University of South Florida

Scholar Commons
Graduate Theses and Dissertations Graduate School

2005

Transformational leadership, leader-member

exchange (LMX), and OCB: The role of motives
Patrick W. Connell
University of South Florida

Follow this and additional works at: http://scholarcommons.usf.edu/etd

Part of the American Studies Commons

Scholar Commons Citation

Connell, Patrick W., "Transformational leadership, leader-member exchange (LMX), and OCB: The role of motives" (2005). Graduate
Theses and Dissertations.
http://scholarcommons.usf.edu/etd/2833

This Dissertation is brought to you for free and open access by the Graduate School at Scholar Commons. It has been accepted for inclusion in
Graduate Theses and Dissertations by an authorized administrator of Scholar Commons. For more information, please contact
[email protected].
 Transformational Leadership, Leader-Member Exchange (LMX), and OCB:

The Role of Motives

by

Patrick W. Connell

A dissertation submitted in partial fulfillment

of the requirements for the degree of
Doctor of Philosophy
Department of Psychology
College of Arts and Sciences
University of South Florida

Co-Major Professor: Walter C. Borman, Ph.D.

Co-Major Professor: Tammy D. Allen, Ph.D.
Michael T. Brannick, Ph.D.
Marcia A. Finkelstein, Ph.D.
Toru Shimizu, Ph.D.

Date of Approval:
October 21, 2005

Keywords: organizational concern, prosocial values, impression management, altruism,

conscientiousness, sportsmanship, courtesy, civic virtue

© Copyright 2005, Patrick W. Connel1

 Acknowledgements

This study would not have been possible without the help and support of certain

individuals. First, I would like to thank the many employees who were willing to provide

their feedback as part of this study, as well as the organizations who the saw value in this

line of research. Second, I would also like to thank my committee members, Drs. Tammy

Allen, Michael Brannick, Marcia Finkelstein, and Toru Shimizu, for the valuable insights

and knowledge they provided throughout this entire process. Finally, I owe a wealth of

gratitude to my advisor, Dr. Walter Borman. My graduate career would not have been a

success without his continuous guidance and support.

A special thank you is also owed to my family whose encouragement and support

was invaluable throughout this process. Finally, I would like to thank my partner, Jon, for

his incredible spirit and constant encouragement to pursue my goals.

 Table of Contents

List of Tables iii

List of Figures iv

Abstract v

Chapter 1: Introduction 1
Organizational Citizenship Behavior (OCB) 2
Leadership and OCB 12
Transformational Leadership 12
Leader-member Exchange 23
OCB Motives 31
Moderators and Mediators 34
The Current Study 35

Chapter 2: Method 41
Participants 41
Measures 45
Transformational Leadership 45
Leader-member Exchange 45
OCB 46
OCB Motives 46
Procedure 47

Chapter 3: Results 50
Preliminary Steps and Analyses 50
Variable Descriptives 51
Zero-order Correlations 51
Analysis Approach 58
Hypothesis 1 60
Hypothesis 2 60
Self-reports of OCB 60
Supervisor-reports of OCB 61
Hypothesis 3 64
Self-reports of OCB 64
Supervisor-reports of OCB 65

i
 Hypothesis 4 66
Self-reports of OCB 66
Supervisor-reports of OCB 66
Hypothesis 5 67
Self-reports of OCB 67
Supervisor-reports of OCB 68
Hypothesis 6 68
Hypothesis 7 68
Self-reports of OCB 68
Supervisor-reports of OCB 70
Hypothesis 8 71
Self-reports of OCB 71
Supervisor-reports of OCB 71
Additional Analyses 72
Transformational Leadership, Altruism, and the OC and PV Motives 72
LMX-quality, OCB, and Motives 73
Mediational Analyses from the Supervisors’ Leadership Perspective 76

Chapter 4: Discussion 78
Relationships Among Variables 79
Transformational Leadership, Motives, and OCB 85
LMX-quality, Motives, and OCB 86
Comparative Mediational Effects 89
Different Leadership Perspectives (Subordinate versus Supervisor) 91
Limitations 92
Future Research 94
Conclusions 98

References 99

Appendices 108
Appendix A: Transformational Leadership Inventory (TLI),
Subordinate Version 109
Appendix B: Transformational Leadership Inventory (TLI),
Supervisor Version 111
Appendix C: LMX7, Subordinate Version 113
Appendix D: LMX7, Supervisor Version 115
Appendix E: OCB Measure, Subordinate Version 117
Appendix F: OCB Measure, Supervisor Version 119
Appendix G: Citizenship Motives Scale 121

ii
 List of Tables

Table 1 TLI Transformational Leadership Dimensions 20

Table 2 Participant Demographics (N = 186) 43

Table 3 Supervisor Survey Instructions 48

Table 4 Descriptive Statistics for Study Variables 52

Table 5 Variable Reliabilities and Intercorrelations for Self Ratings

of OCB (N = 131) 53

Table 6 Variable Reliabilities and Intercorrelations for Supervisor Ratings

of OCB (N = 131) 54

Table 7 Variable Reliabilities and Intercorrelations for Self and Supervisor

Ratings of OCB (N = 131) 55

Table 8 Motives Mediator Analysis for Transformational Leadership and

Self-reports of OCB 62

Table 9 Motives Mediator Analysis for Transformational Leadership and

Supervisor-reports of OCB 63

Table 10 Motives Mediator Analysis for LMX-quality and Self-reports of

Altruism 69

Table 11 Motives Mediator Analysis for LMX-quality and Supervisor-reports

of Altruism 71

Table 12 Motives Mediator Analysis for Transformational Leadership and

Self-reports of Altruism 73

Table 13 Motives Mediator Analysis for LMX-quality and Self-reports of

OCB 74

Table 14 Motives Mediator Analysis for LMX-quality and Supervisor-reports

of OCB 75

Table 15 Motives Mediator Analysis Based on the Supervisors’ Perspective

of LMX-quality and OCB 77
iii
 List of Figures

Figure 1 A Mediated Model of the Effects of Transformational

Leadership and LMX-quality on OCB 78

Figure 2 Penner et al.’s (1997) Conceptual Model of OCB 93

iv
 Transformational Leadership, Leader-member Exchange (LMX), and OCB:

The Role of Motives

Patrick W. Connell

ABSTRACT

The purpose of this study was to investigate the impact of employee motives

regarding select leadership-OCB relationships. Based on previous research, it was

hypothesized that the relationships observed between transformational leadership and

various dimensions of OCB would be mediated by subordinate Organizational Concern.

In contrast, the relationship between LMX-quality and subordinate Altruism was

predicted to be either mediated or moderated by subordinate Prosocial Values.

Two hundred and one part-time and full-time employees (subordinates and

supervisors) served as participants in this study, representing a total of 13 organizations

in the Southeast United States. Results were based on a final sample of 131 supervisor-

subordinate pairs. In general, participants responded to questionnaires that measured

transformational leadership, LMX-quality, and OCB Motives (i.e., Prosocial Values,

Organizational Concern, and Impression Management). Both subordinate and supervisor

ratings of OCB were also collected.

Analyses were based upon Baron and Kenny’s (1986) approach for mediation and

moderation, as well as the Aroian version (1944/1947) of the Sobel test (1982). Across

self- and supervisor-reports of OCB, results revealed that the Organizational Concern

v
Motive significantly mediated the relationship between transformational leadership and

various dimensions of OCB (Conscientiousness, Sportsmanship, Courtesy, and Civic

Virtue). Results also supported the Prosocial Values Motive as a partial mediator in the

relationship between LMX-quality and self-reported Altruism. Surprisingly, a stronger

mediating effect was consistently observed for the Organizational Concern Motive across

both leadership styles and all five of Organ’s (1988) OCB dimensions. In contrast, no

evidence was found for either motive with regard to moderation. Results also differed

based on leadership perspective (subordinate versus supervisor).

Taken as a whole, these results suggest that both transformational leadership and

LMX-quality are strongly associated with an employee’s general concern for the

organization. This motive is, in turn, associated with a variety of citizenship behaviors. In

summary, this evidence addresses an important gap in the OCB literature by providing

evidence for an indirect relationship between leadership perceptions and OCB.

vi
 Chapter One

Introduction

Research in the area of organizational citizenship behavior (OCB) has shown a

dramatic increase in the last few years. This trend is illustrated by the rapid growth in

publications dealing with OCB over recent decades, ranging from 13 occurring in the

period from 1983 to 1988, to 122 in the period from 1993 to 1998 (Podsakoff,

MacKenzie, Paine, & Bachrach, 2001). Although research has been extensive in

addressing the numerous antecedents of OCB (e.g., job satisfaction, perceptions of

fairness, personality factors), less attention has been focused on other important areas

related to the construct. One such area is the mechanisms by which certain antecedents

influence citizenship performance, as well as the potential for additional dispositional

variables to moderate antecedent-OCB relationships (Podsakoff et al., 2001).

The purpose of this study is to address this particular gap in the literature by

further investigating the role of motives in relation to OCB. Building on the results of

past research that has found evidence for motives as both moderators and mediators

between certain antecedent variables (both attitudinal and dispositional) and select

dimensions of OCB (Tillman, 1998; Connell & Penner, 2004), the current study explores

whether the effects of certain leadership styles (e.g., transformational leadership, leader-

member exchange) on OCB reflect a similar trend. That is, contingent upon the type of

leadership style and OCB motive explored, it is expected that the relationship between

1
leadership behaviors and OCB is either moderated or mediated by motives. The

following introduction discusses four major areas of research relevant to this hypothesis:

(1) the nature of OCB and its antecedents, (2) transformational leadership theory, (3)

leader-member exchange (LMX) theory, and (4) OCB motives. At the conclusion, these

four streams of research are tied together to form the foundation of the current study.

Organizational Citizenship Behavior (OCB)

Much of the work done on the conceptual framework of OCB is similar to

research carried out by Borman and Motowidlo (1993) and Motowidlo and Van Scotter

(1994). Specifically, these researchers distinguished between two types of job

performance. The first of these is task performance, which they defined as “the

effectiveness with which job incumbents perform activities that contribute to the

organization’s technical core either directly by implementing a part of its technical

process, or indirectly by providing it with needed materials or services” (Borman &

Motowidlo, 1997, p. 99). For example, for a sales manager position, task performance

activities would include keeping track of inventory, scheduling employees, and aiding

and assisting customers. The second type of performance is contextual performance.

Contextual performance includes activities that “shape the organizational, social, and

psychological context that serves as the catalyst for task activities and processes”

(Borman & Motowidlo, 1997, p. 100). Contextual activities are volitional, and include

behaviors that may not be in an employee’s formal job description. Some examples of

contextual performance include cooperating with other employees to accomplish tasks,

working extra hours on a project even though it is not required, or volunteering to

organize social events for the organization. Borman and Motowidlo suggest that

2
contextual performance makes a significant and valued contribution in organizations, and

that in contrast to task performance that is specific to a particular job, contextual

performance is more generalized and can cut across numerous jobs or occupations. In

addition, Borman and Motowidlo (1993) have identified five specific categories of

contextual performance: (1) volunteering to carry out task activities that are not formally

part of the job, (2) persisting with extra enthusiasm when necessary to complete own task

activities successfully, (3) helping and cooperating with others, (4) following

organizational rules and procedures even when its is personally inconvenient, and (5)

endorsing, supporting, and defending organizational objectives.

The construct of OCB is similar to contextual performance. Specifically, OCB

was originally described by Organ (1988) as “individual behavior that is discretionary,

not directly or explicitly recognized by the formal reward system, and that in the

aggregate promotes the effective functioning of the organization” (p.4). Although Organ

initially defined OCB as extra-role behavior (i.e., behavior that is beyond an individual’s

job requirements), he has since acknowledged that the distinction between in-role and

extra-role performance is inherently “muddy” due to the role of supervisor expectations

in the leader-member exchange dyad. More specifically, leader expectations can range

from beliefs that are far below formal job requirements to those that go above and beyond

them (Katz & Kahn, 1966). Thus, agreement on what is extra-role behavior can vary

considerably depending on the source of inquiry (i.e., supervisors, subordinates, or

peers). As a result, Organ has redefined OCB to refer to contextual performance, or

behavior that “shapes the organizational, social, and psychological context that serves as

the catalyst for task activities and processes” (Borman & Motowidlo, 1997, p. 100).

3
Although the essential notions of OCB remain the same, this new conceptualization shifts

the focus from the dichotomy of in-role and extra-role performance to an emphasis on

task and non-task behaviors.

Earlier research investigating OCB identified two main dimensions, Altruism and

Conscientiousness (Smith, Organ, & Near, 1983; Organ & Konovsky, 1989). Later

efforts expanded this framework to include three additional dimensions: Sportsmanship,

Courtesy, and Civic Virtue (Organ & Ryan, 1995). This dimensional structure is

conceptually similar to the five categories used to describe contextual performance, and

is still widely used in research investigating OCB.

The Altruism dimension is used to describe OCB behaviors that are directed

toward members of the organization (Organ, 1988; Smith, Organ, & Near, 1983). This

type of helping behavior can be job-related, such as assisting a co-worker with a specific

project or work task, or non-job-related, such as helping a co-worker or supervisor with a

personal problem.

Conscientiousness (Generalized Compliance) refers to more “impersonal

contributions to the organization” such as excellent attendance, and adherence to

organizational rules and policies (Organ & Ryan, 1995, p.782). These contributions are

not directed at any one person or co-worker, but are indirectly helpful to other members

of the organization (Smith et al., 1983).

The dimension of Courtesy refers to behaviors that are intended to help prevent

problems of coworkers (Organ & Ryan, 1995). These behaviors contribute most

importantly to the smooth functioning of the organization, and involve both formal and

4
informal cooperation among employees (Organ, 1997; George & Brief, 1992; Konovsky

& Organ, 1996).

Sportsmanship refers to “the inclination to absorb minor inconveniences and

impositions accruing from the job without complaints or excessive demands for relief or

redress (Konovsky & Organ, 1996, p. 255). Thus, a person high on this dimension would

not complain about trivial aspects of the job, and would be inclined to think about others’

work problems in addition to his or her own (Konovsky & Organ, 1996).

The last dimension, Civic Virtue, refers to behaviors that represent active

involvement and interest regarding organizational issues, as well as the governance of the

organization as a whole (Organ & Ryan, 1995). This dimension includes behaviors such

as attending meetings, reading and answering company email, keeping informed on

organizational developments, and playing an active role in the overall running of the

organization (Konovosky & Organ, 1996).

In general, researchers have suggested (Borman & Motowidlo, 1993; Organ,

1988; Smith et al., 1983) that OCB can have a beneficial effect on the organization by

“lubricating” such aspects as its “social machinery,” increasing efficiency, and reducing

friction among employees. As suggested earlier by Katz, organizational success is

dependent upon more than just prescribed role behaviors, and creative behavior, such as

OCB, “is vital to organizational survival and effectiveness” (1964, p.132). One of the

ways in which OCB may enhance efficiency is by improving coworker or managerial

productivity (MacKenzie et al., 1993; Organ, 1988; Podsakoff & MacKenzie, 1997). For

example, Podsakoff and MacKenzie (1997) combined the results of four independent

studies and found that “OCB accounted for an average of approximately 19% of the

5
variance in performance quantity, over 18% of the variance in the quality of performance,

about 25% of the variance in financial efficiency indicators, and about 38% of the

variance in customer service indicators” (p. 142). These results provide empirical support

for the assumption that OCB is related to organizational effectiveness. Other suggested

ways in which OCB can affect efficiency include freeing up company resources to be

used for more productive purposes, aiding in the effective coordination of work teams,

and enhancing the ability of organizations to adapt to change (Podaskoff & MacKenzie,

1997).

OCB is also important at the level of the individual employee. This notion is

illustrated through studies that showed that OCB contributed independently to overall

evaluations of employee performance (Motowidlo & Van Scotter, 1994; Borman &

Motowidlo, 1997). In a study by Orr, Sackett, and Mercer (1989), supervisors were

shown to take both prescribed and discretionary behaviors into account when evaluating

employee job performance. Werner (1994) also provided evidence for an interaction

between in-role performance and OCB. Specifically, when employee in-role performance

was shown to be low, overall ratings of performance were also low regardless of the level

of OCB displayed. However, as in-role performance increased, ratings of overall

performance increased more sharply for high OCB employees than for those displaying

average levels of OCB. Although using a somewhat outdated conceptualization of OCB

(i.e. extra-role performance), these findings still strongly suggest that supervisors

consider discretionary behaviors during the performance appraisal process (Werner,

1994). Thus, the notion that OCB is an important component of effective performance,

both at the organizational and employee level, is supported by the OCB literature.

6
 Other research has addressed the antecedents of OCB. These range from

employee perceptions (e.g., Smith et al.,1983; Organ & Ryan, 1995; Konovsky & Organ,

1996; Skarlicki & Latham, 1996) to the personality characteristics associated with this

type of behavior (e.g., Borman & Motowidlo, 1993; Borman, Penner, Allen, &

Motowidlo, 2001). In a meta-analysis that combined the results of 55 studies, Organ and

Ryan (1995) identified a number of variables as antecedents of OCB. One of the primary

variables identified was job attitudes. For example, Organ and Ryan (1995) found that

employee job satisfaction correlated significantly with both the Altruism (.28) and

Generalized Compliance (.28) dimensions of OCB. Other notable findings involved

perceptions of organizational justice and organizational commitment, both of which also

correlated significantly with both the Altruism and Generalized Compliance dimensions.

These results were further confirmed in a recent meta-analysis conducted by Podsakoff et

al. (2001), which also reported significant relationships between each of the antecedents

and both OCB dimensions.

According to the results described above, employees who are satisfied with their

jobs are more likely to engage in altruistic and generalized compliance behaviors than

employees who are less satisfied. Similarly, those employees who possess high levels of

perceived justice or high levels of organizational commitment also tend to perform more

OCBs than employees who display lower levels of each of these antecedents. These

results should be tempered with the fact that other research has reported additional

findings that suggest a slightly less straightforward relationship between these constructs.

It has been suggested, for example, that perceptions of justice may account for the

significant relationship found between job satisfaction and OCB (Moorman, 1991;

7
Williams & Anderson, 1991). More specifically, both Moorman (1991) and Williams and

Anderson (1991) provided evidence that when perceptions of fairness were controlled,

the relationship between job satisfaction and OCB was no longer significant. Evidence

also suggests that the type of commitment experienced by the employee (e.g., affective,

continuance, or normative) plays an important role in relation to the performance of

citizenship behaviors. For example, Organ and Ryan (1995) found significant average

correlations between affective commitment (an emotional attachment to the organization)

and the Altruism (.23) and Generalized Compliance (.30) dimensions of OCB. In

contrast, continuance commitment (feeling committed to the organization because of the

salary or benefits associated with it) showed no significant correlation with either OCB

dimension. Thus, although evidence is generally supportive of the relationship between

job attitudes and OCB, additional research is needed to further refine and clarify the

nature of these relationships.

Some researchers have also suggested that personality may play a role in OCB

(Borman & Motowidlo, 1993; Borman et al., 2001). Personality characteristics will most

likely be expressed in behaviors that involve planful actions, occur over an extended

period of time and a variety of situations, and which are not limited by formal

requirements or characteristics of the situation (Funder, 1995). As described previously,

OCB shares many of these characteristics.

Research in this area has provided mixed results. For example, in the meta-

analysis conducted by Organ and Ryan (1995), relationships between certain personality

characteristics and selected OCB dimensions were examined. Specifically, these

researchers addressed the personality traits of Agreeableness and Conscientiousness

8
taken from the Five-Factor Model of Personality (Costa & McCrae, 1992), as well as

positive affectivity and negative affectivity (Watson & Clark, 1992). Results showed that

among the personality variables examined, only the trait of conscientiousness showed

even moderate correlations with OCB. More specifically, conscientiousness correlated

.22 with the Altruism dimension and .30 with the Generalized Compliance dimension

(these correlations were corrected for criterion unreliability and restriction of range).

Agreeableness was also shown to correlate significantly with both the Altruism

(uncorrected r = .13) and Generalized Compliance dimensions (uncorrected r = .11).

However, these relationships were generally not as strong as those observed for

conscientiousness. Based on these results, Organ and Ryan concluded that with the

exception of conscientiousness, it is unlikely that personality plays a direct role in OCB.

Borman et al. (2001) analyzed research findings since Organ and Ryan’s meta-

analysis and reported more promising results. For example, Neuman and Kickul (1998)

found a significant relationship between conscientiousness and all five of Organ’s OCB

dimensions. In addition, recent studies have found significant support for the relationship

between agreeableness and OCB (e.g., Hense, 2000; McManus & Kelly, 1999; Neuman

& Kickul, 1998; Van Scotter & Motowidlo, 1996), and have even identified additional

variables that may affect citizenship performance, such as locus of control (Motowidlo &

Van Scotter, 1994; Funderberg & Levy, 1997), collectivism (e.g., Moorman & Blakely,

1995; Van Dyne, Vandewalle, Kostova, Latham, & Cummings, 2000; Allen, 1999), and

personal initiative (Facteau, Allen, Facteau, Bordas, & Tears, 2000). Finally, in a recent

meta-analysis by Podsakoff et al. (2001), significant relationships were found between

the Altruism dimension and conscientiousness (r = .22), agreeableness (r = .13), and

9
positive affectivity (r = .15). In addition, conscientiousness (r = .30), agreeableness (r =

.11), and negative affectivity (r = -.12) all correlated significantly with the Generalized

Compliance dimension. Taken together, these results suggest that personality may be

more strongly related to OCB than originally reported by Organ and Ryan.

As can be seen from the wealth of previous studies, research in the area of OCB

has been extensive in covering a wide range of antecedents and outcomes associated with

this construct. Looking across this research, however, certain theoretical and

methodological issues have also surfaced which deserve mention. For example, in a

recent review of the OCB literature, Podsakoff et al. (2001) concluded that common

method variance has had a significant impact on observed OCB relationships reported in

studies where this artifact was not controlled. Specifically, in a review of 11 field studies

dealing with OCB, results revealed that when common method variance was not

controlled, the proportion of performance variance explained by objective performance

averaged 9.5 percent, whereas the amount explained by OCBs averaged 42.9 percent. In

contrast, when common method variance was controlled, the average amount explained

by objective performance averaged 11.3 percent, while the amount explained by OCBs

decreased to an average of 19.3 percent. In light of these results and others displaying the

same general trend, Podsakoff et al. (2001) concluded that although common method

variance can have a significant impact on the relationship between OCB and managerial

judgments, “this bias generally weakens these relationships, it does not eliminate them”

(p. 543).

On the same topic, two related methodological issues include the need to obtain

evidence for the direction of causality between OCB, its antecedents, and outcomes, as

10
well as the need to conceptually distinguish measures of OCB and contextual

performance from other closely related constructs (Podsakoff et al., 2001). Because the

majority of OCB research has been cross-sectional in nature, it is currently not clear

whether OCB is the cause in certain investigated relationships or the effect. Although

certain studies (e.g., Koys, 2001) have provided longitudinal evidence that employee

attitudes (e.g., job satisfaction) and behaviors (e.g., OCB) predict organizational

effectiveness (rather than vice versa), there is still a significant need for longitudinal

research in this area that further addresses the impact of these relationships over time.

Similarly, as a result of many overlapping definitions of OCB-like behavior, it is

necessary to test the discriminant validity of each of these constructs. Although the

majority of research has focused on the relationships between OCB and other constructs,

these relationships may be misinterpreted if the nature of the construct itself (i.e., OCB)

is not truly understood.

Finally, Podsakoff et al.(2001) point out the need to investigate additional

antecedents of OCB. A number of task variables (e.g., task feedback, task routinization,

and intrinsically satisfying tasks), for example, have shown consistent relationships with

OCB across a small number of studies. In addition, certain leadership styles (mainly

transformational leadership and leader-member exchange) have also shown consistent

relationships across all five OCB dimensions. Both of these variables remain grossly

under investigated in the literature, despite their potential as an obvious predictor of

citizenship performance. Similarly, personality variables alternative to the “Big Five”

dimensions also deserve further consideration in relation to OCB. Other-oriented

Empathy (the tendency to experience empathy for, and to feel responsibility and concern

11
about, the well-being of others) and Helpfulness (the self-reported history of engaging in

helpful actions and an absence of egocentric physical reactions to other’s distress), for

example, have correlated significantly with both the Altruism and Generalized

Compliance dimensions across several studies (Penner, Midili, & Kegelmeyer, 1997;

Rioux & Penner, 2001; Connell & Penner, 2004). As cited in Podsakoff et al. (2001), Van

Dyne, Cummings, & Parks (1995) suggest that the propensity to trust, need for

affiliation, and empathic concern might also be worthwhile constructs to explore in the

context of contextual performance. Thus, although research on OCB and its related

constructs is both expansive and comprehensive, certain areas deserve further

clarification and refinement. As mentioned previously, one such area is the impact of

different leadership behaviors on OCB, a suggestion that is expanded upon in the current

study.

Leadership and OCB

Research investigating predictors of employee performance has suggested that

specific types of leadership behaviors are also important to employee task and contextual

performance. Two primary theories of leadership that have shown consistent

relationships with employee performance are Transformational Leadership theory and

Leader-Member Exchange (LMX) theory. Although certain elements are shared between

the two theories, each are considered distinct constructs, and have individually been

shown to predict positive outcomes at both the individual and organizational level.

Transformational Leadership. The origin of transformational leader theory lies

with the work of Burns (1978) who originally proposed two distinct leadership styles

based on his analysis of the behaviors displayed by various political leaders. The first of

12
these, transactional leadership, characterized many of the traditional leadership theories

existing at that time. According to Burns (1978), transactional leadership was based on an

exchange process between leaders and subordinates where rewards were administered to

employees based upon acceptable levels of displayed effort and performance. This type

of leadership was in contrast to transformational leadership, the goal of which was to

encourage followers to transcend their own self-interests and move beyond simple leader-

member transactions for the good of the group or organization (Burns, 1978; Bass, 1985).

Under this type of direction, followers were also expected to gain increased awareness for

valued outcomes as well as their own higher level needs; the end result being a

heightened desire to exceed traditional performance expectations.

Although Burns is credited with the original identification of these two types of

leadership styles, much of the subsequent work on transformational leadership has been

done under the direction of Bass and colleagues (e.g., Bass, 1985; Bass & Avolio, 1990;

Bass, 1997). Bass’s conception of transformational leadership is very similar to that of

Burn’s. However, their perspectives diverge in three main respects. First, Bass’s (1985)

conceptualization makes specific reference to the expansion of the follower’s “portfolio

of needs and wants” which is absent in Burns’s description of the construct (p. 20).

Second, according to Burns, a necessary component of transformational leadership is that

followers are elevated to a goal that is inherently good or positive. Bass does not make

this distinction, and considers all cases where the needs and actions of followers are

“transformed,” regardless of the nature of the intent (e.g., positive or negative), as

examples of transformational leadership. Finally, the most notable distinction between

these two researchers’ perspectives deals with the relationship between transformational

13
and transactional leadership. Burns specifically views these leadership styles as polar

constructs, with transactional leadership on one end of the continuum, and

transformational leadership on the other. In contrast, Bass’s view was that both constructs

are complimentary in nature; a leader may display both transactional and

transformational leadership behaviors to some degree. Overall, transformational

leadership is proposed to augment the effects of transactional leadership in terms of

subordinate performance, a theory now labeled the “augmentation hypothesis”

(Waldman, Bass, & Yammarino, 1990).

In general, research supports Bass’s theory that transformational leadership

enhances the effects of transactional leadership. For example, based on a sample that

included both U.S. Army officers and Fortune 500 managers, Bass (1985) found that

transformational leadership behaviors accounted for significant variance in subordinate

extra-effort and subordinate-rated leader effectiveness above and beyond what was

accounted for by transactional leadership behaviors. Similar results were also reported by

Hater & Bass (1988) using managers at an air delivery service company, as well as by

several additional studies that have investigated the augmentation hypothesis (e.g., Bycio,

Hacket, & Allen, 1995; Waldmen et al., 1990). In general, Bass (1985) conceptualizes the

transactional leader as one who works within the existing culture and constraints of the

organization, placing a higher emphasis on process (e.g., leader-member exchanges) as

opposed to outcomes. The transactional leader also places a premium on maintaining

efficiency, and is most likely to be effective in environments that are stable and

predictable. In contrast, Bass (1985) characterizes the transformational leader as one who

challenges the organization’s systems and culture rather than accepts them. More

14
specifically, transformational leaders are more likely to challenge the status quo by

seeking new and creative ways of accomplishing goals. Transformational leaders also

tend to be less risk avoidant than transactional leaders, and generally emphasize

effectiveness over efficiency. As stated by Avolio and Bass (1988), transformational

leaders attempt to create and shape their environments rather than simply react to the

circumstances that are provided to them.

At the time of Bass’s (1985) theoretical conceptualization of transformational

leadership, no valid measure existed to measure the construct. This led Bass and

colleagues to develop their own measure of transactional and transformational leadership

which they labeled the Multifactor Leadership Questionnaire (MLQ). The development

of the MLQ was based on a review of the literature as well as survey responses provided

by 70 senior executives who were asked to describe the qualities of both transactional and

transformational leaders (Bass, 1985; Bass, Avolio, & Goodheim, 1987). Factor analysis

of the responses suggested a five-factor structure for the measure, which has also been

demonstrated across additional studies (e.g., Hater & Bass, 1988). Three of the factors

identified by the questionnaire were interpreted as transformational, whereas two were

seen as transactional in nature. The first transformational factor was labeled Charisma,

which describes leader behaviors that instill pride, faith, and respect in subordinates,

communicate important issues, and clearly articulate a sense of mission and purpose. The

second transformational factor, Individual Consideration, involves leader behaviors such

as delegating projects to subordinates, showing a concern for follower development by

acting as a coach or mentor, and treating followers with respect and concern. Finally, the

third transformational factor was labeled Intellectual Stimulation, which describes leader

15
behaviors that emphasize subordinate problem solving and the ability of followers to

think creatively. In particular, followers are encouraged to submit their own opinions, and

are not criticized even if their ideas differ from their leader (Bass & Avolio, 1994).

In addition to these transformational factors, two transactional factors, Contingent

Reward and Management-by-Exception, were also identified. The Contingent Reward

factor describes the typical behaviors that embody a transactional leader. That is, this

factor identifies whether the leader rewards subordinate performance that is in

accordance with previous leader expectations. Management-by-Exception, however, is

somewhat more passive in nature. Specifically, this factor describes the leaders tendency

to avoid giving direction to subordinates if their level of performance is satisfactory. In

other words, a leader scoring high on this factor would be likely to simply let his/her

subordinates perform their jobs on their own as long as their level of performance was

considered acceptable.

The MLQ has been used in over 75 research studies, and has been tested in a

variety of organizational settings ranging from manufacturing and military settings to

religious organizations. In addition, respondents have ranged from first-line supervisors

to high-level managers (Lowe, Kroeck, & Sivasubramaniam, 1996). Most studies

involving the MLQ support the distinction between transactional and transformation

behaviors. For example, Bass (1985) revealed that the three transformational factors

(Charisma, Individual Consideration, and Intellectual Stimulation) were all highly

correlated, and that 66 % of the variance in the transformational scale was accounted for

by Charisma.

16
 Recent research has further explored the psychometric structure of the MLQ and

has found somewhat more mixed results. Bycio, Hackett, and Allen (1995), for example,

tested the fit of alternative two- and five-factor MLQ models using a sample of registered

nurses. Based on the confirmatory factor analysis fit indices, they concluded that Bass’s

(1985) five-factor structure was most appropriate to describe the nature of the MLQ.

However, strong correlations observed between the three transformational factors

suggested that a two-factor interpretation may also be plausible. Additional research has

also revealed a high positive correlation between the transformational factors and

Contingent Reward (Lowe et al., 1996), and other studies have shown that Management-

by-Exception sometimes loads on its own unique factor rather than on transactional

leadership. These results are inconsistent with Bass’s original conceptualization of

transformational and transactional leadership. Thus, although the MLQ remains the most

widely used measure of transformational leadership, additional work is warranted

regarding the psychometrics of the measure.

In terms of outcome variables, studies have shown that many of the

transformational leadership facets are associated with a number of positive leader and

subordinate outcomes. For example, in a study by Bycio et al. (1995) each of the

transformational dimensions showed high positive correlations with subordinate extra-

effort, satisfaction with the leader, affective commitment, and ratings of leader

effectiveness. In addition, significant negative relationships were also observed between

each of the transformational factors and the intent to leave the profession, and intent to

leave the job. In general, although the transactional leadership facets showed significant

relationships with a number of outcome variables, these effects were augmented with the

17
presence of transformational leadership behaviors. These results have also been

supported by other independent and meta-analytic studies which have addressed the

relationship between transformational leadership and subordinate outcomes (e.g., Lowe et

al. 1996). In addition to transformational factors, results have also been supportive of the

contingent reward factor across a number of studies. In general, however, these findings

have been weaker and less consistent (Yukl, 1999).

In the most general sense, the above results can be interpreted as evidence for a

positive effect of transformational leadership behaviors on subordinate outcomes.

Although this relationship appears relatively straightforward, other research has

suggested that additional variables may moderate the relationship between

transformational leadership and leader effectiveness. In a recent meta-analysis by Lowe

et al. (1996) including 39 studies involving the MLQ, it was found that the type of

organization (public versus private) and type of criterion (subordinate perceptions versus

organizational measures) moderated the transformational leadership-effectiveness

relationship. Contrary to prediction, a stronger positive relationship was found between

transformational leadership behaviors and leader effectiveness in public as opposed to

private organizations. In addition, significantly higher positive relationships were found

for subordinate perceptions as compared with organizational measures of effectiveness.

Although results did not support the level of the leader (low versus high) as a moderator,

the mean incidence of transformational leadership behavior was significantly higher for

low level as opposed to high level leaders. In addition, transformational leadership

behaviors were more commonly observed in public as opposed to private organizations.

These results are significant because they suggest that the relationship between

18
transformational leadership and leader effectiveness is contingent upon additional factors.

In addition, these findings contradict preconceived notions about transformational

leadership, specifically that the incidence of these behaviors is more prevalent in private

organizations and within upper levels of management.

Another important performance variable found in past studies to be related to

transformational leadership is OCB. Although a number of researchers have investigated

the relationship between transformational leadership and OCB, the majority of the work

in this area is credited to Podsakoff and colleagues (e.g., Podsakoff et al., 1990;

Podsakoff, MacKenzie, & Bommer; 1996; Podsakoff et al., 2001). As an alternative to

the MLQ, Podsakoff developed his own measure of transformational and transactional

leadership labeled the Transformational Leadership Inventory (TLI). Based on a review

of the transformational leadership literature, this measure consists of four first-order

transformational factors (see Table 1): high performance expectations, individualized

support, intellectual stimulation, and a “core” transformational behavior construct. In

addition, one first-order transactional leadership factor, contingent reward behavior, was

also identified. Each of these factors uses individual items as indicators. The only

exception is the “core” transformational construct, which uses individual factor scores for

three separate constructs as indicators: (1) articulating a vision; (2) providing an

appropriate model; and (3) fostering the acceptance of group goals. Initial confirmatory

factor analysis results support the existence of an overall six-factor structure for the

measure (Podsakoff et al., 1990; Podsakoff et al., 1996). However, other studies have

provided support for six first-order transformational behavior dimensions as opposed to

combining three of the constructs into the “core” transformational leadership factor

19
(Podsakoff et al., 1996). Based on these findings, additional research addressing the

dimensional nature of the measure is still warranted using different research samples.

Table 1. TLI Transformational Leadership Dimensions

High Performance Expectations Behavior that demonstrates the leader’s

expectation for excellence, quality, and/or
high performance expectations.
Individualized Support Behavior on the part of the leader that
indicates that he/she respects followers and
is concerned about their personal feelings
and needs.
Intellectual Stimulation Behavior on the part of the leader that
challenges followers to re-examine some of
their assumptions about their work and
rethink how it can be performed.
Articulating a vision Behavior on the part of the leader aimed at
identifying new opportunities for his/her
unit/division/company, and developing,
articulating, and inspiring others with his or
her vision of the future.
Providing an Appropriate Model Behavior on the part of the leader that sets
an example for employees to follow that is
consistent with the values the leader
espouses.
Fostering the Acceptance of Group Goals Behavior on the part of the leader aimed at
promoting cooperation among employees
and getting them to work together toward a
common goal.

Using the TLI, Podsakoff has found significant support for the relationship

between transformational leadership behaviors and OCB. For example, in an independent

study using employees of a petrochemical company, Podsakoff et al. (1990) found a

number of significant relationships between the TLI’s transformational factors and

Organ’s (1988) five OCB dimensions. Most notably, the “core” transformational

20
behavior dimension was found to correlate significantly with Conscientiousness (.27),

Sportsmanship (.20), Courtesy (.23), and Altruism (.22). Similar relationships were also

found with the individualized support dimension, that also correlated significantly across

all OCB dimensions except Civic Virtue. These results were later confirmed in a more

recent study involving corporate employees in both the U.S. and Canada (Podsakoff et

al., 1996). In this study, the subdimensions comprising the “core” transformational

construct were also examined separately, and revealed that each of the three constructs

(articulating a vision, providing an appropriate model, and fostering the acceptance of

group goals) correlated significantly with all OCB dimensions except Civic Virtue.

Since the discovery of these promising results, Podsakoff et al. (2001) have

conducted a meta-analysis examining the effects of transformational leadership on OCB

across studies. Results of this study again revealed significant relationships between each

of the TLI factors (including contingent reward behavior) and OCB. Most notable was

the finding that all TLI factors correlated significantly with the Altruism,

Conscientiousness, Courtesy, and Sportsmanship dimensions, with the majority of

correlations within the .20 to .25 range. In addition, significant (albeit smaller)

relationships were also observed between each of the core transformational constructs

and Civic Virtue, as well as between contingent reward behavior and the Civic Virtue

dimension. Taken together, these results show that transformational leaders have a

consistent positive impact on every form of citizenship behavior. These behaviors range

from OCBs directed at individual members of the organization, to those that are intended

to benefit the organization as a whole. As suggested by Podsakoff and colleagues (2001),

these results should not come as a shock, as the central notion of transformational

21
leadership is to encourage employees to perform above and beyond expectations.

Although studies have provided relatively strong support for the link between

transformational leadership and OCB, little research has provided insight regarding the

actual nature of these effects. That is, the issue of whether the effect of transformational

leadership on citizenship behavior is more direct or indirect in nature has yet to be

determined.

Although transformational leadership has shown impressive validities regarding a

number of positive performance-related outcomes, researchers have also criticized certain

aspects of the theory. In his evaluation of some of the conceptual weaknesses of

transformational leadership theory, Yukl (1999) pointed out that one major flaw has been

the lack of theoretical rational for labeling certain behaviors as transformational. For

example, the MLQ’s individualized consideration scale includes both supporting and

developing behaviors as key constructs. Although there is significant evidence to support

such developmental constructs as coaching and mentoring as predictors of subordinate

performance and self-efficacy, the effect of supporting behaviors on subordinate

motivation and performance has generally been weak (Bass, 1990, Yukl, 1998). Thus, the

rational for the inclusion of supporting behaviors as a core transformational construct is

somewhat unclear. Along similar lines, the high inter-correlation found between

transformational behavior dimensions raises additional concerns about construct validity.

Are these dimensions really distinct, or, in contrast to the theory, does evidence suggest

that they are all measuring the same behaviors?

In addition to doubts about construct validity, Yukl (1999) has also raised a few

other concerns in relation to the theory. One complaint was that there is an over-emphasis

22
on the dyadic process between leader and subordinate. In other words, the emphasis of

transformational leadership theory is too narrow, and should be broadened to include

both group and organizational influence processes. In addition, the theory includes what

Yukl (1999) labels a “heroic leadership bias.” That is, the theory devotes significant

attention to how the actions of the leader impact those of the followers. These theories are

explained without mention of how the influence process may be reciprocal in nature, with

subordinate actions conjointly influencing leader behavior. Finally, Yukl (1999) also

makes specific reference to the theory’s significant ambiguity in its description of the

influence process. Based on the current research, it is still unclear how transformational

leadership behaviors influence subordinate outcomes. According to Yukl (1999), what is

needed is systematic study of how “certain mediating variables relevant to task

performance, such as arousal of motives,” are related to transformational leadership

behaviors and subordinate performance (p. 287). In his opinion, “the theory would be

stronger if the essential influence processes were identified more clearly,” a criticism that

is addressed by the current study (p. 287).

Leader-member Exchange. In addition to transformational leadership, the leader-

member exchange model of leadership (LMX) has also received increasing amounts of

attention by researchers in recent decades. Born from the “Vertical Dyad Linkage”

(VDL) model of leadership, LMX is unique from other leadership theories in that its’

focus is on “the dyadic relationship” between the leader and the subordinate (Gernster &

Day, 1997). In contrast to more traditional theories, which are concerned with identifying

effective leader traits and behaviors, LMX focuses on how the quality of the relationship

23
between a leader and a subordinate can have positive effects at the individual, group, and

organizational level (Gernster & Day, 1997).

Although certain aspects of LMX theory have been altered since its’ original

conception, the general focus of the theory has remained the same throughout the

decades. In a summary of the evolution of LMX, Graen and UhlBien (1995) described

the history of the theory as occurring in four distinct stages. In the first stage, the major

discovery was that leaders develop different relationships with each of their subordinates.

This finding was somewhat revolutionary, as it was predominantly assumed by most

leadership scholars of the time that leaders engaged in similar leadership behaviors across

all of their subordinates (an approach known as the “Average Leadership Style”)

(Schrisheim, Castro, & Cogliser, 1999). Building on the notion of individualized

relationships, the second stage of LMX development focused primarily on the specific

relationship constructs involved in the leader-subordinate dyad. In addition, the

relationships between these constructs and those that were similar/dissimilar to LMX

were also explored. In the third stage, these efforts were expanded to address the specific

leader behaviors that were used to develop individualized “partnerships” with

subordinates (Graen &Uhl-Bien, 1995). Finally, in the fourth stage of evolution, sole

attention on the leader-subordinate dyad was widened to include investigation of how

networks of dyads are organized both inside and beyond organizational boundaries.

According to LMX theory, dyadic relationships are developed through a series of

“exchanges” that occur between the leader and the subordinate over time. For example,

the leader may offer increased job responsibility and flexibility to the subordinate, while

the subordinate may respond by showing increased effort, commitment, or performance

24
(Diensesch & Liden, 1986; Liden & Graen, 1980; Scandura & Graen, 1984). Both parties

invest each of their own resources into the relationship, which serves to shape the overall

quality of the relationship over time (Bauer & Green, 1996).

In a formal effort to describe the evolution of leader-subordinate exchange

relationships, Graen and Scandura (1987) provided a three-phase model of LMX

development. In the first phase, Role-taking, a key component is perspective taking. That

is, both the leader and subordinate learn to view work-related issues from the perspective

of both parties. In the next phase, Role-making, the focus is shifted to the development of

trust between leader and subordinate. Special emphasis is also given to how leader and

subordinate actions influence their own attitudes and behaviors. Finally, in the last stage

of Role-routinization, efforts are made to incorporate the behaviors learned in the first

two phases (e.g., perspective taking, trust-building) into the “routine” of the relationship

between leader and subordinate. In this last stage, the goal is that these behaviors should

become automatic during exchanges between the two parties, leading to an overall high-

quality leader-subordinate relationship.

Because the resources available to both leaders and subordinates are limited, it is

inevitable that a leader’s relationships with his or her subordinates will range on a

continuum from low to high quality (Bauer & Green, 1996; Liden et al., 1993). Those

subordinates who engage in higher quality exchanges with their supervisor are termed the

“in-group,” and usually receive special benefits and opportunities from the leader such as

specialized attention, favorable assignments, and career planning support (Deluga, 1998).

In contrast, those subordinates classified in the “out-group” tend to have lower quality

relationships with their supervisors, typically characterized by less attention and

25
“restricted levels of reciprocal influence and support” (Deluga, 1998, p. 190). Although

LMX theory emphasizes the existence of differential relationships between leaders and

subordinates, there is still some disagreement as to what elements actually constitutes a

“high” and “low” quality relationship. In one of the earliest attempts at describing the

theory of LMX, Graen (1976) proposed that LMX was an exchange relationship

consisting of three dimensions: competence, interpersonal skill, and trust. In contrast,

Cashman, Dansereau, Graen, and Haga (1976) argued that LMX was based solely on two

constructs: attention and sensitivity. Mirroring the initial disagreement characterizing the

early development of LMX, later efforts to describe the dimensionality of the construct

was plagued by a similar lack of consensus among researchers. Dienesch and Liden

(1986), for example, proposed that LMX was comprised of three dimensions: perceived

contribution, loyalty, and affect. In contrast, Graen and Uhl-Bien argued that a

combination of respect, trust, and mutual obligation comprise the LMX construct.

Although the dimensionality of LMX is still somewhat in question, six content

subdomains have surfaced as the most prominent across studies (Schriesheim et al.,

1999). These include: mutual support, trust, liking, latitude, attention, and loyalty. In

general, high levels of support, trust, liking, latitude, attention, and loyalty characterize

high-quality LMX relationships, whereas low quality exchanges are typified by lower

levels of each of these subdimensions.

A driving force behind the major interest in LMX theory has been the numerous

significant relationships found between LMX and both performance-related and

attitudinal outcomes. For example, results of several independent studies have shown that

higher quality exchanges between leaders and subordinates are predictive of higher

26
performance ratings (Liden, Wayne, & Stilwell, 1993), increased objective performance

(Graen, Novak, & Sommerkamp, 1982), higher organizational commitment (Nystrom,

1990), and higher overall job satisfaction (Graen, et al., 1982). In addition, meta-analytic

studies have found similar results. In a recent meta-analysis by Gernester and Day (1997)

summarizing the results of 79 independent studies, the quality of the relationship between

the leader and subordinate was found to be significantly correlated with objective

performance (.11), supervisor ratings of performance (.30), satisfaction with supervision

(.71), overall job satisfaction (.50), organizational commitment (.42), role conflict (- .31),

role clarity (.43), and member competence (.28). In addition, although LMX was not

found to significantly correlate with turnover (- .04), a significant relationship was found

with turnover intention (-.31).

In addition to the large number of positive task-related performance outcomes

(Dansereau, Graen, & Haga, 1975; Dienesch & Liden, 1986; Graen & Scandura, 1987),

high quality leader-member exchanges have also been associated with increased non-task

related activities such as OCB (e.g., Deluga, 1994; Settoon, Bennett, & Liden, 1996;

Wayne & Green, 1993). Specifically, a number of studies have found that the quality of

the relationship between leader and subordinate is predictive of subordinate OCB, at both

the aggregate and subdimensional level. For example, in a study by Wayne and Green

(1993) involving 73 nurses and their supervisors, results showed that the nurses who had

higher quality relationships with their supervisors engaged in significantly more altruistic

OCBs (e.g., assisting a supervisor or co-worker with their work, helping others who have

been absent) than those with lower quality relationships with their supervisors. In

contrast, no significant relationship was found between LMX-quality and Generalized

27
Compliance behaviors (e.g., not taking undeserved breaks or time-off, being punctual). In

another study by Tansky (1993), employee perceptions of the quality of the

supervisory/subordinate relationship were significantly correlated to all five OCB

dimensions. In addition, after controlling for a number of demographic variables (e.g.,

years of education, sex, supervisory position, and age), the quality of the

supervisory/subordinate relationship accounted for an additional 13% of the variance in

the Altruism dimension, 14% in Conscientiousness, 9% in Sportsmanship, 9% in

Courtesy, and 13% in Civic Virtue. Finally, the results of Organ and Ryan’s meta-

analysis (1995) support the suggestion that the quality of leader-subordinate relationships

influence OCB. In their review of OCB studies dealing with leader supportiveness, they

found an average correlation of .32 between leader supportiveness and the Altruism

dimension, as well as an average correlation of .35 with the Generalized Compliance

dimension.

Although these results are encouraging, research has most consistently supported

the relationship between perceptions of LMX-quality and both Altruism and overall

OCB. In a recent meta-analysis summarizing the empirical correlates of OCB, Podsakoff

et al. (2001) found a .36 corrected correlation between perceptions of LMX-quality and

Altruism, and a .30 corrected correlation with overall OCB. Taken together, these results

suggest that one by-product of high quality exchanges between leader and subordinate is

altruistic behaviors aimed toward both the supervisor and other co-workers. These

findings precipitate the question, however, of why these behaviors occur?

The primary explanation for the relationship found between LMX-quality and

OCB lies in the framework of social exchange and reciprocity. As suggested by Blau

28
(1964), social exchanges are based on a foundation of trust, with the expectation that acts

of goodwill from one party will be reciprocated by the other. When certain gestures made

on behalf of the supervisor are perceived positively by the recipient (i.e., the

subordinate), these actions evoke feelings of subordinate obligation. In response, the

subordinate engages in increased functional behavior (e.g., task performance, OCB) as a

means of fulfilling the perceived obligation. Research has shown that individuals seek to

reciprocate in ways that will be clearly recognized by the other party in the relationship

(Blau, 1964; Gouldner, 1960). In addition, it has been found that employees view both

task and contextual performance as appropriate forms of reciprocation within the context

of a work environment (e.g., Katz & Kahn, 1966; Levinson, 1965; Foa & Foa, 1980;

Moorman, 1991). Based on this framework, it seems appropriate to assume that high

quality relationships between leader and subordinate are characterized by gestures of

goodwill that are perceived positively by the subordinate (e.g., favorable job assignments,

increased responsibility). In return for these benefits, the subordinate is motivated to

respond in kind, and does so by engaging in altruistic behaviors aimed at benefiting the

supervisor and/or other employees of the organization.

Although research has consistently been supportive of the relationships between

LMX and both performance and attitudinal outcomes, researchers have also raised

significant theoretical and methodological concerns in relation to LMX. One of the

primary criticisms is that the evolution of the theory has included multiple iterations of

LMX definitions that have been “confusing and sometimes appear to be contradictory”

(Schrisheim, et al., 1999). More importantly is the fact that there have been no

explanations for why the theory has evolved over the years, or why particular changes in

29
the nature of the construct have been adopted. Although it is important for a theory to

evolve, it is critical to have adequate theoretical justification accompanying any changes

that are made. Unfortunately, this latter point was not strictly observed during the

development of LMX theory.

A related theoretical issue is the level of analysis adopted by LMX theory.

Although the VDL model (the premise of LMX) and early LMX frameworks focused on

both leader and subordinate behaviors within the relationship dyad, later research has

shown a departure from this level of analysis. As reviewed by Schrisheim et al. (1999),

Graen and Uhl-Bien (1995) argue that the relationship itself should be the primary focus,

as opposed to either the leader or the subordinate. Taking an opposing perspective,

Dansereau et al. (1995) suggest that this approach introduces ambiguity into LMX theory

by deeming any level of analysis appropriate, as long as the “relationship” between the

two parties remains the focus. In general, it is important for a theory to specify upfront

the level of analysis at which a phenomenon is expected to exist so that the theory,

measurement, and data-analytic techniques may be aligned and accurate results may be

acquired (Klein, Dansereau, & Hall, 1994). Thus, based on these two criticisms, it seems

that more theoretical work is needed to further clarify both the foundation and focus of

LMX theory.

One final criticism of LMX theory deals with the variety of scales that been

developed to measure the construct. Overall, many different measures have been used to

operationalize LMX, with various scales ranging from 2 to 25 items (Schrisheim, et al.,

1999). In addition, as noted by Schrisheim et al. (1999), the rationale for choosing these

measures was frequently not provided, and some “were modified from existing measures

30
without adequate psychometric testing” (p. 94). Because different studies used different

scales, it has also been unclear whether mixed results are due to the construct itself or the

method by which it was measured (Gernster & Day, 1997). Although psychometric

evidence is supportive of the seven-item LMX7 (the predominantly used LMX measure),

future LMX research should focus on increased scale validation as well as efforts to

increase the content validity of existing measures (Schrisheim et al., 1999).

OCB Motives

In addition to leadership variables, it has also been suggested that in order to

understand the causes of OCB, one must identify the motives that underlie these actions

(Rioux & Penner, 2001). Although other researchers have previously investigated the role

of motives in relation to OCB (e.g., Bolino, 1999), Penner and his colleagues were the

first to address the area from a functional perspective. The basic idea behind the

functional approach is that people engage in certain behaviors (e.g., OCB) because these

actions serve some need or purpose for them (Borman & Penner, 2001). However,

different people may engage in the same behavior for different reasons. For example, one

employee may stay late after work to help a co-worker because he/she generally enjoys

helping other people. Another employee may engage in the same type of behavior, not

because he/she enjoys helping others, but because of concern for the welfare of the

organization. In each case, the person’s behavior is the same. However, the motives

behind these behaviors are different depending upon the needs of the individual.

A large portion of the support for taking a functional approach to OCB comes

from research on a related phenomenon, volunteerism. Volunteering is defined by

Hanson (1991) as a “form of formal planned helping” that involves aiding others usually

31
through organizations such as churches, schools, hospitals, and service organizations

(also see Allen, 1982). This type of behavior is similar to OCB in that both kinds of

behavior: (1) are considered to be long-term phenomenon; (2) are preceded by thought

and planning; (3) occur in an organizational context; and (4) involve a choice to help

made on behalf of the individual. In general, researchers applying a functional approach

to volunteerism (Clary & Snyder, 1991; Omoto and Snyder, 1995; Omoto, Snyder, &

Berghuis, 1993; Penner and Finkelstein, 1998; Clary & Orenstein, 1991) have found

strong support for the view that the reasons for volunteering can vary based on the needs

of the individual. As a result, this research has served as a springboard for work on OCB

motives.

Recently, Rioux and Penner (2001) applied a functional explanation to OCB.

Specifically, they identified three primary OCB dimensions or motives. The first of these

was called Prosocial Values, that describes OCB that is motivated by a desire to help

others and be accepted by them. As suggested by Rioux (1998) this motive is especially

important to those who have a need to be liked by their co-workers, and who place a

heavy emphasis on maintaining relationships. The second motive is Organizational

Concern, or engaging in OCB out of a need to show commitment to the organization.

This motive also allows an employee to increase and expand his/her knowledge of the

organization as well as gain increased work experience (Tillman, 1998). Finally, the third

motive is called Impression Management, and describes OCB that is motivated by a

desire to avoid negative evaluation by others or to gain material rewards. As suggested by

Rioux (1998), certain individuals are greatly concerned with how they are perceived by

32
others. Thus, engaging in OCB is a way to avoid being perceived as lazy or irresponsible,

which can even lead to certain monetary rewards such as raises or promotions.

In their study of OCB, Rioux and Penner administered a scale that measured each

of these three motives to a group of municipal employees. In addition, they obtained self-

, peer-, and supervisor-ratings of two dimensions of OCB: Altruism and Generalized

Compliance. Results showed that motives correlated significantly across all three types of

ratings. More specifically, the Prosocial Values Motive was shown to correlate most

strongly with the Altruism dimension, while the Organizational Concern Motive was

shown to correlate most strongly with the Generalized Compliance dimension. Results

also showed significant correlations between the Organizational Concern Motive and

procedural justice (.44), mood (.49), and Other-Oriented Empathy (.27). Similarly, the

Prosocial Values Motive was found to significantly correlate with procedural justice

(.24), mood (.21), Other-Oriented Empathy (.46), and Helpfulness (. 31). Independent

evidence of these relationships was provided by Forde (2000), using a sample of working

college students. Taken together, these results show that motives are, in fact, related to

certain aspects of OCB and its antecedents. Thus, it is possible that motives may play an

important role in the prediction of this type of prosocial behavior.

Building on the results of Rioux and Penner, recent research has investigated the

role of motives in relation to some of the antecedents of OCB. For example, Tillman

(1998) found that both the Prosocial Values Motive and Organizational Concern Motive

moderated the relationship between perceptions of procedural justice and OCB. More

specifically, the relationship between procedural justice and OCB was the strongest for

those individuals high on these two motives, and weakest for those scoring low. In

33
addition, the Organizational Concern Motive was shown to moderate the relationship

between conscientiousness and OCB. That is, the relationship between conscientiousness

and OCB was the strongest for those individuals high on this motive, and weakest for

those scoring low.

Expanding on these efforts, Connell & Penner (2004) investigated whether

motives could perhaps mediate the relationship between certain antecedents and

dimensions of OCB. Across both self- and peer-reports of OCB, results provided strong

evidence for the Organizational Concern Motive as the primary mediator between the

Generalized Compliance dimension of OCB and three antecedents: affective

commitment, procedural justice, and conscientiousness. In addition, the relationship

between Other-oriented Empathy and the Altruism dimension was partially mediated by

both the Organizational Concern and Prosocial Values motives. These results were

important from both a theoretical and practical perspective because they revealed that the

influence of certain antecedent variables on select dimensions of OCB were, at least in

part, accounted for by motives. Most notable, however, was the finding that different

motives mediated different antecedent-OCB relationships.

Moderators and Mediators

Because moderator and mediator variables are sometimes confused, a brief

discussion of the differences between them may be in order. As described by Baron and

Kenny (1986), moderators can be described as qualitative or quantitative variables that

affect the direction and/or strength of the relationship between a predictor (independent)

variable and a criterion (dependent) variable. That is, the relationship between an

independent and dependent variable differs based on the level of the moderator variable.

34
In contrast, a variable functions as a mediator when its presence accounts for the

relationship between the predictor and the criterion. More simply, the mediator serves as

“the general mechanism” through which the predictor influences the criterion (Baron, &

Kenny, 1986, p. 1173). Thus, the primary distinction between these two variables is that

moderators specify when certain effects will be observed between variables, while

mediators indicate how or why such effects are observed.

The Current Study

The purpose of the current study is to further investigate the role of motives in

relation to both transformational leadership and LMX behaviors and OCB. Various

researchers (e.g., Podsakoff et al., 2001; Yukl, 1999) have suggested that future studies

should address possible mediators in the leader influence process. In addition, two other

streams of research provide the foundation for the current study. The first steam includes

those studies that have found a positive relationship between both transformational

leadership and LMX behaviors and various dimensions of OCB (e.g., Altruism,

Conscientiousness). In general, this research suggests that increased levels of either

transformational or LMX behaviors is associated with increased citizenship performance

among subordinates. In addition, evidence implies that the effects of certain leadership

behaviors (e.g., transformational leadership) may indirectly affect OCB via other

constructs (e.g., trust) (Podsakoff et al., 1990), and that certain variables may also

moderate the relationship between leadership behavior and OCB. The second stream

suggests that motives play an important role in relation to OCB, specifically serving as

both moderators and mediators of antecedent-OCB relationships. Based on these results,

35
we contend that it may be worthwhile to investigate whether additional variables (e.g.,

motives) moderate or mediate the relationship between leadership behaviors and OCB.

The specific goal on the current study is to expand on previous studies (e.g.,

Tillman, 1998; Connell & Penner, 2004) that have found evidence for motives as both

moderators and mediators between select antecedent variables and both the Altruism and

Generalized Compliance dimensions of OCB. Similar to the previous studies, positive

relationships are predicted to exist between additional antecedent variables (e.g.,

transformational and LMX behaviors) and OCB motives. In addition, the Organizational

Concern motive is expected to mediate the relationship between transformational

leadership and OCB, while the Prosocial Values Motive is predicted to either mediate or

moderate the relationship between LMX-quality and OCB.

Hypothesis 1. There will be a positive relationship observed between supervisor

transformational leadership behavior and subordinate Organizational Concern.

One of the basic tenets of transformational leadership is to encourage followers to

transcend their own self-interests and move beyond simple leader-member transactions

for the good of the organization (Bass, 1985). This aspect of transformational leadership

seems synonymous with the promotion of organizational concern among employees. Due

to the conceptual overlap observed between these two constructs, it is expected that

increases in transformational leadership behavior will be associated with increases in the

Organizational Concern Motive among subordinates.

Hypothesis 2. Subordinate Organizational Concern will mediate the relationship

between supervisor transformational leadership behavior and subordinate

Conscientiousness (i.e., Generalized Compliance).

36
 Previous research has shown that transformational leadership is positively related

to the Conscientiousness dimension of OCB (e.g. Podsakoff et al., 2001). Other research

has also revealed a link between the Organizational Concern Motive and the

Conscientiousness dimension (Rioux & Penner, 2001). On the basis of these

relationships, we propose that leaders who engage in transformational leadership

behaviors will have subordinates who participate in behaviors that are indirectly helpful

to other members of the organization. Assuming support for Hypothesis 1, this

relationship will be mediated by the subordinate’s concern for the welfare of the

organization. More specifically, the performance of transformational leadership behaviors

will be associated with subordinate Organizational Concern, which, in turn, will be linked

to OCB that is beneficial to the organization as whole.

Hypothesis 3. Subordinate Organizational Concern will mediate the relationship

between supervisor transformational leadership behavior and subordinate Civic Virtue.

In addition to the Generalized Compliance dimension, transformational leadership

has correlated positively with the Civic Virtue dimension in past studies. Other research

has also revealed a positive relationship between the Organizational Concern Motive and

the Civic Virtue dimension (Rioux & Penner, 2001). We propose that leaders who engage

in transformational leadership behaviors will have subordinates who show active

involvement and interest regarding organizational issues, as well as the governance of the

organization as a whole (Organ & Ryan, 1995). Assuming support for Hypothesis 1, we

also predict that this relationship will be mediated by the employees’ concern for the

welfare of the organization. That is, the performance of transformational leadership

37
behaviors will be associated with a general concern for the organization among

subordinates, which, in turn, will be linked to subordinate Civic Virtue.

Hypothesis 4. Subordinate Organizational Concern will mediate the relationship

between supervisor transformational leadership behaviors and subordinate Courtesy.

Evidence has shown that transformational leadership behaviors correlate

positively with the Courtesy dimension of OCB. Other research has also revealed a

positive relationship between a generalized concern for the organization and the Courtesy

dimension (Rioux & Penner, 2001). It is predicted that leaders who engage in

transformational leadership behaviors will have subordinates who actively help prevent

problems among coworkers by engaging in both formal and informal cooperation with

other employees (Organ & Ryan, 1995; Organ, 1997; George & Brief, 1992; Konovsky

& Organ, 1996). Assuming support for Hypothesis 1, this relationship is expected to be

mediated by employees’ concern for the welfare of the organization. That is, the

performance of transformational leadership behaviors will be associated with subordinate

Organizational Concern, which, in turn, will be linked with cooperation among

employees.

Hypothesis 5. Subordinate Organizational Concern will mediate the relationship

between supervisor transformational leadership behaviors and subordinate

Sportsmanship.

Previous research has shown that transformational leadership behaviors correlate

positively with subordinate Sportsmanship. In addition, other research has revealed a

positive relationship between the Organizational Concern Motive and the Sportsmanship

dimension (Rioux & Penner, 2001). We propose that leaders who engage in

38
transformational leadership behaviors will have subordinates who tend to absorb minor

inconveniences about their jobs without complaint, and who also tend to consider other

employees’ work problems in addition to their own (Konovsky & Organ, 1996).

Assuming support for Hypothesis 1, this relationship is predicted to be mediated by the

subordinate’s concern for the welfare of the organization. That is, the performance of

transformational leadership behaviors will be associated with subordinate Organizational

Concern, which, in turn, will be linked to Sportsmanship behaviors among subordinates.

Hypothesis 6. There will be a positive relationship between LMX-quality and

subordinate Prosocial Values.

As explained previously, LMX theory is based on the existence of beneficial

exchange relationships between leaders and their followers. In theory, high levels of

support, trust, liking, and loyalty characterize a high quality relationship, whereas lower

levels of each of these variables typify a low-quality relationship. Based on these

characteristics, we propose that LMX-quality will be positively associated with

subordinate Prosocial Values. That is, a high-quality relationship between a leader and a

subordinate will be positively related with the subordinate’s desire to help others within

the organization and to be accepted by them. In contrast, the desire to help others within

the organization will not be salient among subordinates who hold lower quality

relationships with their supervisor.

Conceptually, it is possible to conceive of the Prosocial Values Motive as both a

moderator and mediator of the relationship between LMX-quality and the Altruism

dimension of OCB. As such, the remaining two hypotheses address the role of Prosocial

Values from both a moderating and mediating perspective.

39
 Hypothesis 7. Subordinate Prosocial Values will mediate the relationship

between LMX-quality and subordinate Altruism.

It is predicted that a possible mechanism through which LMX impacts Altrusim is

the Prosocial Values Motive. More specifically, it is proposed that high LMX-quality

arouses among subordinates the motivation to help others and be accepted by them

(Hypothesis 6), which, in turn, is associated with OCB directed toward individual

members of the organization. This prediction is supported by research that has shown a

positive link between the Altruism dimension and both LMX-quality (e.g., Podsakoff et

al., 2001) and the Prosocial Values Motive (Rioux & Penner , 2001).

Hypothesis 8. Subordinate Prosocial Values will moderate the relationship

between LMX-quality and subordinate Altruism

In addition to Hypothesis 7, it is predicted that the relationship between LMX-

quality and Altruism is contingent upon the extent that the Prosocial Values Motive is

possessed by the subordinate. In other words, a relatively strong positive relationship

between LMX-quality and subordinate Altruism is proposed to exist when the level of

subordinate Prosocial Values is high. In contrast, when the level of subordinate Prosocial

Values is low, a much smaller relationship is predicted to exist between these two

variables. This research is again contingent upon Hypothesis 6, but is also based on the

positive relationships observed between the Altruism dimension and both LMX-quality

(e.g., Podsakoff et al., 2001) and the Prosocial Values Motive (Rioux & Penner , 2001).

40
 Chapter Two

Method

Participants

This study included responses from 201 employees working either part-time or

full-time at one of 13 organizations located in the Southeastern United States. In total, 62

employees who completed the survey responded as a supervisor while 139 responded as a

subordinate. The surveys were distributed to a total of 306 employees (118 supervisors,

188 subordinates) yielding an overall response rate of 66 percent. In addition, the

individual response rates were 53 percent for the supervisors and 74 percent for the

subordinates.

The initial sample of 201 employees was screened using a number of criteria.

First, because the objective of this study was to include the perspective of both

supervisors and their subordinates, both parties were asked to submit responses to the

survey. This approach allowed for the eventual collection of supervisor-subordinate pairs.

After submitting their responses, if a given supervisor could not be paired with a

subordinate response, the supervisor was eliminated from the sample. This was also the

case for any subordinates who could not be paired with a supervisor.

Employees were also eliminated if they failed to answer more than ten percent of

the items included in a scale, or if their responses to the survey appeared questionable. In

order to identify “questionable” responders, the standard deviation for each of the

41
measures (with the exception of the LMX7) was calculated for each of the participants

(supervisors and subordinates). If a participant provided the same response for every

question in a particular measure (SD = 0), that participant’s responses were further

examined to determine if they should be eliminated from the sample. Based on this

criterion, a participant was only eliminated if their responses appeared reasonably

suspect.

Finally, supervisors were instructed to rate only those subordinates whom they

had supervised for at least four months. Thus, if a supervisor indicated that they had

supervised a particular subordinate for less than 4 months, that supervisor-subordinate

pair was eliminated from the sample.

Using the above criteria, 7 supervisors and 9 subordinates were eliminated from

the initial sample. Thus, the final sample was comprised of 55 supervisors and 131

subordinates (i.e., 131 supervisor-subordinate pairs). The demographic characteristics of

the final sample are displayed in Table 2.

Overall, the final sample contained slightly less males than females (43 % versus

57 %, respectively). In addition, most of the employees were of White ethnicity (82 %).

Sixty-five percent reported that they had been employed with their organization for at

least 3 years, while only 11 percent reported that they had been with the organization for

6 months or less. It should be noted that no employees were included in the final sample

who reported less than 4 months experience with their current place of employment.

Finally, this sample consisted mostly of full-time employees (88 %), and over half (55 %)

described their position as managerial/professional.

42
Table 2. Participant Demographics (N = 186)

Supervisors Subordinates Total

(%) (%) (%)
N = 55 N = 131 N = 186

Gender

Male 53 39 43

Female 47 61 57

Not specified 0 0 0

Race/National Origin

American Indian/ 0 0 0
Alaska Native

Asian 0 5 3

Hispanic/Latino 2 3 3

Black/African American 9 7 8

Native Hawaiian/Pacific Islander 0 1 1

White 86 82 82

Other Ethnicity 2 3 3

Not Specified 2 1 1

Months employed with the organization

Less than 6 months 6 13 11

Between 6 and 11 months 6 6 6

Between 12 and 35 months 6 24 18

36 months or longer 84 57 65

Not Specified 0 1 1

43
Table 2. Continued

Supervisors Subordinates Total

(%) (%) (%)
N = 55 N = 131 N = 186
Position Description

Managerial/Professional 87 41 55

Administrative 6 11 9

Clerical 2 3 3

Technical 6 19 15

Other 0 25 18

Not Specified 0 1 1

Employment Status

Part-time 4 20 15

Full-time 96 80 88

Not specified 0 0 0

Overall, responses were collected from a total of 13 organizations. In general,

each of these organizations could be classified as a small businesses (less than 500

employees). The breakdown of responses across industries was as follows: Public

Administration (17 %), Professional, Scientific, and Technical Services (50 %),

Educational Services (17 %), Utilities (2 %), Health Care/Social Assistance (3 %), Retail

Trade (2 %), and Food Service (8 %). As seen by these statistics, this sample is slightly

biased in favor of Professional, Scientific, and Technical Service organizations. Given the

variety of different industries that participated, however, it can be argued that this sample

is still reasonably representative of the current work force.

44
Measures

A number of accepted measures were used to evaluate the constructs relevant to

this study. Each of these measures is briefly described below. In addition, the means,

standard deviations, and reliabilities observed for each scale are also displayed in Tables

4 -7.

Transformational Leadership. Podsakoff et al.’s (1990) Transformational

Leadership Inventory (TLI) was used to measure transformational leadership behaviors in

this study. This scale consisted of 22 items, and measured six dimensions of

transformational leadership: articulating a vision, providing an appropriate model,

fostering the acceptance of group goals, high performance expectations, providing

individualized support, and intellectual stimulation. Although previous research supports

the hypothesized six-factor structure (e.g., Podsakoff et al., 1996), three of the

dimensions have been found to be highly correlated (articulating a vision, providing an

appropriate model, and fostering the acceptance of group goals). As such, these three

factors are sometimes combined to represent a “core” transformational leadership

construct. Internal consistency reliabilities for each of the dimensions range from .82 to

.87. In addition, the TLI has shown impressive validities with related constructs across

several studies (e.g., Podsakoff et al., 1990; Podsakoff et al., 1996; Podsakoff et al.,

2001). For the purposes of this study, each of the TLI dimensions were combined to

create an overall index of transformational leadership. Specifically, this index was

created by summing the individual dimension scores for each participant.

Leader-Member Exchange (LMX). LMX-quality was assessed using a modified

version of the LMX7 (Graen, Novak, & Sommerkamp, 1982). This measure is by far the

45
most frequently used LMX measure, and is recommended by Graen & Uhl-Bien (1995)

as the standard measure of LMX. The LMX7 consisted of seven items, and asked the

respondent to indicate their answer to each item using a five-point Likert scale.

Psychometric evidence for the measure provided by Gerstner & Day (1997) indicates

internal consistency reliabilities are in the range of .79 to .89. In addition, acceptable

validities for the measure have also been observed across multiple studies (e.g., Gernster

& Day, 1997; Podsakoff et al., 2001).

OCB. A modified version of the scales developed by Podsakoff and MacKenzie

(1989) was used to measure OCB. This measure consisted of 24 items, and measured all

five OCB dimensions identified by Organ (1988): Altruism, Conscientiousness,

Sportsmanship, Courtesy, and Civic Virtue. Respondents were asked to indicate the

degree to which they agreed with each item using a seven-point Likert scale. This scale

ranged from (1) “strongly disagree” to (7) “strongly agree.” Numerous studies using the

scale have shown that the measure possesses good validity as well as acceptable internal

consistency reliability (e.g., MacKenzie, Podsakoff, & Fetter, 1991; Moorman, 1991;

Moorman, Niehoff, & Organ, 1993). Confirmatory factor analysis results also provide

evidence for the five-factor structure of the measure (Podsakoff et al., 1990).

OCB Motives. Participant motives for engaging in OCB was measured using

Rioux and Penner’s (2001) Citizenship Motives Scale (CMS). This scale consisted of 30

items, and measured all three of the motives for engaging in OCB: (1) Prosocial Values,

(2) Organizational Concern, and (3) Impression Management. Respondents were asked to

rate on a scale ranging from “not at all important” to “extremely important” how

influential each item was in their decision to engage in OCB. Psychometric evidence

46
provided by Rioux and Penner (2001) reports internal consistency reliabilities above .80,

and test-retest reliabilities for each of the 2 factors above .70. In addition, the same three-

factor structure for the measure has been replicated across 3 diverse samples.

Procedure

The majority of survey responses were collected and maintained via an online

survey system. Approximately one week prior to the study, each participant received a

brief introduction email that provided: (1) a short description of the study; (2) the time

required to complete the survey (i.e., approximately 20 minutes); (3) a statement of

assurance that each of their responses would be held confidential; and (4) contact

information for the primary researcher. On the scheduled date of administration, each

supervisor was sent an additional email containing the link to the online survey.

After accessing the survey, each supervisor was asked to enter a unique six-digit

code of their own choosing. Next, they were instructed to enter the email addresseses of

up to four of their subordinates using the criteria presented in Table 3. After this

information had been entered, the supervisors were directed to the rest of the survey

which included items taken from the TLI, LMX7, and Podsakoff’s OCB measure. Thus,

each supervisor provided: (1) ratings of their own transformational leadership behaviors;

(2) an estimate of the quality of their relationship with each of the subordinates they

listed; and (3) ratings of each subordinate’s OCB behaviors. Each supervisor was

required to rate their subordinates using the same order that was used on the first page of

the survey.

The online survey system was designed such that after the supervisor entered the

email addresses of his/her subordinates, an email containing a link to the subordinate

47
Table 3. Supervisor Survey Instructions

1. Please select subordinates that you have supervised at least 4 months.

2. Please select subordinates that feel you can provide accurate information about. For

example, if you have supervised a particular subordinate for more than 4 months, but

feel that you are not familiar enough with their behavior to provide accurate feedback,

please do not include them in your final selection.

3. Please try to select employees that, as a group, represent a range of performance (e.g.,

excellent, fair, and poor). In other words, try not to select all high performers or all

low performers.

4. Finally, please select only those subordinates that work at least 20 hours per week.

version of the survey was immediately sent to each of the subordinates that were listed.

The subordinate version included the TLI, the LMX7, Podsakoff’s OCB measure, and the

CMS. Thus, each subordinate provided: (1) ratings of their supervisor’s transformational

leadership behaviors; (2) an estimate of the quality of their relationship with their

supervisor; (3) ratings of their own OCB behaviors; and (4) their own responses to the

CMS. The six-digit code created by each subordinate’s respective supervisor also

appeared with the subordinate’s final set of responses. In addition, a random code of 3

numbers ending with 1, 2, 3, or 4 automatically appeared immediately after this six-digit

code. Taken together, these numbers served as the mechanism by which the subordinates’

responses were matched-up with their supervisors.

48
 After each survey was completed, responses were sent to a secure database that

was only accessible to the principal investigator. A reminder email was also distributed to

all participants encouraging them to complete the survey if they had not already done so

by the specified date.

Because not all employees who were willing to participate in the study had access

to the Internet at their place of employment, paper and pencil versions of the survey were

also distributed. As a result, 27 percent of the sample completed the survey using this

method. The procedure for these participants was basically the same as the one used for

the online survey participants. In most cases, however, the supervisor was responsible for

distributing surveys to each of his/her subordinates. Each participant was also provided

with a self-addressed, stamped envelope, and was instructed to mail completed versions

of the survey back to the Psychology Department at the University of South Florida.

49
 Chapter Three

Results

Preliminary Steps and Analyses

Before formal testing of the hypotheses, further examination of the final sample

was conducted. Specifically, if a respondent failed to answer an item included in a scale,

the missing value was replaced with the median value of the scale. This procedure was

used when the proportion of missing items for the scale was not above ten percent. As a

whole, this procedure was applied to approximately one percent of the sample. In

addition, no participants included in the final sample failed to respond to more than ten

percent of the items for any given scale.

Further analyses were also conducted to determine if it was appropriate to pool

the responses from those participants who completed the online version of the survey

with those who took the paper and pencil version. Specifically, a Box’s M test was

performed to assess if variation between each of the variables were the same for the

different groups. Results of this test were significant (χ2 = 197.99, p < .01), suggesting

that caution should be taken when pooling the covariance matrices associated with each

group. Although this is a cause for concern, it should be noted that the number of

participants who completed the paper-and-pencil version of the survey (35) was

significantly smaller in comparison to those who completed the online version (96). In

addition, Stevens (2002) provides statistical evidence that Box’s M is extremely sensitive

to normality. Therefore, it is possible that a lack of normality may have caused this result,

50
as opposed to unequal covariance matrices in the population. Combined with the fact that

both groups underwent almost identical procedures, it was determined that the results of

the Box’s M test alone did not merit preventing the pooling of these two groups.

Variable Descriptives

The means, standard deviations, and reliabilities for each of the measures

included in this study are displayed in Tables 4 - 7. Internal consistency estimates ranged

from a low of .66 (self-reported Conscientiousness) to a high of .96 (subordinate rated

transformational leadership). In general, these estimates indicate that adequate reliability

was observed for each of the measures. In addition, ratings were similar across measures

for both subordinates and supervisors. That is, responses tended to occur toward the high

end of the scale for both groups. Finally, although some variation was observed, the

standard deviations tended to be somewhat small for the majority of the measures.

It was also observed that each supervisor provided feedback on an average on

2.24 subordinates. The only exception was one supervisor, who provided feedback on ten

subordinates. In addition, the median amount of time that a supervisor reported

supervising a subordinate was 24 months.

Zero-order Correlations

As shown in Tables 5 -7, a number of significant relationships were observed

between the variables included in this study. Beginning with transformational leadership,

subordinate reports of transformational leadership correlated significantly with all five

self-report dimensions of OCB. These correlations ranged from .27 (p < .01) for the

Courtesy dimension to .33 (p < .01) for the Altruism dimension. A similar trend was

observed using supervisor-reports of OCB, although these correlations were somewhat

51
Table 4. Descriptive Statistics for Study Variables

Subordinate Supervisor
Mean SD Min Max Mean SD Min Max

Transformational Leadership 123.05 21.85 48.00 154.00 126.91 13.61 88.00 151.00

LMX-quality 28.11 5.35 13.00 35.00 29.15 3.79 21.00 35.00

Prosocial Values 37.34 7.84 13.00 50.00 -- -- -- --

Organizational Concern 39.03 8.36 13.00 50.00 -- -- -- --

Impression Management 31.50 8.64 10.00 50.00 -- -- -- --

Altruism 29.68 3.50 19.00 35.00 28.69 5.20 9.00 35.00

Conscientiousness 29.65 3.90 19.00 35.00 29.15 5.22 9.00 35.00

Sportsmanship 28.64 5.15 8.00 35.00 27.15 6.70 8.00 35.00

Courtesy 30.10 3.20 20.00 35.00 28.66 4.96 13.00 35.00

Civic Virtue 21.98 3.94 7.00 28.00 21.20 4.15 11.00 28.00

smaller in magnitude. Specifically, subordinate reports of transformational leadership

correlated significantly with four of the five supervisor-reported dimensions:

Conscientiousness (.20, p <.05), Sportsmanship (.21, p < .05), Courtesy (.19, p < .05),

and Civic Virtue (.18, p <.05).

With regard to OCB motives, significant relationships were found between

subordinate reports of transformational leadership and both Organizational Concern (.44,

p < .01) and Prosocial Values (.24, p < .01). Although both of these relationships were

significant, transformational leadership showed a significantly stronger association with

the Organizational Concern Motive [t(130) = 3.16, p < .01)]. In contrast, no relationship

was observed between subordinate reported transformational leadership and the

Impression Management Motive (.05, p > .05).

52
 Table 5. Variable Reliabilities and Intercorrelations for Self Ratings of OCB (N = 131)

Variable 1 2 3 4 5 6 7 8 9 10 11 12

1. Transformational Leadership (subordinate) (.96)

2. Transformational Leadership (supervisor) .16 (.92)

3. LMX-quality (subordinate) .77** .05 (.90)

4. LMX-quality (supervisor) .19* .37** .26** (.79)

5. Prosocial Values Motive .24** .01 .28** -.08 (.92)

53
6. Organizational Concern Motive .44** .15 .39** .18* .70** (.94)

7. Impression Management Motive .05 .06 .08 -.20* .52** .39** (.89)

8. Altruism .33** -.06 .34** .04 .56** .68** .16 (.76)

9. Conscientiousness .31** .02 .24** .16 .32** .55** .10 .54** (.66)

10. Sportsmanship .29** -.01 .21* .04 .27** .36** -.07 .36** .44** (.81)

11. Courtesy .27** .00 .20* -.09 .54** .55** -.13 .63** .48** .47** (.73)

12. Civic Virtue .32** .05 .37* .10 .49** .61** .16 .57** .49** .35** .53** (.71)

* Statistically significant at the .05 alpha level. ** Statistically significant at the .01 alpha level.
 Table 6. Variable Reliabilities and Intercorrelations for Supervisor Ratings of OCB (N = 131)

Variable 1 2 3 4 5 6 7 8 9 10 11 12

1. Transformational Leadership (subordinate) (.96)

2. Transformational Leadership (supervisor) .16 (.92)

3. LMX-quality (subordinate) .77** .05 (.90)

4. LMX-quality (supervisor) .19* .37** .26** (.79)

5. Prosocial Values Motive .24** .01 .28** -.08 (.92)

6. Organizational Concern Motive .44** .15 .39** .18* .70** (.94)

54
7. Impression Management Motive .05 .06 .08 -.20* .52** .39** (.89)

8. Altruism .14 .27** .22* .50** .18* .34** -.08 (.85)

9. Conscientiousness .20* .32** .14 .44** .07 .24** -.05 .80** (.83)

10. Sportsmanship .21* -.01 .19* .19* .09 .17 -.06 .42** .53** (.88)

11. Courtesy .19* .10 .23** .32** .14 .25** -.06 .71** .73** .72** (.86)

12. Civic Virtue .18* .24** .23** .36** .29** .43** .01 .62** .53** .49** .55** (.75)

* Statistically significant at the .05 alpha level. ** Statistically significant at the .01 alpha level.
 Table 7. Variable Reliabilities and Intercorrelations for Self and Supervisor Ratings of OCB (N = 131)

Variable 1 2 3 4 5 6 7 8 9 10

1. Altruism (self) (.76)

2. Conscientiousness (self) .54** (.66)

3. Sportsmanship (self) .36** .44** (.81)

4. Courtesy (self) .63** .48** .47** (.73)

5. Civic Virtue (self) .57** .49** .35** .53** (.71)

55
6. Altruism (supervisor) .29** .33** .21** .19* .33** (.85)

7. Conscientiousness (supervisor) .14 .37** .19* .17 .25** .80** (.83)

8. Sportsmanship (supervisor) .08 .19* .29** .18* .16 .42** .53** (.88)

9. Courtesy (supervisor) .18* .25** .26** .18* .24** .71** .73** .73** (.86)

10. Civic Virtue (supervisor) .32** .28** .22* .32** .44** .62** .53** .49** .55** (.75)

* Statistically significant at the .05 alpha level. ** Statistically significant at the .01 alpha level.
 Surprisingly, results revealed that subordinate reports of transformational

leadership were not significantly related to supervisor reports (.16, p >.05). Supervisor

reports of transformational leadership also failed to correlate significantly with any of the

three OCB Motives. In contrast, significant relationships were observed between

supervisor reports of transformational leadership and three of the five supervisor-reported

OCB dimensions: Altruism (.27, p <.01), Conscientiousness (.32, p < .01), and Civic

Virtue (.24, p < .05).

In general, similar findings were observed with regard to perceptions of LMX-

quality. Specifically, subordinate perceptions of LMX-quality correlated significantly

with all five self-reported OCB dimensions. These correlations ranged from .20 (p < .05)

for the Courtesy dimension to .37 (p < .01) for the Civic Virtue dimension. In addition,

significant relationships were also observed between subordinate LMX perceptions and

four of the five supervisor-reported OCB dimensions: Altruism (.22, p < .05),

Sportsmanship (.19, p < .05), Courtesy (.23, p < .01), and Civic Virtue (.23, p < .01).

Similar to transformational leadership, subordinate perceptions of LMX-quality were also

significantly related to both the Organizational Concern (.39, p < .01) and Prosocial

Values Motives (.28, p < .01). However, although slightly higher in magnitude, the

association between LMX-quality and the Organizational Concern Motive was not

significantly stronger in comparison to the Prosocial Values Motive. Finally, no

significant relationship was found between subordinate perceptions of LMX-quality and

the Impression Management Motive (.08, p > .05).

Unlike perceptions of transformational leadership, subordinate reports of LMX-

quality were significantly related to supervisor reports (.26, p < .01). In addition,

56
supervisor reports of LMX correlated significantly with both the Organizational Concern

(.18, p < .05) and Impression Management Motives (-.20, p < .05), as well as the five

supervisor reported dimensions of OCB: Altruism (.50, p < .01), Conscientiousness (.44,

p < .01), Sportsmanship (.19, p < .05), Courtesy (.32, p < .01), and Civic Virtue (.36, p <

.01). Finally, perceptions of LMX-quality and transformational leadership correlated

significantly across both sources, subordinate (.77, p < .01) and supervisor (.37, p < .01).

Similar to previous research, a strong association was observed between the

Organizational Concern Motive and Prosocial Values Motive (.70, p < .01). In addition,

the Impression Management Motive showed significant correlations with both the

Organizational Concern (.39, p < .01) and Prosocial Values (.52, p < .01) Motives.

Significant relationships were also observed between all five subordinate reported OCB

dimensions and both Prosocial Values and Organizational Concern. Specifically, these

correlations ranged from .36 (p < .01) to .68 (p < .01) for the Organizational Concern

Motive, and from .27 (p < .01) to .56 (p < .01) for the Prosocial Values Motive.

Regarding supervisor reports of OCB, the Organizational Concern Motive correlated

significantly with four of the five dimensions: Altruism (.34, p < .01), Conscientiousness

(.24, p < .01), Courtesy (.25, p < .01), and Civic Virtue (.43, p < .01). In contrast, the

Prosocial Values Motive correlated significantly with only two OCB dimensions:

Altruism (.18, p < 05) and Civic Virtue (.29, p < .01). No significant relationships were

found between the Impression Management Motive and any of the five OCB dimensions.

It is interesting to note that both the Organizational Concern and Prosocial Values

Motives correlated significantly stronger with the Altruism dimension (in comparison to

the Conscientiousness dimension) across both self [t(130) = 2.19, p <.05] and supervisor

57
[t(130) = 1.98, p < .05] reports. This finding is significant in the context of previous

research on motives, and will be addressed later in the Discussion section of this paper.

Finally, Table 7 also displays the correlations between each of the five OCB

dimensions across both subordinate and supervisor reports. Each dimension correlated

with itself across sources within the range of .18 (p < .05) for the Courtesy dimension to

.44 (p < .01) for the Civic Virtue dimension.

Analysis Approach

In the case of each mediational hypothesis (Hypotheses 2, 3, 4, 5, and 7) Baron

and Kenny’s (1986) procedure involving a series of four regression equations was used to

test for mediation. In the first equation, the relevant OCB dimension was regressed onto

the antecedent variable. In the second equation, the relevant OCB motive, or predicted

mediator, was regressed onto the antecedent variable. In the third equation, the relevant

OCB dimension was regressed onto the relevant OCB motive. Finally, in the fourth

equation, the relevant OCB dimension was regressed onto both the OCB motive and

antecedent variable, with the relevant OCB motive being entered first into the equation.

At each stage, the beta coefficients were examined for significance. Mediation

occurred when four criteria were met: (1) the antecedent variable was shown to

significantly affect the relevant OCB dimension (equation 1); (2) the antecedent variable

was shown to significantly affect the relevant motive (equation 2); (3) the relevant motive

was shown to significantly affect the relevant OCB dimension (equation 3); and (4) the

effect of the antecedent variable on the relevant OCB dimension was significantly less in

the fourth equation than in the first. If one or more of the specified criteria were unmet,

then mediation was said not to have occurred.

58
 In cases where partial mediation was observed using Baron and Kenny’s (1986)

test, the Aroian version (1944/1947) of the Sobel test (1982) was also performed to

further test for the significance of the mediation effect. In general, the Sobel test

determined the significance of the intervening variable effect by first calculating the

product of the path coefficient associated with the independent variable and the mediator

(α) and the path coefficient associated with the mediator and the dependent variable (β).

The product of these two terms, α β, was then divided by its standard error, and compared

to a standard normal distribution (see MacKinnon, Lockwood, Hoffman, West, & Sheets,

2002, for a review).

In the case of Hypothesis 8, the moderating effect of the Prosocial Values Motive

on the relationship between LMX-quality and OCB was also tested using Baron and

Kenny’s (1986) procedure. It was assumed that the effect of LMX-quality on Altruism

would change linearly with respect to the moderator. To test this hypothesis, the

dependent variable (Altruism) was regressed onto: (1) the independent variable (LMX-

quality), (2) the predicted moderator (Prosocial Values), and (3) the product of these two

variables (LMX-quality and Prosocial Values). Moderation was indicated by the

significance of the beta-weight associated with the product term while controlling for the

individual effects of the independent and moderator variables.

Finally, in the cases of each mediational hypothesis, separate analyses were

performed using the Organizational Concern Motive and Prosocial Values Motive. This

approach was adopted due to the finding that both motives significantly correlated with

transformational leadership, LMX-quality, and various dimensions of OCB. Both of these

motives are discussed in terms of their comparative mediational effects later in this paper.

59
 Also, unless otherwise mentioned, all analyses investigating the relationship

between leadership perceptions (i.e., transformational leadership or LMX-quality) and

additional criterion measures (e.g., motives, OCB) were based on subordinate perceptions

of leadership. This second approach was justified based on the finding that subordinate

reports of transformational leadership and LMX-quality correlated more consistently, in

comparison to supervisor reports of leadership, with both the motives and self- and

supervisor-reports of OCB. A separate section is devoted to the comparison of

subordinate versus supervisor-reported results.

Hypothesis 1

The first hypothesis predicted a significant, positive relationship between

supervisor transformational leadership behavior and subordinate Organizational Concern.

As shown in Table 5, results supported this prediction using subordinate reports of

transformational leadership. Specifically, a .44 correlation (p < .01) was observed

between subordinate ratings of transformational leadership behavior and the

Organizational Concern Motive. In contrast, when transformational leadership was rated

by the leaders themselves, these perceptions were not significantly related with the

Organizational Concern Motive (r = .15, p > .05).

Hypothesis 2

Self-reports of OCB. The second hypothesis predicted that subordinate

Organizational Concern would mediate the relationship between supervisor

transformational leadership behavior and subordinate Conscientiousness. Using Baron

and Kenny’s (1986) approach, the first two regression equations revealed significant

relationships between transformational leadership and both the Conscientiousness

60
dimension (β = .31, p < .01) and the Organizational Concern Motive (β = .44, p < .01). In

addition, it was shown in the third equation that the Organizational Concern Motive

significantly affected the Conscientiousness dimension (β = .55, p < .01). When

Conscientiousness was regressed onto both the Organizational Concern Motive and

transformational leadership, respectively, the beta-weight associated with

transformational leadership showed a decrease from .31 (p < .01) in the first equation

(which included transformational leadership alone) to .09 (p > .05) in the fourth equation

(which included the Organizational Concern Motive and transformational leadership,

respectively). Based on Baron and Kenny’s (1986) criteria, these results provided

relatively strong evidence for full mediation (see Table 8).

Because significant relationships were also observed between transformational

leadership, the Prosocial Values Motive, and Conscientiousness, this same series of steps

was performed using subordinate Prosocial Values as the predicted mediator. As can be

seen in Table 8, the beta-weight associated with transformational leadership showed a

decrease from .31 (p < .01) in the first equation to .25 (p < .01) in the fourth equation.

These results provide evidence for partial mediation, and were also confirmed by results

of the Sobel test (z = 2.00, p < .05).

Supervisor-reports of OCB. As shown in Table 9, support was also found for

Hypothesis 2 using supervisor-reports of Conscientiousness. Specifically, the first two

regression equations revealed significant relationships between transformational

leadership and both the Conscientiousness dimension (β = .20, p < .05) and the

Organizational Concern Motive (β = .44, p < .01). In addition, it was shown in the third

equation that the Organizational Concern Motive significantly affected the

61
Table 8. Motives Mediator Analysis for Transformational Leadership and Self-reports of
OCB

Dependent Variable: Conscientiousness

Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .31** .31 .10**
Step4: OC Motive, TFL .09 .51** .56 .31**
Dependent Variable: Conscientiousness
Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .31** .31 .10**
Step4: PV Motive, TFL .25** .26** .40 . 16**
Dependent Variable: Sportsmanship
Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .29** .29 .08**
Step4: OC Motive, TFL .16 .29** .39 .15**
Dependent Variable: Sportsmanship
Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .29** .29 .08**
Step4: PV Motive, TFL .24** .22** .36 .13**
Dependent Variable: Courtesy
Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .27** .27 .07**
Step4: OC Motive, TFL .03 .54** .55 .30**
Dependent Variable: Courtesy
Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .27** .27 .07**
Step4: PV Motive, TFL .15 .50** .55 .31**
Dependent Variable: Civic Virtue
Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .32** .32 .10**
Step4: OC Motive, TFL .07 .58** .61 .37**
Dependent Variable: Civic Virtue
Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .32** .32 .10**
Step4: PV Motive, TFL .21** .44** .53 .28**
*p<.05, **p<.01
62
Table 9. Motives Mediator Analysis for Transformational Leadership and Supervisor-
reports of OCB

Dependent Variable: Conscientiousness

Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .20* .20 .04*
Step4: OC Motive, TFL .12 .19 .26 .07*
Dependent Variable: Courtesy
Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .19* .19 .04*
Step4: OC Motive, TFL .10 .21* .27 .07**
Dependent Variable: Civic Virtue
Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .18* .18 .03*
Step4: OC Motive, TFL - .01 .44** .43 .19**
Dependent Variable: Civic Virtue
Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .18* .18 .03*
Step4: PV Motive, TFL .12 .26** .31 .10**
*p<.05, **p<.01

Conscientiousness dimension (β = .24, p < .01). When Conscientiousness was regressed

onto both the Organizational Concern Motive and transformational leadership,

respectively, the beta-weight associated with transformational leadership showed a

decrease from .20 (p < .05) to .12 (p > .05). Thus, these results provided evidence for full

mediation. Results did not support the Prosocial Values Motive as a mediator, however,

as no relationship was found between this motive and supervisor reports of

Conscientiousness (β = .07, p > .05).

In the case of Hypothesis 2, both motives received empirical support as mediators.

However, based on the strength of the mediational effect associated with the

63
Organizational Concern Motive across sources, it can be argued that subordinate

Organizational Concern was the primary mediator in this relationship.

Hypothesis 3

Self-reports of OCB. Similar to the second hypothesis, Hypothesis 3 predicted

that subordinate Organizational Concern would mediate the relationship between

transformational leadership behavior and the Civic Virtue dimension. The first two

regression equations revealed significant relationships between transformational

leadership and both the Civic Virtue dimension (β = .32, p < .01) and the Organizational

Concern Motive (β = .44, p < .01). In addition, it was shown in the third equation that the

Organizational Concern Motive significantly affected the Civic Virtue dimension (β =

.61, p < .01). When Civic Virtue was regressed onto both the Organizational Concern

Motive and transformational leadership, respectively, the beta-weight associated with

transformational leadership showed a decrease from .32 (p < .01) to .07 (p > .05). Similar

to Hypothesis 2, these results again provided evidence for full mediation for the

Organizational Concern Motive.

This same series of steps was also performed with the Prosocial Values Motive.

Again, evidence for partial mediation was found, as the beta-weight associated with

transformational leadership showed a decrease from .32 (p < .05) in the first equation

(which included transformational leadership alone) to .21 (p < .05) in the fourth equation

(which included the Prosocial Values Motive and transformational leadership,

respectively). These results were also confirmed by results of the Sobel test (z = 2.46, p <

.05).

64
 Supervisor-reports of OCB. Analyses based on supervisor-reports of Civic Virtue

revealed similar results. That is, the first two regression equations revealed significant

relationships between transformational leadership and both the Civic Virtue dimension (β

= .18, p < .05) and the Organizational Concern Motive (β = .44, p < .01). In addition, it

was shown in the third equation that the Organizational Concern Motive significantly

affected the Civic Virtue dimension (β = .43, p < .01). When Civic Virtue was regressed

onto both the Organizational Concern Motive and transformational leadership,

respectively, the beta-weight associated with transformational leadership showed a

decrease from .18 (p < .05) to -.01 (p > .05). Based on Baron and Kenny’s (1986) crieria,

these results provided evidence for full mediation.

Similar results were found substituting the Prosocial Values Motive as the

predicted mediator. Specifically, previous analyses revealed significant relationships

between transformational leadership and both the Prosocial Values Motive and

supervisor-reports of Civic Virtue. In addition, when Civic Virtue was regressed onto the

Prosocial Values Motive, this relationship was also found to be significant (β = .29, p <

.01). When Civic Virtue was regressed onto both the Prosocial Values Motive and

transformational leadership, respectively, the beta-weight associated with

transformational leadership showed a decrease from .18 (p < .05) to .12 (p > .05). These

results provided evidence for the Prosocial Values Motive as a full mediator in this

antecedent-OCB relationship.

Thus, although both motives were associated with mediational effects in the case

of this hypothesis, results again supported the Organizational Concern Motive as the

primary mediator. These findings provide empirical evidence for Hypothesis 3.

65
Hypothesis 4

Self-reports of OCB. The fourth hypothesis predicted that subordinate

Organizational Concern would mediate the relationship between supervisor

transformational leadership behavior and subordinate Courtesy. The first two regression

equations revealed significant relationships between transformational leadership and both

the Courtesy dimension (β = .27, p < .01) and the Organizational Concern Motive (β =

.44, p < .01). In addition, it was shown in the third equation that the Organizational

Concern Motive significantly affected the Courtesy dimension (β = .55, p < .01). When

Courtesy was regressed on to both the Organizational Concern Motive and

transformational leadership, respectively, the beta-weight associated with

transformational leadership showed a decrease from .27 (p < .01) to .03 (p > .05). This

finding provides evidence of full mediation.

This same series of steps was also performed with the Prosocial Values Motive. In

contrast to previous results, however, evidence for full mediation was observed.

Specifically, the beta-weight associated with transformational leadership showed a

decrease from .27 (p < .01) in the first equation (which included transformational

leadership alone) to .15 (p > .05) in the fourth equation (which included the Prosocial

Values Motive and transformational leadership, respectively). These results again

conform to the mediational criteria outlined by Baron and Kenny (1986) regarding full

mediation.

Supervisor-reports of OCB. Supervisor-reports of Courtesy were also

significantly associated with transformational leadership (β = .19, p < .05) and the

Organizational Concern Motive (β = .25, p < .01). When Courtesy was regressed onto

66
both the Organizational Concern Motive and transformational leadership, respectively,

the beta-weight associated with transformational leadership decreased to .10 (p > .05).

Thus, support was provided for the Organizational Concern Motive as a full mediator. In

contrast, no relationship was found between the Prosocial Values Motive and supervisor-

reports of Courtesy. As a result, mediational analyses were not conducted with the

Prosocial Values Motive with regard to the Courtesy dimension.

In the case of Hypothesis 4, it was again observed that both the Organizational

Concern and Prosocial Values Motives were associated with mediational effects.

However, the relative reduction in the size of the beta-weights from the first to the fourth

equations again supports the Organizational Concern Motive as the primary mediator.

Thus, these results lend support to Hypothesis 4.

Hypothesis 5

Self-reports of OCB. Finally, results of this study also supported the fifth

hypothesis, which predicted a mediated relationship between supervisor transformational

leadership behavior and subordinate Sportsmanship using the Organizational Concern

Motive. Using Baron and Kenny’s (1986) procedure, the first two regression equations

revealed significant relationships between transformational leadership and both the

Sportsmanship dimension (β =.29, p < .01) and the Organizational Concern Motive (β =

.44, p < .01). The third equation also revealed that the Organizational Concern Motive

significantly affected the Sportsmanship dimension (β = .36, p < .01). When

Sportsmanship was regressed onto both the Organizational Concern Motive and

transformational leadership, respectively, the beta-weight associated with

67
transformational leadership showed a decrease from .29 (p < .01) to .16 (p > .05). These

results again provide evidence for full mediation.

In contrast, mediational support was not found using the Prosocial Values Motive.

Although the beta-weight associated with transformational leadership showed a small

decrease from .29 (p < .01) in the first equation (which included transformational

leadership alone) to .24 (p < .01) in the fourth equation (which included the Prosocial

Values Motive and transformational leadership, respectively), results of the Sobel test did

not support partial mediation (z = 1.81, p > .05).

Supervisor-reports of OCB. In contrast to the self-report results, no evidence was

found for a correlation between supervisor reports of Sportsmanship and either the

Organizational Concern (.17, p < .05) or Prosocial Values (.09, p > .05) Motives. Due to

the absence of these relationships, mediational analyses were not performed using this

criterion. Thus, Hypothesis 5 was only supported using self-reports of Sportsmanship.

Hypothesis 6

Hypothesis 6 dealt specifically with perceptions of LMX-quality, and predicted

that LMX-quality would correlate significantly with subordinate Prosocial Values. This

hypothesis was supported using subordinate reports of LMX-quality (.28, p < .01). In

contrast, no relationship was observed with regard to supervisor perceptions of LMX-

quality (-.08, p >.05).

Hypothesis 7

Self-reports of OCB. Based on the support found for Hypothesis 6, Hypothesis 7

predicted that subordinate Prosocial Values would mediate the relationship between

LMX-quality and the Altruism dimension. This hypothesis was partially supported, as

68
shown in Table 10. Specifically, the first two regression equations revealed significant

relationships between LMX-quality and both the Altruism dimension (β = .34, p < .01)

and the Prosocial Values Motive (β = .28, p < .01). In addition, it was shown in the third

equation that the Prosocial Values Motive significantly affected the Altruism dimension

(β = .56, p < .01). When Altruism was regressed onto both the Prosocial Values Motive

and LMX-quality, respectively, the beta-weight associated with LMX-quality showed a

decrease from .34 (p < .01) to .20 (p < .01). Because the beta-weight associated with

LMX-quality remained significant in the fourth equation, evidence was provided for

partial mediation. Results of the Sobel test reinforced this conclusion (z = 2.98, p < .01).

As results showed a significant correlation between LMX-quality and the

Organizational Concern Motive, this motive was also tested as a possible mediator. In

contrast to the Prosocial Values Motive, evidence for full mediation was observed.

Specifically, the beta-weight associated with transformational leadership showed a

decrease from .34 (p < .01) in the first equation (which included transformational

leadership alone) to .09 (p > .05) in the fourth equation (which included the

Organizational Concern Motive and transformational leadership, respectively).

Table 10. Motives Mediator Analysis for LMX-quality and Self-reports of Altruism

Dependent Variable: Altruism

Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .34** .34 .12**
Step4: PV Motive, LMX-quality .20** .50** .59 .35**
Dependent Variable: Altruism
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .34** .34 .12**
Step4: OC Motive, LMX-quality .09 .65** .69 .47**
*p<.05, **p<.01
69
 Supervisor-reports of OCB. Results based upon supervisor-reports of Altruism

further supported the Organizational Concern Motive as the primary mediator (see Table

11). Specifically, the first two regression equations revealed significant relationships

between LMX-quality and both the Altruism dimension (β = .22, p < .05) and the

Prosocial Values Motive (β = .28, p < .01). In addition, it was shown in the third equation

that subordinate Prosocial Values significantly affected the Altruism dimension (β = .18,

p < .05). When Altruism was regressed onto both the Prosocial Values Motive and LMX-

quality, respectively, the beta-weight associated with LMX-quality showed a decrease

from .22 (p < .05) to .18 (p < .05). Although this finding provided evidence for partial

mediation based on Baron and Kenny’s (1986) procedure, results of the Sobel test were

not significant (z = 1.29, p > .05). Therefore, these results did not provide strong

evidence for the Prosocial Values Motive as a partial mediator.

In contrast, evidence for full mediation was found for the Organizational Concern

Motive. That is, the beta-weight associated with transformational leadership showed a

decrease from .22 (p < .05) in the first equation (which included transformational

leadership alone) to .10 (p > .05) in the fourth equation (which included the

Organizational Concern Motive and transformational leadership, respectively).

Due to the relative reduction in the size of the beta-weights associated with the

Organizational Concern and Prosocial Values Motives, it can be argued that Hypothesis 7

only received partial support. While results were supportive of the Prosocial Values

Motive as a partial mediator in the relationship between LMX-quality and Altruism,

findings more fully supported subordinate Organizational Concern as the primary

mediator.

70
Table 11. Motives Mediator Analysis for LMX-quality and Supervisor-reports of
Altruism

Dependent Variable: Altruism

Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .22* .22 .05*
Step4: PV Motive, LMX-quality .18* .13 .25 .06*
Dependent Variable: Altruism
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .22* .22 .05*
Step4: OC Motive, LMX-quality .10 .31** .35 .13**
*p<.05, **p<.01

Hypothesis 8

Self-reports of OCB. Alternative to Hypothesis 7, Hypothesis 8 predicted that the

Prosocial Values Motive would moderate the relationship between LMX-quality and the

Altruism dimension. Results did not support this prediction, as the beta-weight associated

with the LMX*Prosocial Values interaction term was not significant (β = -.40, p > .05)

once LMX-quality and the Prosocial Values Motive were accounted for in the regression

equation. Similar results were also observed for the Organizational Concern Motive (β =

-.25, p > .05).

Supervisor-reports of OCB. Hypothesis 8 was also tested using supervisor-

reports of Altruism. Again, results did not support a moderated relationship with regard

to the Prosocial Values Motive. In particular, the beta-weight associated with the

LMX*Prosocial Values interaction term was not significant (β = -.66, p > .05) once

LMX-quality and the Prosocial Values Motive were accounted for in the regression

equation. This effect was also observed using the Organizational Concern Motive (β = -

.80, p > .05).

71
Additional Analyses

Based on additional relationships observed in this study (e.g., LMX-quality and

various dimensions of OCB), further analyses were conducted to learn more about the

connection between leadership, motives, and OCB. The results of these analyses are

described below.

Transformational Leadership, Altruism, and the OC and PV Motives. As

mentioned, a significant correlation (.33, p < .01) was observed between subordinate

perceptions of transformational leadership and self-reports of Altruism. To further

investigate the nature of this relationship, additional mediational analyses were conducted

using both the Organizational Concern and Prosocial Values Motives (see Table 12).

Beginning with the Organizational Concern Motive, results were supportive of

full mediation. Specifically, the beta-weight associated with transformational leadership

showed a decrease from .33 (p < .01) in the first equation (which included

transformational leadership alone) to .04 (p > .05) in the fourth equation (which included

the Organizational Concern Motive and transformational leadership, respectively).

In contrast, results supported the Prosocial Values Motive as a partial mediator.

Specifically, the beta-weight associated with transformational leadership showed a

decrease from .33 (p < .01) in the first equation (which included transformational

leadership alone) to .21 (p < .01) in the fourth equation (which included the Prosocial

Values Motive and transformational leadership, respectively). In addition, results of the

Sobel test supported this conclusion (z = 2.55, p < .05).

72
Table 12. Motives Mediator Analysis for Transformational Leadership and Self-reports
of Altruism

Dependent Variable: Altruism

Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .33** .33 .11**
Step4: OC Motive, TFL .04 .67** .69 .47**
Dependent Variable: Altruism
Independent
Variable β (antecedent) β (mediator) R R2
Step1: TFL .33** .33 .11**
Step4: PV Motive, TFL .21** .51** .60 .36**
*p<.05, **p<.01

LMX-quality, OCB, and Motives. It was also observed that LMX-quality related

positively to all five self-reported OCB dimensions, and, with the exception of

Conscientiousness, all five supervisor-reported dimensions. Additional analyses were

performed to identify the nature of these relationships; specifically, to determine whether

the Organizational Concern or Prosocial Values Motives served as a primary mediator.

The results of each of the regression analyses, including the beta-coefficients and

their associated significance levels, are displayed in Tables 13 and 14. With the exception

of Civic Virtue, full mediation was observed regarding both motives for each of the self-

report OCB dimensions. However, in each case, the relative decrease in the beta-weights

from the first to the fourth equations was greatest for the Organizational Concern Motive.

These results argue for the subordinate Organizational Concern as the primary mediator,

although the Prosocial Values Motive was also associated with significant mediational

effects.

In the case of self-reported Civic Virtue, both motives were associated with

partial mediation. This finding was also supported by results of the Sobel test
73
Table 13. Motives Mediator Analysis for LMX-quality and Self-reports of OCB

Dependent Variable: Conscientiousness

Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .24** .24 .06**
Step4: OC Motive, LMX-quality .03 .54** .55 .31**
Dependent Variable: Conscientiousness
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .24** .24 .06**
Step4: PV Motive, LMX-quality .16 .27** .36 .13**
Dependent Variable: Sportsmanship
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .21* .21 .05*
Step4: OC Motive, LMX-quality .09 .32** .37 .14**
Dependent Variable: Sportsmanship
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .21** .21 .05*
Step4: PV Motive, LMX-quality .15 .23** .31 .10**
Dependent Variable: Courtesy
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .20* .20 .04*
Step4: OC Motive, LMX-quality - .02 .56** .55 .30**
Dependent Variable: Courtesy
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .20** .20 .04*
Step4: PV Motive, LMX-quality .05 .52** .54 .29**
Dependent Variable: Civic Virtue
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .37** .37 .13**
Step4: OC Motive, LMX-quality .15* .55** .62 .39**
Dependent Variable: Civic Virtue
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .37** .37 .13**
Step4: PV Motive, LMX-quality .25** .42** .55 .30**
*p<.05, **p<.01

74
Table 14. Motives Mediator Analysis for LMX-quality and Supervisor-reports of OCB

Dependent Variable: Courtesy

Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .23** .23 .05**
Step2: OC Motive, LMX-quality .16 .19* .29 .08**
Dependent Variable: Civic Virtue
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .23** .23 .05**
Step2: OC Motive, LMX-quality .07 .40** .44 .19**
Dependent Variable: Civic Virtue
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .23** .23 .05**
Step2: PV Motive, LMX-quality .16 .24** .33 .11**
*p<.05, **p<.01

[Organizational Concern Motive (z = 3.98, p < .01), Prosocial Values Motive (z = 2.82, p

< .05)]. Again, however, the largest decrease in the value of the beta-weights was

associated with the Organizational Concern Motive. This result would again support

subordinate Organizational Concern as the primary mediator in this relationship.

Supervisor-reports of OCB displayed a similar trend. As mentioned, LMX-quality

correlated significantly with four of the five OCB dimensions (the exception was

Conscientiousness). Of these, subsequent mediational analyses were performed with the

Courtesy and Civic Virtue dimensions (mediation was not attempted with Sportsmanship,

as supervisor-reports of this dimension failed to correlate with either the Prosocial Values

or Organizational Concern Motive). As can be seen in Table 14, evidence for full

mediation was observed for the Organizational Concern Motive regarding both the

Courtesy and Civic Virtue dimensions. In addition, the Prosocial Values Motive was

found to fully mediate the relationship between LMX-quality and Civic Virtue. In the
75
case of this dimension, however, the relative decrease in the value of the beta-weights

was larger for the Organizational Concern Motive in comparison to the Prosocial Values

Motive.

Mediational Analyses from the Supervisors’ Leadership Perspective

As mentioned, subordinate and supervisor perceptions of leadership displayed

differential relationships with a number of criteria. For example, although subordinate

perceptions of transformational leadership correlated significantly with both the

Organizational Concern and Prosocial Values Motives, these relationships did not exist

when transformational leadership was measured from the perspective of the supervisor.

As such, mediational analyses were only conducted using subordinate reports of

transformational leadership.

The case was somewhat different for LMX-quality. Specifically, significant

relationships were observed between supervisor perceptions of LMX-quality and both the

Organizational Concern and Impression Management Motives, as well as the five

supervisor-reported OCB dimensions. Therefore, mediational analyses were conducted

using the Organizational Concern Motive (no relationship was found between the

Impression Management Motive and supervisor-reported OCB) and supervisor-reported

Altruism, Conscientiousness, Courtesy, and Civic Virtue. Sportsmanship was not

included in these analyses, as the Organizational Concern Motive failed to significantly

correlate with this dimension.

As can be seen in Table 15, the beta-weights associated with LMX-quality

dropped slightly from the first regression equation to the fourth for each of the OCB

dimensions tested. Although these results are evidence for partial mediation according to

76
Baron and Kenny’s (1986) criteria, results of the Sobel test argued against partial

mediation for each dimension (Altruism, z = 1.73, p >.05; Conscientiousness, z = 1.38, p

> .05; Courtesy, z = 1.50, p > .05; Civic Virtue, z = 1.87, p > .05). Thus, given these

results and the relatively small decrease in the value of the beta-weights, it can be argued

that the Organizational Concern Motive was not a mediator in these relationships.

Table 15. Motives Mediator Analysis Based on the Supervisors’ Perspective of LMX-
quality and OCB.

Dependent Variable: Altruism

Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .50** .50 .25**
Step4: OC Motive, LMX-quality .45** .26** .56 .31**
Dependent Variable: Conscientiousness
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .44** .44 .19**
Step4: OC Motive, LMX-quality .41** .16* .47 .22**
Dependent Variable: Courtesy
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .32** .32 .10**
Step4: OC Motive, LMX-quality .28** .20* .37 .14**
Dependent Variable: Civic Virtue
Independent
Variable β (antecedent) β (mediator) R R2
Step1: LMX-quality .36** .36 .13**
Step4: OC Motive, LMX-quality .29** .38** .52 .27**
*p<.05, **p<.01

77
 Chapter Four

Discussion

In general, the goal of this study was to expand upon previous research in the area

of leadership and OCB by further defining relationships between select leadership styles

(i.e., transformational leadership and LMX-quality) and different OCB dimensions. In

contrast to a direct effects model, this study hypothesized that both transformational

leadership and LMX- quality were associated with the arousal of specific motives states

(e.g., Organizational Concern, Prosocial Values), and that these motives, in turn, were

associated with the performance of OCB (i.e., a through mediation and/or moderation).

This hypothesis was taken one step further by suggesting that different motives were

more strongly associated with specific leadership styles, as well as with different OCB

dimensions.

As a whole, results of this study support the mediated model depicted in Figure 1.

That is, support was found across both supervisor and subordinate ratings of OCB that

motives do, in fact, mediate the relationship between select leadership styles and different

OCB dimensions. Contrary to prediction, however, the Organizational Concern Motive

was supported as the dominant mediator across all leadership-OCB relationships.

Specifically, although both motives received support as mediators, the mediational effect

was generally stronger for the Organizational Concern Motive as compared to the

Prosocial Values Motive. Taken together, these findings suggest that both leadership

78
styles are associated with a general concern for the organization, which in turn, is linked

with the performance of OCB.

Altruism

Transformational
Conscientiousness
Leadership

Organizational
Sportsmanship
Concern

Courtesy
LMX-quality

Civic Virtue

Figure 1. A Mediated Model of the Effects of Transformational Leadership and LMX

quality on OCB

It should be noted that Figure 1 displays a slightly oversimplified interpretation of

the results of this study by failing to include the mediational effects observed for the

Prosocial Values Motive. The rational for this model as well as further discussion of the

results are provided below.

Relationships Among Variables

As previously mentioned, a number of significant relationships were observed

among the variables in this study. Most importantly, subordinate reports of

79
transformational leadership and LMX-quality were significantly associated with all five

dimensions of OCB. As a whole, this trend was observed across both self- and

supervisor-reports of OCB, although the correlations were somewhat smaller using the

supervisor OCB ratings. The only exception to this trend were the nonsignificant

correlations found between transformational leadership and supervisor-reported Altruism,

as well as LMX-quality and supervisor-rated Conscientiousness.

In general, these findings support previous research addressing the connection

between transformational leadership, LMX-quality, and OCB (e.g., Podsakoff et al.,

2001). That is, transformational leaders were found to have a positive impact across

every form of citizenship behavior. Similarly, supervisor-subordinate relationships

characterized by high levels of trust, support, liking, and attention were also associated

with increased OCB. This research suggests that these specific leadership styles

encourage employees to engage in informal behaviors that benefit the organization,

ranging from those directed toward individual organizational members (e.g., Altruism,

Courtesy) to those aimed at benefiting the organization as a whole (e.g.,

Conscientiousness, Civic Virtue). This was a key finding that set the stage for

determining the exact nature of these leadership-OCB relationships.

As predicted, significant relationships were also found between both styles of

leadership and OCB motives. Specifically, results showed significant positive

relationships between subordinate perceptions of both leadership styles and the

Organizational Concern and Prosocial Values Motives (lending support to Hypotheses 1

and 6). Transformational leadership, however, showed a significantly stronger

relationship with the Organizational Concern Motive. Although larger in magnitude, the

80
correlation between LMX-quality and the Organizational Concern Motive was not

significantly different from its correlation with the Prosocial Values Motive. Finally, no

relationship was found between subordinate perceptions of either leadership style and the

Impression Management Motive.

These findings suggest that both transformational leadership and LMX-quality are

linked with specific employee motivations. Although also associated with the desire to

help others and be accepted by them, transformational leadership was more strongly

related with an employee’s general concern for the organization’s overall well-being. As

mentioned, this association aligns with a basic tenant of transformational leadership. That

is, encouraging followers to transcend their own self-interests and move beyond simple

leader-member transactions for the good of the organization (Bass, 1985).

In contrast, perceptions of LMX-quality were closely linked with both motive

states. That is, high-quality supervisor-subordinate relationships were positively related

to a desire to help others and be accepted by them, as well as a positive regard for the

organization as a whole. Although not significantly different, perceptions of LMX-quality

did show a slightly stronger association with the Organizational Concern Motive. As with

transformational leadership, this finding suggests that a general concern for the

organization is a primary outcome of a healthy supervisor-subordinate relationship.

It was interesting to note that neither leadership style was associated with

Impression Management (when measured from the perspective of the subordinate). More

specifically, subordinate perceptions of either leadership style were not linked with a

desire to avoid negative evaluation by others or to gain material rewards. This finding is

somewhat encouraging, as it suggests that effective leadership is more likely to evoke

81
more genuine motivations to help the organization. While past studies have shown an

association between all three motives and various types of OCB (e.g., Finkelstein &

Penner, 2004), the Organizational Concern and Prosocial Values motives seem to be a

closer match with conceptions of how effective leadership impacts subordinate

motivations.

In general, past studies on OCB motives had revealed differential relationships

between both the Organizational Concern and Prosocial Values Motive and certain

dimensions of OCB (e.g., Rioux & Penner, 2001; Connell & Penner, 2004). In particular,

the Organizational Concern Motive was consistently most strongly associated with the

Conscientiousness dimension, while the Prosocial Values Motive was more strongly

related to the Altruism dimension. The current study observed a slightly different trend.

That is, of the two OCB dimensions mentioned (Altruism and Conscientiousness), both

the Organizational Concern and Prosocial Values Motive were found to correlate

significantly stronger with the Altruism dimension across both subordinate and

supervisor reports of OCB. In addition, both the Organizational Concern and Prosocial

Values Motive correlated most strongly with the Civic Virtue dimension (.43, .29,

respectively, p <.01) using supervisor-reports of OCB. Currently, we have not found a

convincing explanation for why this occurred. For example, restriction of range alone

could not have accounted for this effect, as the degree of variability was similar across

both motives and OCB dimensions. Regardless, the finding that both motives consistently

correlated with each of the five OCB dimensions reinforces their role as important

antecedents of OCB.

82
 With regard to supervisor perceptions of leadership, findings were somewhat less

consistent when compared with subordinate perceptions. For example, it was somewhat

surprising to observe the lack of agreement between subordinate and supervisor

perceptions of transformational leadership (r = .16, p > .05). Similarly, although the

correlation between subordinate and supervisor perceptions of LMX-quality was

significant (.27, p < .01), this relationship was surprisingly small considering both parties

were (in theory) rating the same relationship.

Although perceptions of transformational leadership are predominantly measured

from the subordinate perspective, other research has compared the perceptions of LMX-

quality across leaders and followers. For example, in their meta-analytic review of LMX

theory, Gernster and Day (1997) found an uncorrected correlation of .29 between leader

and member perceptions of LMX-quality. This finding suggests that a certain amount of

disagreement between leader and subordinate perceptions of leadership is not unusual, as

was the case in the current study. As such, it is important to include both parties’

perspectives when investigating the overall effects of leadership on both subordinate and

organizational outcomes. From a practical perspective, this finding also demonstrates that

a leader should make the effort to understand how his/her behaviors are being perceived

by the employees that he/she supervises.

Along similar lines, supervisor perceptions of leadership also showed different

relationships with a number of criterion variables. Not surprisingly, supervisor

perceptions of both leadership styles most consistently correlated with supervisor reports

of OCB (with the exception of transformational leadership and both the Sportsmanship

and Courtesy dimensions). However, in contrast to transformational leadership,

83
supervisor reports of LMX-quality correlated significantly with both the Organizational

Concern and Impression Management Motives. This finding is especially interesting for

the Impression Management Motive, which was not significantly related to subordinate

perceptions of LMX-quality. It is not completely clear why leader perceptions of LMX-

quality would be a better indicator of employee Impression Management than

subordinate perceptions. Conversely, it could perhaps be argued that an employee’s

tendency to impression manage somehow influences their supervisor’s view of their own

leadership ability. Regardless of the direction of this relationship, future research is

needed to further clarify the relationship between Impression Management and

subordinate and supervisor perceptions of LMX-quality.

Finally, significant correlations were also observed between each of the OCB

dimensions across rating sources. In general, the size of these correlations suggests that,

although related, subordinate and supervisor perceptions of OCB do not completely

overlap. This finding again argues for the use of multiple sources regarding OCB

research. In addition, better agreement was observed for certain types of OCB versus

others. For example, OCB directed towards the organization as a whole (e.g., Civic

Virtue, Conscientiousness) tended to show higher correlations across sources than OCB

directed towards individual members (e.g., Altruism, Courtesy). This finding may be

attributed to the increased visibility associated with certain types of OCB. For example,

serving on a formal committee within the organization may be more visible to a

supervisor than assisting another co-worker during a typical workday.

84
Transformational Leadership, Motives, and OCB

Overall, significant support was found for subordinate Organizational Concern as

a primary mediator in the relationship between subordinate perceptions of

transformational leadership and OCB. This likewise provided support for Hypotheses 1 –

5. As mentioned, subordinate perceptions of transformational leadership correlated

significantly with the Organizational Concern Motive (.44). In addition, evidence for full

mediation was found for the Organizational Concern Motive across self- and supervisor-

reports of Conscientiousness, Courtesy, and Civic Virtue, as well as self-reports of

Sportsmanship. Finally, additional analyses also revealed that the Organizational Concern

Motive was associated with full mediational effects with regard to self-reports of

Altruism.

Taken together, these findings provide important evidence for the notion that the

relationship between perceptions of transformational leadership and OCB may not be

direct in nature. Rather, as predicted, a mediated model that includes employee

motivations seems more descriptive (see Figure 1). Based on the results of this study, it

can be argued that subordinate perceptions of transformational leadership are associated

with a general concern for the organization. These feelings are, in turn, linked with a

number of different types of OCB. Thus, evidence supports the notion that subordinate

Organizational Concern serves as the underlying mechanism by which transformational

leadership influences OCB.

It should be noted that evidence was provided for the Prosocial Values Motive as

a partial, and in some cases, full mediator regarding transformational leadership and all

five OCB dimensions. However, in each case, the relative decrease in the beta-weights

85
from the first regression equation (including transformational leadership alone) to the

fourth equation (including transformational leadership and the relevant motive,

respectively) was larger for the Organizational Concern Motive than the Prosocial Values

Motive. These results suggest that subordinate perceptions of transformational leadership

are associated with two specific motive states: a desire to help others and be accepted by

them, and a general concern for the organization’s well-being. However, in terms of their

relative mediational effects regarding different types of OCB, evidence supports the

Organizational Concern Motive as the primary underlying mechanism.

As a whole, these results clearly identify a third variable (Organizational

Concern) as the primary underlying mechanism by which transformational leadership

influences employee OCB. In addition, the mediational effects associated with the

Organizational Concern Motive were roughly the same across most forms of citizenship

performance. This finding is significant, as it supports the theoretical notion that

transformational leaders promote a general positive regard for the organization among

their followers. Most importantly, this study addresses a current gap in the literature by

describing how certain leadership behaviors impact OCB. As depicted here, without

feelings of Organizational Concern, the effects of transformational leadership on

subordinate OCB are unlikely to be realized.

LMX-quality, Motives, and OCB

This study also predicted that the relationship between LMX-quality and the

Altruism dimension would be mediated and/or moderated by employee motives. In the

case of Hypothesis 7, it was predicted that the Prosocial Values Motive would serve as

the underlying mechanism by which LMX-quality influenced subordinate Altruism.

86
Results were supportive of the Prosocial Values Motive as a partial mediator using self-

reports of Altruism (no support was found using supervisor reports). However, stronger

evidence was observed for the Organizational Concern Motive. Specifically, results

provided evidence of full mediation for the Organizational Concern Motive across both

sources of OCB ratings (self and supervisor).

In the case of Hypothesis 8, it was predicted that the Prosocial Values Motive

would also moderate the relationship between LMX-quality and subordinate Altruism. In

other words, a relatively strong positive relationship between LMX-quality and

subordinate Altruism was proposed to exist when the level of subordinate Prosocial

Values was high. In contrast, when the level of subordinate Prosocial Values was low, a

much smaller relationship was predicted to exist between these two variables. Results did

not support this prediction using either self- or supervisor-reports of OCB. In addition,

nonsignificant results were also observed with the Organizational Concern Motive.

Although these last set of hypotheses did not receive strong support, their results

are still useful regarding the effects of relationship quality on OCB. Specifically, this

evidence suggests that LMX-quality is significantly associated with both the

Organizational Concern and Prosocial Values Motives, and that a slightly stronger

relationship exists regarding the Organizational Concern Motive. These motivations are,

in turn, linked with OCB directed toward individual members of the organization. Thus,

similar to transformational leadership, Organizational Concern serves as the general

mechanism by which LMX-quality impacts this particular OCB dimension. Although

contrary to prediction, this result is not surprising based on the relatively strong

correlation found between transformational leadership and LMX-quality (.77, p < .01). In

87
addition, both leadership styles showed larger correlations with the Organizational

Concern Motive as compared with the Prosocial Values Motive.

With regard to the role of either the Organizational Concern or Prosocial Values

Motive as a moderator, it is unlikely that either motive is associated with the

hypothesized effects. More specifically, the strength of the relationship between LMX-

quality and subordinate Altruism was not contingent upon the level of Prosocial Values

(or Organizational Concern) reported by the subordinate. Although a small sample size

likely contributed to this result, the nonsignificant p-values were large, and were unlikely

to increase even given a larger sample size ( i.e., N > 200). Although contrary to

prediction, it seems both the Organizational Concern and Prosocial Values Motives are

more suited to the role of mediators. However, future research employing larger sample

sizes is still needed to further clarify these relationships.

Although results were not strongly supportive of Hypotheses 7 and 8, additional

analyses involving LMX-quality, motives, and additional dimensions of OCB provided

more encouraging results. Specifically, with the exception of Civic Virtue, full mediation

was observed for both motives (e.g., Prosocial Values and Organizational Concern)

regarding each of the self-reported OCB dimensions. For Civic Virtue, both motives were

associated with partial mediation. In the case of supervisor-reports of OCB, results were

also similar. For the dimensions that were tested, Courtesy and Civic Virtue, evidence for

full mediation was observed for the Organizational Concern Motive. Full mediation was

also observed for the Prosocial Values Motive in the case of Civic Virtue.

Taken together, these results somewhat mirror those found with transformational

leadership. That is, across both rating sources and numerous dimensions of OCB,

88
evidence was provided that supports the Organizational Concern Motive as the primary

mediator between LMX-quality and subordinate OCB. As depicted in Figure 1, leader-

member relationships characterized by high levels of support, trust, liking, and latitude

are associated with a general concern for the organization. These feelings, in turn, are

linked with different types of subordinate OCB. This result was somewhat contrary to

prediction, as the Prosocial Values Motive was predicted to play a more substantial role

in these relationships. However, based on the strong correlation found between

perceptions of transformational leadership and LMX-quality, it is not surprising that these

two leadership approaches, although arguably unique, would display similar relationships

with the OCB motives.

Comparative Mediational Effects

Overall, the analyses described above lend support to the Organizational Concern

Motive as the primary mediator in the relationships between perceptions of both

leadership variables and various dimensions of OCB. Two main pieces of evidence

support this argument. First, although in the case of most of the hypotheses, the Prosocial

Values Motive received support as a partial mediator, the relative decrease in the beta-

weights associated with the leadership predictor (e.g., transformational leadership, LMX-

quality) from the first regression equation to the fourth was generally larger for the

Organizational Concern Motive than for the Prosocial Values Motive. Based on this

evidence, it can be argued that the Organizational Concern Motive played a more

dominant role in these leadership-OCB relationships.

Second, partial correlations observed between the leadership variables, both

motives, and the five OCB dimensions further support this argument. That is, in order to

89
address the relative effect of each motive, partial correlations were computed between the

Prosocial Values Motive and both transformational leadership and LMX-quality while

controlling for the effects of the Organizational Concern Motive. In both cases, the partial

correlation between the Prosocial Values Motive and either transformational leadership

or LMX-quality was nonsignificant (-.10, .02, respectively). Results were similar when

partial correlations were calculated between the Prosocial Values Motive and both self-

and supervisor-reports of OCB. Specifically, when the effect of the Organizational

Concern Motive was controlled, the correlation between the Prosocial Values Motive and

each of the five OCB dimensions was nonsignificant (the only exception was self-reports

of Courtesy). In contrast, this effect was not observed with the Organizational Concern

Motive when the influence of the Prosocial Values Motive was controlled.

These findings provide further support for the Organizational Concern Motive as

the primary mediator. Specifically, it is possible that any mediational effects observed

with the Prosocial Values Motive were due to its overlap with the Organizational

Concern Motive. Alternatively, it was also postulated that the Prosocial Values Motive

may have acted as a suppressor variable. In other words, including this variable in the

regression equation along with the other predictors (i.e., either transformational

leadership or LMX-quality, and the Organizational Concern Motive) may have helped to

explain additional variance in OCB. This hypothesis was also tested using multiple

regression. However, results did provide evidence for this effect with regard to any of the

OCB dimensions.

90
Different Leadership Perspectives (Subordinate versus Supervisor)

It should be noted that the results of this study differed substantially depending on

the source of leadership inquiry. For example, one interesting finding was that

mediational effects were only observed with transformational leadership when these

behaviors were measured from the perspective of the subordinate. As mentioned,

supervisor perceptions of transformational leadership failed to correlate significantly with

either the Organizational Concern, Prosocial Values, or Impression Management motives.

Thus, mediational analyses could not be performed using this particular set of predictors.

Combined with the finding that subordinate reports of transformational leadership failed

to significantly correlate with supervisor reports (.16, p > .05), these results suggest that a

leader’s view of their own transformational leadership behaviors is not necessarily an

accurate indicator of their subordinates’ performance motivations.

In the case of LMX-quality, significant relationships were observed between

supervisor-reports of leadership and both the Organizational Concern and Impression

Management Motives, as well as all five supervisor-reported OCB dimensions. However,

evidence was not supportive of either motive as a mediator in these leadership-OCB

relationships. As mentioned, these results are in stark contrast to those associated with

subordinate-reports of LMX-quality, which found substantial evidence for the

Organizational Concern Motive as a primary mediator. These different outcomes again

highlight the need to include both parties perspectives in leadership research. Also, in

comparison to leader perceptions of transformational leadership, this study suggests that a

leader’s assessment of LMX-quality can be a significant predictor of what motivates a

subordinate on the job.

91
Limitations

This study helped to shed light on a number of important relationships between

transformational leadership, LMX-quality, motives, and OCB. That being said, certain

limitations should also be acknowledged. First and foremost, as with most studies

investigating OCB, this study was cross-sectional in nature. As a result, it is difficult to

make causal inferences regarding the relationship between leadership, motives, and OCB.

Future studies should incorporate more longitudinal designs, so that the influence of both

leadership and motives on OCB may be examined over time. In addition, more

experimental approaches should also be used, allowing more insight regarding the issue

of causality.

A second potential weakness of this study was that the supervisors selected the

subordinates included in this study. This approach is in contrast to either the researcher

randomly selecting the subordinates for each supervisor, or the supervisor providing

ratings on each of the employees that they supervise (which was the case in some

instances). The danger with the approach used in this study is that the supervisors could

have inadvertently biased these results by selecting only those employees who perform

exceptionally well on the job. Thus, the sample would have only included employees

who tended to report high levels of OCB, as well as more favorable ratings of

transformational leadership and LMX-quality. We attempted to counter this threat by

including a statement in the supervisor’s set of instructions that reminded them to select

employees that represented a range of performance. In other words, each supervisor was

instructed not to select all high performers or all low performers.

92
 Examination of the descriptives and distributions for each scale indicated that the

instructions were moderately successful at incorporating an acceptable amount of

variance into the sample. Although a number of significant correlations were observed

across variables, the range of scores observed for each scale tended to be somewhat

small. In addition, the distributions were bimodal for some of the measures. That is, the

majority of scores tended to occur both towards the middle and at the high end of the

range for these particular scales. However, although it could be argued that this sample

displayed a slight positive bias with regard to leadership perceptions and OCB, we would

argue that this trend is not significant enough to discredit the results of this study.

Finally, research has consistently demonstrated that reports of OCB differ

depending on the source. For instance, employees tend to exaggerate the frequency of

their own behaviors, or may monitor these behaviors while in the presence of a

supervisor. The present study addressed this concern by including both self- and

supervisor-ratings of OCB. However, it could be argued that the inclusion of co-worker

ratings would have provided a more comprehensive perspective. This criticism is justified

by the tendency of co-workers to have closer and more frequent contact with an

employee when compared to a supervisor, which allows them more opportunities to

observe the occurrence of OCB. However, supervisor ratings generally have been found

to be more objective in comparison to co-worker ratings. This observation has been

credited to factors such as friendship, which may sometimes bias co-worker perceptions.

Thus, although the inclusion of supervisor-reports of OCB was clearly a more

comprehensive approach than relying on self-reports alone, the inclusion of co-worker

93
reports would have arguably provided an even more complete representation of the

subordinates’ OCB.

Future Research

Based on what this study has revealed, a number of future directions should be

taken to further expand on these results. First, these findings have significant bearing on

traditional antecedent-OCB models. Penner et al.’s (1997) conceptual model of OCB (see

Figure 2), for example, argues that short-term OCB (i.e., intermediate OCB) is influenced

by organizational variables, job attitudes, mood on the job, prosocial orientation, and

motives for OCB. However, as time passes, individuals who engage in high levels of

OCB began to identify with the role of the “good organizational citizen.” That is,

organizational citizenship becomes a component of their “role identity” within the

organization. Penner and his colleagues argue that the development of this type of

personal identity is important because it becomes the mechanism by which each of the

variables mentioned above affects “enduring” or long-term OCB. That is, over time, the

direct effects of these variables on OCB are significantly reduced, and are instead

transferred through the individual’s “role identity” as a good organizational citizen.

Combined with the work of other researchers (Tillman,1998; Connell & Penner,

2004; Finkelstein & Penner, 2004), results of the study reconceptualize the manner in

which motives are expected to influence OCB. Specifically, in addition to serving as

antecedents, these prior studies suggest that the Prosocial Values and Organizational

Concern Motives both moderate and mediate the relationship between certain antecedent

variables (e.g., procedural justice, affective commitment) and OCB. In addition,

Finkelstein and Penner (2004) determined that the development of citizen role identity is

94
linked with the acquisition of motives, and that these motives influence the performance

of OCB (rather than the reverse).

Organizational
Variables

Job
Attitudes

Mood on Intermediate Citizen Role Enduring

the Job OCB OCB
Identity

Prosocial
Orientation

Motives for
OCB

Figure 2. Penner et al.’s (1997) Conceptual Model of the Causes of OCB

The current study expanded upon these results by providing evidence for similar

relationships among additional antecedents (i.e., transformational leadership, LMX-

quality) and different dimensions of OCB. That is, perceptions of both leadership styles

were primarily associated with subordinate Organizational Concern, which in turn, was

linked with a variety of short-term OCB behaviors. Combined with previous studies,

these results clearly demonstrate a need to revisit Penner et al’s OCB model. Future

research should continue to expand upon its theoretical tenants, as well as similar models

95
of OCB. As seen with the current study, additional dispositional variables and outcomes

of OCB should be investigated to help further clarify the nature of these relationships.

On a related note, efforts should also be directed towards more underemphasized

antecedents of OCB. As mentioned, one such area that has shown significant potential are

task characteristics. For example, in his recent review of the OCB literature, Podsakoff et

al. (2001) demonstrated that task feedback, task routinization, and intrinsically satisfying

tasks each displayed significant correlations with Altruism, Conscientiousness,

Sportsmanship, Courtesy, and Civic Virtue across multiple studies. Specifically, task

feedback and intrinsically satisfying tasks displayed significant positive relationships,

while task routinization displayed significant negative relationships. Building on the

results of this study, future efforts should address whether employee motives play any

significant role in these relationships. For example, it may be the case that all three task

variables are significantly related with subordinate Organizational Concern. That is, the

amount of satisfaction and feedback associated with a particular task could be positively

related to one’s concern for the organization, while the routine nature of a task could

likewise contribute to a lack of organizational concern. As with perceptions of

transformational leadership and LMX, the Organizational Concern Motive could also

potentially serve as the underlying mechanism by which these task characteristics

influence OCB. In addition, similar effects may also be observed with other leadership

variables (e.g., Supportive Behavior) and antecedents (e.g., role ambiguity, role conflict,

perceived organizational support), each of which have shown significant relationships

with OCB and also share a theoretical connection with the Organizational Concern

Motive.

96
 In sum, the areas mentioned above are currently underresearched in the OCB

literature. The results of the current study provide a theoretical basis for uncovering what

mechanisms may be involved in their relationships with citizenship performance.

At this time, research addressing the effects of cultural differences on OCB is

somewhat limited. As proposed by Podsakoff et al. (2001), these potential effects could

vary from the types of OCB that are performed in organizations, to the strength of the

relationships observed between select antecedents and different dimensions of OCB.

Similarly, employee motivations for engaging in OCB may also vary by culture. For

example, it is reasonable to assume that a greater importance may be placed on the

Organizational Concern Motive in a culture that is primarily collectivistic versus

individualistic. This hypothesis is based on the tendency of collectivist cultures to value

the success of the group as a whole (e.g., the organization), whereas individualistic

cultures tend to emphasize personal success as the ultimate reward. Similarly, an entirely

different set of motives may be appropriate for cultures that are dissimilar to the United

States. That is, cultural values may play a significant role in shaping an employees’

performance motivations. However, until efforts are made to incorporate cultural nuances

into OCB research, the knowledge surrounding OCB, including its drivers and

organizational impacts, will only be generalizable to Western societies.

Finally, better statistical techniques are needed in the investigation of antecedents

and OCB. Structural equation modeling (SEM), for example, has the ability to test the

plausibility of an entire model as it applies to a given data set. Such an approach is

advantageous because the relative effects of multiple variables can be tested at the same

time while also accounting for the effects of measurement error (Byrne, 1998). This is in

97
contrast to multiple regression techniques, which are limited to examining portions of a

model one at a time. Because of these advantages, it has also been argued that SEM is a

more effective means of testing cause and effect relationships between variables.

However, a drawback of using SEM is that it requires rather large sample sizes to test for

these effects. This disadvantage aside, the use of SEM would help to clarify the accuracy

of such models as Penner et al.’s (1997) model of OCB. Applying such a technique

would provide a more comprehensive perspective of the antecedents of OCB, and would

showcase the relative effect of each antecedent in combination with an entire set of

predictors. Based on these obvious advantages, it is recommended that future research on

leadership and OCB utilize more SEM techniques so that more accurate inferences can be

made regarding the leader influence process and citizenship performance.

Conclusions

This study provides empirical evidence that employee motives play a significant

role in the relationships between two specific leadership variables (transformational

leadership and LMX-quality) and various dimensions of OCB. In general, consistent

support was found for subordinate Organizational Concern as a significant mediator in

these relationships across self- and supervisor-reports of OCB. However, the prediction

that subordinate Prosocial Values moderated the relationship between LMX-quality and

subordinate Altruism was not supported. These findings are significant in the context of

current OCB research, as they provide insight regarding the nature of certain antecedent-

OCB relationships. From a practical standpoint, they also highlight the need to consider

employee motives as key determinants of employee citizenship performance.

98
 References

Allen, K. (1982). Americans volunteer: 1981. Voluntary Action Leadership, 21-23.

Allen, T.D. (1999). Mentoring others: Mentor dispositions and desired protégé
characteristics. Paper presented at the 14th Annual Meeting of Society of
Industrial and Organizational Psychology, Atlanta, GA, April.

Aroian, L. A. (1944/1947). The probability function of the product of two normally

distributed variables. Annals of Mathematical Statistics, 18, 265-271.

Avolio, B.J., & Bass, B.M. (1988). Transformational leadership, charisma, and beyond.
In J.G Hunt, B.R. Baliga, H.P. Dachler & C.A. Schrisheim (Eds.), Emerging
leadership vistas. Lexington, MA: Lexington Books.

Baron, R.M., & Kenny, D.A. (1986). The moderator-mediator variable distinction in
social psychological research: conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51(6), 1173-1182.

Bass, B.M. (1985). Leadership and performance beyond expectations. New York: Free
Press. Bass, B.M. (1988). The inspirational process of leadership. Journal of
Management Development, 7, 21-31.

Bass, B.M (1990). Handbook of leadership: A survey of theory and research. New York:
Free Press.

Bass, B.M. (1997). Does the transactional-transformational paradigm transcend

organizational and national boundaries? American Psychologist, 52, 130-139.

Bass, B.M., & Avolio, B.J. (1990). Multifactor leadership questionnaire. Palo Alto, CA:
Consulting Psychologist Press.

Bass, B.M., & Avolio, B.G. (1994). Improving organizational effectiveness through
transformational leadership. Thousand Oaks, CA: Sage Publications.

Bass, B.M., Avolio, B.J., & Goodheim, L. (1987). Biography and the assessment of
transformational leadership at the world class level. Journal of Management, 13, 7
– 19.

99
Bauer, T.A., & Green, S. (1996). Development of leader-member exchange: A
longitudinal test. Academy of Management Journal, 39(6), 1538-1567.

Blau, P.M. (1964). Exchange and power in social life. New York: Wiley.

Bolino, M.C. (1999). Citizenship and impression management: Good soldiers or good
actors? Academy of Management Review, 24(1), 82-98.

Borman, W.C., & Motowidlo, S.J. (1993). Expanding the criterion domain to include
elements of contextual performance. In N. Schmitt & W.C. Borman (Eds.),
Personnel Selection in Organizations. San Francisco: Jossey-Bass.

Borman, W.C., & Motowidlo, S.J. (1997). Task performance and contextual
performance: the meaning for personnel selection research. Human Performance,
10(2), 99-109.

Borman, W.C., & Penner, L.A. (2001). Citizenship performance: Its nature, antecedents,
and motives. In B.W. Roberts, & R. Hogan (Eds.), Personality Psychology in the
Workplace (pp. 45-61), Washington, DC: American Psychological Association.

Borman, W.C., Penner, L.A., Allen, T.D., & Motowidlo, S.J. (2001). Personality
predictors of citizenship performance. International Journal of Selection and
Assessment, 9(1-2), 52-69.

Burns, J.M. (1978). Leadership. New York: Harper & Row.

Bycio, P., Hackett, R.D., & Allen, J.S. (1995). Further assessment of Bass’s (1985)
conceptualization of transactional and transformational leadership. Journal of
Applied Psychology, 80(4), 468-478.

Cashman, J., Dansereau, F., Graen, G., and Haga, W.J. (1976). Organizational
understructure and leadership: A longitudinal investigation of the managerial role-
making process. Organizational Behavior and Human Performance, 15, 278-296.

Clary, E.G., & Orenstein, L. (1991). The amount and effectiveness of help: The
relationship of motives and abilities to helping behavior. Personality and Social
Psychology Bulletin, 17, 58-64.

Clary, E.G., & Snyder, M. (1991). A functional analysis of altruism and prosocial
behavior: The case of volunteerism. In M.S. Clark (Ed.), Review of Personality
and Social Psychology: Vol. 12. Prosocial Behavior (pp.119-148), Knobbier,
CA: Sage.

100
Connell, P., & Penner, L.A. (2004). The antecedents of OCB: Motives as mediators.
Paper presented at the meeting of the Society for Industrial and Organizational
Psychology, Chicago.

Costa, P.T., & McCrae, R.R. (1992). NEO PI-R professional manual. Odessa, FL:
Psychological Assessment Resources.

Dansereau, F., Graen, G., & Haga, W.J., (1975). A vertical dyad linkage approach to
leadership within formal organizations: A longitudinal investigation of the role-
making process. Organizational Behavior and Human Performance, 13, 46-78.

Dansereau, F., Yammarino, F.G., & Markham, S.E. (1995). Leadership: The multiple
level approaches. Leadership Quarterly, 6, 97-109.

Deluga, R.J. (1994). Supervisor trust building, leader-member exchange and

organizational citizenship behavior. Journal of Occupational and Organizational
Pyschology, 67, 315-3126.

Deluga, R.J. (1998). Leader-member exchange quality and effectiveness ratings. Group
and Organization Management, 23 (2), 189-216.

Dienesch, R.M., & Liden, R.C. (1986). Leader-member exchange model of leadership: A
critique and further development. Academy of Management Review, 11, 618-634.

Facteau, J.D., Allen, T.D., Facteau, C.L., Bordas, R.M., & Tears, R.S. (2000). Structured
interviewing for OCBs: Construct validity, faking, and the ethics of question type.
Paper presented at the 15th Annual Meeting of the Society for Industrial and
Organizational Psychology, New Orleans, LA. April.

Finkelstein , M.A., & Penner, L.A., (2004). Predicting organizational citizenship

behavior: Integrating the functional and role identity approaches. Social Behavior
and Personality, 32(4), 383-398.

Foa, U.G., & Foa, E.B. (1980). Resource theory: Interpersonal behavior as exchange. In
K.J. Gergen, M.S. Greenberg, R.H. Willis (Eds.). Social Exchange: Advances in
Theory and Research, New York: Plenum Press.

Forde, D.S.(2000). Correlates of organizational citizenship behavior motives. Honors

thesis, University of South Florida, Tampa, Florida.

Funder, D.C. (1995). Two cheers for the big five! Psychological Inquiry, 5, 125-127.

Funderberg, S.A., & Levy, P.E. (1997). The influence of individual and contextual
variables on 360-degree feedback system attitudes. Group and Organization
Management, 22, 210-235.

101
George, J.M., & Brief, A.P. (1992). Feeling good-doing good: A conceptual analysis of
the mood at work-organizational spontaneity relationship. Psychological Bulletin,
112, 310-329.

Gernster, C.R., & Day, D.V. (1997). Meta-analytic review of leader-member exchange
theory: Correlates and construct issues. Journal of Applied Psychology, 6, 827-
844.

Gouldner, A.W. (1960). The norm of reciprocity. American Sociological Review, 25,
165-167.

Graen, G.B. (1976). Role-making processes within complex organizations. In M.D.

Dunnette (Ed.), Handbook of Industrial and Organizational Psychology (pp.
1201-1245). Chicago, IL: Rand McNally.

Graen, G., Novak, M.A., & Sommerkamp, P. (1982). The effects of leader-member
exchange and job design on productivity and job satisfaction: testing a dual
attachment model. Organizational Behavior and Human Performance, 30, 109-
131.

Graen, G.B., & Scandura, T.A. (1987). Toward a psychology of dyadic organizing.
Research in Organizational Behavior, 9, 175-208.

Graen, G.B., & Uhl-Bien, M. (1995). Relationship–based approach to leadership:

Development of leader-member exchange (LMX) theory of leadership over 25
years: Applying a multi-level multi-domain perspective. Leadership Quarterly, 6,
219-247.

Hanson, M.A. (1991). Volunteer motivation. Unpublished manuscript, University of

Minnesota, Minneapolis.

Hater, J.J., & Bass, B.M. (1988). Superior’s evaluations and subordinates’ perceptions of
transformational and transactional leadership. Journal of Applied Psychology, 73,
695-702.

Hense, R.L. (2000). The Big Five and contextual performance: Expanding person-
environment fit theory. Unpublished manuscript. University of South Florida,
Tampa, Florida.

Katz, D. (1964). The motivational basis of organizational behavior. Behavioral Science,

9, 131-133.

Katz, D., & Kahn, R.L. (1966). The social psychology of organizations. New York:
Wiley.

102
Klein, K.J., Dansereau, F., & Hall, R.J., (1994). Levels issues in theory development,
data collection, and analysis. Academy of Management Review, 19, 195-229.

Konovsky, M.A., & Organ, D.W. (1996). Dispositional and contextual determinants
of organizational citizenship behavior. Journal of Organizational Behavior, 17(3),
253-266.

Koys, D.J. (2001). The effects of employee satisfaction, organizational citizenship

behavior, and turnover on organizational effectiveness: A unit-level, longitudinal
study. Personnel Psychology, 54(1), 101-114.

Levinson, H. (1965). Reciprocation: The relationship between man and the organization.
Administrative Science Quarterly, 9(4), 370-390.

Liden, R.C. & Graen, G. (1980). Generalizability of the vertical dyad linkage model of
leadership. Academy of Management Journal, 23, 451-465.

Liden, R.C., Wayne, S.J., & Stilwell, D. (1993). A longitudinal study on the early
development of leader-member exchanges. Journal of Applied Psychology, 78,
662-674.

Lowe, K.B., Kroeck, K.G., & Sivasubramaniam, N. (1996). Effectiveness correlates of

transformational and transactional leadership: A meta-analytic review of the MLQ
literature. Leadership Quarterly, 7, 385-425.

Mackenzie, S.B., Podsakoff, P.M., & Fetter, R. (1993). The impact of organizational
citizenship behavior on evaluations of salesperson performance. Journal of
Marketing, 57(1), 70-80.

MacKinnon, D.P., Lockwood C.M., Hoffman, J.M., West, S.G., & Sheets, V. (2002). A
comparison of methods to test mediation and other intervening variable effects.
Psychological Methods, 7(1), 83 –104.

McManus, M.A., & Kelly, M.L. (1999). Personality measures and biodata: Evidence
regarding their incremental predictive value in the life insurance industry.
Personnel Psychology, 52, 137-148.

Moorman, R.H. (1991). Relationship between organizational justice and organizational

citizenship behaviors: Do fairness perceptions influence employee citizenship?
Journal of Applied Psychology, 91, 845-855.

Moorman, R.H., & Blakely, G.L., (1995). Individualism-collectivism as an individual

difference predictor of organizational citizenship behavior. Journal of
Organizational Behavior, 16, 127-142.

103
Moorman, R.H., Niehoff, B.P., & Organ, D.W. (1993). Treating employees fairly and
organizational citizenship behavior: Sorting the effects of job satisfaction,
organizational commitment, and procedural justice. Employee Responsibilities
and Rights Journal, 6, 209-225.

Motowidlo, S.J., & Van Scotter, J.R. (1994). Evidence that task performance should be
distinguished from contextual performance. Journal of Applied Psychology, 79,
475-80.

Neuman, G.A. & Kickul, J.R. (1998). Organizational citizenship behaviors: Achievement
orientation and personality. Journal of Business Psychology, 13, 263-279.

Nystrom, P.C. (1990). Vertical exchanges and organizational commitments of American

business managers. Group and Organizational Studies, 15, 296-312.

Omoto, A., & Snyder, M. (1995). Sustained helping without obligation: Motivation,
longevity of service, and perceived attitude change among AIDS volunteers.
Journal of Personality and Social Psychology, 68, 671-687.

Omoto, A.M., Snyder, M., & Berghuis, J.P. (1993). The psychology of volunteerism: A
conceptual analysis and a program of action research. In J.B. Pryor & G. Reeder
(Eds.), The Social Psychology of HIV Infection (pp. 333-356). Hillsdale, NJ:
Lawrence Erlbaum Associates, Inc.

Organ, D.W. (1988). Organizational citizenship behavior: The good soldier syndrome.
Lexington, MA: Lexington.

Organ, D.W. (1997). Organizational citizenship behavior: It’s construct clean-up time.
Human Performance, 10, 85-97.

Organ, D.W., & Konovsky, M. (1989). Cognitive versus affective determinants of

organizational citizenship behavior. Journal of Applied Psychology, 74, 157-164.

Organ, D.W., & Ryan, K. (1995). A meta-analytic review of attitudinal and dispositional
predictors of organizational citizenship behavior. Personnel Psychology, 48, 775-
802.

Orr, J.M., Sackett, P.R., & Mercer, M. (1989). The role of prescribed and nonprescribed
behaviors in estimating the dollar value of performance. Journal of Applied
Psychology, 74, 34-40.

Penner, L.A., & Finkelstein, M.A. (1998). Dispositional and structural determinants of
volunteerism. Journal of Personality and Social Psychology, 74(2), 525-537.

104
Penner, L.A., Midili, A.R., & Kegelmeyer, J. (1997). Beyond job attitudes: A personality
and social psychology perspective on the causes of organizational citizenship
behavior. Human Performance, 10(2), 111-131.

Podsakoff, P.M., & MacKenzie, S.B. (1989). A second generation measure of

organizational citizenship behavior. Indiana University.

Podsakoff, P.M., & MacKenzie, S.B. (1997). Impact of organizational citizenship

behavior: A review and suggestions for future research. Human Performance, 10,
133-151.

Podsakoff, P.M., MacKenzie, S.B., Paine, J.B., & Bachrach, D.G. (2001). Organizational
citizenship behaviors: A critical review of the theoretical and empirical literature
and suggestions for future research. Journal of Management, 26(3), 513-563.

Podsakoff, P.M., MacKenzie, S.B., & Bommer, W.H. (1996). Transformational leader
behaviors and substitutes for leadership as determinants of employee satisfaction,
commitment, trust, and organizational citizenship behaviors. Journal of
Management, 22(2), 259-298.

Podsakoff, P.M., MacKenzie, S.B., Morrman, R.H., & Fetter, R. (1990).

Transformational leader behaviors and their effects on followers’ trust in leader,
satisfaction, and organizational citizenship behaviors. Leadership Quarterly, 1(2),
107-142.

Rioux, S.M. (1998). Assessing individual motives for engaging in organizational

citizenship behaviors: A functional approach. Unpublished Dissertation,
University of South Florida, Tampa, Florida.

Rioux, S.M., & Penner, L.A. (2001). The causes of organizational citizenship behavior: A
motivational analysis. Journal of Applied Psychology, 86(6), 1306-1314.

Scandura, T.A., & Graen, G.B. (1984). Moderating effects of initial leader-member
exchange status on the effects of leadership intervention. Journal of Applied
Psychology, 69, 428-436.

Schriesheim, C.A., Castro, S. L., & Cogliser, C.C. (1999). Leader-member exchange
(LMX) research: A comprehensive review of theory, measurement, and data-
analytic practices. Leadership Quarterly, 10(1), 63-113.

Settoon, R.P, Bennett, N., & Liden, R.C. (1996). Social exchange in organizations:
Perceived organizational support, leader-member exchange, and employee
reciprocity. Journal of Applied Psychology, 81, 219-227.

105
Skarlicki, D.P. & Latham, G.P. (1996). Increasing citizenship behavior within a labor
union: A test of organizational justice theory. Journal of Applied Psychology, 81,
161-169.

Smith, C.A., Organ, D.W., & Near, J.P. (1983). Organizational citizenship behavior: Its
nature and antecedents. Journal of Applied Psychology, 68, 653-663.

Stevens, J.P. (2002). Applied multivariate statistics for the social sciences. Mahwah, NJ:
Lawrence Erlbaum Associates, Publishers.

Tansky, J.W. (1993). Justice and organizational citizenship behavior: What is the
relationship? Employee Responsibilities and Rights Journal, 6, 195-207.

Tillman, P.T.E. (1998). In search of moderators of the relationship between predictors of

organizational citizenship behavior and organizational citizenship behavior: The
case of motives. Unpublished masters thesis. University of South Florida, Tampa,
Florida.

Van Dyne, L., Cummings, L.L., & Parks, J.M. (1995). Extra-role behaviors: In pursuit of
construct and definitional clarity (A bridge over muddied waters). In L.L.
Cummings & B.M. Staw (eds.), Research in Organizational Behavior (Vol. 17):
215-285. Greenwich, CT: JAI Press.

Van Dyne, L.V., Vandewalle, D., Kostova, T., Latham, M.E., & Cummings, L.L., (2000).
Collectivism, propensity to trust, and self-esteem as predictors of organizational
citizenship in a non-work setting. Journal of Organizational Behavior, 21, 3-23.

Van Scotter, J.R., & Motowidlo, S.J., (1996). Interpersonal facilitation and job dedication
as separate facets of contextual performance. Journal of Applied Psychology, 81,
525-531.

Waldman, D.A., Bass, M.A., & Yammarino, F.J. (1990). Adding to contingent reward-
behavior: The augmenting effect of charismatic leadership. Group and
Organizational Studies, 15, 381-394.

Watson, D., & Clark, L.A. (1992). Negative affectivity: The disposition to experience
aversive emotional states. Psychological Bulletin, 96, 465-490.

Wayne, S.J., & Green, S.A. (1993). The effects of leader-member exchange on employee
citizenship and impression management behavior. Human Relations, 46, 1431-
1440.

Werner, J.M. (1994). Dimensions that make a difference: Examining the impact of in-role
and extra-role behaviors on supervisory ratings. Journal of Applied Psychology,
79(1), 98-107.

106
Williams, L.J., & Anderson, S.E. (1991). Job satisfaction and organizational commitment
as predictors of organizational citizenship and in-role behaviors. Journal of
Management, 91, 601-617.

Yukl, G. (1998). Leadership in organizations (4th ed.). Englewood Cliffs, N.J.: Prentice
Hall.

Yukl, G. (1999). An evaluation of conceptual weaknesses in transformational and

charismatic leadership theories. Leadership Quarterly, 10(2), 285-305.

107
Appendices

108
 Appendix A: Transformational Leadership Inventory (TLI)

Subordinate Version

Below is a set of statements that may or may not describe your supervisor’s behavior at
work. Using the scale below, please indicate the extent to which you agree (or disagree)
that each statement is descriptive of your SUPERVISOR.

1 2 3 4 5 6 7
Strongly Disagree Somewhat Neutral Somewhat Agree Strongly
Disagree Disagree Agree Agree

1. Is always seeking new opportunities for the unit/department/organization.

2. Paints an interesting picture of the future for our group.

3. Has a clear understanding of where we are going.

4. Inspires others with his/her plans for the future.

5. Is able to get others committed to his/her dream of the future.

6. Leads by “doing” rather than simply “telling.”

7. Provides a good model to follow.

8. Leads by example.

9. Fosters collaboration among work groups.

10. Encourages employees to be “team players.”

11. Gets the group to work together for the same goal.

12. Develops a team attitude and spirit among his/her employees.

13. Shows that he/she expects a lot from us.

14. Insists on only the best performance.

15. Will not settle for second best.

16. Acts without considering my feelings.

109
 Appendix A (Continued)

17. Shows respect for my personal feelings.

18. Behaves in a manner that is thoughtful of my personal needs.

19. Treats me without considering my personal feelings.

20. Has provided me with new ways of looking at things which used to puzzle me.

21. Has ideas that have forced me to rethink some of my own ideas that I have never

questioned before.

22. Has stimulated me to think about old problems in new ways.

110
 Appendix B: Transformational leadership Inventory (TLI)

Supervisor Version

Below is a set of statements which may or may not describe your behavior at work.
Using the scale below, please indicate the extent to which you agree (or disagree) that
each statement is descriptive of YOUR behavior as a leader.

1 2 3 4 5 6 7
Strongly Disagree Somewhat Neutral Somewhat Agree Strongly
Disagree Disagree Agree Agree

1. I am always seeking new opportunities for the unit/department/organization.

2. I paint an interesting picture of the future for our group.

3. I have a clear understanding of where we are going.

4. I inspire others with my plans for the future.

5. I am able to get others committed to my dream of the future.

6. I lead by “doing” rather than simply “telling.”

7. I provide a good model to follow.

8. I lead by example.

9. I foster collaboration among work groups.

10. I encourage employees to be “team players.”

11. I get the group to work together for the same goal.

12. I develop a team attitude and spirit among my employees.

13. I show that I expect a lot from my employees.

14. I insist on only the best performance.

15. I will not settle for second best.

16. I act without considering my employees’ feelings.

111
 Appendix B (Continued)

17. I show respect for my employees’ personal feelings.

18. I behave in a manner that is thoughtful of my employees’ personal needs.

19. I treat my employees without considering their personal feelings.

20. I have provided my employees’ with new ways of looking at things which used to

puzzle them.

21. I have ideas that have forced my employees’ to rethink some of their own ideas

that they have never questioned before.

22. I have stimulated my employees to think about old problems in new ways.

112
 Appendix C: LMX7

Subordinate Version

Using the scales presented below, please answer each of the following statements.

1. Do you know where you stand with your supervisor…do you usually know how

satisfied your leader is with what you do?

1 2 3 4 5
Rarely Occasionally Sometimes Fairly Often Very Often

2. How well does your supervisor understand your job problems and needs?

1 2 3 4 5
Not a Bit A Little A Fair Amount Quite a Bit A Great Deal

3. How well does your supervisor recognize your potential?

1 2 3 4 5
Not at All A Little Moderately Mostly Fully

4. Regardless of how much formal authority he/she has built into his/her position,

what are the chances that your supervisor would use his/her power to help you

solve problems at work?

1 2 3 4 5
None Small Moderate High Very High

5. Again, regardless of the amount of formal authority your supervisor has, what are

the chances that he/she would “bail you out,” at his/her expense?

1 2 3 4 5
None Small Moderate High Very High

113
 Appendix C (Continued)

6. I have enough confidence in my supervisor that I would defend and justify his/her

decision if he/she were not present to do so?

1 2 3 4 5
Strongly Disagree Neutral Agree Strongly
Disagree Agree

7. How would you characterize your working relationship with your supervisor?

1 2 3 4 5
Extremely Worse Than Average Better Than Extremely
Ineffective Average Average Effective

114
 Appendix D: LMX7

Supervisor Version

Using the scales presented below, please answer each of the following statements.

1. Does your subordinate know where they stand with you…do they usually know

how satisfied you are with what they do?

1 2 3 4 5
Rarely Occasionally Sometimes Fairly Often Very Often

2. How well do you understand your subordinate’s job problems and needs?

1 2 3 4 5
Not a Bit A Little A Fair Amount Quite a Bit A Great Deal

3. How well do you recognize their potential?

1 2 3 4 5
Not at All A Little Moderately Mostly Fully

4. Regardless of how much formal authority you have built into your position, what

are the chances that you would use your power to help solve your subordinate’s

problems at work?

1 2 3 4 5
None Small Moderate High Very High

5. Again, regardless of the amount of formal authority you have, what are the

chances that you would “bail your subordinate out,” at your own expense?

1 2 3 4 5
None Small Moderate High Very High

115
 Appendix D (Continued)

6. My subordinate has enough confidence in me that they would defend and justify

my decision if I was not present to do so?

1 2 3 4 5
Strongly Disagree Neutral Agree Strongly
Disagree Agree

7. How would you characterize your working relationship with your subordinate?

1 2 3 4 5
Extremely Worse Than Average Better Than Extremely
Ineffective Average Average Effective

116
 Appendix E: OCB Measure

Subordinate Version

Below is a set of statements which may or may not describe YOUR behavior at work.
Using the scale below, please indicate the extent to which you agree (or disagree) with
each statement.

1 2 3 4 5 6 7
Strongly Disagree Somewhat Neutral Somewhat Agree Strongly
Disagree Disagree Agree Agree

1. My attendance at work is above the norm.

2. I help orient new people even though it is not required.

3. I consume a lot of time complaining about trivial matters.

4. I am one of my supervisor’s most conscientious employees.

5. I help others who have heavy work loads.

6. I am the classic “squeaky wheel” that always needs greasing.

7. I attend meetings that are not mandatory, but are considered important.

8. I keep abreast of changes in the organization.

9. I tend to make “mountains out of molehills.”

10. I obey company rules and regulations even when no one is watching.

11. I try to avoid creating problems for coworkers.

12. I am mindful of how my behavior affects other people’s jobs.

13. I always focus on what’s wrong, rather than the positive side.

14. I read and keep up with organizational announcements, memos, and so on.

15. I willingly help others who have work related problems.

16. I attend functions that are not required, but help the company image.

117
 Appendix E (Continued)

17. I do not abuse the rights of others.

18. I consider the impact of my actions on coworkers.

19. I help others who have been absent.

20. I do not take extra breaks.

21. I am always ready to lend a helping hand to those around me.

22. I believe in giving an honest day’s work for an honest day’s pay.

24. I always find fault with what the organization is doing.

118
 Appendix F: OCB Measure

Supervisor Version

Below is a set of statements which may or may not describe your subordinate’s behavior
at work. Using the scale below, please indicate the extent to which you agree that each
statement is descriptive of your SUBORDINATE.

1 2 3 4 5 6 7
Strongly Disagree Somewhat Neutral Somewhat Agree Strongly
Disagree Disagree Agree Agree

1. Attendance at work is above the norm.

2. Helps orient new people even though it is not required.

3. Consumes a lot of time complaining about trivial matters.

4. Is one of my most conscientious employees.

5. Helps others who have heavy work loads.

6. Is the classic “squeaky wheel” that always needs greasing.

7. Attends meetings that are not mandatory, but are considered important.

8. Keeps abreast of changes in the organization.

9. Tends to make “mountains out of molehills.”

10. Obeys company rules and regulations even when no one is watching.

11. Tries to avoid creating problems for coworkers.

12. Is mindful of how his/her behavior affects other people’s jobs.

13. Always focuses on what’s wrong, rather than the positive side.

14. Reads and keeps up with organizational announcements, memos, and so on.

15. Willingly help others who have work related problems.

16. Takes steps to try to prevent problems with other workers.

119
 Appendix F: (Continued)

17. Attends functions that are not required, but help the company image.

18. Does not abuse the rights of others.

19. Considers the impact of his/her actions on co-workers.

20. Helps others who have been absent.

21. Does not take extra breaks.

22. Is always ready to lend a helping hand to those around him/her.

23. Believes in giving an honest day’s work for an honest day’s pay.

24. Always finds fault with what the organization is doing.

120
 Appendix G: Citizenship Motives Scale (CMS)

During the course of the workday people often engage in prosocial or helpful behaviors.
These behaviors are not a required part of the job and they are not formally rewarded
(e.g., more money). Yet these behaviors are very important and help the organization
function smoothly. Examples of such behavior include:

- helping coworkers with a heavy workload - not taking long lunches or breaks

- touching base with others before initiating action - keeping informed of changes in the
organization

- attending functions that aren’t mandatory - not complaining over small things

People are motivated to engage in these kinds of behavior by many different things.
Below is a list of motives that may influence people to engage in these behaviors. For
each motive listed, please indicate HOW IMPORTANT that motive is for YOU to
engage in these kinds of behaviors at work. Please see the scale below and darken in the
number corresponding to your response.

Use the following scale to indicate your answer:

1 2 3 4 5
Not at all Slightly Important Very important Extremely
important important important

1. Because I have a genuine interest in my work.

2. Because I feel it is important to help those in need.

3. To make myself more marketable to other organizations.

4. So that others will see me as helpful.

5. Because I want to be fully involved in the company.

6. To get a good raise.

7. In order to keep my job.

8. Because I am concerned about other people’s feelings.

9. Because I want to be a well-informed employee.

121
 Appendix G (Continued)

10. To have fun with my co-workers

11. To get a promotion.

12. So that others will like me.

13. Because I care what happens to the company.

14. Because I like interacting with my co-workers.

15. So that others will think of me as supportive.

16. Because the organization values my work.

17. Because I want to help my co-workers in any way I can.

18. Because I feel pride in the organization.

19. Because I can put myself in other people’s shoes.

20. Because I want to understand how the organization works.

21. Because I believe in being courteous to others.

22. So that others will think highly of me.

23. To keep up with the latest developments in the organization.

24. Because it is easy for me to be helpful.

25. So that I don’t get laid off.

26. Because I am committed to the company.

27. To get to know my co-workers better.

28. Because the organization treats me fairly.

29. To be friendly with others.

30. So that others will think I pull my weight.

122