Leadership Styles and Conflict Management Styles: An Exploratory Study

Submitted to Regent University

School of Leadership Studies

In partial fulfillment of the requirements

for the degree of

Doctor of Philosophy in Organizational Leadership

Angelia Denise Stanley

August 2004
 UMI Number: 3140603

Copyright 2004 by
Stanley, Angelia Denise
All rights reserved.

________________________________________________________

UMI Microform 3140603

Copyright 2004 ProQuest Information and Learning Company.

All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
____________________________________________________________

ProQuest Information and Learning Company

300 North Zeeb Road
PO Box 1346
Ann Arbor, MI 48106-1346
ii
 iii

Abstract

This dissertation has explored the relationship between leadership styles and

conflict management styles. A review of the literature validates distinctions

among leadership styles that can be measured by the Multifactor Leadership

Questionnaire (MLQ). Furthermore, the literature unequivocally reveals that

different conflict management styles exist and can be assessed using the

Thomas-Kilmann Conflict Mode Instrument (TKI). The literature implies, but does

not explicitly state, that a relationship may exist between leadership styles and

conflict management styles. Data were collected from 99 leaders who completed

the MLQ and the TKI. Each leader received a score on Transactional Leadership,

a score on Transformational Leadership, and a score on Laissez-faire

Leadership. This study used the scores on the leadership styles as measures of

the “degree” (or intensity) of Transactional Leadership, the “degree” of

Transformational Leadership, and the “degree” of Laissez-faire Leadership that

each leader exhibits. This study used TKI scores as measures: of the extent to

which a leader is Competing, the extent to which a leader is Collaborating, the

extent to which a leader is Compromising, the extent to which a leader is

Avoiding, and the extent to which a leader is Accommodating. Canonical

correlation analysis shows that no statistically significant relationship exists

between the set of leadership styles (as assessed by the MLQ) and the set of

conflict management styles (as assessed by the TKI). In all tests the data fail to

reject the null hypothesis that the set of MLQ scores is not related to the set of

TKI scores. Although none of the bivariate correlations between the three MLQ
 iv

scores and the five TKI scores are statistically significant, some correlations

among the MLQ variables are statistically significant. For example, Laissez-faire

Leadership and Transformational Leadership are negatively correlated and the

correlation is significantly different from zero. Generally speaking, this means that

those with high (low) scores on Laissez-faire Leadership have low (high) scores

on Transformational Leadership. Future research on leadership styles will be

needed to explore the implications of the current study and to corroborate any

implied relationship that may exist between leadership styles and conflict

management styles.
 v

Dedication

This dissertation is dedicated to my mom and dad, Colleen P. Stanley and

Kenneth D. Stanley, for their love and support.

 vi

Acknowledgements

Thanks to Dr. Barry A. Love for his help with my research method and

statistical analysis; to Dr. Felicia Mitchell for her research assistance, editing, and

for listening to me, believing in me, and not giving up on me; to Dr. Jim Duchamp

for retrieving numerous articles and books for me; to Patty Greany, Jane

Caldwell, and David Baber for assistance in obtaining books and articles for my

research via interlibrary loan; to Lisa Eskridge for scanning documents and

encouraging me; to Dr. Chris Fielitz for helping me stay focused; and to Ann

Fleenor for providing a listening ear and encouragement.

Recognition also is due to my colleague Prof. Rick Neil, who provided

moral support for the multi-tasking necessary to complete a Ph.D. while

employed at Emory & Henry College. Rick passed away before we could share

the joy of completing our degrees this fall.

Appreciation is expressed to the Academic Council of Emory & Henry

College for a Mellon Summer Grant to support my research.

Appreciation is expressed to my advisor, Dr. Bruce Winston, for his

patience and assistance. I also appreciate the helpful suggestions of my other

committee members, Dr. Mihai Bocarnea and Dr. Gail Longbotham.

 vii

Table of Contents

Leadership Styles and Conflict Management Styles: An Exploratory Study ...................... i

Dedication ........................................................................................................................... v
Acknowledgements............................................................................................................ vi
Table of Contents.............................................................................................................. vii
List of Tables and Figures.................................................................................................. ix
Chapter I – Introduction...................................................................................................... 1
Theory and Variables.................................................................................................. 3
Relationships of the Variables .................................................................................... 5
Hypothesis................................................................................................................... 5
Method and Analysis .................................................................................................. 7
Scope........................................................................................................................... 7
Limitations of the Study.............................................................................................. 8
Chapter 2 - Literature Review……………………………………………………………..9
Significance of Conflict and Conflict Management ................................................... 9
Personality Type and Conflict Management Style ................................................... 14
Situational Leadership and Conflict Management Style .......................................... 17
Theory and Variables................................................................................................ 17
Leadership................................................................................................................. 18
Transactional Leadership .......................................................................................... 20
Transformational Leadership .................................................................................... 22
Laissez-faire Leadership ........................................................................................... 24
Conflict Management Styles..................................................................................... 24
Summary of the Literature ........................................................................................ 28
Chapter 3 – Method .......................................................................................................... 29
Hypothesis................................................................................................................. 29
Data Collection ......................................................................................................... 31
Multifactor Leadership Questionnaire ...................................................................... 32
Thomas-Kilmann Conflict Mode Instrument............................................................ 34
Sample Population & Frame..................................................................................... 35
Analysis..................................................................................................................... 35
Justification and Support for Use of Canonical Correlation..................................... 38
Summary of Research Design................................................................................... 39
Chapter 4 – Results ........................................................................................................... 40
Data Collection ......................................................................................................... 40
Characteristics of Leaders in the Sample.................................................................. 40
Data Collection Procedures....................................................................................... 41
Data Coding and Analysis Procedures...................................................................... 42
Description of Sample Data ...................................................................................... 43
Canonical Correlation Analysis ................................................................................ 50
Statististical Assumptions of the Analysis................................................................ 51
Deriving the Canonical Functions and Their Significance ....................................... 54
Interpreting the Canonical Variates .......................................................................... 56
Summary of Statistical Analysis............................................................................... 61
 viii

Chapter 5 – Discussion ..................................................................................................... 62

Exploring Leadership Styles and Conflict Management Styles ............................... 62
Observations/Limitations.......................................................................................... 65
Implications of Findings ........................................................................................... 66
Recommendations for Future Research .................................................................... 68
References......................................................................................................................... 70
Appendix........................................................................................................................... 76
Demographic Questionnaire ..................................................................................... 76
Letter to Emory & Henry Alums .............................................................................. 77
Letter to Leaders Participating in this Study............................................................. 78
Permission to Cite Conflict Handling Modes Figure (Figure 1)............................... 79
MLQ Duplication Permission ................................................................................... 80
Permission to Include Sample Items from MLQ ...................................................... 81
Sample Items from MLQ .......................................................................................... 82
Permission to Include Sample Items from TKI ........................................................ 83
Sample Items from TKI ............................................................................................ 85
 ix

List of Tables and Figures

Figure 1. Five Conflict-Handling Modes.............................................................. 25

Figure 2. Scatterplots of TKI Scores Versus MLQ Scores................................. 86
Figure 3. Scatterplots of MLQ1 Variate with MLQ Scores ................................. 91
Figure 4. Scatterplots of MLQ1 Variate with TKI Scores .................................... 92
Figure 5. Scatterplots of TKI1 Variate with TKI Scores....................................... 94
Figure 6. Scatterplots of TKI1 Variate with MLQ Scores ................................... 96
Figure 7. Scatterplots of TKI and MLQ Variates ................................................ 97

Table 1 ................................................................................................................. 6
Table 2 ............................................................................................................... 45
Table 3 ............................................................................................................... 46
Table 4 ............................................................................................................... 47
Table 5 ............................................................................................................... 47
Table 6 ............................................................................................................... 53
Table 7 ............................................................................................................... 54
Table 8 ............................................................................................................... 54
Table 9 ............................................................................................................... 55
Table 10 ............................................................................................................. 57
Table 11 ............................................................................................................. 57
Table 12 ............................................................................................................. 57
Table 13 ............................................................................................................. 58
Table 14 ............................................................................................................. 59
Table 15 ............................................................................................................. 59
Table 16 ............................................................................................................. 59
Table 17 ............................................................................................................. 59
 1

Chapter I – Introduction

Leaders inevitably face conflict, whether desirable or not, within their

organizations (Darling & Walker, 2001; Thomas, 1992). Zigarelli (2002) reports

83 percent of leaders agree that “it’s their job to resolve conflict among their

employees quickly” (p. 4). Dubrin (2004) states leaders “devote about 20 percent

of their time to dealing with conflict” (p. 386) and managers “spend roughly nine

workweeks a year resolving employee personality clashes” (p. 386). Rahim

(1981) examined 195 syllabi for courses in organizational behavior in Master of

Business Administration (MBA) programs and generated a list of 65 topics based

on frequency of content analysis in these syllabi. Rahim reported that leadership

was the first most frequently mentioned topic and conflict ranked as the fifth most

frequently mentioned topic (p. 585). Thus, it is no surprise that leadership and

conflict management have received and continue to receive significant attention

in research studies.

Marion (1995) examined the relationship between personality type and

conflict management style and asserted that “insight into the way individuals

manage conflict provides a framework for greater understanding of the dynamics

of conflict and thus a useful tool for improved management” (p. 3). Marion used

the Myers-Briggs Type Indicator (MBTI) to assess personality type and the

Thomas-Kilmann Conflict Mode Instrument (TKI) to evaluate conflict

management style. Marion concluded “the findings from this study suggest that

there is indeed a relationship between individual personality type and conflict

management style” (p. 9). Marion’s findings support previous research by

 2

Kilmann and Thomas (1975) and Mills, Robey, and Smith (1985) that also found

correlations between some elements of the MBTI and the TKI.

Romero (1983) examined the relationship between situational leadership

and conflict management style. The LEAD (Leader Effectiveness and

Adaptability Description) Self instrument was developed by Hersey and

Blanchard (1969a, 1969b, 1974, 1982,1996) while the two conflict style

instruments were developed specifically for Romero’s study. Romero concluded

that “principals do not change their leadership or conflict style as situations

change” (p. 125). Holt (1986) also examined situational leadership and conflict

management style. Holt also used the LEAD-Self Instrument, but Holt used the

TKI instead of the two conflict management style instruments used by Romero.

Holt concluded “there were no statistically significant relationships between

leadership styles and conflict management techniques in the sample studied” (p.

73).

Marion’s study (1995) used the MBTI and the TKI while Romero’s (1983)

study used the LEAD-Self Instrument and two other conflict style instruments.

Holt (1986) used the LEAD-Self Instrument and the TKI. No study has specifically

investigated the relationship, if any, between the three leadership styles

(transactional, transformational, and laissez-faire) measured by leaders’

responses to the Multifactor Leadership Questionnaire (MLQ) and the five conflict

management styles (competing, collaborating, compromising, avoiding, and

accommodating) measured by leaders’ responses to the TKI. This dissertation

seeks to bridge this gap in the literature.

 3

Theory and Variables

The literature (Bass, 1985; Bass & Avolio, 1994; Burns, 1978; Hartog,

Muijen, & Koopman,1997; Yammarino, Spangler, & Bass, 1993) implies, but

does not test or expressly state, that different leadership styles (transactional,

transformational, and laissez-faire) will show different conflict management

styles. Burns describes transactional leadership purely as an exchange occurring

"when one person takes the initiative in making contact with others for the

purpose of an exchange of valued things" (p. 19). The theory of transactional

leadership implies that leaders will enter into bargained-for exchanges with their

followers. Therefore, the theory of transactional leadership implies a positive

relationship between leaders’ scores on transactional leadership and their scores

on competing. Furthermore, based on this notion of a bargained-for exchange,

the theory of transactional leadership also implies a positive relationship between

leaders’ scores on transactional leadership and their scores on compromising.

Since transactional leaders want to enter into an exchange with their followers,

transactional leaders are not likely to avoid demands by their followers nor

automatically give in and accommodate the demands of their followers. Thus, the

theory of transactional leadership implies negative relationships between leaders’

scores on transactional leadership and their scores on avoiding and leaders’

scores on transactional leadership and their scores on accommodating.

Bass (1985) advocates that transformational leaders mentor and motivate

followers, encouraging the followers to become leaders. Bass and Avolio (1994)

state:
 4

Transformational leadership is seen when leaders:

1. Stimulate interest among colleagues and followers to view their

work from new perspectives.
2. Generate awareness of the mission or vision of the team and
organization.
3. Develop colleagues and followers to higher levels of ability and
potential.
4. Motivate colleagues and followers to look beyond their own
interests toward those that will benefit the group. (p. 2)

Transformational leaders will work together with their followers, reminding the

followers of the importance of the team and the organization. Thus, the theory of

transformational leadership implies a positive relationship between leaders’

scores on Transformational Leadership and their scores on Collaborating and a

negative relationship between leaders’ scores on Transformational Leadership

and their scores on Avoiding.

Transactional and transformational leaders can be described as active

leaders, acting to prevent problems from occurring in their organizations and

acting to solve problems (Yammarino et al., 1993). Hartog et al. (1997)

distinguish between these active forms of leadership and the “extremely passive

laissez-faire leadership” (p. 21), noting that the laissez-faire leader “is inactive,

rather than reactive or proactive” (p. 21). Laissez-faire leaders “avoid decision

making and supervisory responsibility” (p. 21). Thus, the theory of laissez-faire

leadership implies a positive relationship between leaders’ scores on laissez-faire

leadership and their scores on avoiding and a negative relationship between

leaders’ scores on laissez-faire leadership and their scores on collaborating.

 5

Relationships of the Variables

This study investigated the relationships of two sets of variables,

leadership styles and conflict management styles. Leadership styles are those

measured by leaders’ responses to the Multifactor Leadership Questionnaire

(MLQ). Specifically, these are: (a) leaders’ scores on transactional leadership, (b)

leaders’ scores on transformational leadership, and (c) leaders’ scores on

laissez-faire leadership. Based on his/her responses to the MLQ, each leader

received a score from 0 to 48 for transactional leadership, a score ranging from 0

to 80 for transformational leadership, and a score from 0 to 16 for laissez-faire

leadership.

Conflict management styles are those measured by leaders’ responses to

the TKI. These are: (a) leaders’ scores on the competing style of conflict

management, (b) leaders’ scores on the collaborating style of conflict

management, (c) leaders’ scores on the compromising style of conflict

management, (d) leaders’ scores on the avoiding style of conflict management,

and (e) leaders’ scores on the accommodating style of conflict management.

Based on his or her responses to the TKI, each leader received a score ranging

from 0 to 12 for each of the five conflict management styles (competing,

collaborating, compromising, avoiding, and accommodating).

Hypothesis

Table 1 shows the expected relationships (‘+’ for positive, ‘-‘ for negative,

and ‘0’ for no relationship) as posited by the literature on transactional,

transformational, and laissez-faire leadership styles and the five conflict

 6

management styles. For example, the literature implies positive relationships

between leaders’ scores on Transactional Leadership and their scores on

Competing and Compromising. The literature also implies negative relationships

between leaders’ scores on Transactional Leadership and their scores on

Avoiding and Accommodating. The theory of transformational leadership implies

a positive relationship between leaders’ scores on Transformational Leadership

and their scores on Collaborating and a negative relationship between leaders’

scores on Transformational Leadership and their scores on Avoiding. The theory

of laissez-faire leadership implies a negative relationship between leaders’

scores on Laissez-faire Leadership and their scores on Collaborating and a

positive relationship between leaders’ scores on Laissez-faire Leadership and

their scores on Avoiding.

Table 1

Expected Relationships of the Variables

TKI MLQ Variables

Regression
Variables TR TF LF
1 Competing + 0 0
2 Collaborating 0 + -
3 Compromising + 0 0
4 Avoiding - - +
5 Accommodating - 0 0

One hypothesis, based on these expected relationships posited by the

literature, was addressed in this dissertation. This hypothesis relates to whether

or not there is a relationship between leaders’ scores on the three leadership

styles (transactional, transformational, and laissez-faire) as measured by the

 7

MLQ and leaders’ scores on the five conflict management styles (competing,

collaborating, compromising, avoiding, and accommodating) as measured by the

TKI. Specifically,

H01: The set of MLQ scores is not related to the set of TKI scores.

Method and Analysis

In order to test this hypothesis, two research instruments (the MLQ and

the TKI) were administered to leaders. The MLQ provided a score for each

leader on transactional leadership, transformational leadership, and laissez-faire

leadership. The TKI provided a score for each leader on each of the following five

conflict management styles: competing, collaborating, compromising, avoiding,

and accommodating. The data (scores from the MLQ and the TKI) was entered

into SPSS. All three of the MLQ variables (the scores on transactional

leadership, the scores on transformational leadership, and the scores on laissez-

faire leadership) were used as well as the five TKI variables (competing,

collaborating, compromising, avoiding, and accommodating). Canonical

correlation analysis was used to examine the relationship between the scores on

the leadership styles (transactional, transformational, and laissez-faire) and

scores on the conflict management styles (competing, collaborating,

compromising, avoiding, and accommodating).

Scope

This study explored the relationships between leadership styles

(measured by leaders’ responses to the MLQ) and conflict management styles

(measured by leaders’ responses to the TKI). After leaders in the sample

 8

completed both questionnaires, the questionnaires were scored; and each leader

in the sample received a score on each of three leadership styles and each of the

five conflict management styles. Specifically, from his or her responses to the

MLQ, each leader received a score ranging from 0 to 80 for transformational

leadership, a score from 0 to 48 for transactional leadership, and a score from 0

to 16 for laissez-faire leadership.

From his or her responses to the TKI, each leader received a score

ranging from 0 to 12 for each of the five conflict management styles (competing,

collaborating, compromising, avoiding, and accommodating). Based on my

review of the literature, no study has specifically investigated the relationships, if

any, between leadership styles (transactional, transformational, and laissez-

faire), measured by leaders’ responses to the MLQ, and conflict management

styles (competing, collaborating, compromising, avoiding, and accommodating),

measured by leaders’ responses to the TKI.

Limitations of the Study

There are two specific factors that may have influenced the results of this

study. First, the sample size was 99 leaders. Therefore, the results may not be

generalized to all leaders without additional research efforts. The second factor

that may influence the results of this study is the self-reporting nature of the two

research instruments used in this study. Thus, it is possible that participants

chose responses that do not accurately reflect their behavior. This limitation is

not unique to my study, but applies to all research studies that use instruments of

a self-reporting nature.
 9

Chapter 2 – Literature Review

This chapter reviews previous research related to the four areas that are

significant to this study. First, to offer a background for the current study’s focus

on conflict management, an introduction to conflict and conflict management is

provided. Second, the chapter presents a relevant overview of the research on

personality type and conflict management styles in order to set up the

connections among conflict management, personality types, and leadership

styles. Third, research related to situational leadership and conflict management

style is summarized. Finally, this chapter provides an overview of articles

pertaining to the theory and variables underlying the hypothesis in this

dissertation. Leadership styles addressed include the following: transactional

leadership, transformational leadership, and laissez-faire leadership. Conflict

management styles include these styles: competing, collaborating,

compromising, avoiding, and accommodating.

Significance of Conflict and Conflict Management

Darling and Walker (2001) assert that leaders inevitably face conflict,

whether desirable or not, within their organizations. They point to individuals’

different values and situations as the creators of conflict. Darling and Walker

define conflict as “a situation in which two or more individuals operating within a

unit appear to be incompatible” (p. 230). They advocate “when such conflict is

recognized, acknowledged and managed in a proper manner, personal and

organizational benefits will accrue” (p. 230). In their research regarding effective

conflict management, Darling and Walker emphasize that leaders should view
 10

conflict as growth opportunities for individuals as well as their organizations.

Furthermore, they assert that leaders have “a dominant behavior style that is

reflected in how that individual works, interacts, and communicates with others”

(p. 232). If Darling and Walker are correct, then a relationship may exist between

leadership styles and conflict management styles. This dissertation examined

whether or not a relationship exists between these variables.

Zigarelli (2002) notes that 83 percent of leaders agree that “it’s their job to

resolve conflict among their employees quickly” (p. 4). His research also states

that 55% of these leaders report they “strongly agree” with the following

statement: “It is a high priority for me to serve the needs of my employees” (p. 4).

An additional 33% of his respondents indicated “moderate” agreement with this

particular statement. Thus, a total of 88% of the respondents in his survey

indicated that serving employees is a “high priority” (p. 4).

This concept of leaders serving employees or having a “concern for

people” has been around a long time before Zigarelli’s study. In the 1960s,

Robert Blake and Jane Mouton (1964, 1978) developed “The Managerial Grid.”

The Grid’s horizontal axis represents a “concern for production” while the vertical

axis represents a “concern for people.” Blake and Mouton’s Grid “provides a

framework for understanding the assumptions managers make as they compete

or cooperate with one another” (1964, p. 15). Blake and Mouton (1985) assert

that “conflict can be either disruptive and destructive or creative and constructive,

depending on how it is handled” (p. 3). This dissertation explores whether or not

leadership styles are related to conflict management styles.

 11

Dubrin (2004) states that, leaders “devote about 20 percent of their time to

dealing with conflict” (p. 386). Dubrin notes that a study by Accountemps reveals

that “managers spend roughly nine workweeks a year resolving employee

personality clashes” (p. 386). In addition to resolving conflict, leaders have many

other important tasks. Bennis and Nanus (1997) conducted a study consisting of

ninety leaders (p. 18) to obtain a better understanding of leadership: “the most

studied and least understood topic of any in the social sciences” (p. 19). Prior to

the research, they proclaimed, “leadership is like the Abominable Snowman,

whose footprints are everywhere but who is nowhere to be seen” (p. 19). Thus,

they embarked on their research journey and interviewed ninety individuals,

including many CEOs from Fortune’s top-200 list (p. 23). The median age of the

leaders in their sample was 56 and the average income (not including “perks”) of

those leaders was $400,000 (p. 23). Their research method included

observations, interviews, and exploratory dialogues.

Bennis and Nanus (1997) state “organizations cannot be successful

without effective leadership” (p. 19). They assert that leaders are to effectively

and efficiently employ human and financial resources to help organizations

achieve their goals. These goals are developed from the organization’s mission

statement/purpose. Leaders not only see where the organization currently is with

regard to its mission statement and goals, but they also have a plan to help the

organization achieve its goals and its purpose. This is known as vision. Bennis

and Nanus adamantly declare, “vision animates, inspirits, transforms purpose

into action” (p. 29). They proclaim, “management of attention through vision is
 12

the creating of focus. All ninety people interviewed had an agenda, an

unparalleled concern with outcome. Leaders are the most results-oriented

individuals in the world, and results get attention” (p. 26). After approximately 200

pages of discussion, Bennis and Nanus conclude:

The absence or ineffectiveness of leadership implies the absence of

vision, a dreamless society, and this will result, at best, in the maintenance

of the status quo or, at worst, in the disintegration of our society because

of lack of purpose and cohesion. (p. 220)

Ford (1991), after working closely for thirty-one years with renowned

leader and evangelist Billy Graham, states, “vision is the very stuff of

leadership—the ability to see in a way that compels others to pay attention” (p.

99). He explains that other tasks of a leader include, but are not limited to,

strategizing, empowerment, serving, sustaining, and resolving conflict. Ford

notes that “in recent years conflict resolution has become a major field of study”

(p. 253). He describes Jesus of Nazareth as a model leader and notes that Jesus

dealt “creatively and flexibly with various conflict situations” (p. 254). Ford

emphasizes that today’s leaders must also know how to deal with conflict, and

specifically, how to handle “divisions within modern society, how to tone down

the rhetoric, how to search for acceptable compromises, how to help opposing

parties understand how the other perceives the problems, and how to open the

way for creative solutions” (p, 253).

Ford (1991), in his discussion of how leaders handle conflict, explicitly

states, “some of us tend to be constitutional belligerents because we have a

 13

deep need to dominate” (p. 257) while “others of us tend to be constitutional

pushovers because of a need to please and be accepted” (p. 257). Ford’s

research clearly affirms the existence and the significance of different conflict

resolution styles. Ford does not explain why some leaders have a need to

dominate and why others have a need to please. This dissertation builds upon

Ford’s research. If Ford is correct in his assertion that some leaders have a need

to dominate their followers while others have a need to please their followers,

then why do these differences exist? This leads one to wonder whether or not

these differences can be attributed to leadership styles. This dissertation

expands upon Ford’s research and examines the relationship, if any, between

leadership styles and conflict management styles.

Rahim (1981) examined 195 syllabi for courses in organizational behavior

in Master of Business Administration (MBA) programs and generated a list of 65

topics based on frequency of content analysis in these syllabi. He randomly

selected one thousand faculty members (from a list in the Organizational

Behavior Division of the Academy of Management) of graduate schools in the

United States that had programs in Business Administration. Rahim requested

that these faculty members send him their syllabi for their courses in

organizational behavior. He received 195 usable syllabi and analyzed them in

order to produce a list of topics covered in these courses. Rahim reported that

leadership was the most frequently mentioned topic and conflict ranked as the

fifth most frequently mentioned topic (p. 585). Thus, it is no surprise that

leadership and conflict management have received and continue to receive

 14

significant attention in research studies. According to Rahim’s research, this

dissertation examines the first and fifth most frequently mentioned topics that

graduate students will study in their organization behavior courses in MBA

programs.

Personality Type and Conflict Management Style

Marion (1995) examined the relationship between personality type and

conflict management style. She used the Myers-Briggs Type Indicator (MBTI) to

assess personality type and the Thomas-Kilmann Conflict Mode Instrument (TKI)

to evaluate conflict management style. Marion describes the MBTI as “a

personality typology instrument spawned from the psychological type theories of

Carl Jung and centering on the two functions of perception and judgment” (p. 15).

Marion explains that the MBTI “consists of four scales each with two polar

dimensions” (p. 15). The first scale is attitudes and is measured by extraversion

or introversion. The second scale is manner/style and is measured by judgment

or perception. The third scale is perceptions and is measured by sensing or

intuition. The fourth scale is judgments and is measured by thinking or feeling.

Marion asserts “insight into the way individuals manage conflict provides a

framework for greater understanding of the dynamics of conflict and thus a useful

tool for improved management” (p. 3). Marion’s assertion provides support for

this research.

In addition to examining personality type, Marion (1995) also examined

conflict management style. As previously stated, she used the TKI to assess

conflict management style. The TKI “is designed to assess an individual’s

 15

behavior in conflict situations—that is, situations in which the concerns of two

people appear to be incompatible” (Thomas & Kilmann, 2002, p. 7). The

instrument contains “30 pairs of statements describing possible behavioral

responses” (p. 1) and participants select the response that best characterizes

their behavior. Marion’s sample consisted of 161 senior administrators in

community colleges. She ran bivariate correlations for each of the personality

dimensions and each of the conflict management styles. Marion concluded “the

findings from this study suggest that there is indeed a relationship between

individual personality type and conflict management style” (p. 9). This

dissertation emerges from the literature, most notably Marion’s finding that there

is a relationship between personality type and conflict management style. If there

is a relationship between personality type and conflict management style, then it

leads one to wonder if there is a relationship between leadership style and

conflict management style.

Kilmann and Thomas (1975) and Mills et al., (1985) also found

correlations between some elements of the MBTI and the TKI. Kilmann and

Thomas conducted a study (of 76 male students in a graduate course in

Behavioral Science for Management) using the MBTI and various conflict-

handling instruments, including the TKI. After analyzing their data, Kilmann and

Thomas conclude, “the results suggest that the Jungian functions related to

judging (thinking vs feeling) and the type of enactment (introverted vs

extraverted) are significantly related to an individual’s conflict-handling behavior

(p. 971). Marion (1995) states that her study “confirms most of the Kilmann-
 16

Thomas findings with those scoring high on feeling tending to be less assertive

and more cooperative in conflict management” (p. 137). She also concludes,

“there is, in the conflict process, a central role for the individual with his/her

perceptions values, beliefs, and attitudes” (p. 22). Marion emphasizes “that a

more positive framework for conflict management is provided through learning

and understanding the differences of individuals and their behavior preferences

and styles” (p. 22-23). If Marion is correct, then this dissertation may provide data

to contribute to a more positive framework for conflict management. This

dissertation differs from Marion’s study because the current study explores

leadership styles (instead of personality styles) and conflict management styles

Mills et al. (1985) also examined the relationships between personality

dimensions (as measured by the MBTI) and conflict-handling styles (as

measured by the TKI). Their results supported the research of Kilmann and

Thomas (1975). Mills et al. asked 199 project management personnel to

complete the MBTI and the TKI. After data analysis, the authors conclude, “in

evaluating our hypotheses, we find support for the hypotheses that Thinking and

Feeling are associated with the distributive dimension of conflict handling” (p.

1141). Walton and McKersie (1965) explain that the five conflict-handling styles

can be characterized as two dimensions: distributive and integrative. Mills et al.

explains that a distributive dimension “assumes trade-offs between competing

and cooperating” (p. 1136) while an integrative dimension “assumes that both

parties can increase their satisfaction as an outcome of conflict” (p. 1136).

 17

Situational Leadership and Conflict Management Style

Romero (1983) examined the relationship between the four styles of

situational leadership and conflict management styles. She gave the Lead-Self

Instrument (Hersey & Blanchard, 1969a, 1969b, 1974, 1982, 1996) and two

conflict inventories developed specifically for her study to 28 secondary school

principals. Romero found that “principals do not change their leadership or

conflict style as situations change” (p. 125). This dissertation differs from

Romero’s study in that this dissertation examines the relationship between

leadership styles (as measured by the MLQ) and conflict management styles (as

measured by the TKI). Romero neither used the MLQ nor the TKI.

Holt (1986) conducted a study similar to Romero’s (1983). She explored

the relationship between leadership styles and conflict management styles. Holt

used the Lead-Self Instrument and the TKI in her study of 156 middle managers

in various hospitals in North Dakota. After data analysis, Holt concluded “there

were no statistically significant relationships between leadership styles and

conflict management techniques in the sample studied” (p. 73). This dissertation

differs from Holt’s study because this dissertation uses the MLQ to assess

leadership styles while Holt used the LEAD-Self Instrument to assess leadership

styles.

Theory and Variables

This section provides an overview of the theory and variables underlying

the hypothesis in this dissertation. Leadership styles addressed include the

following: transactional leadership, transformational leadership, and laissez-faire

 18

leadership. Conflict management styles include these styles: competing,

collaborating, compromising, avoiding, and accommodating.

This dissertation explores the relationships between leadership styles

(measured by leaders’ responses to the MLQ) and conflict management styles

(measured by leaders’ responses to the TKI). The literature (Bass, 1985; Bass &

Avolio, 1994; Burns, 1978; Hartog et al., 1997; Yammarino et al., 1993) implies,

but does not test or expressly state, that different leadership styles (transactional,

transformational, and laissez-faire) will show different conflict management styles

(competing, collaborating, compromising, avoiding, and accommodating).

One hypothesis, based on the expected relationships posited by the

literature, was addressed in this dissertation. This hypothesis relates to whether

or not there is a relationship between leaders’ scores on the three leadership

styles (transactional, transformational, and laissez-faire) as measured by the

MLQ and leaders’ scores on the five conflict management styles (competing,

collaborating, compromising, avoiding, and accommodating) as measured by the

TKI. In order to answer this question the following null hypothesis was tested:

H01: The set of MLQ scores is not related to the set of TKI scores.

Leadership

Burns (1978) has written extensively about leadership. He emphasized

that “leadership over human beings is exercised when persons with certain

motives and purposes mobilize, in competition or conflict with others, institutional,

political, psychological, and other resources so as to arouse, engage, and satisfy

the motives of followers” (p. 18). Burns explains that leadership is different than
 19

power, noting that “to control things—tools, mineral resources, money, energy—

is an act of power, not leadership, for things have no motives. Power wielders

may treat people as things. Leaders may not” (p. 18). He also makes the

assertion that “all leaders are actual or potential power holders, but not all power

holders are leaders” (p. 18).

According to Burns (1978):

The crisis of leadership today is the mediocrity or irresponsibility of so

many of the men and women in power, but leadership rarely rises to the

full need for it. The fundamental crisis underlying mediocrity is intellectual.

If we know all too much about our leaders, we know far too little about

leadership. We fail to grasp the essence of leadership that is relevant to

the modern age and hence we cannot agree even on the standards by

which to measure, recruit, and reject it. Is leadership simply innovation—

cultural or political? Is it essentially inspiration? Mobilization of followers?

Goal setting? Goal fulfillment? Is a leader the definer of values? Satisfier

of needs? If leaders require followers, who leads whom from where to

where, and why? How do leaders lead followers without being wholly led

by followers? Leadership is one of the most observed and least

understood phenomena on earth. (p. 2)

This dissertation examined the relationship between leadership styles and

conflict management styles, specifically considering the hypothesis stemming

from expected relationships as posited by the literature on transformational,

 20

transactional, laissez-faire leadership styles and the five conflict management

styles.

Transactional Leadership

Leadership behavior can be examined to assess leadership style. Bass

and Avolio (1994) explain that leaders can be described as transactional,

transformational, or laissez-faire. Burns (1978) describes transactional leadership

purely as an exchange occurring "when one person takes the initiative in making

contact with others for the purpose of an exchange of valued things" (p. 19). After

this transaction is complete, the relationship ceases to exist. Burns further states,

"A leadership act took place, but it was not one that binds the leader and follower

together in a mutual and continuing pursuit of a higher purpose" (p. 20).

In their discussion of transactional leadership, Bass and Avolio (1994)

observe that transactional leaders respond to follower performance in two

particular ways (rewarding or disciplining). "Transactional leadership depends on

contingent reward (CR) or the more negative active or passive forms of

management-by-exception (MBE-A or MBE-P)" (p. 4). Although CR is more

effective than management-by-exception, neither form of transactional leadership

is considered as effective as transformational leadership (p. 4).

The theory of transactional leadership implies that leaders will enter into

bargained-for exchanges with their followers. This implies a positive relationship

between leaders’ scores on transactional leadership and their scores on

competing. This concept of a bargained-for exchange also implies a positive

relationship between leaders’ scores on transactional leadership and their scores

 21

on compromising. Furthermore, since transactional leaders “desire” to enter into

an exchange with their followers, transactional leaders are not likely to avoid

demands by their followers nor automatically give in and accommodate the

demands of their followers. Thus, the theory of transactional leadership implies

negative relationships between leaders’ scores on transactional leadership and

their scores on avoiding and leaders’ scores on transactional leadership and their

scores on accommodating.

The literature presented in this dissertation depicts transactional leaders

as those who bargain. The theory of transactional leadership also implies that

transactional leaders will be assertive. In Thomas’ (1976) graphic illustration (see

Figure 1.) of the five conflict management styles, competing is shown as

extremely assertive and compromising is shown as assertive.

Since the theory of transactional leadership postulates that transactional

leaders are assertive and they enter into bargained-for exchanges, it is

reasonable to consider whether or not there is a positive relationship between

leaders’ scores on transactional leadership and their scores on competing.

Based on the theory of transactional leadership, it is also reasonable to consider

whether or not a positive relationship exists between leaders’ scores on

transactional leadership and their scores on compromising. Since transactional

leaders are assertive, it is reasonable to consider whether or not there is a

negative relationship between leaders’ scores on transactional leadership and

their scores on avoiding. Finally, transactional leadership theory asserts that

transactional leaders do not give in, but instead bargain or negotiate. Thus, it is
 22

reasonable to consider whether or not there is a negative relationship between

leaders’ scores on transactional leadership and their scores on accommodating.

In order to answer these questions, canonical correlation will be used to

investigate the relationship between the scores on the leadership styles

(transactional, transformational, and laissez-faire) and scores on the conflict

management styles (competing, collaborating, compromising, avoiding, and

accommodating).

Transformational Leadership

Bass (1985) presented his theory (including models and measurement

factors) of transformational leadership, advocating that when transformational

leadership is present, followers respect and trust the leader. This motivates

followers to perform at higher levels. The transformational leader mentors

followers and encourages the followers to become leaders. Bass also identified

four primary factors of transformational leadership: idealized influence,

inspirational motivation, intellectual stimulation, and individualized consideration.

Bass, Avolio, and their colleagues continued to refine this theory and they

developed an instrument called the Multifactor Leadership Questionnaire (Bass &

Avolio, 1990, 1995). The MLQ intercorrelates the four transformational factors to

provide a score for transformational leadership. Furthermore, the MLQ

specifically distinguishes between transformational leadership and transactional

leadership. Schmidt (1993) notes that Bass and Avolio used the intercorrelation

of the transformational factors in the MLQ “to discover that if leaders were only

transactional, their organizations were seen as less effective" (p. 63).

 23

Bass (1985) advocates that transformational leaders mentor and motivate

followers, encouraging the followers to become leaders. Bass and Avolio (1994)

state:

Transformational leadership is seen when leaders:

1. Stimulate interest among colleagues and followers to view their

work from new perspectives.
2. Generate awareness of the mission or vision of the team and
organization.
3. Develop colleagues and followers to higher levels of ability and
potential.
4. Motivate colleagues and followers to look beyond their own
interests toward those that will benefit the group. (p. 2)

Transformational leaders will work together with their followers, reminding the

followers of the importance of the team and the organization. The theory of

transformational leadership postulates that transformational leaders will

collaborate to resolve conflict as opposed to avoiding conflict. Thus, the theory of

transformational leadership implies a positive relationship between leaders’

scores on transformational leadership and their scores on collaborating and a

negative relationship between leaders’ scores on transformational leadership and

their scores on avoiding.

According to the theory of transformational leadership, transformational

leaders work together with their followers. If this is true, then it is reasonable to

consider whether or not a positive relationship exists between leaders’ scores on

transformational leadership and their scores on collaborating. If transformational

leaders desire to work with others and collaborate to resolve conflict, as opposed

to avoiding conflict, then it is reasonable to consider whether or not there is a

 24

negative relationship between leaders’ scores on transformational leadership and

their scores on avoiding.

Laissez-faire Leadership

Yammarino et al. (1993) explain that transactional and transformational

leaders can be described as active leaders, acting to prevent problems from

occurring in their organizations and acting to solve problems. Hartog et al. (1997)

distinguish between these active forms of leadership and the “extremely passive

laissez-faire leadership” (p. 21), noting that the laissez-faire leader “is inactive,

rather than reactive or proactive” (p. 21). Laissez-faire leaders “avoid decision

making and supervisory responsibility” (p. 21). Since the theory of laissez-faire

leadership implies that laissez-faire leaders are inactive and passive, as opposed

to proactive, it is logical to assume that laissez-faire leaders will score high on

avoiding and low on collaborating. Thus, the theory of laissez-faire leadership

implies a positive relationship between leaders’ scores on laissez-faire leadership

and their scores on avoiding and a negative relationship between leaders’ scores

on laissez-faire leadership and their scores on collaborating.

Conflict Management Styles

Blake and Mouton (1964) developed an extremely useful tool called “The

Managerial Grid” that can be used for classification of conflict management

styles. The Grid’s horizontal axis represents a “concern for production” while the

vertical axis represents a “concern for people.” Blake and Mouton’s Grid

“provides a framework for understanding the assumptions managers make as

they compete or cooperate with one another” (p. 15).

 25

Thomas (1976) reinterpreted The Grid and developed a two-dimensional

model that illustrates five conflict management styles: competing, collaborating,

compromising, avoiding, and accommodating. Kilmann and Thomas (1975)

assert that “one of the advantages of this classification scheme is that the five

specific modes reflect several more basic dimensions of interpersonal conflict

behavior” (p. 971). Thomas’ (1976) two-dimensional model is shown in Figure 1.

Figure 1. Five Conflict-Handling Modes.

Figure 1 is adapted from “Conflict and Conflict Management,” by K. Thomas,

1976, p. 900, in M. Dunnette’s (Ed.) The Handbook of Industrial and

Organizational Psychology (Chicago: Rand McNally, pp. 889-935). This figure is

reprinted with permission.

 26

Thomas’ (1976) model uses these two dimensions: assertiveness and

cooperativeness. Thomas and Kilmann (1974, 2002) define assertiveness as “the

extent to which the individual attempts to satisfy his or her own concerns” (p. 7).

They define cooperativeness as “the extent to which the individual attempts to

satisfy the other person’s concerns” (p. 7). These two dimensions, represented

by the vertical and horizontal axes, are used to illustrate the five conflict-handling

modes: competing, collaborating, compromising, avoiding, and accommodating.

The five conflict-handling modes are explained by Thomas and Kilmann

(2002) as follows:

Competing is assertive and uncooperative—a power-oriented mode.

When competing an individual pursues his or her own concerns at the

other person’s expense, using whatever power seems appropriate to win

his or her position—the ability to argue, rank, economic sanctions, and so

on. Competing might mean standing up for your rights, defending a

position you believe is correct, or simply trying to win. (p. 8)

Accommodating is unassertive and cooperative—the opposite of

competing. When accommodating, an individual neglects his or her own

concerns to satisfy the concerns of the other person; there is an element

of self-sacrifice in this mode. Accommodating might take the form of

selfless generosity or charity, obeying another person’s order when you

would prefer not to, or yielding to another’s point of view. (p. 8)

Avoiding is unassertive and uncooperative. When avoiding, an individual

does not immediately pursue his or her own concerns or those of the other
 27

person. He or she does not address the conflict. Avoiding might take the

form of diplomatically side-stepping an issue, postponing an issue until a

better time, or simply withdrawing from a threatening situation. (p. 8)

Collaborating is both assertive and cooperative—the opposite of avoiding.

When collaborating, an individual attempts to work with the other person

to find a solution that fully satisfies the concerns of both. It involves

digging into an issue to identify the underlying concerns of the two

individuals and to find an alternative that meets both sets of concerns.

Collaborating between two persons might take the form of exploring a

disagreement to learn from each other’s insights, resolving some condition

that would otherwise have them competing for resources, or confronting

and trying to find a creative solution to an interpersonal problem. (p. 8)

Compromising is intermediate in both assertiveness and cooperativeness.

When compromising, the objective is to find an expedient, mutually

acceptable solution that partially satisfies both parties. Compromising falls

on a middle ground between competing and accommodating, giving up

more than competing but less than accommodating. Likewise, it

addresses an issue more directly than avoiding but doesn’t explore it in as

much depth as collaborating. Compromising might mean splitting the

difference, exchanging concessions, or seeking a quick middle-ground

position. (p. 8)
 28

Summary of the Literature

In summary, a review of the literature validates distinctions among

leadership styles that can be measured by the MLQ. Furthermore, the literature

unequivocally reveals that different conflict management styles exist and can be

assessed using the TKI. The literature pertaining to leadership styles and conflict

management styles implies, but does not explicitly state, that a relationship may

exist between leadership styles and conflict management styles.

 29

Chapter 3 – Method

A review of the literature validates distinctions among leadership styles

clearly exist and can be measured by the MLQ. Furthermore, the literature

unequivocally reveals that different conflict management styles exist and can be

assessed using the TKI. The literature pertaining to leadership styles and conflict

management styles implies, but does not explicitly state, that a relationship may

exist between leadership styles and conflict management styles.

This chapter presents the research method for this study. First, the

chapter provides an overview of the hypothesis. Second, the data collection

method is explained. Third, an overview of the two research instruments (the

MLQ and the TKI) is discussed. Fourth, the sample population and frame is

presented. Finally, the method of analysis of the data is discussed.

Hypothesis

Table 1 shows the expected relationships (‘+’ for positive, ‘-‘ for negative,

and ‘0’ for no relationship) as posited by the literature on transactional,

transformational, and laissez-faire leadership styles and the five conflict

management styles. One hypothesis, based on the expected relationships

posited by the literature, is addressed in this dissertation.

The hypothesis in this study stems from the theories of transactional

leadership, transformational leadership, and laissez-faire leadership. Since the

theory of transactional leadership postulates that transactional leaders are

assertive and they enter into bargained-for exchanges, is there a positive

relationship between leaders’ scores on transactional leadership and their scores

 30

on competing? Is there also a positive relationship between leaders’ scores on

transactional leadership and their scores on compromising? Since transactional

leaders are assertive, is there a negative relationship between leaders’ scores on

transactional leadership and their scores on avoiding? Since transactional

leadership theory asserts that transactional leaders do not give in, but instead

bargain or negotiate, is there a negative relationship between leaders’ scores on

transactional leadership and their scores on accommodating?

According to the theory of transformational leadership, transformational

leaders work together with their followers. If this is true, then is there a positive

relationship between leaders’ scores on transformational leadership and their

scores on collaborating? If transformational leaders desire to work with others

and collaborate to resolve conflict, as opposed to avoiding conflict, then is there a

negative relationship between leaders’ scores on transformational leadership and

their scores on avoiding?

Since the theory of laissez-faire leadership implies that laissez-faire

leaders are inactive and passive, as opposed to proactive, it is logical to assume

that laissez-faire leaders will score high on avoiding and low on collaborating.

Thus, the theory of laissez-faire leadership implies a positive relationship

between leaders’ scores on laissez-faire leadership and their scores on avoiding

and a negative relationship between leaders’ scores on laissez-faire leadership

and their scores on collaborating.

In order to answer these questions the following null hypothesis was

tested:
 31

H01: The set of MLQ scores is not related to the set of TKI scores.

This hypothesis relates to whether or not there is a relationship between

leaders’ scores on the three leadership styles (transactional, transformational,

and laissez-faire) as measured by the MLQ and leaders’ scores on the five

conflict management styles (competing, collaborating, compromising, avoiding,

and accommodating) as measured by the TKI. This hypothesis, based on

expected relationships posited by the literature (and shown in Table 1), is tested

in this dissertation.

Data Collection

The hypothesis in this study required the collection of data regarding

leadership styles, specifically leaders’ scores on transactional leadership,

transformational leadership, and laissez-faire leadership. The hypothesis also

required the collection of data concerning leaders’ conflict management styles,

specifically leaders’ scores on competing, collaborating, compromising, avoiding,

and accommodating. After leaders in the sample completed both questionnaires,

the questionnaires were scored; and each leader in the sample received scores

on each of the three leadership styles and scores on each of the five conflict

management styles.

The data pertaining to leadership styles was collected using leaders’

responses to the Multifactor Leadership Questionnaire (MLQ). The data

pertaining to conflict management styles was collected using leaders’ responses

to the Thomas-Kilmann Conflict Mode Instrument (TKI).

 32

This study investigated the relationships of leadership styles to conflict

management styles. Leadership styles are those measured by leaders’

responses to the MLQ. Specifically, these are: (a) leaders’ scores on

transactional leadership, (b) leaders’ scores on transformational leadership, and

(c) leaders’ scores on laissez-faire leadership. Based on his/her responses to the

MLQ, each leader received a score from 0 to 48 for transactional leadership, a

score ranging from 0 to 80 for transformational leadership, and a score from 0 to

16 for laissez-faire leadership.

Conflict management styles are those measured by leaders’ responses to

the TKI. These are: (a) leaders’ scores on the competing style of conflict

management, (b) leaders’ scores on the collaborating style of conflict

management, (c) leaders’ scores on the compromising style of conflict

management, (d) leaders’ scores on the avoiding style of conflict management,

and (e) leaders’ scores on the accommodating style of conflict management.

Based on his or her responses to the TKI, each leader received a score ranging

from 0 to 12 for each of the five conflict management styles (competing,

collaborating, compromising, avoiding, and accommodating).

Multifactor Leadership Questionnaire

The MLQ (Bass & Avolio, 1995, 2000) was chosen for this study because

it distinguishes among leadership styles. Specifically, the MLQ measures the

constructs of transformational leadership, transactional leadership, and laissez-

faire leadership. The MLQ (Form 5X) contains 45 questions and allows

respondents to rate leadership behaviors using a 5-point Likert-type scale.

 33

During the years 1997 to 2000, “the latest version of the MLQ, Form 5X, has

been used in nearly 200 research programs, doctoral dissertations and masters

theses around the globe” (Bass & Avolio, 2000, p. 2). Kirnan and Snyder (1995),

in their evaluation of the instrument, state that the MLQ "is designed to be used

at all levels of leadership" (p. 651). Kirnan and Snyder also conclude: "the MLQ

stands apart from other measures of leadership in its sound psychometric

properties" (p. 654).

In addition to distinguishing among transactional leadership,

transformational leadership, and laissez-faire leadership, the MLQ measures the

four transformational leadership factors (idealized influence, inspirational

motivation, intellectual stimulation, and individualized consideration); the two

transactional leadership factors (contingent reward and management-by-

exception); and the nonleadership factor, laissez-faire. Bessai (1995) states, "one

of the major strengths of the questionnaire seems to be the empirical support it

provides for the new paradigm of leadership that distinguishes between

transactional and transformational leadership" (p. 651). Bass (1997) states:

Evidence supporting the transactional—transformational leadership

paradigm has been gathered from all continents except Antarctica—even

offshore the North Sea. The transactional—transformational paradigm

views leadership as either a matter of contingent reinforcement of

followers by a transactional leader or the moving of followers beyond their

self-interests for the good of the group, organization, or society by a

transformational leader. The paradigm is sufficiently broad to provide a

 34

basis for measurement and understanding that is as universal as the

concept of leadership itself. Numerous investigations (field studies, case

histories, management games, interviews, and laboratory studies) point to

the robustness of the effects of transformational and charismatic

leadership. (p. 130)

The MLQ scores provided data needed for canonical correlation to

examine the relationships, if any, between leadership styles and conflict

management styles. The MLQ has been used extensively in various research

studies by corporations and individuals. Bessai (1995) states that the instrument

has "good construct validity, adequate reliability, and a strong research base" (p.

651). Consequently, it is "strongly recommended" (p. 651) for research purposes.

Thomas-Kilmann Conflict Mode Instrument

Conflict management style was assessed using the TKI. The TKI scores

provided data needed for canonical correlation to examine the relationships, if

any, between leadership styles and conflict management styles. The TKI

(Thomas & Kilmann, 1974, 2002) “is designed to assess an individual’s behavior

in conflict situations—that is, situations in which the concerns of two people

appear to be incompatible” (Thomas-Kilmann Conflict Mode Instrument, 2002, p.

7). The TKI is a forced-choice instrument containing 30 pairs of statements. The

leader chooses the statement in each pair that best characterizes his/her

behavior.

The TKI was chosen for this study because it is “used extensively in

American academic research, training seminars, and organizational intervention

 35

and development” (Volkema & Bergmann, 1995, p. 6). Thomas and Kilmann

(1978) emphasize that the TKI is “specifically designed to minimize the effect of

social-desirability bias” (p. 1141). There is also immense support for external

validity and reliability of the instrument (Kilmann & Thomas, 1977; Kabanoff,

1987).

Sample Population & Frame

This dissertation explores the relationships between leadership styles

(measured by leaders’ responses to the MLQ) and conflict management styles

(measured by leaders’ responses to the TKI). The findings may be generalized to

leaders in the United States.

Leaders for this study were chosen primarily through referral by alumni

from Emory & Henry College. Various Emory & Henry alumni throughout the

United States were contacted and asked to identify several leaders in their

city/town who would be willing to participate in this study. Using alumni to provide

referrals enabled leaders from a wide geographical pool to be included in the

sample. A total of 120 leaders agreed to participate in the study and thus

received a packet containing the research instruments. Only 105 leaders

returned packets. Of the 105 packets returned, six contained incomplete

instruments and were thus not eligible for inclusion in the statistical analysis.

Therefore, the sample size for this study consisted of 99 leaders.

Analysis

This study explored the relationships of two sets of variables, leadership

styles and conflict management styles. Leadership styles are those measured by
 36

leaders’ responses to the MLQ. Specifically, these are: scores on transactional

leadership, scores on transformational leadership, and scores on laissez-faire

leadership. This study did not assign leaders to one of the three leadership

styles. Instead, this study used the scores on the leadership styles as measures

of the “degree” (or intensity) of transactional leadership, the “degree” of

transformational leadership, and the “degree” of laissez-faire leadership that

each leader exhibits. While leadership style (transactional, transformational, and

laissez-faire) is a categorical variable, the same leader exhibits all three styles to

a different degree.

The MLQ provided scores on: TR, TF, and LF. TR is “scores on”

transactional leadership (ranging from 0 to 48) as measured by the sum of the

leader’s responses (from 0 to 4) on the twelve MLQ questions for transactional

leadership. TF is “scores on” transformational leadership (ranging from 0 to 80)

as measured by the sum of the leader’s responses (from 0 to 4) on the twenty

MLQ questions for transformational leadership. LF is “scores on” laissez-faire

leadership (ranging from 0 to 16) on the four MLQ questions for laissez-faire

leadership.

Conflict management styles are those measured by leaders’ responses to

the TKI. These are: scores on competing, scores on collaborating, scores on

compromising, scores on avoiding, and scores on accommodating. This study

did not assign leaders to a specific conflict management style. Instead, this study

used scores as measures of the extent to which a leader is competing, the extent

to which a leader is collaborating, the extent to which a leader is compromising,

 37

the extent to which a leader is avoiding, and the extent to which a leader is

accommodating. While conflict management style (competing, collaborating,

compromising, avoiding, and accommodating) is a categorical variable, the same

leader exhibits all five styles to a different degree.

The TKI provided scores on: competing, collaborating, compromising,

avoiding, and accommodating. The TKI contains 30 questions and leaders

received: a “score on competing” ranging from 0 to 12, a “score on collaborating”

ranging from 0 to 12, a “score on compromising” ranging from 0 to 12,

a “score on avoiding” ranging from 0 to 12, and a “score on accommodating”

ranging from 0 to 12.

In order to test the hypothesis in this research study, two research

instruments (the MLQ and the TKI) were administered to leaders. The MLQ

provided a score for each leader on transactional leadership, transformational

leadership, and laissez-faire leadership. The TKI provided a score for each

leader on each the following five conflict management styles: competing,

collaborating, compromising, avoiding, and accommodating. The scores from the

MLQ and the TKI were entered into SPSS. The MLQ variables (the scores on

transactional leadership, the scores on transformational leadership, and the

scores on laissez-faire leadership) and the TKI variables (competing,

collaborating, compromising, avoiding, and accommodating) were used in the

canonical correlation analysis.

 38

Justification and Support for Use of Canonical Correlation

The research analysis employed canonical correlation to investigate the

relationship between the scores on the leadership styles (transactional,

transformational, and laissez-faire) and scores on the conflict management styles

(competing, collaborating, compromising, avoiding, and accommodating).

Canonical correlation, described by Hotelling (1935), is a statistical technique

that is used to analyze the relationship between two sets of variables. Hair,

Anderson, Tatham, and Black (1998) explicitly state “canonical correlation

analysis is a multivariate statistical model that facilitates the study of

interrelationships among sets of multiple dependent variables and multiple

independent variables” (p. 444). Canonical correlation differs from multiple

regression in that “multiple regression predicts a single dependent variable from

a set of multiple independent variables” (p. 444) and “canonical correlation

simultaneously predicts multiple dependent variables from multiple independent

variables” (p. 444). Furthermore, “canonical correlation identifies the optimum

structure or dimensionality of each variable set that maximizes the relationship

between independent and dependent variable sets” (p. 445).

Hair et al. (1998), commenting on the proper use of canonical correlation,

state “in situations with multiple dependent and independent variables, canonical

correlation is the most appropriate and powerful multivariate technique” (p. 444).

Thompson (1980) also discusses the power of canonical correlation and the

“theoretically rich” (p. 17) results that can be obtained from this procedure. He

emphatically declares:
 39

the neophyte student of canonical correlation analysis may be

overwhelmed by the myriad coefficients which the procedure produces.

This implies a strength of the procedure. Canonical correlation analysis

produces results which can be theoretically rich, and if properly

implemented the procedure can adequately capture some of the complex

dynamics involved in educational reality. (pp. 16-17)

Summary of Research Design

The purpose of this dissertation was to explore the relationship (if any)

between leadership styles and conflict management styles. Data regarding

leadership styles and conflict management styles were obtained from a sample

of 99 leaders. The research analysis employed canonical correlation to

determine whether to reject or fail to reject the null hypothesis that the set of

MLQ scores is not related to the set of TKI scores.

 40

Chapter 4 – Results

This study explores the relationship between leadership styles (measured

by leaders’ responses to the MLQ) and conflict management styles (measured by

leaders’ responses to the TKI). This chapter discusses data collection and then

presents the results found upon examination of the leaders’ responses to the

MLQ and the TKI. The results include general observations of the data gathered

as well as statistical analysis of the data.

Data Collection

In order to test the hypothesis in this study, data regarding leadership

styles, specifically leaders’ scores on transactional leadership, transformational

leadership, and laissez-faire leadership were collected. The hypothesis also

required the collection of data concerning leaders’ conflict management styles,

specifically leaders’ scores on competing, collaborating, compromising, avoiding,

and accommodating. The data pertaining to leadership styles were collected

using leaders’ responses to the Multifactor Leadership Questionnaire (MLQ). The

data pertaining to conflict management styles were collected using leaders’

responses to the Thomas-Kilmann Conflict Mode Instrument (TKI).

Characteristics of Leaders in the Sample

For this study, a leader was defined as a person who guides or instructs

others. Leaders were chosen for this study primarily through referral by alumni

from Emory & Henry College. Various Emory & Henry alumni throughout the

United States were contacted and asked to identify several leaders in their

city/town who would be willing to participate in this study. Using alumni to provide
 41

referrals enabled leaders from a wide geographical pool to be included in the

sample.

This sample included 99 leaders from a wide array of organizations

(Fortune 500 companies, governmental organizations, churches, schools, small

businesses, etc.). Specifically, the sample included accountants, bankers,

doctors, nurses, physical therapists, ministers, attorneys, civic leaders, military

leaders, elementary school teachers, principals, and others.

Data Collection Procedures

A sampler set of the Multifactor Leadership Questionnaire (as well as

permission to reproduce up to 150 copies of the MLQ) was purchased from Mind

Garden and 120 copies of the Thomas-Kilmann Conflict Mode Instrument were

purchased from CPP. Coding of the leadership instruments (the MLQs) and the

conflict management style instruments (the TKIs) was essential not only to

maintain confidentiality, but also to ensure that leaders’ scores on the leadership

instruments could be correlated with their scores on the conflict management

style instruments. The instruments were numbered from 1 to 120 to ensure that

the leadership instruments could be matched up with the conflict management

style instruments. In addition to the research instruments, each leader also

received a Demographic Questionnaire (see Appendix) containing questions

pertaining to the type of organization in which the leader served, the leader’s

gender, the number of months the leader has served in a leadership position, the

number of people in the organization, and the number of people who report to the
 42

leader. The Demographic Questionnaires (and the envelopes that contained the

research instruments for the participants in the study) were also numbered.

A total of 120 packets were prepared for the sample. Each envelope contained a

letter to each leader (see Appendix), briefly explaining the data collection

procedures, a Demographic Questionnaire, a copy of the MLQ, and a copy of the

TKI. These 120 packets were then distributed to leaders. A total of 105 leaders

returned their packets. Of the 105 packets returned, the data collected from 99

respondents were usable for statistical analysis. Six of the respondents left

several questions on the MLQ and/or the TKI unanswered and therefore,

rendered the incomplete instrument unusable for purposes of statistical analysis.

Hair et al. (1998), in their explanation of how to design a canonical correlation

analysis, state that sample size should include “at least 10 observations per

variable to avoid ‘overfitting’ the data” (p. 447). This study contains eight

variables. Thus, this study’s sample size of 99 leaders is adequate for canonical

correlation analysis.

Data Coding and Analysis Procedures

Upon receipt of the packets, the MLQs and the TKIs were scored. Each

leader who returned a completed packet received scores on each of the three

leadership styles and scores on each of the five conflict management styles. The

data (for each of the 99 leaders) were entered into an Excel spreadsheet and

then imported into a software package (SPSS) for statistical analysis. These data

consisted of leader numbers, the leaders’ scores for transactional leadership,

transformational leadership, and laissez-faire leadership, as well as the leaders’

 43

scores on each of the five conflict management styles (competing, collaborating,

compromising, avoiding, and accommodating).

The data collected from the leadership instrument (MLQ) and the conflict

management style instrument (TKI) were imported into SPSS for statistical

analysis to examine whether or not a relationship exists between leadership

styles and conflict management styles. The leadership styles were measured by

the leaders’ scores on the MLQ (specifically, leaders’ scores on transactional

leadership, leaders’ scores on transformational leadership, and leaders’ scores

on laissez-faire leadership). Conflict management style was assessed using the

leaders’ scores on the TKI (specifically, leaders’ scores on the five conflict

management styles: competing, collaborating, compromising, avoiding, and

accommodating).

Description of Sample Data

Table 2 presents the minimum and maximum scores on the MLQ and TKI

for leaders participating in this study. The mean and standard deviation are also

shown. It is important to note that for the MLQ leaders could receive a score for

Transactional Leadership ranging between 0 and 48, a score ranging from 0 to

80 for Transformational Leadership, and a score from 0 to 16 for Laissez-faire

Leadership. In addition, leaders could receive a score ranging from 0 to 12 on

each of the five conflict management styles.

The mean for Transactional Leadership was 24.02 while Transactional

scores could range from 0 to 48. Therefore, the average score was only slightly

above half of the maximum score available, indicating that, on average, the
 44

leaders in this sample did not score high on Transactional Leadership. The mean

for Transformational Leadership was 60.85 while Transformational scores could

range from 0 to 80, indicating that, on average, leaders received 76 percent of

the available points on Transformational Leadership. The mean for Laissez-faire

Leadership was 2.38, indicating that, on average, the leaders in this sample

scored particularly low on this extremely passive leadership style.

The results of this study indicate that, on average, Compromising was the

first most frequently used conflict management style, and on average, Avoiding

was the second most frequently used conflict management style. Holt’s (1986)

study also found Compromising to be the most frequently used conflict

management style (p. 76). Holt reported that Collaborating was the second most

frequently used conflict management style in her study (p. 77). The results of my

study indicate that, on average, Avoiding was the second most frequently used

conflict management style and, on average, Collaborating was the third most

frequently used conflict management style.

Table 2 also indicates that some leaders scored the maximum possible on

Competing, Collaborating, Compromising, and Avoiding, but not a single leader

participating in this study scored higher than 10 of a maximum of 12 points on

Accommodating. It is also important to note that some leaders scored zero on

Competing and some leaders scored zero on Accommodating. These results

also indicate that some leaders rely heavily on Compromising and Avoiding.
 45

Table 2

Sample Statistics for Variables

N Minimum Maximum Mean Std. Deviation

MLQ Scores

Transactional 99 10 39 24.02 5.225

Transformational 99 40 77 60.85 7.945

Laissez-Faire 99 0 10 2.38 2.175

TKI Scores

Competing 99 0 12 4.22 2.940

Collaborating 99 2 12 6.41 2.286

Compromising 99 3 12 7.18 2.168

Avoiding 99 2 12 6.76 2.250

Accommodating 99 0 10 5.35 2.512

Tables 3 and 4 present the frequency distributions of the MLQ and TKI

scores. Almost all of the Transactional scores were between 16 and 32 on a

scale from 0 to 48, symmetrical about a midpoint of 24. In contrast, the scores for

Transformational and Laissez-faire Leadership were skewed. Table 3 shows that

92% of the Transformational scores were between 48 and 72 on a scale from 0

to 80. Almost all of the Laissez-faire scores were less than or equal to six on a

scale from 0 to 16. A comparison of the TKI frequencies in Table 4 with those for

MLQ scores in Table 3 shows that all five of the TKI scores were more evenly

distributed across their possible range from 0 to 12.

 46

Table 3

Distribution of MLQ Scores

Transactional Transformational Laissez-faire

Score % Cumulative Score % Cumulative Score % Cumulative

% % %

0 0.0 0.0 0 0.0 0.0 0 19.2 19.2

4 0.0 0.0 4 0.0 0.0 2 40.4 59.6
8 0.0 0.0 8 0.0 0.0 4 25.3 84.9
12 2.0 2.0 12 0.0 0.0 6 10.1 95.0
16 3.0 5.0 16 0.0 0.0 8 4.0 99.0
20 22.2 27.2 20 0.0 0.0 10 1.0 100.0
24 25.3 52.5 24 0.0 0.0 12 0.0
28 25.3 77.8 28 0.0 0.0 14 0.0
32 19.2 97.0 32 0.0 0.0 16 0.0
36 2.0 99.0 36 0.0 0.0
40 1.0 100.0 40 1.0 1.0
44 0.0 44 1.0 2.0
48 0.0 48 4.0 6.0
52 9.1 15.1
56 12.1 27.2
60 25.3 52.5
64 16.2 68.7
68 13.1 81.8
72 12.1 93.9
76 3.0 96.9
80 3.0 99.9
 47

Table 4

Distribution of TKI Scores

Accommodating Avoiding Compromising Collaborating Competing

Score % Cum % Cum % Cum % Cum % Cum
% % % % %

0 2.0 2.0 0.0 0.0 0.0 0.0 0.0 0.0 6.1 6.1
2 12.1 14.1 4.0 4.0 0.0 0.0 5.1 5.1 27.3 33.4
4 27.3 41.4 12.1 16.1 15.2 15.2 18.2 23.3 27.3 60.7
6 29.3 70.7 29.3 45.4 21.2 36.4 27.3 50.6 15.2 75.9
8 13.1 83.8 30.3 75.7 35.4 71.8 30.3 80.9 13.1 89.0
10 16.2 100.0 20.2 95.9 24.2 96.0 18.1 99.0 9.0 98.0
12 0.0 4.0 99.9 4.0 100.0 1.0 100.0 2.0 100.0

Table 5

Bivariate Correlations Between Variables

TR TF LF Compet. Collab. Compro. Avoiding Accom.

MLQ Scores

TR 1.000
TF .110 1.000
LF .359** -255* 1.000

TKI Scores
Competing .101 .045 -.061 1.000
Collaborating -.114 .086 -.032 -.003 1.000
Compromising -.120 .157 -.084 -.328** -.137 1.000
Avoiding -.008 -.154 .017 -.337** -.379** -.357** 1.000
Accommodating .099 -.133 .150 -.576** -.477** -.031 .134 1.000

** Correlation is significant at the 0.01 level (2-tailed).

* Correlation is significant at the 0.05 level (2-tailed).
 48

Table 5 shows that none of the bivariate correlations between the three

MLQ scores and the five TKI scores are statistically significant; however, Table 5

does show that some correlations among the MLQ variables are statistically

significant. For example, LF and TR are positively correlated and the correlation

is highly significantly different from zero. This can be interpreted to mean, in

general, those with high (low) scores on Laissez-faire Leadership have high (low)

scores on Transactional Leadership. Table 5 also shows that LF and TF are

negatively correlated and the correlation is significantly different from zero.

Generally speaking, this means that those with high (low) scores on Laissez-faire

leadership have low (high) scores on Transformational leadership.

The data presented in Table 5 also show that 6 correlations among the

TKI variables are statistically significant. The correlation between Compromising

and Competing is highly significantly different from zero. The correlation between

Avoiding and Competing as well as the correlations between Avoiding and

Collaborating and Avoiding and Compromising are also highly significantly

different from zero. The correlations between Accommodating and Competing

and Accommodating and Collaborating are also highly significantly different from

zero.

None of the bivariate correlations (shown in Table 5) between the three

MLQ scores and the five TKI scores are statistically significant. These results are

somewhat surprising. Since Transactional Leadership theory implies that

transactional leaders will enter into “bargained for” exchanges, it would seem

logical that Transactional Leadership would correlate positively with Competing

 49

and Compromising while correlating negatively with Avoiding. Two of these three

expected correlations were found with respect to Transactional Leadership

theory. The correlation between Transactional Leadership and Competing was

positive and the correlation between Transactional Leadership and Avoiding was

negative; however, these correlations were not statistically significant. It is

interesting to note that although Transactional Leadership theory implies a

positive correlation between Transactional Leadership and Compromising, the

correlation (although not statistically significant) in this sample was negative.

Transformational Leadership theory implies that transformational leaders

will work together with their followers. The theory also implies that

transformational leaders are active, not passive. Therefore, a positive correlation

was expected between Transformational Leadership and Collaborating and a

negative relationship between Transformational Leadership and Avoiding. Table

5 does show a positive correlation between Transformational Leadership and

Collaborating and a negative correlation between Transformational Leadership

and Avoiding; however, these correlations were not statistically significant.

Laissez-faire leadership theory implies that laissez-faire leaders are

passive, not active, and will avoid making decisions. The theory thus implies a

negative relationship between Laissez-faire Leadership and Collaborating and a

positive relationship between Laissez-faire Leadership and Avoiding. A negative

correlation was found between Laissez-faire Leadership and Collaborating as

expected and a positive correlation was found between Laissez-faire Leadership

 50

and Avoiding. Once again, however, the correlations were not statistically

significant in this sample.

Canonical Correlation Analysis

To see if the set of three MLQ scores is related to the set of five TKI

scores, a canonical correlation was performed. The objectives of the analysis are

as follows:

1. To determine the magnitude of the relationship between the set of MLQ

scores and the set of TKI scores, or, conversely, if the two sets of scores are
independent.
2. To explain the relationships that exist between the set of MLQ scores
and the set of TKI scores by measuring the relative contribution of each variable
to the canonical functions (Hair et. al., 1998, p. 447).

In this analysis two canonical variates—one for the MLQ scores and one

for the TKI scores—are defined as linear combinations of their respective scores.

For example, the three MLQ scores imply a canonical variate MLQ1, where

MLQ1 = α1TR +α2TF +α3LF. (1)

The five TKI scores imply a canonical variate TKI1, where

TKI1 = β1Avoid +β2Accomm +β3Compro +β4Collab +β5Compet. (2)

For the first canonical root, the α and β coefficients are selected to give the

maximum correlation between MLQ1 and TKI1.

 51

Statististical Assumptions of the Analysis

Canonical correlation assumes that the variables are linearly related,

homoscedastic, and not multicollinear. Furthermore, the validity of hypothesis

tests requires approximate normality. To test the validity of these assumptions,

several tests and analyses were performed. For multicollinearity Table 5 shows

that there is one significant pairwise correlation among the MLQ scores and

some significant pairwise correlations among the TKI scores; however, in all

cases the correlations are below .80. The highest--between Accommodating and

Competing--is only -0.576. Licht (1998) explains that multicollinearity can be a

problem if the “correlations of r > .80” (p. 45). All of the correlations are below

.80.

Table 6 gives the results of the Shapiro-Wilks tests for normality of the

MLQ and TKI scores. We fail to reject normality at the .05 significance level in all

cases but that of scores for the Laissez-faire leadership style and the Competing

conflict management style. Hair et al., (1998) state:

Canonical correlation analysis can accommodate any metric variable

without the strict assumption of normality. Normality is desirable because

it standardizes a distribution to allow for a higher correlation among the

variables. But in the strictest sense, canonical correlation analysis can

accommodate even nonnormal variables if the distributional form (e.g.,

highly skewed) does not decrease the correlation with other variables. (p.

448)
 52

Since the correlations in both of these cases are very high and transforming

Laissez-faire would require using logs that are not defined for almost twenty

percent (as shown for LF scores of zero in the frequency distribution of MLQ

Scores presented in Table 3) of the sample, these variables were left unchanged.

Tabachnick and Fidell (1996) support this decision and explain the reasoning

behind nontransformations as follows:

Although data transformations are recommended as a remedy for outliers

and for failures of normality, linearity, and homoscedasticity, they are not

universally recommended. The reason is that an analysis is interpreted

from the variables that are in it and transformed variables are sometimes

harder to interpret. (p. 80)

To test for linearity between the TKI scores and the MLQ scores, between

the canonical variates and their respective set of variables, and between the

canonical variates themselves, scatterplots were examined. Figure 2 shows the

scatterplots between each TKI score and each MLQ score. There are no signs of

a non-linear relationship among these variables.

 53

Table 6

Tests for Normality of Variables

Variable r p

MLQ Scores

TR .9958 0.87255
TF .9984 0.97984
LF .9713 0.00001

TKI Scores

Competing .9833 0.00076

Collaborating .9948 0.16727
Compromising .9944 0.13307
Avoiding .9986 0.76885
Accommodating .9944 0.50694

Figures 3 and 5 give the scatterplots between the canonical variates and

their respective sets of variables. None of these plots indicate a non-linear

relationship or heteroscedasticity. The apparent decrease in dispersion in the

scatterplot of MLQ1 and LF is an artifact of the data caused by the eight

observations in the rightward third of the plot. (Thus, all of the LF scatterplots

should be interpreted with caution.) Figures 4 and 6 give the scatterplots

between the canonical variates and their opposite set of variables. Again, none of

these plots indicate a non-linear relationship or heteroscedasticity. Finally, for

completeness, Figure 7 plots the relationships between the canonical variates for

the three functions that have been extracted.

 54

Deriving the Canonical Functions and Their Significance

The canonical correlation analysis has three canonical functions because

there are three MLQ scores. Results of significance tests for the canonical

correlations between the set of five TKI scores and the three MLQ scores are

shown in Tables 7 and 8. The first canonical correlation is only .298 and in all

tests the data fail to reject the null hypothesis that all three canonical correlations

are zero. Furthermore, the F-tests in Table 8 confirm the statistical insignificance

of all of the canonical correlations. The sample offers no evidence of a

relationship between the three MLQ scores and the five TKI scores.

Table 7

Measures of Overall Model Fit for Canonical Correlation Analysis

Canonical Canonical Canonical

Function Correlation R2 F Statistic p

1 .298 .089 .928 .533

2 .183 .033 .647 .737
3 .146 .021 .674 .570

Table 8

Multivariate Tests of Significance

Approximate
Statistic Value F Statistic p

Wilks’ lamba .86183 .92834 .533

Pillai’s Trace .14369 .93568 .525
Hotelling’s Trace .15400 .92058 .542
Roy’s Greatest Root .08900
 55

Table 9 gives the results of a redundancy analysis for the three canonical

functions. A redundancy index is calculated to account for the fact that “a

relatively strong canonical correlation may be obtained between two linear

composites (canonical variates), even though these linear composites may not

extract significant portions of variance from their respective sets of variables”

(Hair et al., 1998, p. 451). In effect, the index weights the canonical correlation

with the correlation between the variate and its set of variables by multiplying the

square of the canonical correlation by the average of

Table 9

Redundancy Analysis

Standardized Variance of the TKI Variables Explained by

Their Own The Opposite

Canonical Variate Canonical Variate
(Shared Variance) (Redundancy)
Canonical Cumulative Canonical Cumulative
2
Function Percentage Percentage R Percentage Percentage

1 .247 .247 .089 .022 .022

2 .112 .359 .033 .004 .026
3 .218 .577 .021 .005 .031

Standardized Variance of the MLQ Variables Explained by

Their Own The Opposite

Canonical Variate Canonical Variate
(Shared Variance) (Redundancy)
Canonical Cumulative Canonical Cumulative
Function Percentage Percentage R2 Percentage Percentage

1 .341 .341 .089 .030 .030

2 .290 .631 .033 .010 .040
3 .037 .668 .021 .008 .048
 56

the squared correlations between the variate and its set of variables. A high

redundancy index requires both a high canonical correlation (i.e., a high

percentage of the variance in the dependent variate explained by the

independent canonical variate) and a high amount of shared variance (i.e., a high

percentage of the variance in the variate explained by its set of variables) (Hair

et.al., 1998, pp. 451-452).

The redundancy indices in Table 9 are very low for the first TKI variate

(.022) and the first MLQ variate (0.030). In both cases, the low indices result from

the low canonical R2 and the low shared variances (.247 for the TKI variate and

.341 for the MLQ variate). The redundancy results for the second and third

canonical functions are even weaker.

Interpreting the Canonical Variates

Although the canonical correlations are insignificant, the raw and

standardized coefficients for the canonical variates are reported in Tables 10

through 13. The raw canonical coefficients in Table 10 can be used to write

equations for each of the three canonical variates implied from the five TKI

scores. For example, the five TKI scores imply a canonical variate TKI1 for the

first canonical function, and using the coefficients in Table 10,

TKI1 = .729 COMPETING + .488 COLLABORATING + .352 COMPROMISING +

.675 AVOIDING + .839 ACCOMMODATING.

Furthermore, the data in Table 12 can be used to write equations for the three

canonical variates implied from the three MLQ scores. The following equation is

created using the coefficients in Table 12:

 57

MLQ1 = 0.147 TR – 0.100 TF – 0.048 LF.

Table 10

Raw Canonical Coefficients (Weights) for TKI Scores

Variable TKI1 TKI2 TKI3

COMPETING .729 2.909 -.918

COLLABORATING .488 2.696 -1.272
COMPROMISING .352 2.900 -1.097
AVOIDING .675 2.989 -1.434
ACCOMMODATING .839 2.563 -1.055

Table 11

Standardized Canonical Coefficients (Weights) for TKI Scores

Variable TKI1 TKI2 TKI3

COMPETING 2.143 8.552 -2.700

COLLABORATING 1.116 6.162 -2.907
COMPROMISING .763 6.289 -2.379
AVOIDING 1.518 6.726 -3.228
ACCOMMODATING 2.109 6.439 -2.651

Table 12

Raw Canonical Coefficients (Weights) for MLQ Scores

Covariate MLQ1 MLQ2 MLQ3

TR .147 .065 .135

TF -.100 -.037 .080
LF -.048 -.517 -.024
 58

Table 13

Standardized Canonical Coefficients (Weights) for MLQ Scores

Covariate MLQ1 MLQ2 MLQ3

TR .771 .340 .706

TF -.794 -.294 .639
LF -.105 -1.124 -.052

Tables 11 and 13 give the standardized coefficients for TKI scores and

MLQ scores. Standardized coefficients can be used to show the relative

importance of each variable to the canonical variate. In this case, scores on

Competing and Accommodating were the most important for their first canonical

variate and scores on Compromising were the least important. Table 13 shows

that scores on Transformational and Transactional Leadership were the most

important for their first canonical variate and scores on Laissez-faire Leadership

were the least important.

Tables 14 through 17 show the canonical loadings and cross-loadings for

the three functions. “Because canonical weights are typically unstable,

particularly in instances of multicollinearity, owing to their calculation solely to

optimize the canonical correlation, the canonical loading and cross-loadings are

considered more appropriate… [to rank the relative contribution of variables to

their canonical variate]” (Hair et al., 1998, p. 458).

 59

Table 14

Canonical Loadings: Correlations between TKI Scores and Their Canonical

Variates

Variable TKI1 TKI2 TKI3

COMPETING .163 .493 .705

COLLABORATING -.513 -.153 -.165
COMPROMISING -.699 .041 .138
AVOIDING .384 .127 -.723
ACCOMMODATING .557 -.527 -.159

Table 15

Canonical Loadings: Correlations between MLQ Scores and Their Canonical

Variates

Covariate MLQ1 MLQ2 MLQ3

TR .645 -.096 .758

TF -.683 .030 .730
LF .374 -.927 .039

Table 16

Cross-Loadings: Correlations between TKI Scores and MLQ Canonical Variates

Variable MLQ1 MLQ2 MLQ3

COMPETING 0.04825 0.08999 0.11463

COLLABORATING -0.15299 -0.02781 -0.01703
COMPROMISING -0.020872 0.00760 0.03758
AVOIDING 0.114705 0.02334 -0.10737
ACCOMMODATING 0.166321 -0.09634 -0.05484

Table 17

Cross-Loadings: Correlations between MLQ Scores and TKI Canonical Variates

Covariate TKI1 TKI2 TKI3

TR 0.192578 -0.01732 0.110824

TF -0.20358 0.00547 0.106335
LF 0.111393 -0.16927 0.005726
 60

The correlations between canonical variate TKI1 and the five TKI variables

in Table 14--the loadings--point to the scores on collaborating, compromising,

and accommodating as the most important factors for conflict management

variate. Table 15 shows that Transactional and Transformational leadership

scores are more highly correlated with the MLQ variate than the Laissez-faire

leadership scores are. Recall that the relatively low loadings shown in Tables 14

and 15 contributed to the poor redundancy values exhibited in Table 9.

The low cross-loadings in Tables 16 and 17 are low correlations between

the TKI scores and the MLQ variates (in Table 16) and low correlations between

the MLQ scores and the TKI variates (in Table 17). These correlations imply that

very low percentages of the variance in the variables are explained by the

opposite variate. For example, from the information in the first row and column of

Table 16, only about 0.2 percent (.04825 squared) of the variance in the scores

on the Competing conflict management style is explained by the MLQ variate.

The last row and first column of the same table shows that only 2.8 percent

(0.166321 squared) of the variation in the scores on the Accommodating style is

explained by the MLQ variate for the first function.

Table 17 shows similary weak cross-relationships. The highest cross-

loading of -0.20358 between scores on Transformational leadership and the first

TKI variate shows that the variate only explains about 4 percent of the variation

in that leadership style. The small correlations between the original variables and

their opposite canonical variate confirm the insignificant relationships among the
 61

three pairs of canonical variates shown in Tables 7 and 8 and contribute to the

low redundancy values in Table 9.

Summary of Statistical Analysis

Canonical correlation analysis shows that no statistically significant

relationship exists between the set of leadership styles (as assessed by the

MLQ) and the set of conflict management styles (as assessed by the TKI).

Results of significance tests for the canonical correlations between the set of five

TKI scores and the three MLQ scores are shown in Tables 7 and 8. The first

canonical correlation is only .298 and in all tests the data fail to reject the null

hypothesis that all three canonical correlations are zero. The F-tests in Table 8

confirm the statistical insignificance of all of the canonical correlations. The

sample offers no evidence of a relationship between the three MLQ scores and

the five TKI scores.

 62

Chapter 5 – Discussion

This study explores the relationship between leadership styles, as

assessed by the Multifactor Leadership Questionnaire (MLQ), and conflict

management styles, as assessed by the Thomas-Kilmann Conflict Mode

Instrument (TKI). It discusses transactional, transformational, and laissez-faire

leadership and presents the theoretical implications of these leadership styles

upon conflict management styles. The study’s research design explains the data

gathering process. The results section presents the data in a meaningful format

and explains how canonical correlation was used to analyze the relationship

between the MLQ scores and the TKI scores. This study clearly demonstrates

that using leadership theories to make predictions regarding the relationship

between leadership styles and conflict management styles can be extremely

complex. Finally, recommendations for future research efforts are provided.

Exploring Leadership Styles and Conflict Management Styles

Table 5 shows that none of the bivariate correlations between the three

MLQ scores and the five TKI scores are statistically significant; however, Table 5

does show that some correlations among the MLQ variables are statistically

significant. For example, LF and TR are positively correlated and the correlation

is highly significantly different from zero. This can be interpreted to mean, in

general, those with high (low) scores on Laissez-faire Leadership have high (low)

scores on Transactional Leadership. Table 5 also shows that LF and TF are

negatively correlated and the correlation is significantly different from zero.

Generally speaking, this means that those with high (low) scores on Laissez-faire
 63

leadership have low (high) scores on Transformational leadership. The positive

correlation between LF and TR and the negative correlation between LF and TF

found in this study confirm the results found by Bass and Avolio (2000) in their

examination of the intercorrelations among MLQ scores, using data from 2,154

leaders.

The results of this study also confirm some of Holt’s (1986) findings. Holt

used the LEAD-Self Instrument and the TKI to examine leadership styles and

conflict management styles among hospital mgmt personnel. Holt concluded

“there were no statistically significant relationships between leadership styles and

conflict management techniques in the sample studied” (p. 73). The LEAD-Self

(developed by Hersey & Blanchard) focuses on task behavior and relationship

behavior and classifies leadership styles as: participating, selling, delegating, and

telling. This study differed from Holt’s in two specific ways: (a) this study used the

MLQ (not the LEAD-Self) to assess leadership styles and (b) this sample was not

limited to leaders in hospital management. Although this study used a different

research instrument and included leaders from a variety of occupations, the

results of this study corroborate Holt’s finding that no statistically significant

relationship exists between leadership styles and conflict management styles.

While some of the results in this study confirm some of Holt’s (1986)

findings, other results of this study differ significantly from Holt’s previous

findings. Holt found Compromising to be the most frequently used conflict

management style in her study. Collaborating was second, Competing was third,

Avoiding was fourth, and Accommodating was fifth. The current study’s results
 64

concur with Holt’s findings regarding Compromising to be the most frequently

used conflict management style; however, the rankings for second, third, fourth,

and fifth most frequently used conflict management styles all differ from Holt’s

study. The current study found that Avoiding was the second most frequently

used conflict management style, Collaborating was third, Accommodating was

fourth, and Competing was fifth in the sample studied. These findings may be the

result of the inclusion of leaders from an extensive vocational range or the use of

a different research instrument for assessing leadership styles.

The results of this study show support for six of expected bivariate

correlations (although not statistically significant) as implied by the leadership

theories between leadership styles and conflict management styles. Table 1

shows the expected relationships (‘+’ for positive, ‘-‘ for negative, and ‘0’ for no

relationship) as posited by the literature on transactional, transformational, and

laissez-faire leadership styles and the five conflict management styles. Table 5

provides data showing the following relationships: (a) Transactional Leadership

and Competing were positively correlated, (b) Transactional Leadership and

Avoiding were negatively correlated, (c) Transformational Leadership and

Collaborating were positively correlated, (d) Transformational Leadership and

Avoiding were negatively correlated, (e) Laissez-faire Leadership and Avoiding

were positively correlated, and (f) Laissez-faire Leadership and Collaborating

were negatively correlated. Although the literature implies that leadership styles

may be related to conflict management styles, this study was not able to

corroborate that inference at a level of statistical significance. This study clearly

 65

demonstrates that using leadership theories to make predictions regarding

relationships between leadership styles and conflict management styles can be

extremely complex.

Observations/Limitations

This study did not find a statistically significant relationship between

leadership styles and conflict management styles. Although there may be many

reasons that explain why a statistically significant relationship was not found in

the sample studied, four specific reasons include (a) the self-reporting nature of

the research instruments, (b) the possibility that leaders may behave differently

(i.e., depart from their normal leadership style) in conflict situations, (c) the

organizational culture/climate that the leader is currently working in, and (d) the

possibility that the leadership theories and literatures may inadequately describe

the traits and behaviors of leaders.

With regard to the self-reporting nature of the research instruments, it is

possible that leaders’ responses on the research instruments may have reflected

how they perceive themselves behaving or how they would like to behave, rather

than their actual behavior. It is also possible that leaders may depart from their

normal leadership style when faced with conflict. A leader may have answered

the questions on the MLQ with regard to their leadership behavior in situations

not involving conflict. Organizational climate/culture may also influence the

conflict management styles of leaders. Organizational culture can also influence

how conflict is resolved. Some organizations may encourage leaders to

compromise while other organizations may encourage leaders to avoid conflict.

 66

It is possible that leadership style may not be the predominant factor that affects

conflict management style.

The possibility also exists that leadership theories may inadequately

describe the traits and behaviors of leaders. For example, Burns (1978)

describes transactional leaders as those who negotiate and bargain. The

leadership actions of negotiating and bargaining logically lead to compromising.

Thus, the theory of Transactional Leadership implies that transactional leaders

will resolve conflict through an act of Comprising. It is logical that a positive

relationship should exist between leaders’ scores on Transactional Leadership

and their scores on the Compromising style of conflict management. The results

of this study do not show a positive correlation; instead, the results show a

negative correlation between leaders’ scores on Transactional Leadership and

their scores on the Compromising style of conflict management. Therefore, the

possibility exists that leadership theories need further refinement.

Implications of Findings

According to Figure 1, use of the Compromising conflict management

style indicates that when resolving conflict, the leader is moderately concerned

with his or her own needs and moderately concerned with the needs of others.

The Compromising style of conflict management is used when the leader is

somewhat assertive and somewhat cooperative. The Competing style is used

when the leader is extremely assertive and extremely uncooperative. The

Collaborating style is used when the leader is extremely assertive and extremely

cooperative. The Avoiding style is used when the leader is extremely unassertive
 67

and extremely uncooperative. Finally, the Accommodating style is used when the

leader is extremely unassertive yet extremely cooperative.

Chapter 4 of this study presents minimum, maximum, and mean scores

for 99 leaders (participants in this exploratory study) on transactional,

transformational, and laissez-faire leadership and the five conflict management

styles: competing, collaborating, compromising, avoiding, and accommodating.

Tables 3 and 4 provide frequency distributions for leadership styles and conflict

management styles. This study found Compromising (see Table 2) to be the first

most frequently used conflict management style. Compromising results in a

partial win-win situation for both parties; however, Collaboration may provide

better results for both parties.

Table 2 also shows that Avoiding is the second most frequently used

conflict management style. This may reflect a leader’s desire to obtain more

information before making a decision. This could also indicate that a leader

prefers to avoid conflict. It is possible that leaders in this study use the Avoiding

style too frequently and the Collaborating style too infrequently. Use of the

Collaborating conflict management style provides the ultimate win-win situation

for both parties. Due to the high scores on Avoiding, it is evident that leaders do

not realize the inherent dangers of overuse of this conflict management style.

Leaders who scored in the high range on Avoiding should consider the possibility

that their organizations may be suffering as a result of their own lack of input

(Thomas & Kilmann, 2002, p. 14). Leaders who scored in the high range on

Avoiding should also ask: “Are decisions on important issues sometimes made
 68

by default” (p. 14)? Organizations should provide leaders with training programs

or workshops to address strategies for dealing with conflict. Organizations should

encourage leaders to take advantage of opportunities such as seminars and

workshops that focus on conflict management strategies. Organizations should

critically examine the conflict management goals and techniques (modus

operandi) promoted in their organizations. It is imperative that leaders be

properly trained in conflict management strategies in order to help the

organization achieve its mission.

Recommendations for Future Research

Future studies exploring leadership styles and conflict management styles

should consider using additional variables such as level of leadership, occupation

(accountant, attorney, doctor, nurse, teacher, policeman, fireman, minister, etc.),

and organizational culture/climate. These additional variables may provide more

insight regarding the relationship between leadership styles and conflict

management styles. Future studies using the MLQ could also use the

subcategories of the leadership styles as opposed to the three summary scores.

For example, the MLQ provides a summary score for Transformational

Leadership, but one could also examine the individual scores for the five

components of Transformational Leadership. The five components of

Transformational Leadership are as follows: (a) Idealized Influence (Attributed),

(b) Idealized Influence (Behavior), (c) Inspirational Motivation, (d) Intellectual

Stimulation, and (e) Individualized Consideration (Bass & Avolio, 1995, 2000).
 69

Additional insight regarding leadership styles and conflict management styles

may help leaders more efficiently and more effectively resolve conflict.
 70

References

Bass, B. M. (1985). Leadership and performance beyond expectations. New

York: The Free Press.

Bass, B. M. (1997). Does the transactional/transformational leadership 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 Psychologists Press.

Bass, B. M., & Avolio, B. J. (Eds.). (1994). Introduction. In B.M. Bass, & B.J.

Avolio (Eds.), Improving organizational effectiveness through

transformational leadership (pp. 1-9). Thousand Oaks, CA: Sage

Bass, B. M., & Avolio, B. J. (1995, 2000). Multifactor leadership questionnaire.

Redwood City, CA: Mind Garden, Inc.

Bennis, W., & Nanus, B. (1997). Leaders: The strategies for taking charge. New

York: Harper & Row.

Bessai, F. (1995). Review of the Multifactor Leadership Questionnaire. In J. C.

Conoley & J. C. Impara (Eds.), The twelfth mental measurements

yearbook (pp. 650-654). Lincoln: University of Nebraska - Buros Institute

of Mental Measurements.

Blake, R. R., & Mouton, J. S. (1964). The managerial grid. Houston: Gulf.

Blake, R. R., & Mouton, J. S. (1978). The new managerial grid (2nd ed.).

Houston: Gulf.
 71

Blake, R. R., & Mouton, J. S. (1985). The managerial grid III (3rd ed.). Houston:

Gulf.

Burns, J. M. (1978). Leadership. New York: Harper & Row.

Darling, J. R., & Walker, W. E. (2001). Effective conflict management: use of the

behavioral style model. Leadership & Organization Development Journal,

22(5), 230-242.

DuBrin, A. J. (2004). Leadership (4th ed.). New York: Houghton Mifflin.

Ford, L. (1991). Transforming leadership: Jesus' way of creating vision, shaping

values, & empowering change. Downers Grove, IL: InterVarsity Press.

Hair, J. F., Jr.. Anderson, R. E., Tatham, R. L., & Black, W. C. (1998).

Mulitvariate data analysis (5th ed.). Upper Saddle River, NJ: Prentice Hall.

Hartog, D.,Muijen, J., & Koopman, P. (1997). Transactional versus

transformational leadership: An analysis of the MLQ. Journal of

Occupational and Organizational Psychology, 70, 19-34.

Hersey, P., & Blanchard, K. H. (1969a). Life cycle theory of leadership. Training

& Development Journal, 23, 26-34.

Hersey, P., & Blanchard, K. H. (1969b). Management of organizational behavior.

Englewood Cliffs, NJ: Prentice-Hall.

Hersey, P., & Blanchard, K. H. (1974). So you want to know your leadership

style? Training & Development Journal, 28(2), 22-37.

Hersey, P., & Blanchard, K. H. (1982). Leadership style: Attitudes and

behaviors. Training & Development Journal, 36(5), 50-52.

 72

Hersey, P., Blanchard, K. H., & Johnson, D. E. (1996). Management of

organizational behavior: Utilizing human resources (7th ed.). Upper

Saddle River, NJ: Prentice-Hall.

Holt, T.L. (1986). The relationship between leadership styles and conflict

management techniques as reported by hospital middle management

personnel in eight of the larges hospitals in North Dakota (Doctoral

dissertation, The University of North Dakota, 1986). Dissertation

Abstracts International, 47 (10A), 3761.

Hotelling, H. (1935). The most predictable criterion. Journal of Experimental

Psychology 26, 139-142.

Kabanoff, B. (1987). Predictive validity of the MODE conflict instrument. Journal

of Applied Psychology, 72(1), 160-163.

Kilmann, R. H. and Thomas, K. W. (1975). Interpersonal conflict-handling

behavior as reflections of Jungian personality dimensions. Psychological

Reports, 37, 971-980.

Kilmann, R. H., & Thomas, K. W. (1977). Developing a forced-choice measure of

conflict-handling behavior: The “MODE” instrument. Educational and

Psychological Measurement, 37, 309-325.

Kilmann, R. H., & Thomas, K. W. (1978). Four perspectives on conflict

management: An attributional framework for organizing descriptive and

normative theory. Academy of Management Review, 3(1), 59-68.

Kirnan, J. P., & Snyder, B. (1995). Review of the Multifactor Leadership

Questionnaire. In J. C. Conoley & J. C. Impara (Eds.), The twelfth mental

 73

measurements yearbook (pp. 651-654). Lincoln, NE: Buros Institute of

Mental Measurements - The University of Nebraska-Lincoln.

Licht, M. H. (1998). Multiple regression and correlation. In L. G. Grimm & P. R.

Yarnold (Eds.), Reading and understanding multivariate statistics (pp. 19-

64). Washington, DC: American Psychological Association.

Marion, L. A. (1995). Conflict management and personality type among

community college executives (Doctoral dissertation, State University of

New York at Albany, 1995). Dissertation Abstracts International, 56 (05A),

1606.

Mills, J., Robey, D., & Smith, L. (1985). Conflict-handling and personality

dimensions of project-management personnel. Psychological Reports, 57,

1135-1143.

Rahim, M. A. (1981). Organizational behavior courses for graduate students in

business administration: Views from the tower and battlefield.

Psychological Reports, 49, 583-592.

Romero, M. G. (1983). The relationship of conflict management style to the

leadership style of secondary school principals in their role as middle

managers (Doctoral dissertation, University of Colorado, 1983).

Dissertation Abstracts International, 44 (10A), 2983.

Schmidt, J. E. (1993). Transformational leadership: The relationship between

consciousness, values, and skills (Doctoral dissertation, Western Michigan

University, 1993). Dissertation Abstracts International, 54 (11A), 4057.

 74

Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd ed.).

New York: HarperCollins.

Thomas, K. (1976). Conflict and conflict management. In M. D. Dunnette (Ed.),

Handbook of industrial and organizational psychology (pp. 889-935).

Chicago: Rand McNally.

Thomas, K. W. (1992). Conflict and negotiation processes in organizations. In M.

D. Dunnette & L. M. Hough (Eds.), Handbook of industrial and

organizational psychology (pp. 651-717). Palo Alto, CA: Consulting

Psychologists Press.

Thomas, K. W., & Kilmann, R. H. (1974, 2002). Thomas-Kilmann conflict MODE

instrument. Tuxedo, NY: Xicom.

Thomas, K. W., & Kilmann, R. H. (1975). The social desirability variable in

organizational research: An alternative explanation for reported findings.

Academy of Management Journal, 18(4), 741-752.

Thomas, K.W., & Kilmann, R.H. (1978). Comparison of four instruments

measuring conflict behavior. Psychological Reports, 42, 1139-1145.

Thompson, B. (1980). Canonical correlation: Recent extensions for modelling

educational processes Boston, MA: 64th Annual Meeting of the American

Educational Research Association. (ERIC Document Reproduction

Service No. ED199269)

Volkema, R. J., & Bergmann, T. J. (1995). Conflict styles as indicators of

behavioral patterns in interpersonal conflicts. Journal of Social

Psychology, 135(1), 5-15.

 75

Walton, R. E., & McKersie, R. B. (1965). A behavioral theory of labor

negotiations. New York: McGraw-Hill.

Yammarino, F., Spangler, W., & Bass, B. (1993). Transformational leadership

and performance: A longitudinal investigation. Leadership Quarterly, 4(1),

81-102.

Zigarelli, M. A. (2002). Ordinary people, extraordinary leaders. Gainesville, FL:

Synergy.
 76

Appendix

Demographic Questionnaire

1. Type(s) of organization(s) in which you serve as a leader (e.g., industry,

nonprofit, governmental) ________________________________

2. Title of leader __________________________

3. Gender of leader ____________

4. Approximate number of months served in leadership position(s) _________

5. Approximate size of organization(s) in $ _____________

6. Number of people in organization(s) ___________

7. Number of people who report to you __________

 77

Letter to Emory & Henry Alums

June 2, 2004

Dear

Thank you for agreeing to help me in the data gathering process of my

dissertation.

I need your help in identifying leaders for my study. Please select leaders in your
workplace, church, or community. For purposes of this study, a leader is a
person who guides or instructs others.

Enclosed are several small white envelopes. Each of these small white
envelopes contains research instruments that the leaders will complete. Please
give each leader one white envelope and ask them to complete the research
instruments, seal the envelope, and return the envelope to you. I have also
included instructions for the leaders in their envelope.

If leaders do not return these to you in a few days, please be sure to follow up
with the leader to ensure his/her completion of the research instruments. The
leaders should return the white envelopes to you by Friday, June 11, 2004. After
you have received the completed instruments in the small white sealed
envelopes, please put all of the small white envelopes into the large green and
white envelope and return to me. Please mail the large green and white
envelope to me as soon as possible, preferably no later than Monday, June 14,
2004.

If you have any questions, please e-mail me at dstanley@ehc.edu or call me at

276.944.6187.

Thanks again for your help.

Sincerely,

A. Denise Stanley, MACCT, CPA

 78

Letter to Leaders Participating in this Study

June 2, 2004

Dear leader,

Thank you for your willingness to participate in my research study.

I am a doctoral candidate examining the relationship (if any) between leadership

styles and conflict management styles. Currently, I am in the process of
gathering data that will be used in the research analysis section of my
dissertation.

Please complete the three forms enclosed (Demographic Questionnaire,

Multifactor Leadership Questionnaire, and the Thomas-Kilmann Conflict Mode
Instrument). Then, put the completed forms back in the white envelope provided.
Please seal the envelope and return it to the person who gave it to you by Friday,
June 11, 2004.

Please note that all of your responses are confidential. While instruments and
envelopes are coded with numbers, these codes will only be used to link the
multiple instruments being administered.

Please note that completion of these forms is considered your consent to

participate in this study. If you have any questions, please e-mail me at
dstanley@ehc.edu or call me at 276.944.6187.

Thanks again for your participation in my research study.

Sincerely,

A. Denise Stanley, MACCT, CPA

 79

Permission to Cite Conflict Handling Modes Figure (Figure 1)

 80

MLQ Duplication Permission

 81

Permission to Include Sample Items from MLQ

 82

Sample Items from MLQ

 83

Permission to Include Sample Items from TKI

84
 85

Sample Items from TKI

 86

Competing 10

10 20 30 40
TR

10
Competing

40 50 60 70 80
TF

10
Competing

0 5 10
LF

Figure 2. Scatterplots of TKI Scores Versus MLQ Scores

 87

12

Collaborating

10 20 30 40
TR

12
Collaborating

40 50 60 70 80
TF

12
Collaborating

0 5 10
LF

Figure 2. (Continued)
 88

12
11
10
9
Compromising

8
7
6
5
4
3

10 20 30 40
TR

12
11
10
9
Compromising

8
7
6
5
4
3

40 50 60 70 80
TF

12
11
10
9
Compromising

8
7
6
5
4
3

0 5 10
LF

Figure 2. (Continued)
 89

12

Avoiding

10 20 30 40
TR

12
Avoiding

40 50 60 70 80
TF

12
Avoiding

0 5 10
LF

Figure 2. (Continued)
 90

10

Accommodating

10 20 30 40
TR

10
Accommodating

40 50 60 70 80
TF

10
Accommodating

0 5 10
LF

Figure 2. (Continued)
 91

-1

-2
MLQ1

-3

-4

-5

10 20 30 40
TR

-1

-2
MLQ1

-3

-4

-5

40 50 60 70 80
TF

-1

-2
MLQ1

-3

-4

-5

0 5 10
LF

Figure 3. Scatterplots of MLQ1 Variate with MLQ Scores

 92

-1

-2
MLQ1

-3

-4

-5

0 5 10
Competing

-1

-2
MLQ1

-3

-4

-5

2 7 12
Collaborating

-1

-2
MLQ1

-3

-4

-5

3 4 5 6 7 8 9 10 11 12
Compromising

Figure 4. Scatterplots of MLQ1 Variate with TKI Scores

 93

-1

-2
MLQ1

-3

-4

-5

2 7 12
Avoiding

-1

-2
MLQ1

-3

-4

-5

0 5 10
Accommodating

Figure 4. (Continued)
 94

20.5

19.5

18.5
TKI1

17.5

16.5

15.5
0 5 10
Competing

20.5

19.5

18.5
TKI1

17.5

16.5

15.5
2 7 12
Collaborating

20.5

19.5

18.5
TKI1

17.5

16.5

15.5
3 4 5 6 7 8 9 10 11 12
Compromising

Figure 5. Scatterplots of TKI1 Variate with TKI Scores

 95

20.5

19.5

18.5
TKI1

17.5

16.5

15.5
2 7 12
Avoiding

20.5

19.5

18.5
TKI1

17.5

16.5

15.5
0 5 10
Accommodating

Figure 5. (Continued)
 96

20.5

19.5

18.5
TKI1

17.5

16.5

15.5
10 20 30 40
TR

20.5

19.5

18.5
TKI1

17.5

16.5

15.5
40 50 60 70 80
TF

20.5

19.5

18.5
TKI1

17.5

16.5

15.5
0 5 10
LF

Figure 6. Scatterplots of TKI1 Variate with MLQ Scores

 97

20.5

19.5

18.5
TKI1

17.5

16.5

15.5
-5 -4 -3 -2 -1 0
MLQ1

87

86

85
TKI2

84

83

82

81
-5 -4 -3 -2 -1 0
MLQ2

-33

-34

-35
TKI3

-36

-37

5 6 7 8 9 10
MLQ3

Figure 7. Scatterplots of TKI and MLQ Variates