Individual differences in trust development: Assessing the effects of trustor attributes on trust building, stability, and dissolution

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

INDIVIDUAL DIFFERENCES IN TRUST DEVELOPMENT:
ASSESSING THE EFFECTS OF TRUSTOR ATTRIBUTES ON TRUST BUILDING,
STABILITY, AND DISSOLUTION

by

Hyung-Woo Lee

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(POLICY, PLANNING, AND DEVELOPMENT)

May 2011

Copyright 2011 Hyung-Woo Lee
ii

DEDICATION

To Jesus Christ, who crucified himself for me, and to whom I owed everlasting love.
iii

ACKNOWLEDGEMENTS

Sometimes, I was not very confident that I could get this work done. Some of the
memories of the last few years are something that I would never want to recall. Getting
through those painful moments in life, however, I have learned how deeply Jesus Christ
loves me. And that is far more precious for me than what the degree will afford me.
It would not have been possible for me to finish this study without supports of
people around me. I would like to thank Professor Peter Robertson. He has always
encouraged and challenged me throughout the doctoral program. It was my privilege to
work with him, and I want those wonderful experiences with him to continue in the future
as well. I also want to thank Professor Robert Myrtle. I have been inspired by his
supportive and encouraging character and acute professional insight. I am also grateful
for the helps of Professor Kathleen Wilber. Her advice greatly improved the quality of
this study.
I would like to give thanks to my family as well. I thank my parents, Du-Hwan
Lee and Nam-Young Kim, for their love, supports, and sacrifices for me. I would also
like to say thank you to my wife, Iris, and my lovely daughter, Michelle. You were the
very reason that I had to survive all this.

iv

TABLE OF CONTENTS

Dedication ii

Acknowledgements iii

List of Tables vi

List of Figures viii

Abstract ix

Chapter One: Introduction 1
Background of the Study 1
Significance of the Study 6
Overview of the Remaining Chapter 10

Chapter Two: Conceptual Foundations 12
Trust 12
Trustworthiness 20
Propensity to Trust and Generalized Expectancy 21
Cognitive Closure 26
Phases of Trust Development 30
Chapter Summary 32
v

Chapter Three: Hypotheses 34
The Building Phase 34
The Stability Phase 37
The Dissolution Phase 42

Chapter Four: Methods 49
Research Site, Sample, and Survey Procedure 49
Measures 52
Research Design and Analysis 62

Chapter Five: Results 69
Overview of the Data 70
Model Testing 77
Chapter Summary 90

Chapter Six: Discussion 91

References 98

Appendix: Questionnaires 115

vi

LIST OF TABLES

Table 1. Concept of Propensity to Trust in This Study 33
Table 2. Comparison between Generalized Expectancy and Cognitive Closure 33
Table 3. Items in Scale 60
Table 4. Reliability of the Measures 70
Table 5. Descriptive Statistics and Correlation Matrix for Full Sample 71
Table 6. Sample Profile Information 71
Table 7. Descriptive Statistics for Sub-samples 72
Table 8. Frequency Table for Dichotomized Positive and Negative Experiences 74
Table 9. T-tests for Comparing the Probability of Experiences 75
Table 10. Results of ANOVA on Trust Change 76
Table 11. Descriptive Statistics and Correlation Matrices for Sub-samples 78
Table 12. Results of HLM on the Level of Initial Trust 80
Table 13. Descriptive Statistics and Correlation Matrices for Sub-samples 82
Table 14. Results of Logit HLM on Probability of Negative Experiences 83
vii

Table 15. Descriptive Statistics and Correlation Matrices for Sub-samples 84
Table 16. Results of HLM on Trust 2 for stranger-relationships 86
Table 17. Results of HLM on Trust 2 for friend-relationships 87
Table 18. Summary of Hypotheses Testing 90
viii

TABLE OF FIGURES

Figure 1. Proposed Causal Model in the Building Phase 37
Figure 2. Proposed Causal Model in the Stability Phase 42
Figure 3. Proposed Causal Model in the Dissolution Phase 48
Figure 4. Ego-network Data Structure in the Multilevel Analysis 64
Figure 5. Moderator Effect of the Need for Structure 89

ix

ABSTRACT

Since fostering trust is critical for the well-beings of today ’s organizations,
managers need to understand how trust is formed and develops. Considering that
interpersonal trust is a result of the confluence among three different factors, namely,
attributes of trustor, trustee, and the situation; understanding how trustor attributes
interact with trustee attributes and with situation attributes to determine the level of trust
in interpersonal relationships is critical. However, since scholars have assumed that
propensity to trust should be cross-situationally consistent, few studies have investigated
the effects of diverse dispositional qualities of the trustor on trust in different situations.
This study aims to fill that gap in the literature. Focusing on two dispositional factors,
generalized expectancy and need for cognitive closure, this study investigates how these
two factors come into play in three different trust development phases: building, stability,
and dissolution. The result of this study provides some practical implication for managers
as to how to help newcomers in organizations build trust with their colleagues and how to
maintain trusting relationships over time. The implication for studies on trust repair is
also discussed.

CHAPTER ONE: INTRODUCTION

Background of Study
Trust has been hailed as an important facilitator of virtually all types of
cooperative interaction (Gambetta, 1988). It has been a popular topic among scholars in
various social science disciplines. For political scientists and sociologists, trust is often
described as a quality of culture embedded in a society. Fukuyama (1995) argued that
trust in a society is linked to economic performance, suggesting that particular cultural
attributes in some societies are more conducive to fostering trust and ‗spontaneous
sociability‘ than others, which partially explains why some countries achieve greater
economic progress than others. Specifically, he argued that the sectarian Protestantism in
the U.S., the German apprenticeship system, and a network of voluntary organizations in
Japan contribute to high trusting culture, and in turn, economic success in these countries,
while intense familism in such countries as China and France supplants it.
Some scholars, such as Putnam (1993), conceived trust as a characteristic of the
group in which a norm of reciprocity is shared, and as a kind of social capital that
benefits people in the group just as other tangible resources (e.g., physical or financial
capital) do. Many valuable resources, such as information, are often embedded in a
network of people, and trust enhances the network member‘s accessibility to these
resources. It also facilitates coordination and cooperation for mutual benefit. These
scholars tend to focus on the linkage between trust and political outcomes. Putnam (1993)
demonstrated how two Italian communities with different levels of trust among those
1

voluntarily interacting on common issues also differed in terms of their collective ability
to avoid conflict, foster cooperation, and achieve common goals. Examining data from
the General Social Survey across more than twenty years, Brehm and Rahn (1997) found
that trust and civic engagement exist in a tight reciprocal relationship, each supporting the
other. Gittell and Vidal (1998) argue that building social capital is a useful component of
community development efforts.
Among organization scholars, trust is generally viewed as an interpersonal
phenomenon, that is, one person‘s positive expectation about another. They mostly seek
to investigate the linkage between trust and the outcomes at the individual level. Trust is
found to influence organizational citizenship behavior (Deluga, 1995; McAllister, 1995;
Robinson, 1996), the level of job satisfaction (Driscoll, 1978; Muchinsky, 1977; Rich,
1997), organizational commitment (Flaherty & Papas, 2000; Whitener, 2001), turnover
intention (Brashear, Manolis, & Brooks, 2005), and job performance (Oldham, 1975;
Rich, 1997). Although there are a few organizational studies that focuses on outcomes at
the collective level (Davis, Schoorman, Mayer, & Tan, 2000; Simons & Roberson, 2003),
organizational trust is clearly distinguished from trust in political science or sociology
literature since it was conceptualized as individuals‘ perception of trustworthiness in
interpersonal relationships. In organizational studies focusing on collective outcomes,
trust is measured by aggregating individual levels of trust, rather than conceived as
certain cultural aspect in a society.
Trust is particularly important when two or more organizations work together for
mutual benefit and/or common goal. In this case, the involvement of participants and
2

their mutual cooperation are often voluntary, not obligatory. Thus, trust is an essential
mechanism for coordinating their interactions and resolving potential conflicts. Astley
and Fombrun (1983) demonstrated that it is easier to form collective strategies when
there is a robust trust between organizations, and Lundin (2007) found that trust
positively affects cooperation when organizations have similar objectives. Trust is a
valuable facilitator of collective learning (Dodgson, 1993) that enables collaborative
inter-organizational networks to implement creative and innovative solutions through
which to pursue their objectives. Trust facilitates shared learning by encouraging
participants to disclose and exchange fine-grained information transparently (Hamel,
1991; Muthusamy & White, 2005; Sako, 1992; Uzzi, 1996).
The number of management studies that directly address the topic of
organizational trust has increased exponentially over the decade, whether public or
business management. In the journals published by the Academy of Management alone,
two trust studies appeared in the 1970s, two in the 1980s, 24 in the 1990s, and 31 in the
2000s
1
. In Public Administration Review, the numbers of trust studies were one in the
1970s, zero in the 1980s, three in the 1990s, and seven in the 2000s.
2
The recent
popularity of trust as a research topic appears fueled by changes in the way scholars view
organizations in the post-bureaucratic era, and by changes of environment in which

1
The numbers here are based on the search for the articles that contain ‗trust‘ in the title and are published
in Academy of Management Review, Academy of Management Journal, and Academy of Management
Perspective.

2
The numbers here are based on the search for the articles that contain ‗trust‘ in the title and are published
in PAR.

3

contemporary organizations are operating. First, trust is the key to the success of today‘s
devolutionary and participative organizational changes. Since it is generally believed that
trust leads to voluntary cooperation among people, fostering trust is more appropriate and
effective for managing people than the use of bureaucratic rewards and sanctions
(Bradach & Eccles, 1989; Zaheer & Venkatraman, 1995). In addition, the lack of trust
within an organization is a cause of excessive micromanagement that undermines
efficiency, flexibility, and adaptability (Behn, 1995). Second, as scholars have recognized
that organizations are interconnected with a diverse array of other organizations whose
cooperation is critical for their survival (Oliver, 1990; Schermerhorn, Jr., 1975), the
importance of inter-organizational trust has increasingly been highlighted (Agranoff &
McGuire, 2001; Newell & Swan, 2000; Ring & Van de Ven, 1994; Zaheer, McEvily, &
Perrone, 1998)
Especially for the field of public management, the importance of trust research is
emphasized by the growing need for collaboration with nongovernmental entities. Many
of today‘s government functions have been performed in collaboration with business and
nonprofit organizations (Agranoff & McGuire, 2003; La Porte, 1996; O‘Toole, 1997).
Consequently, how to ensure the effectiveness of the inter-organizational collaboration
has been one of the biggest issues in the field of public policy and management. Many
public administration studies on this topic are focused on how to design interaction
structures among participants to ensure successful collaboration. To briefly review these
studies, one group of researchers has investigated how government agencies can
effectively manage contracts with public service providers. For instance, Shetterly (2000)
4

discussed and examined what design elements of contract influence service performance.
Brown and Potoski (2003a) listed the techniques government agencies can adopt to
monitor service providers in situations where the risk of contract failure is elevated.
Brown and Potoski (2003b) also maintained that government needs to invest in enhancing
the capacity to assess feasibility of the bidding, to implement the contract, and to evaluate
the performance of vendors. Another stream of researches approaches the issue from a
broader perspective of networks. Unlike in the contracting-out literature, this line of
researches assumed that the relationship between actors can be best described as
cooperation and mutual dependence, rather than principal-agent relationship. The typical
research focus for these studies is how to design the optimal governance structure
promoting successful collaboration under the given conditions of interdependence.
Examples of the conditions that need to be considered are the nature of resource sharing
(Huang & Provan, 2006), the sectoral characteristics of participant organizations
(Herranz, 2008), density of trust in the network, the number of participants, goal
consensus, and the need for network-level competencies (Provan & Kenis, 2008).
Commonly overlooked in both lines of literature are the interpersonal dynamics
(e.g., trust) among key decision makers. Of course, institutional design may help guide
the interactions among individuals, and lead to fruitful collaboration (Ring & Van de Ven,
1994). However, the importance of micro level factors cannot be discounted since it is
human decision making in daily context that constitutes inter-organizational collaboration.
Indeed, many public management scholars have agreed that interpersonal process matters.
Sitkin & Roth (1993) argued that, even when two organizations are bound by legal
5

contract (e.g., contracting out relationships), legalistic constraints can neither preclude
nor attenuate the importance of trusting relationships. Chen (2008) also emphasized the
mediating role of the quality of collaboration process (e.g., joint decision making, trust
building, etc.) between the preconditions of partnership (e.g., resource acquisition,
organizational legitimacy, etc.) and the collective outcome. More specifically, Ospina and
Yaroni (2003) found that, even two parties representing institutionally adversarial
positions in a zero-sum situation (i.e., labor vs. management) were able to find a shared
interest and develop a cooperative relationship as interpersonal trust increased through
daily interactions. Despite the importance of interpersonal trust, studies on how trust as
an interpersonal phenomenon forms and develops are in short supply. This study is aimed
to explore the factors affecting interpersonal trust formation, which will help to fill that
gap in the literature.

Significance of the Study
Given the significance of trust for managing today‘s organizations, knowledge of
antecedents to the emergence of trust, and the mechanism by which those antecedents
give rise to interpersonal trust, is essential for managers to be effective. In response to
this compelling need, management scholarship has explored the antecedents to trust.
Compatible with research in interpersonal perception more generally, trust is determined
by three factors representing different components of a dyadic relationship: attributes of
the perceiver, the target, and the relationship (Kenny, 1994). This three-factor scheme is
useful in summarizing how this specific topic has been explored by management
6

researchers. The early social psychologists studying trust tended to view trust as a trait
of the perceiver (trustor), which is more or less stable across situations and relationships.
Rotter (1971) was interested in how one‘s propensity to trust forms. He stated that an
individual‘s tendency to trust is built through his/her past experiences in similar situations.
As positive or negative experiences accumulate, people learn what they can expect from
general others (i.e. human beings). Thus, his notion of generalized expectancy is related
to beliefs about human nature (Wrightsman, 1991). According to Rotter, trust in a given
relationship is a result of transferring this generalized expectancy to the current trustee.
Other characteristics of trustors, such as agreeableness (Mooradian, Renzl, & Maztler,
2006) and cognitive style (Sorrentino, Holmes, Hanna, & Sharp, 1995), are also found to
be related to trusting tendency.
On the other hand, most organizational researchers have viewed attributes of the
target (trustee) as the key determinants of trust. Reflecting the premise that people tend to
trust someone who has particular characteristics, researchers have endeavored to identify
the characteristics that give rise to perceptions of trustworthiness. A number of
characteristics have been listed as factors of trustworthiness, such as, good intentions
toward the trustor, competence, benevolence, honesty, reliability and integrity (Butler &
Cantrell, 1984; Cook & Wall, 1980; Larzelere & Huston, 1980). Based on an extensive
review of studies on trustworthiness, Mayer, Davis, & Schoorman (1995) maintained that
three factors, ability, integrity, and benevolence, are the most frequently identified in the
literature and explain a major portion of trustworthiness. Numerous other management
researches have focused on the situational antecedents that promote the emergence of
7

trust. Based on the premise that trust develops within relationships characterized by
intimacy and frequent interactions (Putnam, 1993), organizational scholars have
demonstrated particular interest in specifying the contextual factors of trust, such as,
familiarity (Gulati, 1995), communication media (Jarvenpaa & Leidner, 1999),
contractual agreement (Das & Teng, 1998), and presence of clear rules specifying actors‘
roles (Ring & Van de Ven, 1994).
Despite all this research exploring a variety of antecedents to trust, the current
trust literature is still insufficient to provide an adequate knowledge base for managers,
for two reasons. First, the significance of trustors‘ individual differences is
underestimated by management scholars. Although considerable attention has been given
to contextual factors and trustworthiness, we know relatively less about what constitutes
people‘s propensity to trust and how it influences the level of trust in relationships. The
propensity to trust has not been fully explored since the work of Rotter (1967). In the
field of organization and management, most studies tend to only briefly mention Rotter‘s
idea of generalized expectancy, with no further exploration of the diverse factors shaping
trusting propensity other than a trustor‘s past experiences. Little empirical work has
investigated the impact of trustor‘s attributes on trust.
This trend probably reflects the common belief that propensity to trust has little
practical implication for managing organizations. Scholars tend to believe that the
knowledge of propensity to trust is not useful as a basis for managerial action (Kramer,
1999). In contrast, the managerial utility of trustworthiness and situational guarantees has
been appreciated. Scholars tend to believe that we can readily display our trustworthiness
8

to recover trust in relationship (Tomlinson & Mayer, 2009), that we can use our
knowledge of trustworthiness to screen out untrustworthy employees and select only
trustworthy ones (Kipnis, 1996) or to train managers to enhance managerial trustworthy
behavior (Whitener, Brodt, Korsgaard, & Werner, 1998), that we can improve overall
level of trust within an organization if we specify the rules as to what roles each actor
plays (Ring & Van de Ven, 1994), and that we can set up a contractual agreement to build
trust (Das & Teng, 1998).
To the contrary, we generally assume that there is nothing a manager can do to
control one‘s propensity to trust because it is deeply rooted in one‘s past experiences
(Rotter, 1971) which sometimes date back to one‘s relationship with parents in
infanthood (Erikson, 1963). However, Rotter‘s (1967) theory of generalized expectancy
implies that managers can influence propensity to trust. He viewed trust as a conscious
rational decision, rather than an impulsive manifestation of deep-seated psychological
needs. He suggested that the reason why someone demonstrates behavioral consistency
(i.e., personality) is because the positive consequences one has experienced in a particular
situation motivate him/her to behave in the same way as s/he had previously (Rotter,
1954). Since generalized expectancy is something that builds through a diversity of
relationships with numerous others, it is neither innate nor determined very early in
childhood. Therefore, managers can foster trust among their employees by reinforcing the
positive outcomes of trust, which may ultimately change generalized expectancy at some
point in time.
Second, existing research on antecedents to trust is too fragmented and simplistic
9

to provide managers with a comprehensive view of how trust forms and changes
(Lewicki & Bunker, 1995). Since the level of trust in a given relationship is determined
by interactions among the three factors (attributes of trustor, trustee, and context in which
they interact), we need to build and test a three-part model of trust that takes into account
all three factors at the same time (Hardin, 2004; Kramer, 1999). Yet very little research
has modeled and tested the interactions between two or more factors.
This study addresses these gaps (i.e., inattention to propensity to trust and to the
interaction among the three types of antecedents to trust) in the literature. To this end, I
examine the impacts of attributes of trustors on the level of trust in social context. While
the focus of this study lies primarily on the effect of trustor attributes on trust, I examine
how these impacts interplay with other attributes of trustees (e.g., types of
trustworthiness) and situations (e.g., different phases in trust development).

Overview of the Remaining Chapters
In the following chapter, I will establish the conceptual foundations of this study.
First, I will review how trust has been conceptualized in a variety of disciplines, and
clarify the concept of trust that this study intends to investigate. Second, the concept of
propensity to trust will also be reviewed and redefined according to the purpose of this
study. Third, two theories which account for why individuals differ in their propensity to
trust, theory of generalized expectancy and theory of cognitive closure, will be introduced
and discussed. Fourth, different phases of trust development will be discussed. In the
hypotheses chapter, three causal models underlying this study will be proposed,
10

suggesting a number of hypotheses to test. Each of the models tests the effects of
dispositional factors, drawn from the two theories, on trust in three distinctive phases of
trust development in relationships. Testing all these models will provide a bigger picture
of how trustors with different dispositions differ in their patterns of trust development. In
the methods chapter, the data collection strategy, data structure, research design, and
analytic strategy will be explained. In the last two chapters, the results of the analyses
will be presented and the implications of the study will be discussed.
11

CHAPTER TWO: CONCEPTUAL FOUNDATIONS

The purpose of this chapter is to establish a conceptual foundation for the models
and hypotheses this study aims to investigate. Toward this end, I will first clarify the
concept of trust examined in this study. Second, I will conceptualize propensity to trust as
the effects of dispositional traits on trust in diverse situations, rather than a cross-
situationally consistent trusting tendency. Then, I will identify the dispositional traits that
are likely to influence trust in different phases of the trust development process, drawing
from two theories: the theory of generalized expectancy and the theory of cognitive
closure. The hypothesis chapter that follows will provide a rationale for why specific
dispositional traits are likely to affect trust in three specific phases of trust development.

Trust
The definition of trust has been a controversial issue (Hosmer, 1995; Mayer et al.,
1995; Rousseau, Sitkin, Burt, & Camerer, 1998; Sitkin & Roth, 1993; Weick, 2008). As
briefly reviewed in the Introduction chapter, numerous scholars have provided a diverse
array of trust definitions. On one hand, this diversity suggests the universal importance of
trust across diverse disciplines as well as the wide applicability of trust studies. On the
other hand, however, the inconsistency regarding the concept of trust may serve as an
obstacle to theory development, since the findings of one study can be irrelevant to, or
have limited implications for, another study when the two are based on different
12

conceptions of trust. In this vein, it is necessary to clarify the concept of trust examined in
this study. As Longino (1990) maintained,
Background assumptions can also lead us to highlight certain aspects of a
phenomenon over others, thus determining the way it is described and the kind of
data it provides … The background assumptions that determine evidential
reasoning are those that emerge from the transformative interrogation by the
scientific community. This means that community values may well remain
embedded in scientific reasoning and research programs.

This quote explains the dissonance in trust conceptions very well. As Rousseau et al.
(1998) suggested, scholars from multiple disciplines basically talk about the same
phenomena, but they describe it in different ways because they have different
assumptions and orientations. Thus, in order to clarify the concept of trust in this study,
the epistemological assumptions and theoretical orientations underlying these varying
conceptualizations will be discussed, an understanding of which will provide a rationale
for the definition of trust used in this study.
According to Coleman (1990), the primary interest of sociology is to explain
behaviors of collectives, not individuals. In essence, while psychologists are interested in
exploring why individuals behave differently, sociologists seek to answer why those who
are embedded in a same group or society behave similarly, or why those from different
groups behave differently. Sociologists seek to explain the reason for this by looking at
the characteristics of the social system (e.g., institutions). The similarity or difference in
behaviors, as a function of the overarching social system, is at the center of sociology
theories.
Hence, when making a conceptual definition, sociologists tend to focus on the
function that sociologists are interested in (i.e., what it does), rather than content (i.e.,
13

what it is). This is manifested in the definition of social capital. For example, Coleman
(1990) stated,
[S]ocial capital is defined as its function … [It is comprised of] a variety of
different entities having two characteristics in common: [The ingredients of social
capital] all consist of some aspect of social structure, and they facilitate certain
actions of individuals who are within the structure (p.302).

Fukuyama (1995) defined it as ―the existence of a certain set of informal values or norms
shared among members of a group that permit cooperation among them‖ (p.378). The
causal relationship between the characteristics of a social system and the behavior of
individuals in groups or society are reflected in these definitions. The content of
institution is also defined in a similar manner:
[S]tructures and mechanisms of social order and cooperation governing the
behavior of a set of individuals within a given human collectivity … identified
with a social purpose and permanence, transcending individual human lives and
intentions, and with the making and enforcing of rules governing cooperative
human behavior.
3

These concepts are used to account for the difference between societies (e.g., social
capital and collective performance), or the similarity of behaviors within a society (e.g.,
institutional isomorphism).
Likewise, the conceptualization of trust in sociology literature is guided by what
it does, rather than what it is. Although sociologists may generally acknowledge that trust
is an individual‘s positive expectation toward others, they rarely conceptualize it as such.
They tend to mention the expectation aspect of trust in passing (Mollering, 2006),
treating it as just a mediating variable between the two variables of their primary interest:

3
Stanford Encyclopedia of Philosophy (http://plato.stanford.edu/entries/social-institutions/)

14

social structure and collective behavior (e.g., Granovetter, 1985). Instead, they tend to
conceptualize trust as a characteristic of a social system (e.g., Fukuyama, 1995). Putnam
(1993) mentioned trust interchangeably with social norm, viewing both as ingredients of
social capital. Some scholars treated it as a ‗functional‘ equivalent and/or alternative to
institutions such as norms, contracts, or power (Bachman, 2006; Luhman, 1979).
Sociologists with a micro approach do conceptualize trust as individual expectation. But
they are still more interested in the function of that expectation. For instance, Luhman
(1979) conceptualized trust as the expectation that reduces social uncertainty. Bradach
and Eccles (1989) conceptualized it as the expectation that reduces the fear of
opportunistic behavior. These scholars used the concept of individual‘s positive
expectation in order to explain the micro-mechanism by which social system influences
the behavior of individuals in that society, the function in which they are ultimately
interested.
The conceptualization of trust as a choice behavior spans multiple disciplines,
including economics (Williamson, 1993) and political theory (Hardin, 2004). Scholars
(e.g., Kramer, 1999) tend to believe that trust as a choice behavior is rooted in the
cooperative choice in the Prisoner‘s Dilemma situation. Although this conception of trust
often confronts a criticism that ―when trust is justified by expectation of positive
reciprocal consequences, it is simply another version of economic exchange, as is clear
from treatments of trust as reputation in repeated games‖ (March & Olsen, 1989: p. 27),
they find the rationale for such conception of trust in Axelrod‘s (1984) simulation
research. They almost invariably cite Axelrod (1984) as arguing that trust evolves in
15

reiterated interaction in which proper incentive to maintain the interaction is in place. In
particular, Hardin (2004) included Axelrod‘s five suggestions regarding how to promote
cooperative choice among selfish individuals (in Ch. 7 of his book) into his unique
definition of trust as encapsulated interest. However, it is important to note that Axelrod‘s
interpretation of his own work undermines the validity of the conception of trust as
rational choice, rather than buttresses it. He pronounced that trusting another party is not
the condition that leads people to make a cooperative choice in the PD situation:
Likewise, there is no need to assume trust between the players: the use of
reciprocity [Tit For Tat] can be enough to make defection unproductive…The
foundation of cooperation is not really trust, but the durability of the relationship.
When the conditions are right, the players can come to cooperate with each other
through trial and error…whether the players trust each other or not is less
important in the long run than whether the conditions are ripe for them to build a
stable pattern of cooperation with each other (Axelrod, 1984; p.174; 182).

It is important to note here that the purpose of this review is not to argue against
the notion of trust as choice behavior entirely. Conceptualizing trust as a choice behavior
in PD game has some value. The PD game can reasonably be used as an experimental
research design to see whether one player trusts another player when other conditions are
properly controlled for (Berg, Dickhaut, & McCabe, 1995). My intention here is to clarify
that the philosophical assumption underlying the practice of conceptually defining trust
as choice behavior is not relevant to the rational choice perspective. Note that, based on
the assumption that human choices are driven by self-interest, Axelrod (1984) argued that
trust is not necessary to promote cooperation if incentive to do so is sufficient. In fact, it
is the unique epistemological assumption of behavioral psychology of the 1950‘s that
underlies the conceptual definition of trust as choice behavior. Deutsch (1958) conducted
16

the iterated PD experiment about 30 years before Axelrod did. Based on the context of
the PD game, he defined trust as follows:
[A]n individual may be said to have trust in the occurrence of an event if he
expects its occurrence and his expectation leads to behavior which he perceives to
have greater negative motivational consequences if the expectation is not
confirmed than positive motivational consequences if it is confirmed (p. 266).

This definition was molded by a typical stimulus-response scheme of behaviorism:
making the cooperative choice (response) when the possible loss is bigger than the
possible gain (stimuli: perceiving the payoffs in the PD game).
Deutsch‘s (1958) conceptual definition is somewhat driven by his operational
definition. He noted, ―[i]n this paper I have attempted to define ‗trust‘ and some other
related concepts in a way which does not violate everyday usage and yet is sufficiently
precise to permit experimental work with these concepts‖ (p.278). According to the norm
within today‘s scientific community, this is not a very sound practice. In order to
understand why such practice was acceptable in his era, a brief explanation of the
intellectual history in the field of psychology is in order. Behaviorism had been a
prevailing paradigm in the field of psychology before the surge of the cognitivist
revolution starting in the late 1950‘s (Gardner, 1986). Behaviorists were one of the most
radical proponents of the positivist scientific method: they typically do not believe in or
define unobservable theoretical constructs. The Operational Behaviorists among them,
most notably Edward C. Tolman, began to accept the idea that cognitive process is at
work (Tolman, 1966), and would conceptually define the theoretical cognitive process by
the observable response to stimuli (Bouton, 2007). Such practice of defining a concept is
supported by Bridgman‘s Operationalism which argued that a concept exists to the extent
17

that you can operationalize it (Bridgman, 1927).
4
Considering the time when Deutsch‘s
work was published, and considering the somewhat distinctive way in which he defined
trust, it would not be unreasonable to argue that the notion of trust as choice behavior
originated from the epistemological idea that a psychological construct, such as trust,
may exist, but cannot be defined unless you operationalize it as an observable behavior.
In contrast, today‘s studies of organizational behavior, whether experimental or
field research, are based on a totally different epistemological assumption, that of
cognitivism. The major tenets of cognitivism are that a cognitive process is the
precondition that leads to a certain behavior, that a cognitive process can be effectively
explained by discrete components (e.g., trust) and the relationships among them, and that
this process can be investigated by scientific methods (Tudela, 2004). In today‘s scientific
methodology, the conceptual definition and the operational definition are two separate
processes: researchers generally begin empirical study by establishing a conceptual
definition of a psychological concept, which becomes the foundation of a subsequent
operational definition. Hence, conceptual definitions play a much more important role for
OB studies under cognitivism, than under operational behaviorism or sociology, since it
serves as the first step to delineate the unknown and essentially unobservable mental
process into a number of discrete and measurable components. As opposed to operational
behaviorism under which operationalization determines conceptual definition (Tolman
1966), OB studies based on cognitivism are concerned more with conceptual definitions,
than operational definitions: the validity of an operational definition depends on whether

4
Stanford Encyclopedia of Philosophy (http://plato.stanford.edu/entries/operationalism/)

18

it measures the concept it is intended to measure, faithful to the conceptual definition
(Campbell, 1959), not vice versa.
5

Most scholars agree that trust is fundamentally a psychological state (Kramer,
1999; Kramer & Lewicki, 2010). Scholars in this tradition have identified a number of
sub-components common across numerous psychological concepts of trust, such as
willingness to take a risk (Mayer et al., 1995; Rousseau, et al., 1998), positive
expectations based on positive recognition of trustworthiness (Das & Teng, 1998;
Robinson, 1996), and emotion or affect (McAllister, 1995). For this study, I define trust
as a trustor ’ s positive expectation that a trustee will perform a behavior that bears
positive consequences to the trustor. I considered two things when proposing this
definition of trust. First, this study is based on the cognitivist perspective, just as is most
of today‘s organizational behavior research. Thus, I will not define trust as a choice
behavior. I conceive a risk-taking choice in a PD situation as one possible behavioral
consequence of trust, but not as trust itself. Second, I am interested in examining the
psychological process in which trust is developed, to be specific, antecedents to trust as
cognitive and psychological constructs. The concept of trust related to social structure is
not relevant to trust examined in the study. The topic of the relationship between social
structure and behavioral similarity in a group falls beyond the scope of this study.
Trust as a positive expectation is probably the most parsimonious, most widely
used (Rousseau et al., 1998), and most important (Kramer & Lewicki, 2010) conception

5
Likewise, some trust researchers are concerned that the operational definitions of trust available in the
literature do not faithfully represent the conceptual definitions (Lewicki, Tomlinson, & Gillespie, 2006).

19

of trust. Whatever philosophical assumptions they adopt, and whatever theoretical
orientation they pursue, there is a clear overlap among scholars, which is that trust entails
positive expectations toward others. Even Deutsch‘s definition includes the expectation
element. Moreover, sociologists tend to consider trust as an expectation, although often
mentioned only in passing or trivialized in pursuit of sociological theory building.
Hosmer (1995) extensively reviewed diverse trust conceptions from multiple disciplines
and, based on commonalities across those disciplines, suggested a comprehensive
definition of trust as,
[T]he expectation by one person, group, or firm of ethically justifiable behavior-
that is, morally correct decisions and actions based upon ethical principles of
analysis-on the part of the other person, group, or firm in a joint endeavor or
economic exchange (p.399).

In this definition, trust as a positive expectation plays a pivotal role. As for the concept
of trust as a willingness to take a risk (Mayer et al., 1995; Rousseau, et al., 1998), trust as
a positive expectation is also indispensible as well. The concept of willingness to take a
risk itself does not stand alone as a legitimate representation of trust because it conveys
many other meanings than trust such as a risk orientation or benefits associated with the
choice (March & Shapira, 1987).

Trustworthiness
Many scholars have identified the characteristics of trustees that lead them to be
perceived as trustworthy, and these can be categorized into three main factors: ability,
intention, and affect. Trustworthiness as ability refers to the extent to which a trustee is
20

perceived to be capable of fulfilling the trustor‘s expectations regarding the trustee.
Butler (1991), Cook and Wall (1980), Deutsch (1960), Good (1988), Jones, James, and
Bruni (1975), Kee and Knox (1970), Lieberman (1981), Sitkin and Roth (1993) all
include ability on their list of trustworthiness factors. Trustworthiness as intention refers
to the extent to which a trustee is perceived as being willing to fulfill the trustor‘s
expectations. This element is expressed in numerous terms, such as trustworthy intentions
(Cook & Wall, 1980), intention to produce (Deutsch, 1960), motives (Kee & Knox, 1970),
integrity and benevolence (Mayer et al., 1995), and intentionality (Doney, Cannon, &
Mullen, 1998). Trustworthiness as affect refers to whether a trustee is a likable person,
corresponding to affect-based trust (McAllister, 1995). Although trust as defined for this
study is essentially a cognitive phenomenon, a trustor‘s affect toward a trustee is still an
integral part of trustworthiness since positive affect can influence cognitive social
judgments of others (Dunn & Schweitzer, 2005; Forgas, 1995). Many favorable
characteristics of a trustee, such as discreetness and receptivity (Butler, 1991), personal
attraction (Giffin, 1967), and caring (Mishra, 1996) have been identified as elements of
trustworthiness in the literature.

Propensity to Trust and Generalized Expectancy
Scholars have used multiple terms to describe the difference among trustors in
their tendency to trust others. McKnight, Cummings, and Chervany (1998) used the term,
trusting stance, meaning that ―one believes that, regardless of whether people are reliable
or not, one will obtain better interpersonal outcomes by dealing with people as though
21

they are well-meaning and reliable‖ (p.477). They argued that trusting stance creates a
stable personal tendency of a trustor. The attitude toward people in general, such as belief
about human nature (Wrightsman, 1991), faith in humanity (McKnight et al., 1998), or
generalized expectancy (Rotter, 1971), is a frequently cited attribute of a trustor that
makes him/her tend to trust more than others. Some people use the term dispositional
trust (Kramer, 1999) or propensity to trust (Mayer et al., 1995) to describe one‘s general
tendency of trusting which is assumed to be stable across diverse situations.
All these terms are rooted in Rotter‘s (1971) theory of generalized expectancy
(GE), which provides a comprehensive theoretical framework accounting for how one‘s
attitude toward people in general (i.e., GE in his term) is shaped, as well as how this
general attitude becomes consequential for one‘s trusting tendency. Rotter begins with his
definition of trust: an expectancy that ―the word, promise, verbal or written statement of
another individual or group can be relied upon‖ (1967: p.651). And he asserted that 1)
people make generalizations as to what they can expect from general others (i.e. human
beings), based on what they learned from their previous relationships with specific others,
and 2) this generalized expectancy is projected to new people whom the trustor is
interacting with in similar trust-related situations. In other words, a trustor‘s positive
generalized expectancy (GE) is an antecedent to the emergence of trust, and mediates
between a trustor‘s past experience and his/her trust with current trustees. Rotter (1971)
suggested that, since trust is essentially the transfer of GE, people will display a ‗stable
tendency‘ to consistently trust or distrust regardless of different situations or relationships.
Based on Rotter‘s argument, the diverse terms for a trustor‘s tendency to trust commonly
22

emphasize the cross-situational consistency. For instance, McKnight et al. (1998) stated,
―a person exhibits a disposition to trust to the extent that she or he demonstrates a
consistent tendency to be willing to depend on others across a broad spectrum of
situations and persons‖ (p.477). Mayer et al. (1995) stated that ―propensity to trust is
proposed to be a stable within-party factor that will affect the likelihood the party will
trust‖ (p.715).
However, I argue that we need to rethink whether it is really necessary to require
cross-situational consistency when discussing propensity to trust. A view of disposition
as cross-situationally consistent behavior or attitude is a tenet of the pure dispositional
approach (Kenrick & Funder, 1988). Early clinical psychologists made the case for a pure
traits model based on their observations that, among people with similar traits, there is a
striking consistency in pathological behaviors across diverse situations (Alston, 1975;
Argyle & Little, 1972). However, the dominant approach now is to reject the pure trait
model. Numerous scholars have critiqued this model (see Kenrick & Funder, 1988 for a
review) by documenting the evidence running counter to the premise of behavioral
consistency (Mischel, 1968). The current consensus is that behavioral research focusing
on individual disposition should view disposition as ―tendencies to respond to situations,
or classes of situations in a particular, predetermined manner‖ (House, Shane, & Herold,
1996: p. 205), since behaviors are embedded in social situations in which dispositional
factors are inseparable from situational factors (Ross & Nisbett, 1991), and because
different aspects of disposition influence behavior in different situations (House et al.,
1996; Schneider, 1983).
23

Conceptualizing propensity to trust according to the assumption of cross-
situational consistency is not only dated, but also undermines the managerial significance
of understanding individual‘s trusting tendency. Kramer (1999) maintained that, while
management scholars acknowledge that individual differences in trusting tendency exist,
they have not expressed much interest in investigating these differences since the
managerial implications are limited. An emphasis on a stable, cross-situational propensity
to trust leads readily to a focus on which person is more trusting, rather than on which
traits of a person make him/her more trusting. In turn, the main prescription for managers
based on a purely dispositional approach is to select employees based on their
dispositions (Davis-Blake & Pfeffer, 1989; Staw & Ross, 1985). Davis-Blake and
Pfeffer (1989) argued, however, that implementing this prescription is cumbersome
because discriminating based on disposition can be legally or ethically questionable.
Moreover, if managers cannot exercise discretion over who their employees are, the
dispositional approach is irrelevant for management.
As an alternative to cross-situational consistency, ‗coherence‘ has been suggested
as a requirement for a dispositional approach to behavioral study. Schneider (1983)
maintained that coherence ―suggests that individuals may exhibit neither absolute nor
relative consistency yet still be predictable, because the way they are inconsistent from
situation to situation is consistent (reliable) for them [the actor]‖ (p.5). Ross and Nisbett
(1991) also suggested that,
Behavioral consistency and the lack of it both represent understandable if not
always predictable consequences of the dynamic relationship existing between the
particular individual and that individual‘s social environments … The approach
that is called for must be idiographic in spirit … we will need to know different
24

things about different people in order to appreciate the distinctiveness and
coherence in their behavior and in order to anticipate when, and how, their
behavior will prove consistent and predictable (p.15).

Based on the coherence principle, this study proposes an alternative conception of
propensity to trust, namely, the attributes of a trustor that lead him/her to trust people
more than others do. Unlike the conventional notion of propensity to trust, the effects of
trustor attributes may not necessarily result in a consistently higher tendency to trust on
the part of the trustor, because the trustor attributes that matter in one situation may not
be influential in other situations (Schneider, 1983). The proposed conception of
propensity to trust is compatible with the conception of trustworthiness defined as the
attributes of trustees that lead them to be trusted by others. When scholars discuss
trustworthiness, they do not focus on which person is more trusted across a diversity of
situations. Instead, the discussion of trustworthiness focuses on which ‗attributes‘ of a
trustee constitute trustworthiness. Thus, a person may not be consistently trusted by other
people, because some of his/her traits are more relevant than others depending on the
situation. For instance, Mayer et al. (1995) asserted that integrity is more important than
competence early in a relationship, such that competence may not be correlated with the
level of trust when people first get to know each other. Moreover, benevolence may not
be a relevant factor of trustworthiness in an arms-length relationship. Thus, if a person is
not capable in a certain domain, he/she may not be trusted in that specific domain,
regardless of his/her benevolence. Likewise, conceptualizing propensity to trust as the
trustor attributes leading them to trust more than others implies that the effects of each
single trustor attribute will differ depending on situation.
25

Not only is the alternative conception of propensity to trust suggested in this
study theoretically relevant, but it also suggests meaningful practical implication for
managers. In reality, organizations are composed of people with diverse personalities. In
order to help managers maintain trusting relationships with various types of subordinates
or to foster trust among them within their work units, we need more information than just
which person is more trusting than others. We need to understand which trustor attributes
influence their trust under what circumstances.

Cognitive Closure
Cognitive closure refers to making a definitive judgment when uncertainty and
ambiguity are inevitable. The literature in social psychology has explored what leads to
cognitive closure. Scholars have suggested that cognitive closure is a function of
situational factors such as perceived benefit and cost of achieving cognitive closure, or
the time-pressure that prods perceivers to make quick judgments (Webster & Kruglanski,
1994). On the other hand, cognitive closure is a function of a perceiver‘s disposition to
make a definitive judgment. Kruglanski and Webster (1996) argued that individuals differ
in their psychological need for cognitive closure. People with a strong need for cognitive
closure (NFC) are decisive, prefer predictability and order, and feel greater discomfort for
ambiguity.
The NFC has been demonstrated to affect a variety of cognitive decisions in
diverse contexts, such as education (DeBacker & Crowson, 2008; DeBacker & Crowson,
2009), leadership (Pierro, Cicero, Bonaiuto, & Knippenberg, 2005; Tetlock, 2000),
26

strategic management (Schwenk, 1984), human resource management (Shaheen, 2010),
and organizational behavior (Chirumbolo & Areni, 2010). For instance, DeBacker and
Crowson (2009) noted that understanding students‘ NFC will enhance the performance of
an instructional process and system, by improving our understanding of their learning
process. DeBacker and Crowson (2008) introduced a new instrument to measure NFC
that is better suited to an instructional setting, and they found that the proposed scale is
correlated to variables known to influence motivation and academic achievement. Tetlock
(2000) found that manager‘s NFC, along with conservatism, influences their fundamental
beliefs about human beings and their degree of cognitive bias, and ultimately results in
different leadership styles for coping with subordinates who fail to fulfill organizational
expectations. Shaheen (2010) argued that the consequences of NFC, such as primacy and
recency effects, play an important role in corporate‘s evaluation of job candidates and
hiring decisions. Chirumbolo and Areni (2010) found that individuals‘ NFC moderated
the effect of job insecurity on job performance and mental health.
The need for cognitive closure is as relevant to the topic of trust as to the
aforementioned topics, since the cognitive decision as to a trustee‘s trustworthiness is a
critical basis of the expectation that the trustee will perform the actions bearing positive
consequences for the trustor (i.e., the concept of trust in this study). Mollering (2006)
maintained that trust depends on a trustor‘s leap of faith, ―suspending irreducible social
vulnerability and uncertainty as if they were favorably resolved‖ (p.395). Sorrentino et al.
(1995) conceptualized trust as ―the antithesis of doubt: It is conceptualized as a state of
felt security that marks at least a temporary resolution of feelings of uncertainty‖ (p.314).
27

The presumption underlying these notions of trust is that the available information
regarding other‘s trustworthiness is by no means complete or precise (i.e., ambiguous)
and that the evidence of both trustworthiness and untrustworthiness (Lewicki, McAllister,
& Bies, 1998) often coexist. Therefore, if a trustor highly trusts a trustee, it essentially
reflects the fact that the trustor has extrapolated to some degree from the given
information (Mollering, 2006; Weick, 2008). Hence, trusting may be seen as a result of
‗cognitive closure‘ related to a trustor‘s judgment of a trustee‘s trustworthiness.
The NFC is believed to be nonspecific in nature, in that it can result in any
closure, regardless of its content (e.g., positive or negative). While cognitive closure in a
specific direction can appear as a result of an interaction between need for closure and
situational factors (e.g., the perceived benefit of closure in a certain direction), the NFC
itself does not entail which specific direction. Sorrentino et al. (1995) conducted an
experimental study investigating how trustors‘ cognitive preferences (i.e., the need for
closure) affect trust in romantic relationships, and found that certainty-oriented persons'
need for cognitive closure resulted in either high or low trust, while uncertainty-oriented
people showed a moderate level of trust. This study suggests that the cognitive preference
of a trustor is associated with the degree of trust or distrust, rather than the direction.
Even if a person has strong NFC, it would be hard to predict whether his/her trust will be
higher than those with low need for cognitive closure since the effect of the need for
cognitive closure can take place in either a positive or a negative direction.
The seizing effect and the freezing effect were introduced by Kruglanski and
Webster (1996) to explicate the psychological process by which a person with a high
28

NFC obtains a definitive answer to a question in essentially ambiguous situations: he/she
forms a quick and confident perception (i.e., seizing effect), and at the same time,
perpetuates this first impression (i.e., freezing effect). They asserted that, ―they [people of
strong NFC] may, therefore, wish to terminate this unpleasant state [discomfort for
ambiguity] quickly (the urgency tendency) and keep it from recurring (the permanence
tendency)‖ (1996: p.265). Webster and Kruglanski (1994) developed a 42-item scale of
NFC, which has five sub-constructs: preference for order; preference for predictability;
decisiveness; discomfort with ambiguity; and closed-mindedness. Kruglanski and
Webster (1996) assumed that the two effects happen together within the same person, and
thus proposed the NFC scale as a uni-dimensional measure of NFC, calculated by
aggregating the scores from all five sub-constructs.
However, a number of other scholars have found that the NFC scale is a multi-
dimensional measure. Neuberg, Judice, and West (1997) reported that three of the sub-
constructs (i.e., preference for order, preference for predictability, and discomfort with
ambiguity) represent one component of the NFC, while decisiveness constitutes the other.
Kossowska, Van Hiel, Chun, and Kruglanski (2002) confirmed that the NFC scale
contains these two different components. They conducted a confirmatory factor analysis,
and found that none of the items of the decisiveness component had significant loadings
on the one-factor solution, while those items showed significant loadings on the second
factor when tested with a two-factor solution. Roets, Van Hiel, and Cornelis (2006)
conducted a structural equation model analysis for the NFC scale, and found that the two
factor argument is supported. The first factor is usually labeled need for structure (NFS)
29

and the second is called decisiveness (Roets et al., 2006; Stalder, 2010).
Therefore, Neuberg et al. (1997) suggested that these two components of the
NFC scale should be used separately, rather than as one combined variable. They
proposed that a person high in NFS will tend to maintain previously acquired closure (i.e.,
freezing effect), and that a person high in decisiveness will tend to reach a quick closure
based on a small amount of information, (i.e., seizing effect). Accepting the advice
proposed by Neuberg et al. (1997), in this study, I will test the effects of the two
components separately. Decisiveness will be used to predict the tendency to quickly seize
on a small amount of information to make a judgment (i.e., seizing effect), while need for
structure will be used to predict the tendency to freeze on that information to maintain
cognitive consistency (i.e., freezing effect).

Phases of Trust Development
Rousseau et al. (1998) argued that trust develops through three phases: building
(where trust is formed), stability (where trust already exists), and dissolution (where trust
declines). The building phase refers to when parties meet for the first time and the initial
level of trust is formed. A comprehensive review of trust literature by Lewicki, Tomlinson,
and Gillespie (2006) reveals that, while in early experimental studies the initial trust level
is presumed to be zero, some empirical studies (e.g., Kramer, 1994) show that people
sometimes demonstrated high trust for those with whom they had no interaction history.
This may be explained by a trustor‘s predisposition to trust. Mayer et al. (1995) posited
that each component of a relationship (i.e., trustor-, trustee-, and relationship-attributes)
30

explains a unique part of the variance in the level of interpersonal trust, and asserted that
the effects of trustor attributes will have a dominant influence on trust when dealing with
strangers.
The stability phase refers to the period in which existing trust develops further
through frequent and regular ongoing interactions. Extant trust studies involving this
phase tend to focus on the role of institutional and societal sanctions against opportunism
(Hagen & Choe, 1998) and on the mechanisms mitigating risks (Sheppard & Sherman,
1998) in maintaining an existing level of trust, while there has been less investigation of
how trustor attributes affect the maintenance of existing trust. In the stability phase,
trustors update their initial assessment of another‘s trustworthiness based on what has
happened in the relationships. In actual organizational life, however, it is not always clear
whether or not initial trust is misplaced. Furthermore, the information used to update and
recalibrate trustworthiness is not an objective array of stimuli, but people‘s subjective
interpretations of events (Rotter, 1971). Thus, which events serve to reinforce or change a
trustor‘s perception of a trustee‘s trustworthiness is likely to depend on trustor attributes
affecting their perceptions.
The dissolution phase refers to when trust declines in the face of a trustee‘s
deviant, antisocial, or betrayal behavior (Elangovan & Shapiro, 1998). Trust studies
involving the dissolution phase have focused on repairing trust after it is breached,
primarily emphasizing the attributes of trustees in doing so (e.g., Kim, Ferrin, Copper, &
Dirks, 2004; Tomlinson & Mayer, 2009). However, it is just as important to understand
differences among trustors in terms of the extent to which their initial trust declines in the
31

face of negative experiences, since this will determine how responsive they are to a
trustee‘s efforts to re-establish trust.

Chapter Summary
Trustors‘ propensity to trust is shaped by the effects of multiple dispositional
qualities. Based on two theories, I identified two factors that may influence trust, namely,
GE and NFC. The influences of these dispositions on trust will differ from one situation
to another. This study aims to investigate the influence of these two dispositional factors
on trust in three phases of trust development separately: building, stability, and
dissolution. Table 1 summarizes the concept of propensity to trust proposed for this study,
and compares it to the conventional notion of propensity to trust and to the concept of
trustworthiness. Table 2 summarizes and compares the two trustor attributes identified in
this study: GE and NFC.

32

Table 1. Concept of Propensity to Trust in This Study
Definition Factors (Ingredients)
Sources /
Referents
Proposed
Conception of
Propensity to
Trust
The trustor
attributes that
lead him/her to
trust more than
others
Generalized expectancy Past experience
NFC
Discomfort felt
by uncertainty
Conventional
Conception of
Propensity to
Trust (Rotter,
1967)
A trustor who
consistently trusts
more than others
across diverse
situations
GE Past experience
Conception of
Trustworthiness
(Mayer et al.,
1995)
The traits of a
trustee that lead
him/her to be
perceived as
trustworthy
Integrity
Ability
Benevolence
Stereotype (e.g.,
membership,
gender, etc.)
Interaction
history

Table 2. Comparison between Generalized Expectancy and Cognitive Closure
Nature of Trust How it affects trust?
Generalized
Expectancy
An extension of
generalized expectancy
toward specific others
Trust or distrust:
Factors of generalized
expectancy determine whether
one trusts or distrusts another
Need for
Cognitive Closure
Antithesis of doubt and a
leap of faith:
Trust marks a resolution of
uncertainty for felt security
Degree of trust/distrust:
The need for cognitive closure
will determine ‗how much‘
trust (or distrust) one has
toward others.

33

CHAPTER THREE: HYPOTHESES

For this study, I defined trust as a positive expectation toward another‘s behavior.
The level of trust is determined by the trustor‘s subjective assessment of the probability
(Gambetta, 1988) that the expectation is fulfilled, based on extrapolating from available
information that the trustor sees as relevant (Mollering, 2006; Weick, 2008). Therefore,
the difference in trusting tendency among trustors will depend on which information they
consider relevant in making their trusting judgment, and the effect of that information on
their judgment in each phase of the trust development process.

The Building Phase
The building phase refers to when parties meet for the first time and an initial
level of trust is formed. As discussed in the previous chapter, Rotter‘s (1967) theory of
generalized expectancy explains how trustors can trust others in a new relationship. To
briefly review, he asserted that 1) people make generalizations as to what they can expect
from general others (i.e., human beings), based on what they have learned from their
previous relationships with specific others, and 2) this generalized expectancy (GE) is
‗projected‘ to new people with whom they interact in similar trust-related situations.
Rotter also noted that situational factors come into play when trust is formed:
In social learning theory, expectancies in each situation [trust] are determined not
only by specific experiences in that situation [by situation-specific expectancy]
but also, to some varying degree, by experiences in other situations that the
individual perceives as similar [by generalized expectancy] (Rotter, 1980, p.2).
34

In other words, expectations about a specific person in a specific situation, i.e.,
interpersonal trust, are influenced by their GE toward people in general, as well as by a
situation-specific expectancy (SE) as to what can generally be expected from others in
that specific type of situation.
Both GE and SE are generalized expectancies formed from past experiences.
However, the two are different in terms of the scope of experiences contributing to them,
the content of the expectancies, and the logic of how they influence the level of trust with
a certain trustee. What can be generally expected from general others (GE) reflects one‘s
learning from virtually all interpersonal experiences. In contrast, SE is based on the
interpersonal experiences only in a specific type of situation (e.g., interacting with other
members of one‘s church, or collaborating with others to accomplish a project at work).
When people meet strangers for the first time, they have no clear information
convincing them that a trustee will behave in a way that the trustor expects him/her to
behave. In this situation, trustors will extrapolate from their GE to assess the probability
that their expectation is fulfilled. Johnson-George and Swap (1982) argued that GE
predicts trust in ―highly ambiguous, novel, or unstructured situations, where one‘s
generalized expectancy is all one can rely on‖ (p.1307). The content of GE involves
expectations toward unspecified other people formed by past experiences in relationships
with numerous other people. In other words, people with a positive GE trust others
extrapolating from the information that most people in the past have behaved in a
trustworthy way, while people with a negative SE distrust others extrapolating from the
information that most people in the past have behaved in a distrustful way. Thus, the
35

following hypothesis can be suggested.
Hypothesis 1. In the building (initial trust formation) phase, trustors‘ generalized
expectancy will be positively related to their level of trust.

McKnight et al. (1998) argued that the felt security based on the perception that
the situation is normal will enable trustors to easily trust others in a new organizational
relationship. McKnight and his colleagues listed some conditions supporting the
situational normality belief, such as when expecting that a trustee will act in specific way
in that situation is customary (Baier, 1986), or when they feel that what happens in that
situation is in proper order (Lewis & Weigert, 1985). Another condition for the situational
normality belief is positive SE. Hence, positive SE can make people feel secure in that
type of situation. Furthermore, SE can provide information that trustors can extrapolate
from when assessing the probability that trust will be fulfilled. In other words, trustors
with a positive SE will anticipate that a given trustee will behave in a trustworthy way
because others have behaved in a trustworthy way in that specific situation, while trustors
with a negative SE will anticipate that a given trustee will behave in a distrustful way
because others have behaved in a distrustful way in that specific situation. Thus, the
following hypothesis can be suggested.
Hypothesis 2. In the building (initial trust formation) phase, trustors‘ situation-specific
expectancy is positively related to their level of trust.

Based on these first two hypotheses, Figure 1 below models the effects of the
two trustor attributes (GE and SE) on trust in the building phase.
36

The Stability Phase
As interaction iterates and a relationship develops, people come to know each
other better. Based on first-hand experiences with a trustee, trustors update and
recalibrate their initial assessment of trustworthiness, and the level of trust will change
accordingly. This updating process renders trust dynamic (Currall & Inkpen, 2006; Weick,
2008). The source of updating and recalibrating trustworthiness is not an objective
array of stimuli, but people‘s subjective interpretation of events (Rotter, 1971). Thus,
which events will serve to reinforce or change a trustor‘s perception of a trustee‘s
trustworthiness is likely to depend on attributes of trustors that affect their perceptions.
Brunswik‘s (1956) lens model of social cognition provides a framework for
understanding the process by which trustworthiness is recognized and inferred by a
trustor. The model consists of three components: 1) a reality that is essentially non-
deterministic and equivocal (i.e., objective trustworthiness of a trustee); 2) a number of
cues that are observed by a perceiver (i.e., the events which a trustor uses to infer
H2 (+)
Generalized
Expectancy

The Level of
Trust before
Interacting
(Trust T
1
)
Situational
Expectancy
Figure 1. Proposed Causal Model in the Building Phase
H1 (+)
37

trustworthiness); and 3) a subjective judgment about reality made on the basis of the cues
observed (subjective assessment of trustworthiness). He explained that the inference
about reality that results from this process (i.e., an assessment of trustworthiness) is
determined by two parameters: cue encoding and cue decoding. First, cue encoding refers
to the degree to which reality is expressed in observable cues. Second, cue decoding
refers to the degree to which the observed cues are the basis of the inference a perceiver
makes about reality (Gifford, 1994). Applying this framework to this study, it can be
argued that the cue encoding process represents the degree to which interpersonal events
are perceived as reflecting a trustee‘s trustworthiness or untrustworthiness, while the cue
decoding process represents the degree to which the perceived negative or positive events
have an impact on the perception of trustworthiness or untrustworthiness.
In the stability phase, the initial level of trust increases or decreases depending
on the nature of trustors‘ cue encoding related to trustworthiness. In the process of cue
encoding, the number of events one perceives as positive or negative is likely to depend
on their belief about human nature (GE). Omodei and McLennan (2000) developed a
scale to measure GE based on respondent interpretations of eighteen hypothetical
scenarios. For instance, respondents were given a situation, e.g., ―a workmate complains
to you of a bad headache and asks you to finish a boring task,‖ and then asked how
strongly they agreed with the interpretation, ―the headache is just as bad as he or she says
it is.‖ Another scenario was, ―a workmate accidentally causes a small fire. As you both
put out the fire, this workmate keeps asking if you are burnt,‖ and the proposed
interpretation is, ―the workmate is trying to stop you from realizing that he or she caused
38

the fire.‖ The basic assumption underlying this measure of GE is that people‘s general
belief about human nature is reflected in the interpretation of seemingly neutral
interpersonal events. If someone believes that human beings are generally ―mean, selfish,
malevolent, and/or unreliable and thus should not be depended on to treat one well‖
(p.279) (i.e., negative GE), s/he will interpret the events suggested in the scenarios as
signs of the workmate‘s untrustworthiness. They found that this scenario-based measure
of GE is significantly correlated with Rotter‘s (1967) Interpersonal Trust Scale and with
other measures of GE. From this finding, we can infer that the probability of perceiving
negative events while working with others is associated with one‘s beliefs about human
nature (i.e., GE). Thus, the following hypotheses are proposed.
Hypothesis 3-1. GE is positively associated with the probability that trustors interpret
interpersonal events as indicating trustworthiness

Hypothesis 3-2. GE is negatively associated with the probability that trustors interpret
interpersonal events as indicating untrustworthiness.

People tend to selectively seek out or attend to the events that are consistent with
their prior perceptions, knowledge, beliefs, and attitudes (Fiske & Taylor, 1984;
Greenwald, 1980). Moreover, trustors may assimilate their experiences during
interactions with others to their initial impression by reinterpreting events that they view
as reflecting a trustee‘s level of trustworthiness. If people have experienced positive
events in a particular situation, they are likely to believe that people in general will act in
a trustworthy way in that situation (i.e., positive SE). Their prior assumption will lead
them to interpret interpersonal events in ways compatible with their positive SE. In other
words, people would also be inclined to believe that a specific trustee will also act in
39

trustworthy manner in that situation, just as others did in the past. Likewise, if people
have negative SE, they will be more likely to interpret interpersonal events as signs of a
trustee‘s untrustworthiness. Based on this logic, the two following two hypotheses will be
tested.
Hypothesis 4-1. SE is positively associated with the probability that trustors interpret
interpersonal events as indicating trustworthiness.

Hypothesis 4-2. SE is negatively associated with the probability that trustors interpret
interpersonal events as indicating untrustworthiness.

Kruglanski and Webster (1996) argued that people with strong NFC tend to
display a strong primacy effect of their initial perceptions, suggesting that the degree to
which a trustor is affected by prior perceptions is greater for those with stronger NFC.
Ritcher and Kruglanski (1998) explained the reason for this primacy effect. Generally,
people (regardless of their NFC) tend to use early information as an anchor in making
judgments, and then pay less attention to information acquired later. In addition, high
NFC people have a greater desire to avoid the discomfort caused by new information that
contradicts their initial level of trust. When some events that may be seen as reflecting
untrustworthiness are experienced by high NFC people, it can be expected that they are
more likely than those with weak NFC to selectively attend to events that confirm their
initial trust or to reinterpret them as confirming their initial trust, in order to avoid the
cognitive dissonance caused by interpreting them as running counter to their initial
judgment. Thus, the primacy effect will be stronger for high NFC people. For example,
Kruglanski, Webster, and Klem (1993) found that, in a jury study, people with strong
NFC tend to stick to their initial verdict, even though experimenters attempted to
40

persuade them to change their views later. As discussed in the previous chapter, the
NFS component of the need for cognitive closure scale is related to the tendency to freeze
on one‘s initial perception. Thus, the following hypotheses can be suggested.
Hypothesis 5-1. The positive association between trustors‘ initial trust and the probability
that they interpret interpersonal events positively will be greater for those with a strong
NFS.

Hypothesis 5-2. The negative association between trustors‘ initial trust and the probability
that they interpret interpersonal events negatively will be greater for those with a strong
NFS.

Based on Hypotheses 3-1 through 5-2, Figure 2 below provides the causal model
to be tested in this study regarding the effects of three trustor attributes (i.e., GE, SE, and
NFS) on the probability that they experience positive and negative interpersonal events in
relationships.
41

The Dissolution Phase
Once stable trusting relationships have been established, a manager‘s key role is
to help maintain such relationships, given that trust can be fragile. Trust studies involving
the dissolution phase have focused on repairing trust after it is breached, primarily
emphasizing the attributes of trustees in doing so (e.g., Kim, Ferrin, Copper, & Dirks,
2004; Tomlinson & Mayer, 2009). However, it is just as important to understand
differences among trustors in terms of the extent to which their initial trust declines in the
face of negative experiences, since this will determine how responsive they are to a
trustee‘s efforts to re-establish trust.
Figure 2. Proposed Causal Model in the Stability Phase
(-)
(+)

GE

The Probability
of Positive
Experiences

The Probability
of Negative
Experiences

SE
H4-1
(+)
H3-2(-)
H4-2(-)
H3-1 (+)
H5-1 (+)
H5-2 (-)

Trust T
1

Need for
Structure
(NFS)
42

Mollering (2006) argued that trust depends on a trustor‘s leap of faith,
―suspending irreducible social vulnerability and uncertainty as if they were favorably
resolved‖ (p.395). This leap of faith is more likely to happen for those with a high NFC.
As discussed previously, strong NFC individuals tend to make judgments firmly and
instantly based selectively on a small amount of information that is internally consistent.
People with high NFC tend not to make a balanced judgment accommodating
contradicting information. Hence, once trust is formed in a relationship at the building
phase, high NFC people will try to maintain peace of mind by suspending irreducible
social vulnerability and uncertainty as if they were favorably resolved and focusing only
on the events confirming their initial trust. When interpersonal experiences in the
interaction process subtly reflect the untrustworthiness of colleagues, high NFC trustors
are more likely to reinterpret the events as confirming their initial perception (Hypothesis
5-1).
Then, how will high NFC trustors‘ trust decline differ from that of low NFC
trustors, once any event is accepted by trustors as suggesting the untrustworthiness of a
trustee? To answer this question, we need to think about how high NFC trustors will
process the additional information (i.e., recent negative events) that is inconsistent with
their prior conclusions. First of all, low NFC people will more readily make judgments
considering both prior and additional information more or less proportionately, and that
high NFC people will make judgments overwhelmed by either of the two contradicting
types of information. However, it is not easy to predict which of the two, between the
prior positive information and the recent negative information, they will use to make a
43

trusting judgment, since the literature suggests both can be possible. The first possibility
is that, when interpersonal events are interpreted as negative, high NFC trustors may try
to resolve the tension of cognitive inconsistency by seizing on the recent negative
information and paying less intention to the information that they had used previously to
make their initial judgment. In contrast, they may try to resolve that tension by freezing
on their initial perception, and paying less attention to the newer negative information.
Ritcher and Kruglanski (1998) also noted that there are two possible predictions
regarding the process of impression formation by high NFC people, and identified the
conditions that determine whether the primacy effect of their first impression or the
recency effect of the newly presented information will manifest itself. In an experimental
study, they presented both positive and negative descriptions of a job candidate, and
found that the purpose of reading those descriptions determined which information the
subjects primarily relied on to make a decision whether to hire the candidate. When
subjects read the negative description to assess its grammatical structure and the
coherence of the text, and subsequently read the positive description to develop an
impression of the candidate, the hiring decisions of high NFC people tended to be heavily
influenced by the later (positive) information, compared to low NFC people (i.e., recency
effect). Likewise, when subjects read the positive description for the purposes of judging
grammatical structure and sentence coherence, and later read the negative description in
order to form an impression, the effects of negative (later) information were greater for
high NFC people than for low NFC people. In contrast, when subjects performed the
same tasks but started with the impression formation goal, decisions by high NFC people
44

tended to be influenced by the initial (vs. later) information. Based on these findings, they
concluded that the goals with which perceivers approach an information processing task
is one condition determining whether a primacy or recency effect dominates the
judgments of strong NFC people.
Since the present study is based on field research, the conditions influencing
subjects‘ perceptions cannot be manipulated as Ritcher and Kruglanski (1998) did. Hence,
it would be appropriate to make the investigation of the effects of NFC on trust decline
exploratory. However, several conditions that may influence whether the primacy effect
of initial trust perceptions or the recency effect of new negative information will
dominate can be identified, and under these limited conditions, some hypotheses can be
identified. First, people with strong decisiveness will tend to seize on recent negative
information, while people with strong need for structure will tend to freeze on the prior
positive information. As discussed in the previous chapter, decisiveness and the need for
structure are two separate components of the need for cognitive closure, and each
component accounts for one of two tendencies: decisiveness leads to the tendency to
quickly terminate ambiguity (i.e., seizing effect), and need for structure leads to the
tendency to prevent ambiguity from recurring (i.e., freezing effect). Scholars have found
that the desire to quickly terminate the unpleasant ambiguity (i.e., decisiveness) often
leads people to seize on the most accessible information (Ford & Kruglanski, 1995;
Kruglanski & Webster, 1996; Ritcher & Kruglanski, 1998). So, for example, once
interpersonal events are interpreted by trustors as indicating the untrustworthiness of
trustees in whom they have had high trust, that information is likely to be more accessible
45

to trustors than the information used when making their initial judgment, since it is more
recent. Hence, after interpreting interpersonal events negatively, a decline in trust will be
greater for those highly decisive trustors. In contrast, the desire to avoid the recurrence
of ambiguity or cognitive inconsistency tends to lead trustors with a high need for
structure to stick to their initial judgment that was determined previously, and thus to be
less influenced by recent negative information.
Another condition that matters is whether trustees are friends or strangers. When
strong NFC people experience a negative interpersonal event with people that they have
trusted, they will tend to be disproportionately influenced by either the prior positive
experiences they have had in those relationships or by the negative experience they just
encountered. Which of these two sources of information (i.e., the prior positive
experiences vs. the new negative experience) is more influential will be determined in
part by how many positive experiences the trustor has had in relationship with the given
trustee.
Initial trust generally buffers the effect of negative experiences, such as a breach
of a psychological contract, on subsequent trust levels (Robinson, 1996). However, this
buffering effect may not always happen. Rather, it is contingent on whether the initial
trust is supported by quality evidence. While a trustor may have high trust for both
strangers and friends, the former is more likely to be based on a smaller amount of
information, e.g., a couple of positive experiences that corroborate their pre-existing GE
and SE. In contrast, high trust toward friends is usually based on a greater amount of
prior positive experiences in more diverse situations. Thus, if a negative experience
46

happens in a relationship with a friend, the probability that a trustor will tend to rely more
on these prior positive experiences is greater than if it happens in a relationship with a
stranger. If a negative event is experienced early on in a relationship with a stranger, the
trustor will generally tend to pay more attention to the negative experience, and trust will
quickly decrease.
Therefore, this condition (i.e., strangers vs. friends) determines whether high
NFC people‘s cognitive closure will lead to the primacy effect of initial perception or the
recency effect of recent information that contradicts with the initial perception. In other
words, when people experience negative events in relationships with friends, the rate of
trust deterioration will be generally less than when in relationships with strangers. For
those with a high NFC, the rate will be even less than when trustors are low NFC people.
In contrast, when strong NFC people experience negative events in relationships with
strangers, the rate of trust decrease will be much greater. Based on the predictions
associated with the first condition (i.e., decisiveness vs. need for structure) and second
condition (i.e., strangers vs. friends), the following hypotheses can be suggested.
Hypothesis 6-1. When initial trust was high in a relationship with a stranger, the effect on
a change in trust level of negative experiences during their interaction is amplified by the
trustor‘s decisiveness: the negative relationship between negative experiences and decline
in trust is stronger for those who are decisive.

Hypothesis 6-2. When initial trust was high in relationship with a friend, the effect on a
change in trust level of negative experiences during their interaction is buffered by the
trustor‘s need for structure: the negative relationship between negative experiences and
decline in trust is weaker for those who have high need for structure.

Based on Hypotheses 6-1 and 6-2, Figure 3 presented below suggests the causal
model I am testing in this study regarding the effects of trustor attributes on subsequent
47

level of trust after experiencing negative events in their relationships.

When initial trust
is high in
relationship
between friends
(-)

The Degree of
Trust after
Interaction
(Trust T
2
)
Decisiveness
Need for Structure
H6-1 (+)

Negative
Experiences
during
Interaction
H6-2 (-)
When initial trust
is high in
relationship
between strangers
Figure 3. Proposed Causal Model in the Dissolution Phase
48

CHAPTER FOUR: METHODS

This chapter provides a detailed explanation of the sample and data collection
process, measures, research design and analysis used to examine the hypotheses outlined
in the previous chapter. First, I will explain how data was collected and from whom.
Second, the items used to measure the concepts will be identified. Third, the research
designs for this study will be explained for three different models separately.

Research Site, Sample, and Survey Procedure
Data for this study were collected from graduate students attending classes at the
University of Southern California (USC), Los Angeles, CA, over two semesters, Fall
2009 and Spring 2010. As a part of class activity, respondents formed groups of three to
seven people and collaborated on a final project. Collaboration among students in a
classroom context resembles the way members in other organizational work groups work
together towards common goals. Operation of such work groups, such as task forces,
project groups, and self-managed work teams, is common in today‘s organizations. For
public organizations in particular, it has been seen as a vehicle to introduce an effective
management system without massive restructuring and major operational changes
requiring legislative action (Levine, 1973). Moreover, interagency task group
development is viewed as a more effective management practice than hierarchical
coordination when multiple parties collaborate for developing joint solutions (Agranoff &
McGuire, 2001). Compared to conventional bureaucratic organizational settings,
49

collaboration in work groups is distinctive in many ways. For example, self-managed
work teams – like student groups – have the power to manage their task without outside
authority directing who does what. The members often rotate on taking a leadership role,
depending on the type of task at hand, and the primary role of the leader is to facilitate
collaboration (Nahavandi & Malekzadeh, 1999). Since the context of collaboration in a
classroom setting and in work groups in organizations is similar, an examination of trust
among groups of students working together in project groups should be reasonably
generalizable to an understanding of trust in other kinds of work groups.
Surveys were conducted at two or three different time points during the semester.
All the predictor variables in this study were included in the first survey (T
1
), which was
administered near the beginning of the semester: generalized expectancy towards human
nature (GE) and towards the situation (SE); the need for structure (NFS); and
decisiveness. The second round of the survey (T
2
) was conducted shortly after the groups
were formed and the members had introduced themselves to each other. In the second
survey, respondents were asked to evaluate their initial level of trust toward each other
(Trust1). They were also asked to identify their level of social distance with each
teammate. For some classes, the first and second surveys were conducted at the same
time. Since the first survey was mainly about the trustor‘s traits, which are unlikely to
change much in just a few weeks, it is assumed that whether the first and second surveys
were completed separately or together will not affect the results. What was critical was to
assess the initial levels of trust before the students had a chance to interact together in
their project groups. The third round of the survey was conducted at the end of the
50

semester. In the third survey (T
3
), the respondents were asked to indicate the number of
times they had experienced events during the semester that indicated their teammates
were capable/incapable, demonstrated good/bad intentionality, and were caring and
receptive/not caring and not receptive (i.e., to measure the positive and negative
experiences associated with the factors related to trustee trustworthiness), and to evaluate
their level of trust toward each other once more (Trust2). Each respondent evaluated their
trust, intimacy, and positive and negative events for each teammate separately. Since data
were collected for the dyadic relationships between a respondent and his/her teammates,
the dyads are the units of analysis for this study. The questionnaire used for this study is
attached at the Appendix A.
From a total of 143 students enrolled in the five classes in which data were
collected, 100 students completed the first survey, 92 completed the second survey, and
95 completed the third survey. The response rates for each round of the survey were
70.56 percent, 66.13 percent, and 67.34 percent, respectively. Of these, only the
respondents who completed all three surveys were included in the analyses conducted for
this study. Hence, the number of respondents finally included in the study was 65, for a
final response rate of 46.15 percent. The 65 respondents yielded a total of 311 dyadic
within-group relationships, which provided the dataset for the analysis. The average size
of the groups was 4.79.

51

Measures
Trust
The definition of trust has been a controversial issue (Hosmer, 1995; Mayer et al.,
1995; Rousseau et al., 1998; Sitkin & Roth, 1993; Weick, 2008 among others). Just as the
conceptual definitions of trust are diverse, there is no consensus as to how trust can be
best measured. Each study tends to use a different measure of trust to serve the
idiosyncratic purpose of the research. This is somewhat inevitable since trust is inherently
context-specific (Hardin, 2004; Rousseau et al. 1998). After a review of trust conceptions
at Chapter 2, I defined trust as a trustor ’ s positive expectation that a trustee will perform
a behavior that bears favorable consequences to the trustor. Thus, for this study, I
constructed a six–item scale based on this conceptual definition to measure trust between
teammates in the context of classroom collaboration.
In order to customize the trust scale to be relevant to this context, I first
identified six types of activities respondents might have to perform during their work
together on their final project for the class: 1) choosing partners to work with; 2)
distributing tasks for the project; 3) asking for assistance; 4) empowering someone to
take a leadership role; 5) freely sharing ideas; 6) and monitoring efforts. Then, six
questions were developed to assess respondents‘ expectations regarding their teammates
pertaining to these activities. The questions were worded as follows: 1) If I have to
choose one partner to work with in this class, and if I choose this person, we would be
able to work together in pleasant and productive manner; 2) If I am to distribute the roles
for the final project and let this person do the most important part, he/she will get the job
52

done satisfactorily; 3) If I ask this person for help when I have difficulty doing my role in
the negotiation, he/she will respond with care and attentiveness; 4) If I let this person be a
leader who makes important decisions on behalf of our group, the performance of our
group will be successful; 5) If I freely share my idea with this person, he/she will not
negatively criticize it. Rather, he/she will make constructive comments on it; 6) Even if I
do not monitor this person, he/she will do the best to get the job done.
These items, as well as those used to measure the other variables as described
below, were assessed using a six-point scale: 1) strongly disagree, 2) disagree, 3)
somewhat disagree, 4) somewhat agree, 5) agree, and 6) strongly agree. The rationale for
using a six-point scale, instead of the more common five-point scale, is to prevent
respondents from answering neither disagree nor agree. It was anticipated that this was a
likely possibility in this study since, in the absence of clear evidence regarding others‘
trustworthiness, respondents may be inclined to answer neither disagree nor agree. The
trust measure was calculated as the average of these item scores.
Generalized Expectancy
There are a number of ways to measure the concept of generalized expectancy.
Rotter (1967) developed a 40-item scale, comprised of the expectancies toward 40
different social objects, and aggregates these scores to get a single measure of generalized
expectancy. As Rotter (1980) explained, the ITS includes past experiences in a variety of
interpersonal relationships:
if one is trying to get a good measure of a generalized expectancy, one should
avoid stimuli with which the individual has had a great deal of previous, specific
experience, such as father, mother, lover, and friend, and one should try instead to
sample widely from a large number of social agents… (p.2).
53

The basic idea behind this measure is that people with a positive GE will score
consistently higher across these diverse social objects (i.e., cross-situational consistency).
Other researchers constructed their own trust inventories to directly measure self–stated
consistent tendency to trust. These inventories (e.g., Couch, Adams, & Jones, 1996)
include such questions as ―Basically, I am a trusting person.‖ These measures focus on
the consequence of generalized expectancy (i.e., a stable trusting tendency), rather than
the content of generalized expectancy (i.e., beliefs about human nature). Others have
measured the content of generalized expectancy. Gurtman (1992) aggregated diverse
aspects of trusting attitudes toward general human beings. His measure includes
questions about beliefs as to human nature (Wrightsman & Wuescher, 1974), cynicism
(Kanter & Mirvis, 1989), world assumptions (Janoff-Bulman, 1989), and
machiavellianism (Christie & Geis, 1970; Robinson & Shaver, 1973). Yamagishi (1988)
obtained a score of generalized expectancy using a similar approach (e.g., ―given the
opportunity, people will take advantage of you‖). Since the focus of this study is to test
whether the content of generalized expectancy (i.e., the belief that human beings in
general are good) influences trusting judgments in relationships with current, specific
trustees, a content measure of generalized expectancy is more appropriate. Thus, I
selected five questions from the pool of questions (i.e., the content measure) used by
Yamagishi (1988).
Situational Expectancy
To assess situational expectancy, it is necessary to consider situations similar to
those facing the respondents in this study. A defining feature of this context, similar to
54

groups in other organizational settings, is that people need to cooperate with others to
accomplish common goals. A situation in which people must cooperate with others to
accomplish tasks and achieve objectives distinguishes this context from other trust-
related situations such as romantic relationships, relationships with caregivers, or
relationships with other members of a church or other type of social organization. Just as
a person‘s past experiences form a generalized expectancy that is common across diverse
people, whether they are parents, teachers, etc., the experiences of working with others
towards common goals also builds a general attitude toward cooperative situations
whether it is in the classroom, the office, a sports team, etc.
To measure this situation-specific expectancy, a cooperativeness scale was used
for this study, which captures the respondent‘s attitude toward cooperation and the quality
of past experiences regarding cooperation. This five-item scale was developed by Lu and
Argyle (1991) and includes three questions that assess an individual‘s orientation toward
cooperation (I would rather cooperate with others than compete with others; I prefer to
work independently rather than in a group; and I have enjoyed activities that involve a
high level of cooperating with other people); and two that assess past experience of
cooperation (I have found joint projects with other people very satisfying; I believe that
teamwork is the best way of getting results). These questions have been used in a number
of studies that tested its association with behavioral cooperation (Chatman & Barsade,
1995; Yilmaz & Hunt, 2001), happiness and mental health (Lu & Argyle, 1991), and
bullying behavior in secondary school (Rigby, Cox, & Black, 1997).

55

Need for Structure and Decisiveness
As explained in the discussion of the conceptual foundations for this research,
this study uses two different components of the need for cognitive closure scale which
originally reflected five sub-constructs: preference for order, preference for predictability,
discomfort for ambiguity, decisiveness, and closed-mindedness. According to advice by
Neuberg et al. (1997), three of these sub-constructs (i.e., preference for order, preference
for predictability, and discomfort for ambiguity) were combined to measure the need for
structure, while decisiveness was measured separately. The closed-mindedness sub-
contruct was excluded due to its psychometric weakness (Neuberg et al., 1997).
Furthermore, Houghton and Grewal (2000) proposed a reduced scale to measure the need
for cognitive closure, eliminating some of the original 42 items suggested by Webster and
Kruglanski (1994). Based on the results of a factor analysis, their reduced scale contains
four items for each of the five sub-constructs. For the present research, in order to limit
the number of questions included on the survey, the reduced version of the need for
cognitive closure was used. Thus, the average score on twelve items was used to measure
need for structure and the average score on four items was used to measure decisiveness.
The specific questions used are provided in Table 3.
Experiences regarding Trustworthiness
Initial trust can change as trustors have first-hand experiences with which to infer
a trustee‘s ability, integrity, and affection, whether from observed behavior of trustees or
from an evaluation of collaborative outcomes. In order to develop questions regarding
such experiences in relevant to a classroom context, informal interviews of several
56

graduate students at USC were conducted, and they were asked to identify the conditions
that enable them to trust their teammates, and what makes them distrust their teammates.
These conditions were found to be closely related to the aforementioned three factors of
trustworthiness. For example, the ability to make a contribution to the class project was
considered when trusting; free-riding (intention) was always a concern; and an aggressive
take-charge attitude or disrespect for other‘s opinions (affection) led to distrust.
Based on this input, respondents were asked to answer six questions regarding
their experiences with their teammates that could lead them to perceive a trustee as
trustworthy or untrustworthy. These questions were developed to capture both positive
and negative experiences related to the three aspects of trustworthiness. In particular,
respondents were asked: ―During the final project, how many times have you experienced
the following events in relationship with your teammates?‖ Then, six types of events
were specified, including those indicating that the trustee 1) has proper knowledge and
capability necessary to contribute to the project (ability); 2) is willing to do his/her best to
get the job done (intention); 3) respects others‘ opinions (affection); 4) is incapable of
making a contribution to the project (inability); 5) is unwilling to do his/her own part of
the project (ill-intention); and 6) is disrespectful of others (disaffection). Ultimately,
these variables were coded as binary variables indicating whether or not trustors
experienced these events in their relationships with their teammates, rather than coding
the number of events.
6
These variables were used in the analysis of the second model

6
Many respondents answered this open-ended question by indicating there were ―many,‖ ―multiple,‖ and
―a lot‖ of such experiences. Thus, it was hard to quantify the number of these events, such that I ended up
using a binary variable that simply indicated the presence or absence of any such event.
57

(i.e., the stability phase) as dependent variables, and as predictor variables in the analysis
of the third model (i.e., the dissolution phase).
Social Distance
A question was included on the questionnaire asking respondents to identify the
nature of their relationship with each teammate before they started working together in
their group. This question asked them to choose from among the following options: 1)
we never met before; 2) I know his/her name and face; 3) I talked to him/her a couple of
times; 4) I am somewhat close to him/her; and 5) we are very close to each other. I then
dichotomized this variable by differentiating between friend dyads (those who were
somewhat close or very close to each other) and stranger dyads (those who at most had
talked to each other once or twice). Since the first model was intended to investigate the
formation of initial trust, the analysis of that model focused only on those dyads in which
the relationship between teammates was just beginning, i.e., the stranger dyads. Likewise,
the second model was examined for stranger relationships and friend relationships
separately, since the focus of this study is to examine how initial trust is formed between
strangers and how that initial trust is sustained. The third model was also examined for
stranger relationships and friend relationships separately, since the effects of trustor
attributes (e.g., NFC) on trust dissolution after negative experiences will differ between
stranger relationships and friend relationships.
Control V ariables
The literature suggests that, when people meet strangers, the initial level of trust
can be influenced by readily available signals as to the trustee‘s trustworthiness such as
58

gender (Orbell, Dawes, & Schwartz-Shea, 1994) and ethnicity (Greenwald & Banaji,
1995; McConahay, 1983). Gender and ethnicity can affect initial trust for two reasons:
similarity and stereotyping. Zucker (1986) describes this as ‗character-based trust‘ since a
trustor trusts trustees who are socially similar to him/herself. Thus, I included similarity
between trustor and trustee in terms of gender and ethnicity as control variables. Two
dummy indicators were used to denote whether the trustor is the same gender as the
trustee and whether the trustor is in the same ethnic group as the trustee (coded 1 if the
same, 0 otherwise). In addition to similarity between actors, initial decisions regarding
whether or not to trust a stranger may be based on a categorization process (McKnight et
al., 1998), in which trustees in a favorable category such as male and white are trusted
more readily. Thus, I also controlled for trustee gender and ethnicity. Dummy variables
were used to indicate whether a trustee is female (coded 1, 0 if male), and whether a
trustee is white (coded 1, 0 if non-white). For the dissolution phase models, the initial
trust level may influence the level of trust measured at the end for semester. Thus, initial
level of trust was also included as control variable. Table 3 summarizes the items used for
all the measures included in the analyses.

59

Table 3. Items in Scale
Variable Questions
Trust
If I have to choose one partner to work with in this class,
and if I choose this person, we would be able to work
together in pleasant and productive manner.
If I am to distribute the roles for the final project and let
this person do the most important part, s/he will get the job
done satisfactorily.
If I ask this person for help when I have difficulty doing
my role in the negotiation, s/he will respond with care and
attentiveness.
If I let this person be a leader who makes important
decisions on behalf of our group, the performance of our
group will be successful.
If I freely share my ideas with this person, s/he will not
negatively criticize it. Rather, s/he will make constructive
comments on it.
Even if I do not monitor this person, s/he will do the best to
get the job done.
GE
People are always interested only in their own welfare. (R)
Given the opportunity, people will take advantage of you
when there is a benefit of doing so. (R)
One can avoid falling into trouble by assuming that all
people have a vicious streak. (R)
In dealing with strangers, one is better off to be cautious
until they have provided evidence that they are trustworthy.
(R)
No matter what they say, most people inwardly dislike
putting themselves out to help others. (R)
SE
I would rather cooperate with others than compete with
others.
I prefer to work independently rather than in a group. (R)
I have enjoyed activities that involve a high level of
cooperating with other people.
I have found joint projects with other people very
satisfying.
I believe that teamwork is the best way of getting results.
NFC Decisiveness
I tend to struggle with most decisions. (R)
I would describe myself as indecisive. (R)
I tend to put off important decisions until the last possible
moment. (R)
I usually make important decisions quickly and confidently.
60

Table 3. Continued
NFC
Preference
for Order
I find that a well-ordered life with regular hours suits my
temperament.
I find that establishing a consistent routine enables me to
enjoy life more.
I enjoy having a clear and structured way of life.
I like to have a place for everything and everything in its
place.
Preference
for
Predictability
I dislike unpredictable situations.
I do not like to be with people who are capable of
unexpected actions.
I prefer to socialize with familiar friends because I know
what to expect from them.
I enjoy the uncertainty of going into a new situation
without knowing what might happen. (R)
Discomfort
for
Ambiguity
I feel uncomfortable when someone‘s meaning or intention
is unclear to me.
When I am confused about an important issue, I feel very
upset.
I dislike it when a person‘s statement can mean more than
one thing.
I feel uncomfortable when I do not understand the reason
why an event occurred in my life.
During this final project, how many times have you experienced the events such as
below in relationships with your teammates?
Positive
Experience
Ability Events showing that this person has proper knowledge and
capability necessary to contribute to the negotiation.
Intention Events showing that this person is willing to do his/her best
to get the job done.
Affection Events showing that this person respects others‘ opinions.

Negative
Experience
Ability Events showing that this person is incapable of making a
contribution to the negotiation.
Intention Events showing that this person is unwilling to do his/her
own part of the negotiation.
Affection Events showing that this person is disrespectful of others.

R denotes the variables are reverse coded.

61

Research Design and Analyses
Data Structure
This study conceptualizes trust as a dyadic phenomenon – a trustor‘s positive
expectation toward a specific trustee – which allows a simultaneous examination of the
effects of multiple types of antecedents separately, i.e., the attributes of trustors, trustees,
and their relationships. Wellman and Frank (2001) provided a detailed instruction and a
good example of how to investigate these multiple types (levels) of antecedents for dyad-
level phenomena. Using what is called the ego-centered network structure (Everett &
Borgatti, 2005), they investigated the effects of multitude of predictors of everyday social
support. The predictors in their study were the gender of the recipient (ego) of social
support, the size of the recipient‘s social network, the gender of social support providers
(alters), and the strength of the relationship between social support recipients and
providers. Chua, Ingram, and Morris (2008) also used the ego-network structure to
analyze the effects of multiple types (levels) of antecedents to affect-based and cognitive-
based trust.
The dyadic nature of the data for this study creates a data structure hierarchically
nested within trustors, as demonstrated in Figure 4. As indicated in the table presented
with Figure 4, since each respondent evaluates each teammate in their group, scores on
the measures of the trustor attributes do not vary within trustor, i.e., they are the same for
each trustee described by a given trustor (e.g., 23 for Trustor 1 and 18 for Trustor 2). In
contrast, the scores on the dependent variable do vary within trustor, i.e., each trustee
62

receives a different score (e.g., 14, 17, and 13 from Trustor 1, and 16, 19, 21 from Trustor
2). As a result, the separate trust scores provided by the same trustor are not independent
observations. In the example in the table, the trust scores given by Trustor 2 are
consistently higher than those given by Trustor 1. When the data are nested within higher
units, the cases that belong to the same higher unit tend to score similarly since these
cases share the same characteristics attributable to the focal higher unit. In this case,
conventional inferential statistical analyses are inappropriate. In other words, if the
intercept and coefficients are calculated assuming the data are measured at a single level,
the error terms will be correlated among the cases (i.e., the dyads) nested within trustors.
To resolve this problem, I analyzed the data using ordinal hierarchical linear modeling
(HLM). HLM calculates the parameters at multiple levels separately, based on maximum
likelihood estimates. The analytic process used in HLM repeats an assessment of the
intercept and coefficients until it finds parameter estimates that minimally increase the
likelihood function. That way, it yields unbiased estimates of the coefficients. The
parameters were estimated using the Restricted Maximum Likelihood method (RML),
and all the variables except for the dummy variables were centered around the grand
mean.

63

Figure 4. Ego-network Data Structure in the Multilevel Analysis

Case ID TrustorID TrusteeID Trustor
attributes(IV
)
Trust(DV)
1 1 1 23 14
2 1 2 23 17
3 1 3 23 13
4 2 4 18 16
5 2 5 18 19
6 2 6 18 21

Trust, the dependent variable in this study, is a level 1 variable in the hierarchical
structure, as are the control variables identifying characteristics of the trustees (e.g.,
gender, age, and ethnicity). The principal independent variables are trustor attributes,
which are level 2 variables. It is important to note here that these levels refer to levels in
the data structure (see Figure 4), rather than conceptual levels (i.e., the three levels of
trustor, trustee, relationship). Inclusion of these conceptually distinct levels by itself
would not require the use of HLM. The need to differentiate between levels 1 and 2 in the
data structure is to address the methodological issue, that is, to resolve the problem of
correlation among the cases. If this problem were not a concern, data from all three
conceptual levels (whether it is propensity to trust, trustworthiness, or relationship
Trustor 1 Trustor 2
Trustee 1 Trustee 2

Trustee 3 Trustee 4 Trustee 5 Trustee 6
Trust 4 Trust 5
Trust 2
Trust 6
Trust 3 Trust 1
64

characteristics) could be treated as a single data level and analyzed using conventional
ordinary least squares (OLS) regression.
Model Testing
I tested three models to examine the study hypotheses, each representing one
phase of trust development: building, stability, and dissolution (Rousseau et al., 1998).
The first model examined Hypotheses 1 and 2. The dependent variable was Trust1 and
the independent variables were GE and SE. Since the first model was intended to
investigate the formation of initial trust, this analysis focused only on those dyads in
which the relationship between teammates was just beginning. Thus, for Model 1, I
divided the sample into two groups, namely, stranger-relationships and friend-
relationships. Since the primary focus for this analysis lies on the stranger relationships,
only these were used to test Hypotheses 1 and 2, while the analysis of the friend
relationships is presented for comparison purposes.
The second model tested Hypotheses 3-1 through 5-2. There were six dependent
variables for this model, namely, the dummy variables indicating whether respondents
had experienced positive and negative events regarding three different types of
trustworthiness, namely, ability, intention, and affection. The independent variables for
this model are GE, SE, and Trust1, with NFC serving as a moderator to examine the
interaction effect between Trust1 and the dependent variables. Because the dependent
variables are binary variables, this second model was tested using Logit HLM.
The third model tested Hypotheses 6-1 and 6-2. As discussed earlier, there is no
clear basis for predicting whether NFC will increase or decrease the amount of trust
65

change after experiencing negative events. Thus, as a preliminary analysis, one-way
ANOVA was performed to investigate whether there are any differences in the amount of
trust decline among the groups of low, medium, and high NFC. The results of this
ANOVA provide information regarding how a trustor‘s NFC generally impacts trust
decline when relevant conditions (i.e., initial trust is high between strangers or between
friends) are not specified. For this preliminary analysis, the dependent variable was trust
change, as calculated by subtracting Trust1 from Trust2. A negative value for this variable
means that trust declined over time, while a positive value means that trust increased. The
analysis focused only on those dyadic relationships in which respondents experienced
negative events with a trustee. Out of the 311 dyads in the sample, there were 106
relationships in which negative ability-related events were experienced, 48 relationships
in which negative intention-related events were experienced, 45 relationships in which
negative affection-related events were experienced. These dyads were divided into three
groups representing respondents with low, medium, and high NFC, respectively. The low
NFC group included respondents whose NFC scores were less than the mean value minus
one standard deviation. The medium NFC group consisted of those respondents whose
NFC scores were between one standard deviation above and below the mean. The high
NFC group was comprised of those whose NFC scores were more than one standard
deviation above the mean.
Subsequently, HLM was used to test Hypotheses 6-1 and 6-2 which focus on
relationships between variables in a limited set of conditions, i.e., high trust between
strangers and the effect of decisiveness in the former, and high trust between friends and
66

the effect of need for structure in the latter. The examination of Hypotheses 6-1 and 6-2
seeks to determine whether or not the effects of negative experiences differ depending on
trustors‘ NFC, assuming that high trust has already been developed in a relationship
through the building and stability phases. In order to test the two hypotheses, Trust2 was
used as the dependent variable, and only the dyadic relationships in which Trust1 is high
were included in the analysis. The purpose for doing this was to help insure that the
experience of negative events with a given trustee would create some degree of cognitive
dissonance on the part of the trustor. In experimental studies examining the effects of
NFC on attitude change, conditions can be established in the laboratory that make it very
likely that subjects will face some degree of such dissonance that will in turn invoke the
freezing and/or seizing effect the experimenter is interested in studying (e.g., Krugalnski
et al., 1993; Stalder, 2010). Since the conditions needed to establish such dissonance
could not be manipulated in the context of the field-based research adopted for the
present study, the best way to guarantee that the respondents are confronted with some
unpleasant cognitive discomfort is to focus on those relationships in which a high level of
trust had already been established prior to the experience of negative events that run
counter to the existing trust level. In other words, if Trust1 was high, it can be
reasonably assumed that any negative events experienced during their collaboration will
be perceived as contradicting the positive perceptions already held by the trustors.
Considering the six-point scale used to measure trust, the relationships in which
Trust1 was rated higher than 5.0 were included in the analysis. The number of these high
trust relationships is 144 for strangers, and 70 for friends. The initial trust level (Trust1)
67

was included as a control variable, since it may influence the level of trust measured at
the end of the semester. The independent variables were the three different types (i.e.,
ability, intention, and affection) of negative experiences. The moderating effects of
decisiveness, for Hypothesis 6-1, and need for structure, for hypothesis 6-2, were also
analyzed by use of interaction terms. The cutoff alpha level for this study was .10
throughout.

68

CHAPTER FIVE: RESULTS

In this chapter, I first provide an overview of the data, summarizing information
about the reliabilities of the measures, the distributions of the variables, sample
characteristics, descriptive statistics, and the correlation matrix for the full sample.
Second, I present a number of preliminary analyses to provide an overall picture of the
level of trust in the sample, before investigating the determinants of trust in subsequent
sections in which the three models are tested. This section examines whether trust
increased after collaborating during the semester, whether trust between friends is greater
than trust between strangers, whether there is any difference between strangers and
friends in terms of the probability of positive and negative experiences they had while
working together, and whether, in general, NFC influences the degree of trust decline
when negative experiences are encountered. These analyses contribute to a general
understanding of the trust dynamics in these relationships that is not provided by the
model testing analyses. Third, I test the hypotheses suggested in the Hypotheses chapter.
Since different sub-samples (e.g., strangers vs. friends, high initial trust vs. low initial
trust) are used to test these models, and the variables used for each study are different, it
is necessary to provide the descriptive statistics and correlation matrix for each analysis
separately. Finally, I summarize the results of the hypothesis testing in a table.

69

Overview of the Data
Reliabilities
The measurements of the independent and dependent variables in this study are
reliable. The Cronbach‘s Alpha values for the variables are at least acceptable, ranging
between 0.675 and 0.936. Table 4 provides a summary of the reliability scores.

Table 4. Reliabilities of the Measures
Variable Number of Items Cronbach ’s Alpha
Generalized Expectancy (GE) 5 .675
Situational Expectancy (SE) 5 .806
Decisiveness 4 .758
Need for Structure (NFS) 12 .761
Trust1 6 .924
Trust2 6 .936

Normality of Distributions
Most variables in the sample were distributed normally. However, the
distributions for Trust1 and Trust2 exhibited a negative skew. The skewed distribution for
Trust1 reflects the fact that, on average, most people start with more than a moderate
level of trust prior to working together, rather than with a neutral or distrusting stance.
The skewed distribution of these variables fails to meet the necessary conditions for
running a regression analysis, including HLM, which could bias the results. Thus, these
two variables were transformed using the formula
7
that Tabachnick and Fidell (2007:
p.89) suggested for use when the distribution is J-shaped.

7
New X = 1 / (K – X), where K = a constant from which each score is subtracted so that the smallest score
is 1; usually equal to the largest score +1.
70

Descriptive Statistics
Table 5 provides the descriptive statistics and correlation matrix for the variables
in this study, and Table 6 provides information regarding the profile of the sample.
Table 5. Descriptive Statistics and Correlation Matrix for Full Sample
Variable N Mean Std. Dev. Minimum Maximum
Trust1
1

311 5.108 .875 1.000 6.000
Trust2
1

311 5.309 .873 1.333 6.000
GE
311 3.381 .715 1.800 5.600
SE
311 4.372 .697 2.600 6.000
Decisiveness
311 3.888 .874 1.250 5.500
NFS
311 4.003 .552 2.670 5.583
Trust Change
2

311 .201 .860 -3.167 5.000
1. The statistics for Trust1 and Trust2 are calculated based on the original, not transformed,
variables.
2. Trust Change is also based on the original, not transformed, Trust 1 and Trust 2.
Trust1 Trust2 GE SE Decisiveness NFS
Trust1

Trust2
.516
GE
.123 .091
SE
.165 .251 -.020
Decisiveness
-.090 -.062 -.042 -.208
NFS
Trust Change
.154
-.493
.091
.491
-.199
-.033
.104
.087
-.343
.029 -.065
Bold italic font indicates that the correlation is significant at the .10 level (2-tailed).

Table 6. Sample Profile Information
Average Age
(Std. Dev.)
28.18
(7.977)
Frequency Percent
Gender Male 113 36.3
Female
198 63.7
Ethnicity White 109 21.1
Black 19 3.7
Latino 17 3.3
Asian 85 16.4
Indian 42 8.1
Other 34 6.6
71

Table 6. Continued
Gender
Similarity
1
Different 119 38.3
Same 112 39.2
Missing 70 22.5
Ethnic
Similarity
1
Different 171 55.0
Same 58 18.6
Missing 81 26.4
Social
Distance
2

1 37 11.9
2 36 11.6
3 153 49.2
4 70 22.5
5 15 4.8
Social Distance
(Dichotomized)
Strangers (1,2,3)
226 72.7
Friends (4,5)
85 27.3
1. There is a relatively high level of missing data for gender and ethnic similarity because not
all trustees responded to the survey. For those trustees who did not respond, gender and
ethnicity information was unavailable.
2. As a reminder, the categories for social distance are: 1) never met, 2) know his name and
face, 3) talked to him once or twice, 4) somewhat close to each other, 5) very close to each
other.

Preliminary Analyses
Table 7 provides the descriptive statistics for dyads of stranger-relationships and
those of friend-relationships separately. This information is provided as a basis for the
preliminary analyses.
Table 7. Descriptive Statistics for Sub-samples
Variable N Min Max Mean
Std.
Dev.
Stranger
Relationship
Trust 1 226 2.333 6.000 5.035 .823
Trust 2 226 1.333 6.000 5.178 .930
GE 226 1.800 5.600 3.363 .650
SE 226 2.600 6.000 4.279 .694
Decisiveness 226 1.250 5.500 3.951 .850
NFS 226 2.670 5.583 4.001 .552
Trust Change 226 -3.167 3.667 .144 .849
72

Table 7. Continued
Friend
Relationship
Trust 1 85 1.000 6.000 5.304 .977
Trust 2 85 3.667 6.000 5.657 .571
GE 85 1.800 5.600 3.431 .867
SE 85 3.200 5.600 4.619 .647
Decisiveness 85 1.750 5.000 3.721 .921
NFS 85 2.920 5.500 4.010 .555
Trust Change 85 -1.333 5.000 .353 .874

An independent sample T-test reveals that the average level of initial trust was
significantly higher in relationships between friends than between strangers (5.045 for
strangers vs. 5.304 for friends: t=-2.436 and p=.015). This indicates that the respondents
tend to start their relationships with a considerable amount of trust even when they have
no interaction history with a focal trustee. In over 85 percent of the dyadic relationships
between strangers in the sample, trustors started with trust scores higher than 4.0, which
corresponds to their assessment that they ―somewhat trust‖ the trustee. As for initial trust
in relationships with friends, it is not surprising that the average level of trust prior to
starting their class collaboration is even higher than it is among strangers.
There is slight increase in trust from before to after working together in groups.
The amount of trust change was .133 for stranger-relationships, and .353 for friend-
relationships. A paired sample T-test reveals that these slight increases are statistically
significant for both stranger-relationships (t=-2.541, p=.012) and friend-relationships (t=-
3.722, p=.001), suggesting that the differences in the before and after levels of trust
occurred for reasons other than chance (e.g., trustor‘s propensity, positive and negative
73

experiences, etc.). These results are consistent with the argument (Putnam, 1993) that
interactions between actors generally tends to increase the overall level of trust. An
independent sample t-test on the trust change variable reveals that the amount of trust
increase in stranger-relationships (.144) was not significantly different from that of
friend-relationships (.353) (t=-1.923, p=.055), such that the observed difference in trust
change scores between stranger- and friend-relationships may have happened by chance.
Table 8 presents the number of relationships where the three types of positive
and negative events were experienced. The number of dyads in which trustors
experienced at least one positive event in their relationship with a given trustee is
considerably greater than the number of relationships with no positive events. For
negative experiences, this pattern is reversed. The number of relationships in which
trustors experienced at least one negative event is lower than those with no negative
events. Respondents in this sample experienced many more positive than negative events
in their class collaboration during the semester.
Table 8. Frequency Table for Dichotomized Positive and Negative Experiences

U.S.
Strangers Friends
Yes No Yes No
Positive
Experience
Ability 222 4 81 4
1
Intention 219 7 81 4
1
Affection 219 7 81 4
1
Negative
Experience
Ability 75 151 31 54
Intention 44 182 4 81
Affection 40 186 5 80
1. It was the same four dyads that did not experience any of the three types of positive events.

Table 9 shows the comparison between stranger-relationships and friend-
74

relationships, in terms of the probability of experiencing events corresponding to each of
the six categories. Respondents in the two types of dyads exhibit no significant
differences in their probability of experiencing all three types of positive events. However,
dyads of strangers are significantly more likely to experience two types of negative
events than those involving friends, namely, events regarding intention (t=4.208, p=.000)
and affection (t=3.269, p=.001), with no difference in those regarding ability.
Table 9. T-tests for Comparing the Probability of Experiences
Stranger vs. Friends
1

Positive
Experience
2
Ability 1.188
Intention .683
Affection .683
Negative
Experience
3
Ability -.543
Intention 4.208***
Affection 3.269***

1. Independent sample T-tests were performed for the dichotomized positive and negative event
variables for each of the three types of trustworthiness (ability, integrity, and emotion).
2. A significant positive t-value indicates that the probability that stranger-relationships
experience a positive event is significantly greater than that of friend-relationships, while a
significant and negative t-value indicates that the probability that friend-relationships
experience a positive event is significantly greater than that of stranger-relationships.
3. A significant positive t-value indicates that the probability that stranger-relationships
experience a negative event is significantly greater than that of friend-relationships, while a
significant and negative t-value indicates that the probability that friend-relationships
experience a negative event is significantly greater than that of stranger-relationships.
*p<.10
**p<.05
***p<.01

Table 10 below shows the results of a one-way ANOV A on Trust Change. As
discussed in the Hypotheses chapter, there is no clear basis for predicting whether NFC
will lead to the primacy effect of initial trust or the recency effect of negative experiences.
Thus, this ANOVA result explores which of these two effects is manifested when there
75

are no conditions as in Hypotheses 6-1 and 6-2. Looking at the mean values for each
group, the amount of trust decline (i.e., a negative value on the variable) is greatest for
the high NFS group and lowest for the low NFS group after experiencing all three types
of negative events. Although there was no significant difference in trust change among
the Decisiveness groups after experiencing negative events regarding affection, the high
Decisiveness group demonstrated the most trust decline after both of the other two types
of negative experiences.
Table 10. Results of ANOV A on Trust Change
The Effect of Negative Ability Experience on Trust Change
Group Mean F Sig.
Low Decisiveness
.019 13.369 .000
Medium Decisiveness
.032

High Decisiveness
-1.625

Low NFS
.222 19.112 .000
Medium NFS
.059

High NFS
-1.028

The Effect of Negative Intention Experience on Trust Change
Group Mean F Sig.
Low Decisiveness
-.050 6.268 .004
Medium Decisiveness
-.187

High Decisiveness -1.500

76

Table 10. Continued
Low NFS
.095 15.338 .000
Medium NFS
-.119

High NFS
-1.778

The Effect of Negative Affection Experience on Trust Change
Group Mean F Sig.
Low Decisiveness
-.361 .510 .604
Medium Decisiveness
-.118

High Decisiveness .167

Low NFS
.271 5.848 .006
Medium NFS
-.157

High NFS
-1.500

Model Testing
Model 1. Building Phase
Descriptive Statistics and Correlations
Table 11 provides the descriptive statistics and correlation matrices for the
variables used to test model 1. The correlations among the independent variables are
acceptable, for both stranger-relationships and friend-relationships, ranging from .006
to .271, and from .014 to .359, respectively.
77

Table 11. Descriptive Statistics and Correlation Matrices for Sub-samples
Stranger-relationships

Mean
S.D. 1 2 3 4 5 6
1.Trust1 5.035 .823
2.SameSex .468 .500 .191
3.SameRace .184 .389 .078 .141
4.Woman .676 .469 .063 .253 -.123
5.White .342 .476 .128 -.036 -.019 -.094
6.GE 3.363 .650 .087 .271 -.009 -.030 .032
7.SE 4.279 .694 .221 -.003 .030 .201 -.086 -.098
Bold italic font indicates that the correlation is significant at the .10 level (2-tailed).

Friend-relationships

Mean
S.D. 1 2 3 4 5 6
1.Trust1
5.304 .977

2.SameSex
.603 .493
-.103
3.SameRace
.424 .498
.025 .215
4.Woman
.588 .496
.028 .359 .190
5.White
.415 .497
-.010 .014 -.103 -.268
6.GE
3.431 .867
.172 -.128 -.185 -.148 -.016
7.SE
4.619 .647
-.054 -.093 -.028 -.120 .205 .117
Bold italic font indicates that the correlation is significant at the .10 level (2-tailed).

Model Summary
In the base model of stranger-relationships, which includes no predictors, the
between-unit variance (among trustors) is .034 and the within-unit variance is .022. Thus,
the variance in the level of trust among trustors (i.e., the between-unit variance) relative
to the total variance is .562 – over 56 percent of the variance in scores is explained by
differences among the trustors (although particular factors influencing this variance are
not yet specified in this model). For the base model between friends, the between-unit
variance is .036 and the within-unit variance is .025. Thus, the variance in the level of
trust among trustors relative to the total variance is .616 – almost 62 of the variance is
explained by trustor differences. Conversely, it also indicates that about 40 percent of
78

total variance in both sets of dyads (i.e., 44 percent for stranger-relationships, and 38
percent for friend-relationships) is explained by trustee attributes (i.e. trustworthiness)
and relationship characteristics. Therefore, it suggests that trustor attributes have more
influence than the other two categories of factors. This is consistent with the argument
(Mayer et al., 1995) that propensity to trust is a more significant predictor of trust in the
initial stage of relationship.
The Chi-squares for the base models were also significant (412.66, p=.000 for
stranger-relationships; 148.81, p=.000 for friend-relationships), indicating that the
between-unit variances presented above are significantly different from zero. This finding
indicates that the differences in the initial level of trust among trustors (higher level units)
still need to be explained. Since none of the predictor variables were included in the base
models for the two types of dyads, the significant Chi-square values mean that
differences among trustors do exist and should be examined further.
Tabanich and Fidell (2007) referred to this Chi-square value as ‗intraclass
correlation,‘ and proposed that it can be an indicator of whether HLM is required rather
than OLS analysis. A high intraclass correlation implies that the error terms at level 1 are
correlated with each other, which deflates the p-value for each coefficient. As a result, the
significance of the coefficients calculated using OLS without taking the level 1 error
terms into consideration would be biased, such that HLM is the preferred approach for
further analysis of the data.
Hypothesis Testing
Table 12 presents the HLM results for the level of initial trust between strangers
79

and between friends. First, none of the control variables were significantly related to
initial trust level in either group. Second, Hypothesis 1 (the effect of GE on initial trust) is
supported in the friend-relationships model (t=1.830, p=.076), but not in the model for
stranger-relationships (t=.361, p=.719). In contrast, Hypothesis 2 (the effect of SE on
initial trust) is supported in the stranger-relationships model (t=1.824, p=.074), but not
supported in the friend-relationships model (t= -.295, p=.770).

Table 12. Results of HLM on the Level of Initial Trust
Stranger-Relationships Friend-Relationships
Intercept .564***
(.037)
.456
(.288)
Controls
(level 1)

Same Gender .030
(.037)
-.017
(.054)
Same Race .033
(.042)
.059
(.050)
Woman -.003
(.036)
.064
(.055)
White .053
(.032)
.058
(.050)
Main Effects (level
2)

GE
(H1)
.021
(.047)
.080*
(.044)
SE
(H2)
.073*
(.040)
-.016
(.053)
Variance
Composition
(Random Effect)

Intercept .029 .028
Error term .026 .028
Model
Summary

ᵡ2 221.082*** 84.804***
The numbers in parentheses are standard errors.
*p<.10
**p<.05
***p<.01
80

Model 2. Stability Phase
Descriptive Statistics and Correlation Matrices
Table 13 shows the descriptive statistics and correlation matrices for the
variables used to test model 2. The correlation coefficient between dichotomized negative
intention and negative affection is .797 for stranger-relationships and .889 for friend-
relationships. Studenmund (2006) advised that researchers should drop one of two
correlated variables when they are redundant, but do nothing when the two correlated
variables are theoretically distinctive. In the present case, neither of the variables should
be discarded because negative experiences regarding intentions are conceptually
distinguished from those regarding affection, and thus there is no redundancy. Instead,
each type of negative experiences will be analyzed in a separate model in order to keep
the correlated variables from rendering their coefficients unreliable.
For the friend-relationships, the three types of positive experiences are perfectly
correlated with each other (r=1.0). Furthermore, almost all the dyads in both groups
experienced positive events regarding ability, intention, and affection (see the mean
values for +ability, +intention, +affection for both stranger- and friend-relationships). In
this case, the analysis of the three positive experience variables is likely to be neither
meaningful nor reliable. Thus, the analysis is done only on the three types of negative
experiences (H3-2, H4-2, and H5-2).

81

Table 13. Descriptive Statistics and Correlation Matrices for Sub-samples
Stranger-Relationships

Mean SD 1 2 3 4 5 6 7 8 9
1. +Ability
.982 .132
2. +Intention .969 .174 .363
3. +Affection .969 .174 .170 .263
4. –Ability .332 .472 .023 .072 .072
5. –Intention .195 .397 -.104 -.170 -.041 .579
6. –Affection .177 .383 -.026 -.185 -.051 .436 .797
7. Trust1 5.035 .823 .251 .132 .220 -.053 -.196 -.163
8. SE 4.2 79 .694 -.072 -.057 .094 -.246 -.140 -.140 .221
9. GE 3.363 .650 .065 -.034 .013 .023 -.068 .009 .087 -.098
10. NFS 4.001 .552 .102 .049 .058 .088 -.065 -.1 13 .129 .123 -.199
Bold italic font indicates that the correlation is significant at the .10 level (2-tailed).

Friend-Relationships

Mean SD 1 2 3 4 5 6 7 8 9
1. +Ability
.953 .213
2. +Intention .953 .213 1.000
3. +Affection .953 .213 1.000 1.000
4. –Ability .365 .484 .168 .168 .168
5. –Intention .047 .213 .049 .049 .049 .293
6. –Affection .059 .237 .056 .056 .056 .226 .889
7. Trust1 5.304 .977 .003 .003 .003 .170 -.012 -.173
8. SE 4.619 .647 -.063 -.063 -.063 -.265 -.214 -.194 -.054
9. GE 3.431 .867 -.031 -.031 -.031 .466 -.059 -.067 .172 .1 17
10. NFS 4.010 .555 .147 .147 .147 -.255 -.021 -.095 .212 .054 -.206
Bold italic font indicates that the correlation is significant at the 0.10 level (2-tailed).

Hypothesis Testing
The results of the analysis to test model 2 are presented in Table 14. First,
Hypotheses 3-2 and 5-2 were not significantly supported for any of the three types of
negative experiences. Second, Hypotheses 4-2 was not supported in relationships with
strangers at all. However, it was partially supported in relationships with friends, as the
effects of SE on the probability of experiencing negative ability-related (t=-1.735,
82

p=.091) and affection-related (t=-1.879, p=.068) events were significant.
8

Table 14. Results of Logit HLM on Probability of Negative Experiences
Stranger-Relationships Friend-Relationships
-Ability -Intention -Affection -Ability -Intention -Affection
Intercept - .888 **
(.374)
- 1.735***
(.355)
- 1.756***
(.338)
- 1.068**
(.44 7)
- 3.074***
(.563)
- 3.085***
(.572)
Main
Effects

Trust1 - .292
( .9 82)
- 2.463
(1.168)
- 1.520
(1.11 6)
.267
(1.49 3)
- 1.545
(2.207)
- 2.804
(2.00 7)
GE
(H3-2)
.150
(.554)
- .338
(.554)
.088
(.517)
.897
(.597)
- .417
(.78 1)
- .487
(.807)
SE
(H4-2)
- .663
(.565)
- .341
(.540)
- .379
(.511)
- 1.214*
(.700)
- 1.363
(.836)
- 1.516*
(.807)
Interaction
Effects

Trust1
* NFS
(H5-2)
- .402
(1.71 1)
- .363
(1.957)
.830
(1.912)
- .289
(2.717)
- 5.505
(4.70 2)
1.055
(.54 4)
V ariance
Composition

Intercept 7.091 5.309 4.789 4.257 2.149 1.415

Model
Summary

ᵡ2
179.493*** 146.463*** 138.431*** 62.428*** 22.412 35.713
The numbers in parentheses are standard errors.
*p<.10
**p<.05
***p<.01

Model 3. Dissolution Phase
Descriptive Statistics and Correlations
Table 15 includes the descriptive statistics and correlation matrices for the

8
The effect of SE on the probability of negative intention-related experiences (in friend-relationships) was
nearly significant, with a t-value of -1.629, p=.111.
83

dependent and independent variables included in model 3. As indicated in the table, there
is high multi-collinearity among the independent variables. In the stranger-relationships
sub-sample, the correlation coefficient was .722 between negative experience of intention
and that of affection; in the friend-relationships sub-sample, there was a perfect
correlation (1.000) between negative intention and negative affection experiences. Given
this multi-collinearity, it is necessary to address this issue in order to test the model. As
a solution, the main effects and interaction effects of each type of negative experience
will be tested separately so that they do not interfere with each other in the analysis.

Table 15. Descriptive Statistics and Correlation Matrices for Sub-samples
Stranger-relationships

Mean
S.D. 1 2 3 4 5 6
1. Trust2 5.542 .381
2. Trust1 5.497 .701
.350

3. –Ability .292 .456 -.070 -.099
4. –Intention .153 .361
-.343 -.367 .577

5. –Affection .139 .347
-.264 -.271 .405 .722

6. Decisiveness 3.911 .810
.099 .066 .075 .071
-.074
7. NFS 4.022 .488 -.103
-.152 .028 .003
-.077
-.268
Bold italic font indicates that the correlation is significant at the 0.10 level (2-tailed).

Friend-relationships

Mean
S.D. 1 2 3 4 5 6
1. Trust2 5.662 .328
2. Trust1 5.786 .389 .408
3. –Ability .400 .493 .057 -.038
4. –Intention .043 .204 -.069 -.339 .259
5. –Affection .043 .204 -.069 -.339 .259 1.000
6. Decisiveness 3.707 .932 -.042 -.197 -.049 .219 .219
7. NFS 4.037 .575 .185 .058 -.331 -.065 -.065 -.270
Bold italic font indicates that the correlation is significant at the 0.10 level (2-tailed).
84

Model Summary
In the base model of stranger-relationships, which includes no predictors, the
between-unit variance (among trustors) is .036 and the within-unit variance is 0.020.
Thus, the variance in the level of trust among trustors (i.e. the between-unit variance)
relative to the total variance is .638 – over 63 percent of the variance in scores is
explained by differences among the trustors. For the base model between friends, the
between-unit variance is .005, and the within-unit variance is .025. Thus, the variance in
the level of trust among trustors relative to the total variance is .169 – only about 17
percent of the variance is explained by trustor differences. Conversely, this also indicates
that, for this group, about 83 percent of the total variance is explained by trustee
attributes (e.g., trustworthiness) and relationship characteristics. Furthermore, the Chi-
squares for the base models were significant both for stranger-relationships (293.034,
p=.000) and friend-relationships (43.449, p=.085).
Hypothesis Testing
The results of the test of Hypothesis 6-1 are presented in Table 16. Three
different models were tested, one for each of the three types of negative experiences.
Model 3-1 tests Hypothesis 6-1 for Negative Ability, model 3-2 for Negative Intention,
and model 3-3 for Negative Affection. First of all, the level of Trust 1, the control
variable, significantly influenced the level of Trust 2 for all types of negative events,
indicating that initial trust level explains unique and significant variance in the level of
subsequent trust. Second, all types of negative experiences significantly affected the level
of Trust 2, after controlling for Trust 1. Third, Decisiveness did not significantly
85

moderate the negative relationship between negative experiences and the level of Trust 2,
leaving Hypothesis 6-1 unsupported.

Table 16. Results of HLM on Trust 2 for stranger-relationships
Trust 2 between Strangers
Model 3-1 Model 3-2 Model 3-3
Intercept
.821***
(.036)
.816***
(.029)
.800***
(.028)
Control
Trust1
.430***
(.108)
.296***
(.103)
.382***
(.092)
Main Effects
Negative Ability
-.192**
(.085)

Negative Intention
-.350***
(.057)

Negative Affection
-.232***
(.060)
Moderators
Decisiveness
- .012
(.056)
.003
(.059)
.022
(.060)
Variance Composition
Intercept .036 .034 .028
Error term .015 .011 .016

Model Summary
ᵡ2 403.544*** 441.937*** 314.375***
The numbers in parentheses are standard errors.
*p<.10
**p<.05
***p<.01

The same analysis was performed on the relationships between friends, the
results of which are shown in Table 17. Again, three different models were tested for each
of the three types of negative experiences. Model 3-4 tests Hypothesis 6-2 for Negative
Ability, model 3-5 for Negative Intention, and model 3-6 for Negative Affection. First,
the level of Trust 1, the control variable, significantly influenced the level of Trust 2 for
86

all types of negative events, indicating that initial trust level explains unique and
significant variance in the level of subsequent trust. Second, after controlling for the
effect of initial trust, all three types of negative experiences significantly decreased the
level of Trust 2. Third, Need for Structure significantly moderated the relationships
between negative intention- and affection-related experiences and the level of Trust 2,
suggesting the need for further exploration of the nature of these moderating effects.

Table 17. Results of HLM on Trust 2 for friend-relationships
Trust 2 between Friends
Model 3-4 Model 3-5 Model 3-6
Intercept
.898***
(.029)
.886***
(.022)
.886***
(.022)
Control
Trust 1
.474***
(.1 14)
.423***
(.107)
.423***
(.107)

Main Effects
Negative Ability
- .092*
(.051)

Negative Intention
-.271***
(.091)

Negative Affection
-.271***
(.091)

Moderators
Need for Structure
- .094
(.066)
-.306**
(.1 22)
-.306**
(.122)
Variance Composition
Intercept .008 .006 .006
Error term .018 .017 .017

Model Summary
ᵡ2 54.826*** 49.999** 49.999**
The numbers in parentheses are standard errors.
*p<.10
**p<.05
***p<.01

87

Trustors‘ need for structure amplified the effect of negative experiences
regarding intention and affection on the level of Trust 2. Aiken and West (1991)
suggested that significant interaction effects should be probed further to better understand
the nature of the effects. Thus, Figure 5 demonstrates these effects graphically. (Since the
correlation between these two negative experiences variables is 1.000, only one graph is
presented.) In this graph, the X-axis represents the binary variable of whether
respondents experienced negative events (i.e., 1 if there were any negative experiences,
and 0 otherwise), and the Y-axis refers to the level of Trust 2 (the transformed variable,
which ranges from 0 to 1). The bar on the far left represents the level of Trust 2 for those
respondents from the low NFS group who did not experienced negative events, and the
bar next to it represents those from the high NFS group who did not experience negative
events. The third bar indicates the level of Trust 2 for the low NFS group who did
experience negative events, and the bar on the right represents those from the high NFS
group who did experience negative events. As the graph clarifies, the level of Trust 2
was not any different for the low and high NFS groups who did not experience any
negative events. However, when negative events were experienced, the level of Trust 2
was much smaller for the high NFS group than for the low NFS group.

88

Figure 5. Moderator Effect of the Need for Structure
(Negative Intention and Affection)
0.00 1.00
0.39
0.54
0.69
0.84
0.99
Negative Experience (Intention/Affection)
Trust 2
Low Need for Structure
High Need for Structure

89

Chapter Summary
Table 18 summarizes the results of testing Hypotheses 1 through 6.

Table 18. Summary of Hypotheses Testing
Model
#1
Hypothesis Strangers Friends
H1 The relationship between GE and Trust1 N/S
2
.080
1
H2 The relationship between SE and Trust1 .073

N/S

Model
#2
3
Hypothesis
Positive
Experiences
Negative
Experiences
H3-2
The relationship between GE and positive
experiences
N/S

N/S

H4-2
The relationship between SE and negative
experiences
N/S

-1.214
(Intention)
&
-.1.516
(Affection)
H5-2
The moderator effect of Need for Structure
on the relationship between Trust1 and the
experiences
N/S

N/S

Model
#3
Hypothesis Strangers Friends
H6-1
The moderating effect of decisiveness on the
relationship between negative experiences
and Trust2
N/S

H6-2
The moderating effect of need for structure
on the relationship between negative
experiences and Trust2

-.306

(Intention &
Affection)
1. The numbers represent the coefficients in the models.
2. N/S denotes no significant results.
3. Hypotheses 3-1, 4-1, and 5-1 were not tested due to a problem with the data as discussed
above.

90

CHAPTER SIX: DISCUSSION

The focus of this study was to investigate the effects of trustors‘ attributes on the
level of trust. Specifically, this study identified four different trustor traits as predictors of
trust level: cross-personal generalized expectancy (GE); situational generalized
expectancy (SE); decisiveness; and need for structure (NFS). This study tested the effects
of these four predictors on trust development in three phases, namely, the building,
stability, and dissolution of trust. In this chapter, the findings presented in the previous
chapter will be discussed, along with theoretical and managerial implications of the
findings, the limitation of this study, and suggestions for future research.
First of all, the intra-correlation values for all the models were relatively high.
This indicates that the variance of both Trust 1 and Trust 2 is considerably explained by
the differences among trustors (level 2 variance), rather than by characteristics of trustees
and of their relationship (level 1 variance). Furthermore, the base model chi-square
values were mostly significant. The strong intra-correlations and the significant model
chi-squares suggest that HLM, which takes into account such trustor-level (level 2)
differences, was the appropriate analytical method to use. They also justify the need to
study how trustors‘ attributes impact differences in their propensity to trust, an issue
which has been overlooked in the literature.
Generally speaking, trustors‘ general beliefs about human nature (i.e., GE) did
not influence their level of trust or the probability that trustors find positive or negative
91

signs regarding a trustee‘s trustworthiness. It influenced the level of trust only when they
were about to start working together with friends. In contrast, the situation-specific
expectancy (i.e., SE) significantly influenced the initial level of trust between strangers,
while it has no corresponding impact on the level of trust between friends. One possible
explanation is that, when someone assumes that a given trustee is trustworthy based on
the belief that people in general are trustworthy (i.e., an extension of generalized
expectancy), trusting is based on a ‗rough guess‘ by the trustor. Considering that trust is
essentially an extrapolation from given information, to assume that strangers will act in a
trustworthy manner based on information that unspecified others are generally
trustworthy requires a trustor to take a big leap of faith. Admittedly, there is no guarantee
that even a familiar friend who has been trustworthy in the past will act the same way in
the future. In this regard, numerous scholars (e.g., Mayer et al., 1995) have emphasized
that risk-taking (the leap of faith) is an essential element of trusting. However, the
amount of faith trustors need to have is much greater when trusting strangers based on
GE than when trusting friends based on GE. In contrast, when people trust strangers
based on SE, they extrapolate information formed from experiences with a smaller
number of people in a more specific situation. Thus, the information from positive SE is
likely to be more directly relevant to their trusting judgment. Therefore, the result
indicates that GE helps people to take a small leap of faith, while taking a bigger leap of
faith requires the more specific information reflected in their SE.
There have been mixed empirical findings with regards to this idea (Holmes,
1991; Johnson-George & Swap, 1982; McKnight et al., 1998). For instance, some studies
92

investigating the relationship between GE and behavioral trust in the prisoners‘ dilemma
(PD) game also found no significant relationship (e.g., Glaeser, Laibson, Scheinkman, &
Soutter, 2000). In an effort to explain this inconsistency, McKnight et al. (1998)
speculated that the condition of having no prior interaction history with a specific trustee
(i.e., initial trust between strangers) will lead trustors to rely on GE. However, given that
the PD game experiments are conducted with two people who do not know each other,
McKnight and colleagues‘ speculation does not seem to provide a satisfactory
explanation of the inconsistency. The results of the present study also suggest otherwise. I
argue that the results presented above provide some clues that help to explain this
inconsistency. If the research context required the subjects or respondents to take a big
leap of faith, the relationship would be less likely to be significant. For example, different
designs of the payoff in the PD game experiment can affect whether GE influences the
probability of behaviorally trusting another (i.e., risk-taking behavior). If the possible loss
by one player when trusting and being betrayed by another player is designed to be
twenty dollars in the experiment, the influence of GE on behavioral trust would be much
less than when the possible loss was designed to be five dollars, since losing twenty
dollars requires a bigger leap of faith than losing five dollars.
This result has implications for managers. When managers organize a task-
oriented ad hoc workgroup to address a specific problem that the organization faces, they
need to consider whether the members of a workgroup have successfully collaborated
within their original work units and thus are likely to have positive SE. To ensure the
success of a workgroup, trust building among the members is as important as the quality
93

of the expertise each member brings to the group (Costa, 2003; Erdem & Ozen, 2003),
given the lack of clearly specified rules governing transactions among relatively
autonomous members. Yet, more often than not, many workgroup members are strangers
to each other, such that building trust can be a challenge. Therefore, when organizing
workgroups, managers need to consider whether potential members have experienced
positive consequences of collaboration in their workplaces, in which case they will have
less trouble trusting the members of newly organized teams.
The factors influencing the probability that trustors experience positive and/or
negative interpersonal events were examined, based on the premise that interpersonal
events are not objective but are subject to the subjective interpretation of reality. No such
effect was found in relationships with strangers. However, in relationships with friends,
SE had a significant negative influence on the probability that trustors experienced
negative incidents regarding trustees‘ ability and affect. Furthermore, it was also found in
the preliminary analysis that trustors were significantly more likely to experience
negative events in relationship with strangers than with friends. These results may
indicate that people are more objective in assessing the trustworthiness of strangers,
while they are more subjective in assessing that of friends. In other words, when negative
events occur that may be interpreted as showing the untrustworthiness of friends, trustors
with a positive SE may reinterpret the events depending on their positive prior
experiences in that specific situation. Because of the effect of reinterpreting negative
events in relationships with friends, people are less likely to experience negative events in
these relationships. These results can also be related to the first finding. The first
94

finding was interpreted as indicating that, in order to trust strangers, people may rely
more on directly relevant information (i.e., SE) than remotely related information (i.e.,
GE), because trusting strangers based on the latter requires a bigger leap of faith.
Likewise, reinterpreting equivocally negative experiences in relationships with strangers
as being positive, with no concrete evidence, requires a bigger leap of faith than
reinterpreting those experiences in relationships with friends with whom the trustor has
had a history of interaction. Thus, the subjective reinterpretation of events is more likely
to happen in relationships with friends than with strangers.
The results of the ANOVA preliminary analysis reveal that, once high NFS
trustors perceived a negative event, the degree to which their trust declined was
significantly greater than lower NFS trustors, regardless of the types of negative
experiences. A similar pattern was found for highly decisive trustors. Their trust decline
was greater than that for low decisiveness trustors when the negative experiences had to
do with ability and intention (affection experiences had no such effect). Once a trusting
relationship is established, the key is to maintain it by focusing on managing negative
events. In dealing with people with high NFC – who are decisive, prefer order and
predictability, and feel greater discomfort for ambiguity – a manager should be cautious
not to give them the impression that he/she is incapable or does not have good intentions
toward them. The results pertaining to Hypotheses 6-1 and 6-2 reveal that, although the
effect of negative incidents experienced in relationships with strangers was not
particularly strong for highly decisive people, the trust decline was greater when high
NFS people experienced negative intention- and affection-related incidents in
95

relationships with friends. Thus, once trust is breached in a relationship with high NFS
people, it is likely to be harder to repair the broken trust even if one has had a trusting
relationship with them prior to that point.
This result runs counter to what was hypothesized about these effects. It was
initially hypothesized that, in relationships with friends, strong NFC trustors will undergo
a relatively smaller amount of trust decline after experiencing negative events, since they
freeze on their initial trust in order to maintain psychological comfort. But what was
found instead was a greater amount of trust decrease in the face of negative experiences.
It seems that the effects of prior positive experiences, which supported the high initial
trust, did not suppress the effects of recent negative experiences. As discussed in the
hypotheses chapter, scholars have reported inconsistent findings regarding the effects of
NFC, and we suggested two conditions (i.e., strangers vs. friends, and decisiveness vs.
NFS) as determinants of whether NFC brings about a primacy effect or a recency effect.
It appears that the conditions examined in this study were not determinants of which
effect the NFC brings about. The findings from this study support previous research
suggesting that high NFC induces a recency effect, with trust impacted by the latest
information, rather than those studies supporting a primacy effect.
This study has some limitations that future research needs to address. First, this
research focused on students in a university setting, rather than workers in organizations.
Considering that trust is situation specific, the difference in the contexts in which trustors
interact with each other may lead to different results. The content of SE may vary
depending on the situation. For instance, when a boss needs to empower a subordinate,
96

the relevant SE will have been formed by past experiences with empowering subordinates.
Future research needs to examine whether the findings of this study also apply in other
contexts. Second, the demographic composition of our sample of students is more
homogenous than that of workers in organizations. While the students in our sample were
mostly enrolled in related degree programs, workers in organizations are likely to come
from a variety of different backgrounds with unique skills and expertise with which to
make a unique contribution to organizational tasks. Considering that the assessment of
the ability component of trustworthiness is context specific (Mayer et al., 1995), research
in other kinds of organizations might obtain different results than those found in this
study. Hence, future research exploring the role of trustor propensity to trust should take
place in a broader array of organizations with more diverse samples of employees to see
if the patterns found in this study are generalizable to other contexts. Third, there might
be some interaction effects among trustor attributes that were not considered in this study.
For example, the effects of positive GE on initial trust may be greater for strong NFC
people than for weak NFC people. Research on cognitive closure has demonstrated that
the effect of a more or less equivocal and hypothetical anchoring point (e.g., GE) on
social judgments can be amplified by NFC. Webster (1993) found that perceivers with
strong NFC tend to fall prey to the correspondence bias (Jones, 1979) more easily than
those with a weak need for closure, because perceivers with strong NFC rely heavily on
their initial disposition-based hypothesis in attributing the cause of others‘ actions.
Examining these interaction effects could also produce meaningful theoretical
implications for the trust literature and practical implications for managers.
97

REFERENCES

Agranoff, R., & McGuire, M. 2001. Big questions in public network management
research. Journal of Public Administration Research and Theory, 11: 295-326.

Agranoff, R., & McGuire, M. 2003. Collaborative public management: New strategies
for local government. Washington, DC: Georgetown University Press.

Aiken, L. S., & West, S. G. 1991. Multiple regression: Testing and interpreting
interactions. Thousand Oaks: Sage.

Alston, W. P. 1975. Traits, consistency, and conceptual alternatives for personality theory.
Journal for the Theory of Social Behavior, 5: 17-48.

Argyle, M., & Little, B. R. 1972. Do personality traits apply to social behavior? Journal
for the Theory of Social Behavior, 2: 1-35.

Astley, W. G., & Fombrun, C. J. 1983. Collective strategy: Social ecology of
organizational environment. Academy of Management Review, 8(4): 576-587.

Axelrod, R. 1984. The evolution of cooperation. New York, NY: Basic Books.

Bachman, R. 2006. Trust and power: Towards a sociological theory of organizational
relationships. In R. Bachman, & A. Zaheer (Eds.), Handbook of trust research: 393-
408. Cheltenham, UK and Northampton, MA: Edward Elgar.

Baier, A. 1986. Trust and antitrust. Ethics, 96: 231-260.

Behn, R. D. 1995. The big questions of public management. Public Administration
Review, 55: 313-324.

98

Berg, J., Dickhaut, J., & McCabe, K. 1995. Trust, reciprocity, and social-history. Games
and Economic Behavior, 10(1): 122-142.

Bouton, M. E. 2007. Learning and behavior: A contemporary synthesis. Publisher:
Sinauer Associates, Inc.

Bradach, J. L., & Eccles, R. G. 1989. Price, authority, and trust: From ideal types to plural
forms. Annual Review of Sociology, 15: 97-118.

Brashear, T. G., Manolis, C., & Brooks, C. M. 2005. The effects of control, trust, and
justice on salesperson turnover. Journal of Business Research, 58(3): 241-249.

Brehm, J., & Rahn, W. 1997. Individual-level evidence for the causes and consequences
of social capital. American Journal of Political Science, 41, 999-1023.

Bridgman, P. W. 1927. The logic of modern physics. New York: Macmillan.

Brown, T. L. & Potoski, M. 2003a. Managing contract performance: A transaction costs
approach. Journal of Policy Analysis and Management, 22(2): 275–297.

Brown, T. L. & Potoski, M. 2003b. Transaction costs and institutional explanations for
government service production decisions. Journal of Public Administration
Research and Theory, 13(4): 441-468.

Brunswik, E. 1956. Perception and the representative design of psychological
experiments. Berkeley and Los Angeles: University of California Press.

Butler, J. K. 1991. Toward understanding and measuring conditions of trust: Evolution of
a conditions of trust inventory. Journal of Management, 17: 643-663.

Butler, J. K., & Cantrell, R. S. 1984. A behavioral decision theory approach to modeling
dyadic trust in superiors and subordinates. Psychological Reports, 55: 19-28.

99

Campbell, D. T. 1959. Convergent and discriminant validation by the multitrait-
multimethod matrix. Psychological Bulletin. 81–105.

Chatman, J. A., & Barsade, S. G. 1995. Personality, organizational culture, and
cooperation: Evidence from a business simulation. Administrative Science Quarterly.
40:423-443

Chen, B. 2008. Managing inter-organizational partnerships: Preconditions, processes and
perceived outcomes. Best Papers Proceedings of the Sixty-Sixth Annual Meeting of
the Academy of Management

Chirumbolo, A., & Areni, A. 2010. Job insecurity influence on job performance and
mental health: Testing the moderating effect of the need for closure. Economic and
Industrial Democracy. 31(2): 195-214.

Christie, R., & Geis, F. 1970. Studies in machiavellianism. . San Diego, CA: Academic
Press.

Chua, R. Y ., Ingram, P., & Morris, M. W. 2008. From the head and the heart: Locating
cognition- and affect-based trust in managers‘ professional networks. Academy of
Management Journal, 51(3): 436–452.

Coleman, J. 1990. Foundation of social theory. Cambridge, MA: Harvard University
Press.

Cook, A. J., & Wall, P. D. 1980. New work attitude measures of trust, organizational
commitment and personal need non-fulfillment. Journal of Occupational
Psychology, 53: 39-52.

Costa, A. C. 2003. Work team trust and effectiveness. Personnel Review, 32: 605-622.

Couch, L. L., Adams, J. M., & Jones, W. H. 1996. The assessment of trust orientation.
Journal of Personality Assessment, 67(2): 305-323.

100

Currall, S. C., & Inkpen, A. C. 2006. On the complexity of organizational trust: A multi-
level co-evolutionary perspective and guidelines for future research. In R. Bachman,
& A. Zaheer (Eds.), Handbook of Trust Research: 235-246. Cheltenham, UK and
Northampton, MA: Edward Elgar.

Das, T. K., & Teng, Bing-Sheng. 1998. Between trust and control: Developing confidence
in partner cooperation in alliances. Academy of Management Review, 23(3): 491-512.

Davis, J. H., Schoorman, F. D., Mayer, R. C., & Tan, H. H. 2000. The trusted general
manager and business unit performance: Empirical evidence of a competitive
advantage. Strategic Management Journal, 21: 563-576.

Davis-Blake, A., & Pfeffer, J. 1989. Just a mirage: The search for dispositional effects in
organizational research. Academy of Management Review, 14, 385-400.

DeBacker, T. K. & Crowson, H. M. 2008. Measuring need for closure in classroom
learners. Contemporary Educational Psychology. 33(4): 711-732.

DeBacker, T. K. & Crowson, H. M. 2009. The Influence of need for closure on learning
and teaching. Educational Psychology Review, 21(4): 303-323.

Deluga, R. J. 1995. The relation between trust in the supervisor and subordinate
organizational citizenship behavior. Military Psychology, 7: 1-16.

Deutsch, M. 1958. Trust and suspicion. Journal of Conflict Resolution, 2(4): 265-279.

Deutsch, M. 1960. The effect of motivational orientation upon trust and suspicion.
Human Relation, 13, 123-140.

Dodgson, M. 1993. Learning, trust, and technological collaboration. Human Relations,
46: 77-95.

101

Doney, P. M., Cannon, J. P., & Mullen, M. R. 1998. Understanding the influence of
national culture on the development of trust, Academy of Management Review,
23(3): 601-620

Driscoll, J. W. 1978. Trust and participation in organizational decision making as
predictors of satisfaction. Academy of Management Journal, 21(1): 44-56.

Dunn, J. R., & Schweitzer, M. E. 2005. Feeling and believing: The influence of emotion
on trust. Journal of Personality and Social Psychology, 88(5): 736-748.

Elangovan, A. R. & Shapiro, D. L. 1998. Betrayal of trust in organizations. Academy of
Management Review. 23(3): 547-566.

Erdem, F. & Ozen, J. 2003. Cognitive and affective dimensions of trust in developing
team performance, Team Performance Management, 9:131-135.

Erikson, E. H. 1963. Childhood and Society. New York, NY: W.W.Norton & Company
Inc.

Everett, M.G., and Borgatti, S.P. 2005. Ego-network betweenness. Social Networks.
27(1): 31-38.

Fiske. S. T., & Taylor. S. E. 1984. Social cognition. Reading, MA: Addison-Wesley.

Flaherty, K. E., & Pappas, J. M. 2000. The role of trust in salesperson-sales manager
relationships. Journal of Personal Selling and Sales Management, 20: 271-278.

Ford, T. E., & Kruglanski, A. W. 1995. Effects of epistemic motivations on the use of
accessible constructs in social judgment. Personality and Social Psychology Bulletin,
21, 950-962.

Forgas, J. P. 1995. Mood and judgment: The affect infusion model (AIM). Psychological
Bulletin, 117(1), 39-66.

102

Fukuyama, F. 1995. Trust: The social virtues and the creation of prosperity. Free Press.

Gambetta, D. 1988. Can we trust trust? In D. Gambetta (Ed.), Trust: Making and
breaking cooperative relations: 213-238. New York, NY: Basil Blackwell.

Gardner, H. 1986. The mind's new science: A history of the cognitive revolution New
York: Basic Books.

Giffin, K. 1967. The contribution of studies of source credibility to a theory of
interpersonal trust in the communication department. Psychological Bulletin, 68:
104-120.

Gifford, R. 1994. A lens-mapping framework for understanding the encoding and
decoding of interpersonal dispositions in nonverbal behavior. Journal of Personality
and Social Psychology, 66(2): 398-412.

Gittell, R., & Vidal, A. 1998. Community organizing: Building social capital as a
development strategy. Thousand Oaks, CA: Sage.

Glaeser, E. L., Laibson, D. I., Scheinkman, J. A., & Soutter, C. L. 2000. Measuring trust.
Quarterly Journal of Economics, 115(3): 811-846.

Good, D. 1988. Individuals, interpersonal relations, and trust. In D.G.Gambetta (Ed.),
Trust: Making and breaking cooperative relations, 131-185. New York: Basil
Blackwell.

Granovetter, M. S. 1985. Economic action and social structure. American Journal of
Sociology. 91: 481-510.

Greenwald. A. G. 1980, The totalitarian ego: Fabrication and revision of personal history.
American Psychologist. 35: 603-618.

Greenwald, A. G., & Banaji, M. R. 1995. Implicit social cognition: Attitudes, self-esteem,
and stereotypes, Psychological Review. 102: 4-27.
103

Gulati, R. 1995. Does familiarity breed trust? The implications of repeated ties for
contractual choice in alliances. Academy of Management Journal, 38(1): 85-112.

Gurtman, M. B. 1992. Trust, distrust, and interpersonal problems: A circumplex analysis.
Journal of Personality and Social Psychology, 62: 989-1002.

Hagen, J. M., & Choe, S. 1998. Trust in Japanese interfirm relations: Institutional
sanctions matter. Academy of Management Review. 23(3): 589-600.

Hamel, G. 1991. Competition for competence and interpartner learning within
international strategic alliances. Strategic Management Journal, 12: 83-103.

Hardin, R. 2004. Trust and trustworthiness. New York: Sage.

Herranz, J. 2008. The multisectoral trilemma of network management. Journal of Public
Administration Research and Theory. 18(1): 1–31.

Holmes, J. G. 1991. Trust and the appraisal process in close relationships. In W. H. Jones
& D. Perlman (Eds.), Advances in personal relationships, vol. 2: 57-104. London:
Jessica Kingsley.

Hosmer, L. T. 1995. Trust: The connecting link between organizational theory and
philosophical ethics. Academy of Management Review, 20: 379-403.

House, R.J., Shane, S.A., Herold, D.M. 1996. Rumors of the death of dispositional
research are vastly exaggerated. Academy of Management Review, 21(1), 203-224.

Houton, D. C., & Grewal, R. 2000. Please, let‘s get an answer—Any answer: Need for
consumer cognitive closure. Psychology & Marketing. 17(11):911–934
104

Huang, K. & Provan, K. 2006. Tangibility of resources and patterns of network iteraction
in a publicly-funded health and human service network. Journal of Public
Administration Research and Theory, 17 (3): 435-454.

Janoff-Bulman, R. 1989. Assumptive worlds and the stress of traumatic events:
Applications of the schema construct. Social Cognition, 7: 113-136.

Jarvenpaa, S. L., & Leidner, D. E. 1999. Communication and trust in global virtual teams.
Organization Science, 10(6): 791-815.

Johnson-George, C. & Swap, W. 1982. Measurement of specific interpersonal trust:
Construction and validation of a scale to assess trust in a specific other. Journal of
Personality and Social Psychology, 43: 1306-1317.

Jones, E. E. 1979. The rocky road from acts to dispositions. American Psychologist, 34,
107-117.

Jones, A. P., James, L. R., & Bruni, J. R. 1975. Perceived leadership behavior and
employee confidence in the leader as moderated by job involvement. Journal of
Applied Psychology, 60: 146-149.

Kanter, D. L., & Mirvis, P. H. 1989. The cynical Americans: Living and working in an
age of discontent and disillusion. San Francisco: Jossey-Bass.

Kee, H. W., & Knox, R. E. 1970. Conceptual and methodological considerations in the
study of trust. Journal of Conflict Resolution. 14:357-366.

Kenny, D. A. 1994. Interpersonal perception. New York, NY: The Guilford Press.

Kenrick, D. T. & Funder, D. C. 1988. Profiting from controversy: Lessons from the
Person-Situation Debate. American Psychologist, 43(1), 23-34.

105

Kim, P. H., Ferrin, D. L., Cooper, C. D., & Dirks, K. T. 2004. Removing the shadow of
suspicion: The effects of apology vs. denial for repairing ability- vs. integrity-based
trust violations. Journal of Applied Psychology, 89(1), 104-118.

Kipnis, D. 1996. Trust and technology. In R. M. Kramer, & T. R. Tyler (Eds.), Trust in
organizations: Frontiers of theory and research. Thousand Oaks, CA: Sage.

Kossowska, M., Van Hiel, A., Chun, W. Y ., & Kruglanski, A. W. 2002. The need for
cognitive closure scale: Structure, cross-cultural invariance, and comparison of mean
ratings between European-American and East Asian samples Psychologica Belgica,
42(4): 267-286.

Kramer, R. M. 1994. The sinister attribution error: Paranoid cognition and collective
distrust in organizations. Motivation and Emotion, 18: 199-230.

Kramer, R. M. 1999. Trust and distrust in organizations: Emerging perspectives, enduring
questions. Annual Review of Psychology, 50: 569-598.

Kramer, R. M., & Lewicki, R. J. 2010. Repairing and enhancing trust: Approaches to
reducing organizational trust deficits, Academy of Management Annals. 4(1): 245–
277

Kruglanski, A. W., & Webster, D. M. 1996. Motivated closing of the mind: "Seizing" and
"Freezing". Psychological Review, 103(2): 263-283.

Kruglanski, A. W., Webster, D. M., & Klem, A. 1993. Motivated resistance and openness
to persuasion in the presence or absence of prior information. Journal of Personality
and Social Psychology, 65, 861 -876.

La Porte, T. R. 1996. High reliability organizations: Unlikely, demanding and at risk.
Journal of Contingencies and Crisis Management. 4(2): 60–71.

Larzelere, R. E., & Huston, T. L. 1980. The dyadic trust scale: Toward understanding
interpersonal trust in close relationships. Journal of Marriage and the Family, 42(3):
595-604.
106

Levine, A. L. 1973. Project and task force approaches. Public Administration Review,
33(5): 485

Lewicki, R. J., & Bunker, B. B. 1995. Trust in relationship: A model of development and
decline. In B. B. Bunker, & J. Z. Rubin (Eds.), Conflict, Cooperation and Justice.
San Fransisco: Jossey-Bass Publishers.

Lewicki, R. J., McAllister, D. J., & Bies, R. J. 1998. Trust and distrust: New relationships
and realities. Academy of Management Review, 23(3): 438-458.

Lewicki, Roy J., Tomlinson, Edward C., & Gillespie, Nicole. 2006. Models of
interpersonal trust development: Theoretical approaches, empirical evidence, and
future directions. Journal of Management, 32(6): 991-1022.

Lewis, J. D., & Weigert, A. 1985. Trust as a social reality. Social Forces, 63(4): 967-985

Lieberman, J. K. 1981. The litigious society. New York: Basic Books.

Longino, H. E. 1990. Science as social knowledge: Value and objectivity in scientific
inquiry. Princeton, NJ: Princeton University Press.

Lu, L., & Argyle, M. 1991. Happiness and cooperation. Personality and Individual
Differences. 12(10): 1019-1030.

Luhmann, N. 1979. Trust and Power. Chichester: Willey.

Lundin, M. 2007. Explaining cooperation: How resource interdependence, goal
congruence, and trust affect joint actions in policy implementation. Journal of
Public Administration Research and Theory, 17: 651-672.

March, J. G., & Olson, J. P. 1989. Rediscovering institutions: The organizational basis of
politics. New York, NY: Free Press.

107

March, J. G. & Shapira, Z. 1987. Managerial perspectives on risk and risk taking.
Management Science. 33(11): 1404-1418.

Mayer, R. C., Davis, J. H., & Schoorman, F. D. 1995. An integrative model of
organizational trust. Academy of Management Review, 20(3): 709-734.

McAllister, D. J. 1995. Affect- and cognition-based trust as foundations for interpersonal
cooperation in organizations. Academy of Management Journal, 38(1): 24-59.

McConahay, J. B. 1983. Modern racism and modern discrimination – The effects of race,
racial-attitudes, and context on simulated hiring decisions. Personality and Social
Psychology Bulletin, 125: 255-275.

McKnight, D. H., Cummings, L. L., & Chervany, N. L. 1998. Initial trust formation in
new organizational relationships Academy of Management Review, 23(3): 473-490.

Mischel, W. 1968. Personality and assessment. New York: Wiley.

Mishra, A. K. 1996. Organizational responses to crisis: The centrality of trust. In R. M.
Kramer & T. Tyler (Eds.), Trust in organizations. Newbury Park, CA: Sage.

Möllering, G. 2006. Trust, institutions, agency: towards a neoinstitutional theory of trust.
In Bachmann, R., & Zaheer, A. (eds.) Handbook of Trust Research: 355-376.
Southampton, MA: Edward Elgar.

Mooradian, T., Renzl, B., & Maztler, K. 2006. Who trusts? Personality, trust and
knowledge sharing. Management Learning, 37(4): 523–540.

Muchinsky, P. M. 1977. Organizational communication: Relationships to organizational
climate and job satisfaction. Academy of Management Journal, 20(4): 592-607.

108

Muthusamy, S. K., & White, M. A. 2005. Learning and knowledge transfer in strategic
alliances: A social exchange view. Organization Studies, 26: 415-441.

Nahavandi, A., & Malekzadeh, A. R. 1999. Organizational behavior: The person-
organization fit. Upper Saddle River, NJ: Prentice-Hall.

Neuberg, S. L., Judice, T. N., & West, S. G. 1997. What the need for closure scale
measures and what it does not: Toward differentiating among related epistemic
motives. Journal of Personality and Social Psychology, 72(6): 1396-1412.

Newell, S. & Swan, J. 2000. Trust and inter-organizational networking. Human Relations,
53(10): 1287-1328.

Omodei, M. M., & McLennan, J. 2000. Conceptualizing and measuring global
interpersonal trust and distrust. Journal of Social Psychology, 140(3): 279-294.

Oldham, G. R. 1975. The impact of supervisory characteristics on goal acceptance.
Academy of Management Journal, 18, 461-475.

Oliver, C. 1990. Determinants of interorganizational relationship: Integration and future
directions. Academy of Management Review. 15:241-265.

Orbell, J., Dawes, R., & Schwartz-Shea, P. 1994. Trust, social categories, and individuals:
The case of gender. Motivation and Emotion, 18: 109-128.

Ospina, S. & Yaroni, A. 2003. Understanding cooperative behavior in labor management
cooperation: A theory-building exercise. Public Administration Review, 63(4): 455–
471.

O'Toole, L. J. 1997. Treating networks seriously: Practical and research-based agendas in
public administration. Public Administration Review, 57: 45-52.

109

Pierro, A., Cicero, L., & Bonaiuto, M., & van Knippenberg, D. 2005 Leader group
prototypicality and leadership effectiveness: The moderating role of need for
cognitive closure. The Leadership Quarterly, 16(4): 503-516

Provan, K. G. & Kenis, P. 2008. Modes of network governance: Structure, management,
and effectiveness. Journal of Public Administration Research and Theory, 18 (2):
229-252.

Putnam, R. D. 1993. Making democracy work: Civic traditions in modern Italy. Princeton,
NJ: Princeton University Press.

Rich, G. A. 1997. The sales manager as a role model: Effects on trust, job satisfaction,
and performance of salespeople. Journal of the Academy of Marketing Science,
25(4): 319-328.

Richter, L., & Kruglanski, A. W. 1998. Seizing on the latest: Motivationally driven
recency effects in impression formation. Journal of Experimental Social Psychology.
34(4): 313-329.

Rigby, K., Cox, I, & Black, G. 1997. Cooperativeness and bully/victim problems among
Australian schoolchildren. Journal of Social Psychology. 137(3): 357-368.

Ring, P. S., & Van de Ven, A. H. 1994. Developmental processes of cooperative
interorganizational relationships. Academy of Management Review, 19(1): 90-118.

Robinson, S. L. 1996. Trust and breach of the psychological contract. Administrative
Science Quarterly, 41: 574-599.

Robinson, J. P., & Shaver, P. R. 1973. Measures of social psychological attitudes. Ann
Arbor, MI: Institute for Social Research.

Roets, A., Van Hiel, A., & Cornelis, I. 2006. The dimensional structure of the need for
cognitive closure scale: Relationships with ―Seizing‖ and ―Freezing‖ processes.
Social Cognition. 24(1): 22-45

110

Ross, L., & Nisbett, R. 1991. The person and the situation. New York: McGraw-Hill, Inc.,
Chapter 6.

Rotter, J. B. 1954. Social learning and clinical psychology. New York: Prentice-Hall

Rotter, J. B. 1967. A new scale for the measurement of interpersonal trust. Journal of
Personality, 35(4): 651-665.

Rotter, J. B. 1971. Generalized expectancies for interpersonal trust. American
Psychologist, 26(5): 443-452.

Rotter, J. B. 1980. Interpersonal trust, trustworthiness, and gullibility. American
Psychologist, 35(1): 1-7.

Rousseau, D. M., Sitkin, S. B., Burt, R. S., & Camerer, C. 1998. Introduction to special
topic forum: Not so different after all: A cross-discipline view of trust. Academy of
Management Review, 23(3): 393-404.

Sako, M. 1992. Prices, quality, and trust: Inter-firm relations in Briatain and Japan. New
York, NY: Cambridge University Press.

Schermerhorn, Jr., J. R. 1975. Determinants of interorganizational cooperation. Academy
of Management Journal, 18(4): 846-856.

Schneider, B. 1983. Interactional psychology and organizational behavior. In B. M. Staw
& L.L. Cummings (Eds.), Research in organizational behavior. vol. 5: 1-31.
Greenwich, CT: JAI Press.

Schwenk, C. R. 1984. Cognitive simplification processes in strategic decision-making.
Strategic Management Journal. 5(2): 111–128

Shaheen, J. 2010. The recency and primacy effects in the talent acquisition process.
Journal of Corporate Recruiting Process. 5(3): 13-15.

111

Sheppard, B. H., & Sherman, D. M. 1998. The grammars of trust: A model and general
implications. Academy of Management Review, 23(3): 422-437.

Shetterly, D. R. 2000. The influence of contract design on contractor performance: The
case of residential refuse collection. Public Performance & Management Review,
24(1): 53-68.

Simons, T., & Roberson, Q. 2003. Why managers should care about fairness: The effects
of aggregate justice perceptions on organizational outcomes. Journal of Applied
Psychology, 88(3): 432–443.

Sitkin, S. B., & Roth, N. L. 1993. Explaining the limited effectiveness of legalistic
remedies for trust and distrust. Organization Science, 4(3): 367-392.

Sorrentino, R. M., Holmes, J. G., Hanna, S. E., & Sharp, A. 1995. Uncertainty Orientation
and Trust in Close Relationships: Individual Differences in Cognitive Styles.
Journal of Personality and Social Psychology, 68(2): 314-327.

Stalder, D. R. 2010. Competing roles for the subfactors of need for closure in moderating
dissonance-produced attitude change. Personality and Individual Differences, 48:
775–778.

Staw, B. M., & Ross, J. 1985. Stability in the midst of change: A dispositional approach
to job attitudes. Journal of Applied Psychology. 70: 469-480.

Studenmund, A. H. 2006. Using econometrics: A practical guide. Pearson Education, Inc.

Tabanick, B. G., & Fidell, L. S. 2007. Using Multivariate Statistics. Pearson Education,
Inc.

Tetlock, P. E. 2000. Cognitive biases and organizational correctives: Do both disease and
cure depend on the politics of the beholder? Administrative Science Quarterly. 45:
293-326.

112

Tolman, E. C. 1966. Behavior and psychological man: Essays in motivation and learning.
CA: University of California Press.

Tomlinson, E. C., & Mayer, R. C. 2009. The role of causal attribution dimensions in trust
repair. Academy of Management Review, 34(1): 85-104.

Tudela, P. 2004. Cognitivism. In Spielberger, C. (Eds.) Encyclopedia of Applied
Psychology: 393-402. Maryland Heights, MO: Elsevier, Academic Press.

Uzzi, B. 1996. The sources and consequences of embeddedness for the economic
performance of organizations: The network effect. American Sociological Review,
61: 674-698.

Webster, D. M. 1993. Motivated augmentation and reduction of the overattribution bias.
Journal of Personality and Social Psychology, 65,261-271.

Webster, D. M., & Kruglanski, A. W. 1994. Individual differences in need for cognitive
closure. Journal of Personality and Social Psychology, 67(6): 1049-1062.

Weick, K. E. 2008. Trust: A bigger picture. Academy of Management Review, 33(1): 271-
283.

Wellman, B., & Frank, K. 2001. Network capital in a multi-level world: Getting support
from personal communities. In Nan Lin, Ron Burt, & Karen Cook, Social capital:
Theory and research: 233-274. Chicago: Aldine De Gruyter

Whitener, E. M. 2001. Do high commitment human resource practices affect employee
commitment? A cross-level analysis using hierarchical liner modeling. Journal of
Management, 27: 515-535

Whitener, E. M., Brodt, S. E., Krosgaard, M. A., & Werner, J. M. 1998. Managers as
initiators of trust: An exchange relationship framework for understanding
managerial trustworthy behavior. Academy of Management Review, 23(3): 513-530.

113

Williamson, O. E. 1993. Calculativeness, trust and economic organization. Journal of
Law and Economics. 36: 453-486.

Wrightsman, L. S. 1991. Interpersonal trust and attitudes toward human nature. In J. P.
Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of personality and
psychological attitudes: 373-412. San Diego, CA: Academic.

Wrightsman, L. S., & Wuescher, M. L. 1974. Assumptions about human nature: A social-
psychological approach. Monterey, CA: Brooks/Cole.

Yamagishi, T. 1988. The provision of a sanctioning system in the United States and Japan.
Social Psychology Quarterly, 51(3): 265-271.

Yilmaz, C., & Hunt, S. D. 2001. Salesperson cooperation: The influence of relational,
task, organizational, and personal factors. Journal of Academy of Marketing Science.
29(4): 335-357

Zaheer, A, McEvily, B., & Perrone, V . (1998). Does trust matter? Exploring the effects of
interorganizational and interpersonal trust on performance. Organization Science. 9,
141-159.

Zaheer, A., & Venkatraman, N. 1995. Relational governance as an interorganizational
strategy: An empirical test of the role of trust in economic exchange. Strategic
Management Journal, 16(5): 373-392.

Zucker, L. G. 1986. Production of trust: Institutional sources of economic structure 1849-
1920. In B. M. Staw & L. L. Cummings (Eds.), Research in organizational behavior,
8: 553-111. Greenwich, CT: JAI.

114

APPENDIX: QUESTIONNAIRES

Part I

This research project consists of three rounds of survey. To track your responses at the three
different surveys, it is VERY IMPORTANT to identify who you are.
In order to protect your privacy while we identify respondents, the class roster with ID numbers is
provided.

Please find your ID number from the enclosed roster, and provide the following information
about yourself.

ID number (from the attached roster) : ________
Age : _______ Gender : _______ Ethnicity : _________

In this part of the survey, you are presented with thirty-three statements that require you to
indicate the extent of your agreement with those statements. Please circle the number that
reflects the extent of agreement with that item.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

1. I find that a well-ordered life with regular hours suits my temperament.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

2. I would rather cooperate with others than compete with others.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

3. I tend to struggle with most decisions.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

115

4. I dislike unpredictable situations.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

5. I feel uncomfortable when someone‘s meaning or intention is unclear to me.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

6. People are always interested only in their own welfare.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

7. Even after I have made up my mind about something, I am always eager to consider a
different opinion.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

8. I do not like to be with people who are capable of unexpected actions.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

9. I have believed that teamwork is the best way of getting results.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

10. Given the opportunity, people will take advantage of you when there is a benefit of doing so.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

11. When I am confused about an important issue, I feel very upset.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree
116

12. I would describe myself as indecisive.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

13. When considering most conflict situations, I can usually see how both sides could be right.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

14. I have found joint projects with other people very satisfying.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

15. I find that establishing a consistent routine enables me to enjoy life more.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

16. One can avoid falling into trouble by assuming that all people have a vicious streak.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

17. I prefer to socialize with familiar friends because I know what to expect from them.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

18. I always see many possible solutions to the problems I face.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

19. I prefer to work independently rather than in a group.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree
117

20. I enjoy having a clear and structured way of life.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

21. I tend to put off important decisions until the last possible moment.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

22. In dealing with strangers, one is better off to be cautious until they have provided evidence
that they are trustworthy.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

23. When thinking about a problem, I consider as many different opinions on the issue as
possible.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

24. I dislike it when a person‘s statement can mean more than one thing.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

25. I like to have a place for everything and everything in its place.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

26. No matter what they say, most people inwardly dislike putting themselves out to help others.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

118

27. I usually make important decisions quickly and confidently.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

28. I enjoy the uncertainty of going into a new situation without knowing what might happen.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

29. I feel uncomfortable when I do not understand the reason why an event occurred in my life.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

30. I have enjoyed activities that involve a high level of cooperating with other people.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

31. I would be willing to lend money to my friends when they request it.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

32. I would be willing to lend personal possessions (e.g. CDs, clothes, bicycle, etc.) to my friends
when they need them.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

33. If I lived with a roommate, I would keep my valuables locked in a drawer.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

119
Part II

In this part of survey, you are presented with six statements about your teammates that require you to indicate the
extent of your agreement with those statements. Please circle the number that reflects the extent of agreement with that
item.
1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

Again, it is very important to identify who you are. Be sure to put your ID number below.
Your ID number (from the attached roster) : _______________________
We also need the information as to which teammate you are evaluating in order to construct dyadic relationship.
This dyadic relationship is the unit of analysis for this research. Thus, it is also very important to write your teammates
ID numbers from the attached roster.

120
Please circle the number that reflects the extent of agreement with the statements about your teammates.
1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

Put the ID numbers of your teammates here 
Teammate1 Teammate2 Teammate3 Teammate4 Teammate5

If I have to choose ONE partner to work with in
this class, and if I choose this person, we would
be able to work together in pleasant and
productive manner.
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
5 5 5 5 5
6 6 6 6 6
If I am to distribute the roles for the final project
and let this person do the MOST important part,
he/she will get the job done satisfactorily.
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
5 5 5 5 5
6 6 6 6 6
If I ask this person for help when I have difficulty
doing my role in the final project, he/she will
respond with care and attentiveness.
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
5 5 5 5 5
6 6 6 6 6
121

Put the ID numbers of your teammates here 
Teammate1 Teammate2 Teammate3 Teammate4 Teammate5

If I let this person be a leader who makes
important decisions on behalf of our group, the
performance of our group will be successful.
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
5 5 5 5 5
6 6 6 6 6
If I freely share my ideas with this person, he/she
will not negatively criticize it. Rather, he/she will
make constructive comments on it.
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
5 5 5 5 5
6 6 6 6 6
Even if I do not monitor this person, he/she will
DO THE BEST to get the job done.
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
5 5 5 5 5
6 6 6 6 6

122
Below is the question as to your relationship with your teammates prior to taking this class.

Put your teammates’ IDs here 
Teammate1 Teammate2 Teammate3 Teammate4 Teammate5

Which is the best description of your
relationship with this person?
1) Never met before
2) I know his/her name and face
3) I talked to him/her a couple of times
4) I am somewhat close to him/her
5) We are very close to each other

1
2
3
4
5

1
2
3
4
5

1
2
3
4
5

1
2
3
4
5

1
2
3
4
5
Did you know this person before this semester?
(If yes, answer the two questions below)
Y / N Y / N Y / N Y / N Y / N
If yes,
How long have you known this person
for?
(Answer in months) mo mo mo mo mo
How often did you meet this person?
1) more than twice a week
2) once or twice a week
3) once a month
4) twice a year
5) less than once a year

1
2
3
4
5

1
2
3
4
5

1
2
3
4
5

1
2
3
4
5

1
2
3
4
5

123
Part III

In this part of the survey, you are presented with six statements about your teammates that require you to indicate the
extent to which you agree those statements. Please circle the number that reflects the extent of agreement with that
item.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

Again, it is very important to identify who you are. Be sure to put your ID number from the attached roster in the space
below:

Your ID number (from the attached roster): _______________________

Since we are interested in understanding the relationships between people we need you to identify the persons who are
part of your team. Review the attached roster and put the number associated with your teammates in the space below.

124
Please circle the number that reflects the extent of agreement with the statements about your teammates.

1 2 3 4 5 6
Strongly
Disagree
Disagree Somewhat
Disagree
Somewhat
Agree
Agree Strongly
Agree

Put the ID numbers of your teammates here 
Teammate1 Teammate2 Teammate3 Teammate4 Teammate5

If I have to choose ONE partner to work with in
this class, and if I choose this person, we would
be able to work together in pleasant and
productive manner.
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
5 5 5 5 5
6 6 6 6 6
If I am to distribute the roles for the final project
and let this person do the MOST important part,
he/she will get the job done satisfactorily.
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
5 5 5 5 5
6 6 6 6 6
If I ask this person for help when I have difficulty
doing my role in the final project, he/she will
respond with care and attentiveness.
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
5 5 5 5 5
6 6 6 6 6

125

Put the ID numbers of your teammates here 
Teammate1 Teammate2 Teammate3 Teammate4 Teammate5

If I let this person be a leader who makes
important decisions on behalf of our group, the
performance of our group will be successful.
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
5 5 5 5 5
6 6 6 6 6
If I freely share my ideas with this person, he/she
will not negatively criticize it. Rather, he/she will
make constructive comments on it.
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
5 5 5 5 5
6 6 6 6 6
Even if I do not monitor this person, he/she will
DO THE BEST to get the job done.
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
5 5 5 5 5
6 6 6 6 6

126
Below is the question as to your experiences in relationship with your teammates.

Put your teammates’ IDs here 
Teammate1 Teammate2 Teammate3 Teammate4 Teammate5

During this negotiation project, HOW MANY TIMES have you experienced the events such as below in
relationships with your teammates?

Events showing that this person has proper
knowledge and capability necessary to contribute to
the negotiation.

Events showing that this person is willing to do
his/her best to get the job done.

Events showing that this person respects others’
opinions.

Events showing that this person is incapable of
making a contribution to the negotiation.

Events showing that this person is unwilling to do
his/her own part of the negotiation.

Events showing that this person is disrespectful of
others.

127

Asset Metadata

Creator Lee, Hyung-Woo (author)

Core Title Individual differences in trust development: Assessing the effects of trustor attributes on trust building, stability, and dissolution

Contributor Electronically uploaded by the author (provenance)

School School of Policy, Planning, and Development

Degree Doctor of Philosophy

Degree Program Policy, Planning, and Development

Publication Date 05/05/2011

Defense Date 11/24/2010

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,Personality,team,Trust

Language English

Advisor Robertson, Peter J. (committee chair), Kathleen, Wilber (committee member), Myrtle, Robert C. (committee member)

Creator Email hyungwol@usc.edu,jmuranak@sppd.usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m3915

Unique identifier UC1301044

Identifier etd-Lee-4273 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-472389 (legacy record id),usctheses-m3915 (legacy record id)

Legacy Identifier etd-Lee-4273.pdf

Dmrecord 472389

Document Type Dissertation

Rights Lee, Hyung-Woo

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email uscdl@usc.edu

Abstract (if available)

Abstract Since fostering trust is critical for the well-beings of today's organizations, managers need to understand how trust is formed and develops. Considering that interpersonal trust is a result of the confluence among three different factors, namely, attributes of trustor, trustee, and the situation

Linked assets

University of Southern California Dissertations and Theses

Individua​l difference​s in trust developmen​t: Assessing the effects of trustor attributes on trust building, stability, and dissolutio​n

Individua​l difference​s in trust developmen​t: Assessing the effects of trustor attributes on trust building, stability, and dissolutio​n

Individual differences in trust development: Assessing the effects of trustor attributes on trust building, stability, and dissolution

Individual differences in trust development: Assessing the effects of trustor attributes on trust building, stability, and dissolution