawar and Pillutla 2000; Dean 2004).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
94
The corporate association theoretical approach to social responsibility is
consistent with the associative processing theory advanced here, though that
approach has not considered the way consumers process such associations. Brown
and Dacin (1997) describe corporate associations as the representation of all
information or beliefs that consumers have about a company and its products.
Corporate social responsibility associations have been found to affect company and
product evaluations somewhat differently than ability-related associations (Brown
and Dacin 1997; Sen and Bhattacharya 2001). The differential effect of these two
types of associations may be due to the way consumers process information about
those two types of information. Information about the company’s ability to operate
efficiently and to manufacture quality products may be more likely to be processed
deliberatively, whereas corporate social responsibility information may be more
likely to be processed associatively.
Though the research presented here examines judgments of company
morality in the context of information about an employee’s actions, the results
suggest that information about corporate social responsibility may often be
processed associatively. Research on corporate social responsibility finds that
consumers’ decisions are often related to justifications based on social responsibility
perceptions. However, such research cannot rule out the possibility that such
justifications are endorsed or constructed after the judgments are formed rather than
deliberated prior to forming a judgment of the company. Because associative and
deliberative processing can arrive at convergent judgments of company morality,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
95
evidence that such judgments can be justified does not necessarily indicate that
consumers form deliberated judgments. Associative processing of corporate social
responsibility information may be the rule rather than the exception, as has been
suggested in the literature on moral judgments in psychology.
This research also contributes to the evidence of intuitive moral judgments in
psychology and to the evidence of the generalization of an individual’s morality to
moral judgments of groups. Haidt (2001) has theorized that moral judgments are
more often intuitive than based on logical reasoning, and his conceptualization is
supported by evidence that moral judgments can be formed without rational
justification (Haidt et al. 1993; Tetlock et al. 2000). The evidence presented here
provides further evidence to support this conceptualization, and extends the
theoretical contribution into the realm of judgments of groups. Whereas prior
evidence of intuitive moral judgments has focused on judgments of individuals, this
research suggests that judgments of groups can be intuitive and based on the mere
association of the group with an individual’s morality.
Though research on collective responsibility has demonstrated that groups
can be held responsible for the actions of an individual member, that research has
focused on situations where the group was perceived to have contributed to the
individual’s actions, whether by commission or omission (Lickel et al. 2003). In
contrast, the findings here indicate that the company is not perceived to be
responsible or to have contributed to the employee’s action. Yet, when participants
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
96
were not motivated to process information deliberatively, the employee’s morality
was generalized to judgments of company morality.
This research also suggests that the strength of customer-company
identification is an important variable to consider in marketing. Identification has
been discussed in marketing as if it were an all-or-nothing construct (e.g.,
Bhattacharya and Sen 2003), whereas the literature in psychology recognizes that
there are degrees of identification (Doosje et al. 1998). Low levels of identification
with companies are likely to be very common, if purchasing from or otherwise
supporting or endorsing a positive view of a company creates a degree of
identification. The research presented here suggests that low levels of identification
do not protect companies from negative customer reactions. However, the results
found here do suggest that even consumers with a low level of identification are
motivated to see that connection with the company in a positive light. Though not
sufficiently motivated to deliberatively process information about the company
when it was associated with immoral behavior, customers with a low level of
identification did seem to be willing to give the company as much credit as possible
when it was connected to moral behavior. For example, in Study 2, participants
rated a moral employee as more representative of the company than an immoral
employee.
Customers’ reactions to an immoral employee action were consistently
negative, suggesting that there is likely to be a negativity effect for moral
information associated with a company. There was a larger and more consistent
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
97
negative effect for immoral judgments of an employee than there was a positive
effect for moral judgments. However, a moral employee action did have a
significant positive effect on customers’ judgments and purchase intentions,
especially when the employee was highly representative. This finding is consistent
with the negativity effect that has been found for moral information in impressions
of individuals (Martijn et al. 1992).
In sum, judgments of company morality based on corporate social
responsibility information may more often be made based on associative than
deliberative processing. High levels of customer-company identification, though
sought by companies, are rare, and many judgment contexts and purchase decisions
are not sufficiently involving to motivate consumers to deliberatively process
information about the company’s social responsibility. When there are available
substitutes, consumers are likely to avoid purchasing from a company that has been
associated with immoral behavior, even when there is little logical justification to do
so.
Managerial Implications
That judgments of company morality can be formed associatively suggests
that companies should consider the implications of their links to any and all
individuals and groups that might have moral implications. Companies should focus
on fostering positive moral associations in customers’ minds and avoiding negative
associations, rather than being overly concerned about the absolute logical fit of
social initiatives or with debating a point of perceived immoral behavior. For
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
98
example, a company initiative to provide funding and volunteers to a homeless
shelter would likely create a positive moral association in the minds of consumers,
even if the charity were unrelated to the purpose of the company. It may not be
worth the additional effort to find just the right charity to complement the
company’s economic purpose.
In dealing with a situation of perceived immoral behavior linked to the
company, it seems to be best to focus on severing the association, if possible, or on
shifting attention away from the situation. If it is not possible to sever the
association or to divert attention, and if the judgment of the company’s morality is
not logically justified, companies should explore ways that they might be able to
prime customers with the concept of responsibility or motivate consumers to process
information deliberatively. Though motivating deliberative processing is difficult,
because customers can’t be required to give reasons for their judgments and
decisions, companies might be able to prime more deliberative judgments based on
the framing of their marketing communications. As was observed in Study 3,
priming the concept of the cause of the employee’s immoral action led to more
positive judgments of the company, presumably because customers recognized that
the company was not the cause of the employee’s action. Though more direct
excuses and justifications may be unlikely to work, as evidenced by the failure of the
contribution to the children’s charity in Study 4, presenting information about the
company’s corporate social responsibility record in the past might also induce
participants to examine their judgments of the company’s morality.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
99
In addition, this research suggests that companies should form explicit social
relationships with their employees, in recognition of the fact that they are judged in
much the same way that other social groups are. Thus, companies should foster the
social aspect of their organization, as a support network for employees as well as a
form of enforcement of moral standards (e.g., Paolillo and Vitell 2002). As much as
possible, companies should strive to create a working environment that encourages
moral behavior (e.g., supportive colleagues, less job stress, perceived fairness) both
in and out of their jobs. In addition, when there is a closer social relationship among
employees, there is likely to be a greater perceived risk for immoral behavior, even
in private life, because social support can be withdrawn.
Companies should be aware of the potential for information about an
employee’s actions, even in his or her private life, to affect the bottom line. If
customers are aware of the employee’s action, even the actions of a relatively low-
level employee who is not seen as especially representative of the company as a
whole can have a significant positive effect, or, more likely given the potential for a
negativity effect for moral information, a significant negative effect. In preparing
for this eventuality, companies might be advised to be prepared with a plan of action
if negative publicity related to an employee’s actions does arise. Moral clauses are
common in contracts with celebrity endorsers (Hein 2003), and it might be wise to
extend the range of contracts in which these are included. In addition, a positive
corporate social responsibility record might be considered as insurance against
especially negative consumer reactions (e.g., Caza, Barker, and Cameron 2004;
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
Dawar and Pillutla 2000), as in the absence of any competing associations relevant
to the company’s morality, information about an employee’s immorality seems to
have the potential for a very negative effect on customers.
Limitations
The main limitation of this research is that associative processing cannot be
tested directly. The experiments reported here have aimed to provide converging
evidence that associative processing is likely to be responsible for the judgments
observed. Though consumers’ lack of conscious awareness of the reasoning by
which they arrive at their judgments might be an indicator of associative processing,
this cannot be measured (Sloman 2002). As observed in Study 1, when participants
were asked to report on their reasoning, they changed their judgments, indicating the
extent of reactivity in observing associative processing.
In addition, measures such as reaction time are not guaranteed to indicate the
absence of associative processing, because though associative processing is more
likely to be automatic, this does not necessarily mean that processing times would be
different. A deliberated judgment could occur relatively quickly, especially when
the judgment rule is as well learned as the judgment of responsibility for an immoral
act. Conversely, an associative judgment could occur relatively slowly, if
consumers are consciously contemplating the company’s associations. It is
important to note that though associative processing is more likely than deliberative
processing to occur automatically and unconsciously, the definition of these two
types of processing does not exactly parallel the distinction between automatic and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
101
controlled processing (Sloman 1996). An associative judgment can be consciously
contemplated, though it does not necessarily have to be. Thus, the failure to find
reaction time differences between judgments that are thought to be made
associatively versus deliberatively would not necessarily disprove the distinction.
However, future research could undertake a study of reaction times in situations
where associative versus deliberative judgments are likely to be made.
Another limitation of this research is the limited context in which it was
undertaken. Both the context of the employee’s action in his or her private life, and
the relatively sparse information available about the company in the scenario
context, limit the generalizability of these findings. The context of the employee’s
private action was chosen to provide a context in which associative and deliberative
judgments could be distinguished based on the resulting judgments of company
morality. However, this context limits the confidence with which I can predict the
likelihood that judgments of company morality are made associatively in the more
common situation where moral information is relevant to the company itself.
The lack of realism in the context was necessary to control for other possible
influences on judgments of the company morality, but it also somewhat limits the
generalizability of the results. It seems that this context provides a rigorous test of
the potential for such an effect, given such spare and ambiguous information about
the employee’s moral or immoral behavior (i.e., no details were given about the
specific situation, as in whether it was a young or relatively old child, whether the
abuse was physical or neglectful, whether it was a mild spanking judged to be abuse
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
102
by an overzealous social worker, etc.). However, the lack of information about the
company and the employee is unrealistic. Consumers may be likely to have some
prior knowledge and more well-formed opinions of the company and its products
before learning about the employee’s action.
Future Research
Future research should test the generalizability of these results by
investigating the effect of prior knowledge of the company and by observing effects
for real companies based on information about an employee’s morality. The effects
for companies that produce various types of products should also be observed. A
clothing company was used in this research because it creates the potential for a
public endorsement of the company by wearing its clothes. If a company’s products
are consumed in private, moral information may be less likely to have an effect on
consumers’ purchase intentions.
Future research should also examine the effects of moral information on
potential versus actual customers. The participants in these experiments were
potential customers because they had not actually purchased from the company,
though they had declared a willingness to do so based on positive quality
information. Future research could create more realistic purchase decision situations
in which to study the effects of moral information.
In addition, though the manipulation of low levels of identification used here
has been employed with success in psychology, there may be important differences
between purely social identification and consumer identification that have not yet
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
103
been examined. For example, the function of consumer identification may be more
multi-faceted than the purely social function of interpersonal identification, because
of considerations such as the functions served by the company’s products. Future
research should examine the potential for low levels of identification to be created
by minimal interaction with a company, and under what conditions this might occur.
Future research should also further examine the boundary conditions of the
effect of social responsibility information, in terms of the types of moral issues that
are likely to have the greatest effect, as well as the effect of prior associations
relevant to the company’s morality. Moral issues that are relevant to the company’s
economic purpose may have the potential for the greatest effect on consumers’
reactions. For example, if the company used in these studies had produced clothing
for children, there may have been an even stronger negative effect observed for the
employee actions that were relevant to child abuse. Another issue that could be
explored is the potential for taboo actions to affect judgments of company morality.
Taboo situations might present another opportunity to test the distinction between
associative and deliberative judgments of company morality, while using taboo
actions committed by the company itself. Taboo situations might results in different
judgments of company morality depending on whether consumers are motivated to
process information deliberatively versus associatively. If taboo actions provide a
context in which deliberative and associative judgments of company morality are
distinct, the generalizability of the associative judgment process observed here to
situations involving actions by the company itself could be tested.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
104
Companies that have prior moral associations, based on a positive record of
corporate social responsibility, may be given more of the benefit of the doubt if the
company became associated with an employee’s moral behavior. However, future
research should investigate the potential for prior social responsibility associations to
insulate companies from immoral associations, as well as the potential for an
immoral action that is relevant to the company’s moral positioning in the past to
have an even greater effect on customers’ reactions. Corporate social responsibility
is capable of creating a positive corporate image and brand equity with consumers
(Hoeffler and Keller 2002), but further research on the effects of a moral positioning
is needed. For example, a moral positioning may create the potential for even more
negative reactions to the company if there is an association with immoral behavior
(Caza et al. 2004).
This research suggests that judgments of company morality can be made
associatively, and that companies’ association with an employee’s morality can
affect judgments of the company. Associative judgments can lack justification, can
be difficult to change, and can affect purchase intentions. These consequences make
it important that we understand when companies might be judged immoral by
association.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
105
REFERENCES
Aaker, Jennifer L. (2001), "T Seek Pleasures and 'We' Avoid Pains: The Role of
Self-Regulatory Goals in Information Processing," Journal o f Consumer Research,
28 (Issue 1), 33-49.
Ahluwalia, Rohini (2002), "How Prevalent is the Negativity Effect in Consumer
Environments?," Journal o f Consumer Research, 29 (2), 270-79.
Ahluwalia, Rohini, Robert E. Bumkrant, and H. Rao Unnava (2000), "Consumer
Response to Negative Publicity: The Moderating Role of Commitment," Journal o f
Marketing Research, 37, 203-14.
Arnett, Dennis B., Steve D. German, and Shelby D. Hunt (2003), "The Identity
Salience Model of Relationship Marketing Success: The Case of Nonprofit
Marketing," Journal o f Marketing, 67 (2), 89-105.
Bargh, John A. (1997), "The Automaticity of Everyday Life," in The Automaticity o f
Everyday Life: Advances in Social Cognition, Jr. Robert S. Wyer, Ed. New York:
Guilford Press.
Bargh, John A. and Tanya L. Chartrand (2000), "The Mind in the Middle: A
Practical Guide to Priming and Automaticity Research," in Handbook o f Research
Methods in Social and Personality Psychology, Harry T. Reis and Charles M. Judd,
Eds. New York: US Cambridge University Press.
Bargh, John A. and Melissa J. Ferguson (2000), "Beyond Behaviorism: On the
Automaticity of Higher Mental Processes," Psychological Bulletin Special Issue:
Psychology in the 21st Century, 126 (6), 925-45.
Bergami, Massimo and Richard P. Bagozzi (2000), "Self-Categorization, Affective
Commitment, and Group Self-Esteem as Distinct Aspects of Social Identity in the
Organization," British Journal o f Social Psychology, 39 (4), 555-77.
Bettman, James R., Mary Frances Luce, and John W. Payne (1998), "Constructive
Consumer Choice Processes," Journal o f Consumer Research, 25 (3), 187-217.
Bhattacharya, C. B., Hayagreeva Rao, and Mary Ann Glynn (1995), "Understanding
the Bond of Identification: An Investigation of Its Correlates Among Art Museum
Members," Journal o f Marketing, 59 (4), 46-57.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
106
Bhattacharya, C. B. and Sankar Sen (2003), "Consumer-Company Identification: A
Framework for Understanding Consumers' Relationships with Companies," Journal
o f Marketing, 67 (April), 76-88.
Brewer, Marilynn B. and Wendi L. Gardner (1996), "Who Is This "We"? Levels of
Collective Identity and Self Representations," Journal o f Personality & Social
Psychology, 71 (1), 83-93.
Brown, Tom J. and Peter A. Dacin (1997), "The Company and the Product:
Corporate Associations and Consumer Product Responses," Journal o f Marketing,
61 (1), 68-84.
Cadinu, Maria R. and Myron Rothbart (1996), "Self-Anchoring and Differentiation
Processes in the Minimal Group Setting," Journal o f Personality and Social
Psychology, 70 (4), 661-77.
Caza, Arran, Brianna A Barker, and Kim S Cameron (2004), "Ethics and Ethos: The
Buffering and Amplifying Effects of Ethical Behavior and Virtuousness," Journal o f
Business Ethics, 52 (2), 169-79.
Chaiken, Shelly, Akiva Liberman, and Alice H. Eagly (1989), "Heuristic and
Systematic Processing Within and Beyond the Persuasion Context," in Unintended
Thought, J. S. Uleman and J. A. Bargh, Eds. New York: Guilford Press.
Dawar, Niraj and Madan M. Pillutla (2000), "Impact of Product-Harm Crises on
Brand Equity: The Moderating Role of Consumer Expectations," Journal o f
Marketing Research, 37 (2), 215-26.
De Bruin, Ellen N. M. and Paul A. M. Van Lange (2000), "What People Look for in
Others: Influences of the Perceiver and the Perceived on Information Selection,"
Personality and Social Psychology Bulletin, 26 (2), 206-19.
Dean, Dwane Hal (2004), "Consumer Pereception of Corporate Donations: Effects
of Company Reputation for Social Responsibility and Type of Donation," Journal o f
Advertising, 32 (4), 91-102.
Doosje, Bertjan, Nyla R. Branscombe, Russell Spears, and Antony S. R. Manstead
(1998), "Guilty by Association: When One's Group Has a Negative History,"
Journal o f Personality & Social Psychology, 75 (4), 872-86.
Eagly, Alice H. and Shelly Chaiken (1993), The Psychology o f Attitudes. Orlando,
FL: Harcourt Brace Jovanovich.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
107
Folkes, Valerie S. and Michael A. Kamins (1999), "Effects of Firms' Ethical and
Unethical Actions on Consumers' Attitudes," Journal o f Consumer Psychology, 8
(3), 243-59.
Folkes, Valerie S. and Vanessa M. Patrick (2003), "The Positivity Effect in
Perceptions of Services: Seen One, Seen Them All?," Journal o f Consumer
Research, 30 (1), 125-37.
Folkes, Valerie S. and Y. O. Whang (2003), "Account-Giving for a Corporate
Transgression Influences Moral Judgment: When Those Who "Spin" Condone
Harm-Doing," Journal o f Applied Psychology, 88 (1), 79-86.
Greenwald, Anthony G., Mahzarin R. Banaji, Laurie A. Rudman, Shelly D.
Famham, Brian A. Nosek, and Deborah S. Mellott (2002), "A Unified Theory of
Implicit Attitudes, Stereotypes, Self-Esteem, and Self-Concept," Psychological
Review, 109 (1), 3-25.
Hahn, Minhi, C. Whan Park, and Deborah J. Maclnnis (1992), "The Adaptive
Information Processing Hypothesis: Accounting for the V-Shaped Advertising
Response Function," Journal o f Advertising, 21 (2), 37-46.
Haidt, Jonathan (2001), "The Emotional Dog and its Rational Tail: A Social
Intuitionist Approach to Moral Judgment," Psychological Review, 108 (4), 814-34.
Haidt, Jonathan, Sylvia Helena Roller, and Maria G. Dias (1993), "Affect, Culture,
and Morality, or Is It Wrong to Eat Your Dog?," Journal o f Personality & Social
Psychology, 65 (4), 613-28.
Handelman, Jay M. and Stephen J. Arnold (1999), "The Role of Marketing Actions
with a Social Dimension: Appeals to the Institutional Environment," Journal o f
Marketing, 63 (3), 33-48.
Heider, Fritz (1958), The Psychology o f Interpersonal Relations. Oxford, England:
Wiley.
Hein, Kenneth (2003), "A Broken Field of Marketing Dreams," Brandweek, 44 (27),
6.
Herr, Paul M., Frank R. Kardes, and John Kim (1991), "Effects of Word-of-Mouth
and Product-Attribute Information on Persuasion - an Accessibility-Diagnosticity
Perspective," Journal o f Consumer Research, 17 (4), 454-62.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
108
Hoeffler, Steve and Kevin Lane Keller (2002), "Building Brand Equity through
Corporate Societal Marketing," Journal o f Public Policy & Marketing, 21 (1), 78-89.
Hogg, Michael A. and Deborah J. Terry (2000), "Social Identity and Self-
Categorization Processes in Organizational Contexts," Academy o f Management
Review, 25 (1), 121-40.
Hutchinson, J. Wesley and Joseph W. Alba (1991), "Ignoring Irrelevant Information:
Situational Determinants of Consumer Learning," Journal o f Consumer Research,
18(3), 325-45.
Jain, Shailendra Pratap and Durairaj Maheswaran (2000), "Motivated Reasoning: A
Depth-o f-Processing Perspective," Journal o f Consumer Research, 26 (4), 358.
Kahneman, Daniel (2003), "A Perspective on Judgment and Choice: Mapping
Bounded Rationality," American Psychologist, 58 (9), 697-720.
Lambert, Alan J., Lauren R. Barton, Brian Lickel, and Jennifer Wells (1998), "The
Influence of Group Variability and Processing Goals on the Ease of Making
Judgments about Social Categories," Personality and Social Psychology Bulletin, 24
(8), 807-20.
Lemer, Jennifer S. and Philip E. Tetlock (1999), "Accounting for the Effects of
Accountability," Psychological Bulletin, 125 (2), 255-75.
Lickel, Brian, David L. Hamilton, and Steven J. Sherman (2001), "Elements of a
Lay Theory of Groups: Types of Groups, Relational Styles, and the Perception of
Group Entitativity," Personality and Social Psychology Review, 5 (2), 129-40.
Lickel, Brian, Toni Schmader, and David L. Hamilton (2003), "A Case of Collective
Responsibility: Who Else Was to Blame for the Columbine High School
Shootings?," Personality and Social Psychology Bulletin, 29 (2), 194-204.
Maclnnis, Deborah J. and Bernard J. Jaworski (1989), "Information Processing from
Advertisements: Toward an Integrative Framework," Journal o f Marketing, 53 (4),
1-23.
Martijn, Carolien, Russell Spears, Joop Vanderpligt, and Esther Jakobs (1992),
"Negativity and Positivity Effects in Person Perception and Inference - Ability
Versus Morality," European Journal o f Social Psychology, 22 (5), 453-63.
Mascarenhas, Oswald A. J. (1995), "Exonerating Unethical Marketing Executive
Behaviors: A Diagnostic Framework," Journal o f Marketing, 59 (April), 43-57.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
109
Mikula, Gerold (2003), "Testing an Attribution-of-Blame Model of Judgments of
Injustice," European Journal o f Social Psychology, 33 (6), 793-811.
Morse, Stephen J. (1998), "Excusing and the New Excuse Defenses: A Legal and
Conceptual Review," in Crime and Justice: A Review o f Research, Vol 23, Vol. 23.
Ofir, Chezy and Itamar Simonson (2001), "In Search of Negative Customer
Feedback: The Effect of Expecting to Evaluate on Satisfaction Evaluations," Journal
o f Marketing Research, 38 (2), 170-82.
Paolillo, Joseph G P and Scott J Vitell (2002), "An Empirical Investigation of the
Influence of Selected Personal, Organizational and Moral Intensity Factors on
Ethical Decision Making," Journal o f Business Ethics, 35 (1), 65-74.
Petty, Richard E. and John T. Cacioppo (1986), Communication and Persuasion:
Central and Peripheral Routes to Attitude Change. New York: Springer.
Reed, II, Americus and Karl F. Aquino (2003), "Moral Identity and the Expanding
Circle of Moral Regard Toward Out-Groups," Journal o f Personality and Social
Psychology, 84 (6), 1270-86.
Sen, Sankar and C. B. Bhattacharya (2001), "Does Doing Good Always Lead to
Doing Better? Consumer Reactions to Corporate Social Responsibility," Journal o f
Marketing Research, 38 (2), 225-43.
Sen, Sankar, Zeynep Guerhan-Canli, and Vicki Morwitz (2001), "Withholding
Consumption: A Social Dilemma Perspective on Consumer Boycotts," Journal o f
Consumer Research, 28 (3), 399-417.
Sirdeshmukh, Deepak, Jagdip Singh, and Barry Sabol (2002), "Consumer-Trust,
Value, and Loyalty in Relational Exchanges," Journal o f Marketing, 66 (1), 15-37.
Skowronski, John J. and Donal E. Carlston (1989), "Negativity and Extremity Biases
in Impression-Formation - a Review of Explanations," Psychological Bulletin, 105
(1), 131-42.
— (1987), "Social Judgment and Social Memory - the Role of Cue Diagnosticity in
Negativity, Positivity, and Extremity Biases," Journal o f Personality and Social
Psychology, 52 (4), 689-99.
Sloman, Steven A. (1996), "The Empirical Case for Two Systems of Reasoning,"
Psychological Bulletin, 119 (1), 3-22.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
110
— (2002), "Two Systems of Reasoning," in Heuristics and Biases: The Psychology
o f Intuitive Judgment, Thomas Gilovich and Dale Griffin and Daniel Kahneman,
Eds. New York: Cambridge University Press.
Smith, N. Craig and Elizabeth Cooper-Martin (1997), "Ethics and Target Marketing:
The Role of Product Harm and Consumer Vulnerability," Journal o f Marketing, 61
(3), 1-20.
Stauss, Bemd (1997), "Global Word of Mouth: Service Bashing on the Internet is a
Thorny Issue," Marketing Management, 6 (3), 28-30.
Tajfel, Henri and John C. Turner (1985), "The Social Identity Theory of Intergroup
Behavior," in Psychology o f Intergroup Relations, S. Worchel and W. G. Austin,
Eds. Chicago: Nelson-Hall.
Tangney, June Price, Paula M. Niedenthal, Michelle Vowell Covert, and Deborah
Hill Barlow (1998), "Are Shame and Guilt Related to Distinct Self-Discrepancies? A
Test of Higgins's (1987) Hypotheses," Journal o f Personality & Social Psychology,
75 (1), 256-68.
Tetlock, Philip E. (2002), "Intuitive Politicians, Theologians, and Prosecutors:
Exploring the Empirical Implications of Deviant Functionalist Metaphors," in
Heuristics and Biases: The Psychology o f Intuitive Judgment, Thomas Gilovich and
Dale Griffin and Daniel Kahneman, Eds. New York: Cambridge University Press.
Tetlock, Philip E., Orie V. Kristel, S. Beth Elson, Melanie C. Green, and Jennifer S.
Lemer (2000), "The Psychology of the Unthinkable: Taboo Trade-Offs, Forbidden
Base Rates, and Heretical Counterfactuals," Journal o f Personality and Social
Psychology, 78 (5), 853-70.
Tropp, Linda R. and Stephen C. Wright (2001), "Ingroup Identification as the
Inclusion of Ingroup in the Self," Personality and Social Psychology Bulletin, 27 (5),
585-600.
Wilson, Timothy D. and Nancy Brekke (1994), "Mental Contamination and Mental
Correction: Unwanted Influences on Judgments and Evaluations," Psychological
Bulletin, 116 (1), 117-42.
Wilson, Timothy D., David B. Centerbar, and Nancy Brekke (2002), "Mental
Contamination and the Debiasing Problem," in Heuristics and Biases: The
Psychology o f Intuitive Judgment, Thomas Gilovich and Dale Griffin and Daniel
Kahneman, Eds. New York: Cambridge University Press.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I l l
APPENDIX A: ITEMS USED TO MEASURE INDEPENDENT, DEPENDENT,
AND CONTROL VARIABLES IN STUDIES 1-4
Purchase intentions
Would you purchase clothing for yourself from Company K?
1) definitely would not - definitely would
2) not at all likely - very likely
Quality perceptions
What would you expect the quality of the clothing made by Company K to be?
1) poor quality - excellent quality
2) very low quality - very high quality
Company morality judgment (Haidt 2001; Smith and Cooper-Martin 1997)
Use the following scales to describe the actions of Company K.
1) highly immoral - highly moral
2) very cruel - very compassionate
3) not at all harmful - very harmful (reverse coded)
Employee morality judgment (Haidt 2001; Smith and Cooper-Martin 1997)
Use the following scales to describe the actions of the head designer of Company K.
1) highly immoral - highly moral
2) very cruel - very compassionate
3) not at all harmful - very harmful (reverse coded)
Identification (Reed and Aquino 2003; Tropp and Wright 2001)
1) I am somewhat associated with Company K.
2) I have a sense of connection with Company K.
3) Being a consumer of Company K’s products is related to my sense of who I am.
Open-ended reasons for judgment
1) Why did you answer these questions about the company the way you did?
(asked following the judgment o f company morality)
2) Why did you answer these questions about purchasing from the company the
way you did?
Employee representativeness (Lickel, Hamilton, and Sherman 2001; Lickel,
Schmader, and Hamilton 2003)
1) The assembly-line employee of Company K has some influence on other
employees.
2) The assembly-line employee of Company K has a lot of interaction with other
employees.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
112
3) The assembly-line employee of Company K is similar to most of the other
employees of the company.
4) The assembly-line employee of Company K has a lot of power in the company.
5) The assembly-line employee of Company K is a prestigious (or impressive)
position.
Employee collective responsibility (Lickel, Schmader, and Hamilton 2003)
1) The people who work for Company K are responsible for looking out for each
other.
2) Everyone in Company K is somewhat responsible for the designer’s actions.
3) All the employees of Company K are somewhat accountable for the designer’s
actions.
4) All the employees of Company K deserve some credit for the designer’s actions.
Negative diagnosticitv
1) Employees of Company K are likely to commit moral offenses or bad behavior.
2) Customers of Company K are likely to commit moral offenses or bad behavior.
3) Many people associated with the company are likely to commit moral offenses
or bad behavior.
Positive diagnosticitv
1) Employees of Company K are likely to behave morally.
2) Customers of Company K are likely to behave morally.
3) Many people associated with the company are likely to behave morally.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
113
APPENDIX B: REASONS GIVEN AS JUSTIFICATION FOR COMPANY
MORALITY JUDGMENTS AND PURCHASE INTENTIONS IN THE HIGH
ACCOUNTABILITY CONDITIONS IN STUDY 1
Reasons for company morality judgments: Low identification
1. The actions of one designer does not suggest to me the overall morals of
Company K.
2. It is doubtful that the company’s morality or compassion is affected by this man
or that their actions are really able to affect whether he will commit abuse again.
3. I think it’s difficult for me to judge one way or the other. I think it’s very
compassionate the company kept him/her because if it were any other company,
that person would probably be fired or asked to take a leave of absence. It does
some damage to the company because people might not respect that aspect of the
company and that will influence their purchasing decision.
4. Because the company would still want this designer if he is so successful, so
minimal punishment is likely.
5. Company K should fire head designer after child abuse conviction.
6. The company was compassionate for letting the employee continue to work
there.
7. Company K doesn’t necessarily have anything to do morally with its employee.
8. I’m not sure if their employee’s ethics have anything to do with them hiring one
employee. I think it’s a good company because of the consumer reports rating.
9. Maybe it’s wrong to abuse children but if the designer really has talent to give
customers high quality clothes, I thinks it’s the company’s decision to hire him.
I won’t refuse to buy product from Company K just because they hire a designer
who abuse his child.
10. Child abuse - only might be the other side of one person, if he is convicted of
stealing or suing of copyright, that will be another scenario. By this scenario, I
don’t think it’s related to the design and the clothing.
11. The company decides to keep the designer although he was convicted. The
company may need the designer very much for their business to overlook his
conviction, more profit driven than moral.
12. Because it’s a business, I don’t care what its employees do.
13. This question is too vague for me to judge.
14. Obviously there is no perfect answer for child abuse. Firing the employee
doesn’t stop him/her from abusing, nor does it fix the kid’s problem.
15.1 do not think that Company K has done anything immoral, cruel, or harmful.
We have not been told that they knew their employee has been convicted, and
even if they did know, I do not think it would necessarily indicate otherwise.
16. Keeping employee is not a moral issue because company not involved in
personal lawsuit dealing with his personal life and again no one knows if work
will impact employee’s personal life unless facts presented to the contrary.
17. They aren’t using morals or he would have been fired, yet that doesn’t mean they
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
114
are compassionate. Maybe he makes the best clothes and they are concerned
about their business.
18. My opinion to separate the issues.
19. The company itself is not immoral, they didn’t do anything.
20. Indifferent as to the company.
21.1 cannot really judge the situation, I am removed from it. If I witnessed it
happening, it would impact how I feel. I do not agree with people abusing their
children.
22. She paid her time and this will hopefully not effect her productivity. I’m sure
she is under special watch.
23. Because it is law enforcement’s job to deal with criminals, not the firm’s.
24. Keeping employee is okay because that is his/her personal life. However, it
might affect the image of the company.
25. Slightly moral reconsideration.
26. The company seems to have made a small penalty on a very terrible offense.
27. It doesn’t have any bearing on the position of the company.
28. Gave a person a second chance, yet still took action on the matter.
Reasons for company morality judgments: No identification
1. They didn’t fire the manager who was convicted of child abuse. This is not
immoral, this is about the performance of the manager.
2. The designer is at the origin of the offense. The company shouldn’t be
sanctioned because of the misconduct of one of its employees.
3. What the head designer did was wrong and I personally wouldn’t give my
money to that company. However, it doesn’t really matter to me because I don’t
care about the clothing anyway.
4. Company K is an entity, not what an individual employee’s reputation is,
therefore they are neutral.
5. They made a bad decision in hiring the designer - that was their only problem.
6. The company isn’t at fault. But, they’re compassionate for not firing the
employee.
7. Like I said, some of the clothes I like come from Gucci, Versace, Prada, and
they’re all likely to be on the list of sinners. We just don’t know exactly what it
is.
8 . -
9. My grandma will be wearing clothes by this company. I don’t want that
reputation on the things my grandma wears.
10. They are not immoral for keeping the head designer but they are being
compassionate to the designer by letting him keep his job. As long as the head
designer is working with people it could be harmful because he has an abuse
problem.
11.1 chose to remain neutral for harmfulness and morality and an 8 for
compassionate because they gave another opportunity to the designer.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
115
12. The company itself is not acting (based on the info provided) different from
another company, the one designer’s action doesn’t reflect the company.
13. The company is not at fault.
14. The company is not responsible for the employee’s actions outside of work.
15. It was immoral to rehire the designer but compassionate towards him/her
(although not towards the victim) and was probably harmful towards the
reputation of the company.
16. Because one person is not representative of Company K, which I am being asked
about.
17. It would have been immoral and cruel to fire the employee - keeping them is not
harmful.
18.1 don’t think the company itself is being either moral, compassionate or harmful
by letting the guy still work there if it was his first offense and he could by law.
19. First conviction, on probation now, make sure not again.
20. Although it would turn me off, child molesters need to work too and its not
really the company’s fault.
21. The head designer is looked to as a leader, if the employees follow his or her
lead, who knows what other moral problems there are in the company.
22. Because companies should not support child abusers.
23. If the head designer is conducting immoral behavior in his personal life, this
permeates through the reputation/image of the rest of the company and brings
them down as well.
Reasons for purchase intentions: Low identification
1. The designer’s actions does not effect my decision on whether to buy or not, but
the designer’s actions do suggest a bad influence in the company.
2. His private life has little effect on the clothes.
3. His/her convictions do not pertain to clothing - the business and clothes will still
probably be good - but I might not purchase it because I might not want to be
associated with him/her in any way.
4. Because if I like the clothing, I will buy it, unless I have personal resentment,
which I don’t think I would in this case.
5. Child abuse has no effect on quality.
6. The conviction doesn’t affect the style, design, or durability of the clothes so I
would still buy them.
7. I would still be relatively likely because the personal actions of the designer
don’t change the style of his clothing, therefore it still has the same appeal.
8. I would still possibly purchase clothes. It’s all based on the review, not the
employee.
9. I purchase clothes mainly because I like its design or quality. I won’t consider
too much about the moral corrupt of the head designer.
10. We treasure the idea of head designer, not really the personality. As long as the
clothings are trendy, that will serve the purpose.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
116
11. Since I don’t know the brand, and knowing this information affects my buying
decision minimally.
12.1 don’t always buy what’s popular, but my decisions stay exactly the same. I’ve
been told the consumer reports said the quality was rated well.
13.1 would probably purchase the clothes from the company because it is considered
to be stylish and durable. I don’t think the quality of the clothes has anything to
do with the designer’s legal problems.
14. I’m not concerned about the behavior of one employee. I’m more concerned
about the corporate behavior.
15.1 don’t think the fact that a single employee does something wrong has anything
to do with the quality of the product, and not buying wouldn’t influence the fact
that the person’s a child abuser.
16. Because employee’s personal life shouldn’t impact their work, if company
decided to keep them employed they should be good employee.
17. Knowing the fact about the designer, I would be slightly less likely to buy his
products, but it wouldn’t be an extremely deterring factor. I still want nice
clothes.
18. Outside life unrelated to the ability of designing clothes or the quality of it.
19. Because the quality wouldn’t change, but the reputation of the company will go
down because the head designer is a child abuser.
20. Because the criminal history of the employee should not matter.
21. It has nothing to do with the quality of the clothing. The whole is not the same
as one of its parts - in terms of morality it is irrelevant.
2 2 .1 would still be very interested in purchasing this clothing. The way one
employee acted has no effect on the quality of the clothing.
23. Because one member isn’t representative of the whole.
24. Because what I buy and recommend is a reflection of me. I would be less likely
to tell others that I buy from a child abuse company, but continue to wear the
clothing.
25. Slightly moral reconsideration.
26. Same as before - no change. But then again, I’ve never cared for styles in the
first place.
27. The money doesn’t support this kind of behavior.
28. Stylish clothes that will last.
Reasons for purchase intentions: No identification
1. Because it’s the quality that matters, not the personality of the employee. (It
could be high quality, but suitable for senior citizens.)
2. Because as a lot of old people like the design, it might be a design made for old
people. I am still young for that. But I don’t think it has anything to do with the
quality of the clothing.
3. Personally, when it comes to actually giving my money to a certain company
with such a problem, I wouldn’t want to, though the quality of the clothing has
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
117
NOTHING to do with the personal life of the head designer.
4. Firstly, the clothing has been reviewed to be of good quality by senior citizens so
clearly it is their taste. Secondly, I wouldn’t purchase the clothes in either case
because it is not for my age if a senior citizen is wearing that clothing.
5. My feelings on the quality would not change. The company’s image would be
affected by the head designer’s actions, therefore I would not be inclined to
support that image.
6. It wouldn’t affect the clothing quality, but I would be less likely to support them
financially. Plus, this isn’t a company that makes clothes for my demographic.
7. If ages 65+ ranked it to be top, likely the style is outdated for me.
8. I feel Co. K is tailored to adults.
9. I was unlikely to in the beginning and this incident only helps to solidify my
negative perception.
10.1 think that I probably wouldn’t purchase clothes from here considering senior
citizens like the clothes and the head designer is a child abuser.
11. My answers remain the same as before, I did not consider the fact that the
designer was convicted of child abuse as much is not revealed to me, but I
consider the ratings were given by senior citizens.
12. Since it seems as though these clothes are more appealing to senior citizens I
don’t think I’d be that interested in them, and after hearing of the abuse scandal I
definitely wouldn’t do business with them.
13.1 still don’t see why or why not to purchase.
14. Not in the 65+ age group.
15.1 probably wouldn’t anyway, but hearing about this offense would make it for
sure that I wouldn’t.
16. Because the clothing is for old people.
17.1 wouldn’t like it probably because senior citizens liked the style.
18. Because the clothes would most likely not appeal to me and I wouldn’t want to
necessarily support the convict financially.
19. It is likely the product itself has nothing to do with the off duty behavior of the
head designer.
2 0 .1 wouldn’t feel good about myself if I knew the clothes I wore were made by a
child molester.
21. There are plenty of quality clothing companies that I can choose. I prefer not to
choose one with negative associations.
22. Because I do not support child abusers.
2 3 .1 strongly disagree with child abuse and unless the clothes were absolutely
amazing, I would not purchase!
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
"How will I feel about it?": The asymmetric impact of affective misforecasting on consumer satisfaction
PDF
Perceived cohesiveness of service firms: Consequences for consumers' perceptions of variability in service quality and resolution of inconsistent service quality information
PDF
Losers, winners and hype: Consumer choice of stocks under uncertainty
PDF
Consumer choice under budget constraint: Why consumers overspend
PDF
Adolescent exposure to alcohol advertising: A prospective extension of Strickland's model
PDF
In need of a favorable conclusion: The role of goal -motivated reasoning in consumer judgments and evaluations
PDF
An investigation of repeat visit behavior on the Internet: Models and applications
PDF
Customer satisfaction, customer bargaining power, and financial performance
PDF
An empirical investigation of everyday low price (EDLP) strategy in electronic markets
PDF
Essays on technology innovation and new product development strategy
PDF
Essays on new product preannouncements
PDF
Essays on technological evolution and financial returns to innovation
PDF
Structural models of marketing to multiple agents
PDF
Information diffusion and the boundary of market efficiency
PDF
Information systems capabilities and indirect relational value: Sustaining networked organizations
PDF
Prospective and retrospective examinations of factors related to intention to leave and turnover among public child welfare workers
PDF
Beyond skill development: The effects of training and development on the attitudes and retention of employees
PDF
An empirical examination of customer perceptions of service quality utilizing the extended service quality model under the condition of multiple subunit service providers
PDF
Technological product failure: The consumer coping process
PDF
Loan portfolio information and accounting in banks
Asset Metadata
Creator
Johnson, Allison Rachelle
(author)
Core Title
Immoral by association: Customers' reactions to information about an employee's morality
School
Graduate School
Degree
Doctor of Philosophy
Degree Program
Business Administration
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
business administration, marketing,OAI-PMH Harvest,psychology, social
Language
English
Contributor
Digitized by ProQuest
(provenance)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-453224
Unique identifier
UC11336533
Identifier
3196825.pdf (filename),usctheses-c16-453224 (legacy record id)
Legacy Identifier
3196825.pdf
Dmrecord
453224
Document Type
Dissertation
Rights
Johnson, Allison Rachelle
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
business administration, marketing
psychology, social