 

3 Hzﬂm A
,1 Fur... ..

‘ dunioeuz...

 

:1.
.133...“
‘ .

 

ﬁgéxzfgﬁaﬂt

tbiL'-.3

" N
I )L

This is to certify that the

dissertation entitled

A SOCIAL COGNITIVE PERSPECTIVE OF MOTIVATIONAL AND
SELF-REGULATORY MECHANISMS OF LEADERSHIP
IN FEMALE COLLEGIATE ROWERS

presented by

T. Michelle Magyar

has been accepted towards fulﬁllment
of the requirements for

Ph . D. degree in KIIIESIOIORY

 

 

:igﬂmiggg

ajor professor
Date 0 (MN ‘1! Q0 03...

 

MSU i: an Afﬁrmative Action/Equal Opportunity Institution O~12771

 

 

_ LIBRARY
Michigan State
University

 

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

If}? 0 3 233

 

93 21 U7 "'Fiﬂgrf'zﬁné

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDuopGS-p. 15

A SOCIAL COGNITIVE PERSPECTIVE OF MOTIVATIONAL AND
SELF-REGULATORY MECHANISMS OF LEADERSHIP
IN FEMALE COLLEGIATE ROWERS

By

T. Michelle Magyar

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Kinesiology

2002

ABSTRACT

A SOCIAL COGNITIVE PERSPECTIVE OF MOTIVATIONAL AND SELF-
REGULATORY MECHANISMS OF LEADERSHIP IN COLLEGIATE ROWERS

By

T. Michelle Magyar
The purpose of this investigation was to examine the motivational and self-regulatory
mechanisms of leadership in collegiate rowers. Participants were 367 female
intercollegiate rowers ages 18-37 (M = 19.75). Rowers completed a demographic
questionnaire, a reduced version of the Bem- Sex Role Inventory (R-BSRI; Bem, 1974;
Covey & Feltz, 1991), a modiﬁed version of the Task and Ego Orientation in Sport
Questionnaire (M-TEOSQ; Duda & Nicholls, 1992), and a modiﬁed version of the
Perceived Motivational Climate in Sport Questionnaire (M-PMCSQ-Z; Newton, Duda, &
Yin, 2000). Regulatory mechanisms of leadership were measured with constructs
developed for the purpose of this research to assess leadership skills, and efﬁcacy beliefs
(e. g., leadership efﬁcacy, task self-efﬁcacy, and collective efﬁcacy). Leadership effects
was classiﬁed into two dimensions, performance and motivational. Performance
leadership was operationalized as someone who is considered to be the “go to” person, is
competent, masterful, assertive, conﬁdent, and may lead a boat toward a successful
performance outcome. Successful performance outcome was operationlized as an
improved race time or winning a race. Motivational leadership was operationalized as
someone who encourages teammates to stay tough and work through the pain (i.e., on the
erg or during a race), resolves conﬂict between members of the boat, acts unselﬁshly,
shows concern for others, or helps teammates calm their nerves before testing and

competitions.

The conceptual model was tested using path analysis. Results from this analysis
demonstrated that leader goal orientation and leadership efficacy emerged as the
strongest predictors of leader effectiveness. Speciﬁcally, athletes who reported greater
leader goal orientation and leadership efficacy obtained higher scores on leader
effectiveness from their teammates. Perceptions of mastery motivational climate also
demonstrated a consistent and signiﬁcant relationship with leadership skills in sport.
Leadership efﬁcacy demonstrated the strongest mediating effect between personal and
situational determinants with leader effects. Future research should examine leadership
over the entire course of the season in order to assess the emergent patterns that may

OCCUI’.

DEDICATION

To my family with love.

iv

ACKNOWLEDGMENTS

A special thank you to my advisor and dissertation committee chair Dr. Deborah Feltz.
Thank you to my dissertation committee members, Drs. Carole Ames, Marty Ewing, and
Mark Reckase for providing guidance and direction on this project.

My admiration and gratitude go to the many athletes and coaches who donated their time
and effort to participate in the study.

To my sister Janine, my grandparents, and my best friends Ian, Peggy, Lynette, and Lori,
thank you for your support. Finally, I would like to express my deepest gratitude to my
parents Jeff and Linda. I could not have accomplished this without your love and support.

TABLE OF CONTENTS

LIST OF APPENDICES ......................................................... viii

LIST OF TABLES ................................................................ ix

LIST OF FIGURES ............................................................... xi

GLOSSARY ....................................................................... xii

CHAPTER 1

INTRODUCTION
Nature of the Problem ........................................................ 1
Statement of the Problem ..................................................... 7
Overview of Social-Cognitive Correlates of Leadership ................. 9
Contextual Factors in the Current Study .................................. 17
Conceptual Model of Leadership in Athletes ............................ 18
Research Questions .......................................................... 20
Conceptual Model Hypotheses ............................................. 24
Additional Hypotheses ....................................................... 23
Basic Assumptions ............................................................ 24
Delimitations .................................................................. 25

CHAPTER 2

REVIEW OF LITERATURE
Trait/Personality Perspective ................................................ 26
Behavioral Perspective ........................................................ 28
Situational Perspective ........................................................ 29
Contingency Perspective ..................................................... 31
Transactional/Transformational Perspective .............................. 35
Conclusions ................................................................... 38
Social Cognitive Theory ..................................................... 41
Social Cognitive Theory of Leadership ..................................... 43
Leadership Effects ............................................................. 44
Self-Regulatory Mechanisms of Leadership .............................. 46
Social Cognitive Determinants of Leadership .............................. 50
Gender Roles ................................................................... 51
Gender and Leadership ........................................................ 55
Gender Roles in Sport ......................................................... 58
Achievement Goals ........................................................... 60
Structural Goals ................................................................ 66
Conclusion ...................................................................... 73

CHAPTER 3

METHODOLOGY
Participants ..................................................................... 74
Measures ....................................................................... 76
Procedures ..................................................................... 85
Treatment of the Data ....................................................... 86

CHAPTER 4

RESULTS
Demographic Information .................................................... 88
Descriptive Statistics .......................................................... 89
Leader Identification and Evaluation Patterns ............................ 90
Background Inﬂuences on Perceptions of Leadership ................... 93
Confirmatory Factor Analyses .............................................. 93
Path Analysis .................................................................. 100
Hypothesis 1 ................................................................... 105
Hypothesis 2 ................................................................... 108
Hypothesis 3 ................................................................... 111
Hypothesis 4 ................................................................... 113
Hypothesis 5 ................................................................... l 15
Hypothesis 6 ................................................................... 1 17
Hypothesis 7 ................................................................... 123
Hypothesis 8 ................................................................... 127

CHAPTER 5

DISCUSSION ...................................................................... 133

APPENDICES ..................................................................... 143

TABLES ............................................................................ 165

REFERENCES ................................................................... 177

vii

LIST OF APPENDICES

Page
Appendix A. Demographic Background Questionnaire ...................... 144
Appendix B. Gender Role Orientations ......................................... 146
Appendix C. Goal Orientations .................................................. 147
Appendix D. Motivational Climate ............................................. 148
Appendix E. Task Self-efﬁcacy .................................................. 149
Appendix F. Leader Efﬁcacy .................................................... 150
Appendix G. Leadership Skills .................................................. 151
Appendix H. Performance Leadership Identiﬁcation ........................ 153
Appendix I. Motivational Leader Identiﬁcation ............................... 154
Appendix J. Peer Performance Leader Evaluation ............................ 155
Appendix K. Peer Motivational Leader Evaluation .......................... 156
Appendix L. Self Performance and Motivational Leader Evaluation ...... 157
Appendix M. Crew Eﬂicacy ...................................................... 158
Appendix N. Coach’s Packet ..................................................... 159

viii

LIST OF TABLES

TABLE O-l. Descriptive Statistics of Self-Regulatory Mechanisms

of Leadership .............................................................
TABLE O-2. Descriptive Statistics of Leader Effects ..............................
TABLE 0-3. Pearson Product Moment Correlations Among Leader Evaluation

Scores, Self and Peer Leader Effectiveness Scores and Ranks ......
TABLE O-4. Distribution of Leader Rankings by Seat on

Performance Leadership ................................................

TABLE O-5. Distribution of Motivational Leader Rankings by Seat .............

TABLE O-6. Pearson Product Moment and Bivariate Correlations of Personal
Demographic Determinants and Leadership ...........................

TABLE O-7. Leader Evaluation Fit Indices, Standardized Parameter Estimates,
Standard Error, and T-values .............................................

TABLE O-8. Efﬁcacy Fit Indices, Standardized Parameter Estimates, Standard
Error, and T-values ......................................................

TABLE 0—9. A-LLSQ Fit Indices, Standardized Parameter Estimates, Standard
Error, and T-values ......................................................

TABLE O-lO. M-TEOSQ Fit Indices, Standardized Parameter Estimates,
Standard Error, and T-values ..........................................

TABLE O-l l. M-PMCSQ-2 Fit Indices, Standardized Parameter Estimates,
Standard Error, and T-values ..........................................

TABLE 1. Correlations, Means, and Standard Deviations for Path Model .......
TABLE 2. Direct Effects for Hypothesis 1 ............................................
TABLE 3. Direct Effects for Hypothesis 2 ...........................................
TABLE 4. Direct Effects for Hypothesis 3 ...........................................
TABLE 5. Direct Effects for Hypothesis 4 .............................................

TABLE 6. Direct Effects for Hypothesis 5 ...........................................

Page
166

167

168

169

170

17]

172

173

174

175

176
104
106
109
111
113

115

Page
TABLE 7. Direct, Indirect, and Total Effects of Personal and Situational
Determinants on Team and Leader Effects ................................ 118

TABLE 8. Descriptive Statistics for Athletes by Class Standing
and Previous Leadership Experience ........................................ 125

TABLE 9. Descriptive Statistics for Athlete ranked #1 Performance Leader

versus Athletes Not Ranked #1 in Performance Leadership ............. 128

TABLE 10. Descriptive Statistics for Athlete ranked #1 Motivational Leader
versus Athletes Not Ranked #1 in Motivational Leadership ........... 130
TABLE 11. Correlates of Coach, Peer, and Self Leadership Rankings ............ 132

 

 

F1

fl

LIST OF FIGURES
Page

FIGURE 1. Social Cognitive Conceptual Model of Leadership in Athletes... 2

FIGURE 2. Example 2-factor Conﬁrmatory Factor Analysis Model ............. 95

FIGURE 3. Illustration of Full Hypothesized Model ................................ 102
FIGURE 4. Illustration of Direct Effects for Hypothesis 1 ........................ 107
FIGURE 5. Illustration of Direct Effects for Hypothesis 2 ........................ 110
FIGURE 6. Illustration of Direct Effects for Hypothesis 3 ........................ 112
FIGURE 7. Illustration of Direct Effects for Hypothesis 4 ........................ 114
FIGURE 8. Illustration of Direct Effects for Hypothesis 5 ........................ 116

FIGURE 9. Direct and Indirect Pathways from Agentic Attributes to Peer
Scores of Leader Effectiveness ........................................... 1 19

FIGURE 10. Direct and Indirect Pathways from Agentic Attributes to Self
Scores of Leader Effectiveness ........................................ 120

FIGURE 11. Direct and Indirect Pathways from Agentic Attributes to
Collective Efﬁcacy ....................................................... 121

FIGURE 12. Direct and Indirect Pathways from Communal Attributes to Self
Scores of Leader Effectiveness ........................................ 122

xi

 

 

1m

Ag

be‘

be

ex

50

SU

Le

ex
1.:
te.
Le
di
PE
.11
lea
C0

C0

Glossary
Independent Variables: Hypothesized Determinants
Agentic Attributes: characteristics that are used to describe stereotypic masculine
behavior (e. g., independent, masterﬁrl, assertive, and competent).
Communal Attributes: characteristics that are used to describe stereotypic feminine
behavior (e. g., friendly to others, unselﬁsh, concerned for others, and emotionally
expressive).
Ego Orientation: the dispositional tendency to use normative referenced standards (i.e.,
social comparison) for evaluating personal ability, task difficulty, performance, and
successful experiences in sport.
Leadership Climate: perceptions that depict the extent to which the coach promotes
leadership by encouraging the use of leadership skills and providing leadership
experiences for athletes.
Leadership Experience: previous experience with a leadership role of any kind (e. g.,
team captain, student council).
Leader Orientation: the dispositional tendency to use leader-referenced standards (i.e,
directing group members/others) for evaluating personal ability, task difﬁculty,
performance, and successful experiences in sport.
Mastery Climate: perceptions that reﬂect the extent to which a coach emphasizes
learning and improvement, makes each athlete feel like he or she provides an important
contribution to the success of the team, and encourages all athletes to work together

collectively.

xii

 

 

Per
riVa
CITE
R0)
Rm
for .
T as
pets
SUC(
M cl

Lea

silo!

dim:

dlmf
Ray;
and (

main:

Perfo;

Performance Climate: perceptions that represent the degree to which a coach encourages
rivalry among athletes on the same team, uses punitive tactics in response to performance
errors, and provides differential attention (e. g., favoritism) to athletes.

Rowing Experience: months/years of experience in the sport of rowing.

Rowing Ability: an objective assessment of rowing ability using an individual’s time
for a 2000 meter (2K) rowing ergometer test.

Task Orientation: the dispositional tendency to use self-referenced standards (i.e.,
personal mastery) for evaluating personal ability, task difﬁculty, performance, and
successful experiences in sport.

Mediating Variables: Hypothesized Regulatory Mechanisms

Leadership Efficacy: An individual’s belief in his/her ability to successfully lead the
group to a certain degree or level (Bandura, 1997; Chemers, 1997).
Motivational-Interpersonal Leadership Skills: skills that emphasize interpersonal
dimension of leadership (e. g., console teammates when they are frustrated).

Negative Tactics Leadership Skills: skills that reﬂect a leader’s use of negative tactics
(e.g., encourage teammates to win at all costs).

Performance-Execution Leadership Skills: skills that emphasize performance
dimensions of leadership (e. g., set performance goals for the team).
Respect-Communication Leadership Skills: skills that reﬂect a leader’s use of respect
and communication (e. g. communicate effectively with teammates) to develop and
maintain leadership.

Task Self-Efﬁcacy: an individual’s belief in his/her capability to successfully execute a

performance skill to a certain degree or level (Bandura, 1997).

Dependent Variables: Hypothesized Effects

Collective Efﬁcacy: individual perceptions regarding the boat’s collective belief in its
capability to successfully execute a performance skill to a certain degree or level
(Bandura, 1997).

Evaluation of #1 Motivational Leader: a score that denotes an athlete’s evaluation of
his/her #1 ranked motivational leader’s stability as a leader in the boat.

Evaluation of #1 Performance Leader: 3 score that denotes an athlete’s evaluation of
his/her #1 ranked performance leader’s stability as a leader in the boat.

Motivational Leader: the person who is considered to be inspirational, friendly, acts
unselﬁshly, resolves conﬂict between members of the boat, and encourages teammates to
work together.

Performance Leader: the person who is considered the “go to” person, is competent,
masterful, assertive, conﬁdent, and may lead a group toward a successful performance
outcome.

Peer-Rank Motivational Leader Effectiveness: a score that represents the average of
rankings provided by peers in the boat regarding an athlete’s rank as a motivational
leader

Peer-Rank Performance Leader Effectiveness: a score that represents the average of
rankings provided by peers in the boat regarding an athlete’s rank as a performance
leader (e. g., average rank of 2 out of 9).

Peer-Score Motivational Leader Effectiveness: a score that reﬂects the average of scores
provided by peers in the boat regarding an athlete’s effectiveness as a motivational

leader.

xiv

 

 

Pee;
mer
Sal}
of c
3rd
561,
of <
leg
Sel
his
Sel
his
Se.
ev;
Set
CV2
Su
Crl
cor

Ch

Peer-Score Performance Leader Effectiveness: an averaged score that reﬂects boat
members’ (peers) perceptions of an athlete’s effectiveness as a performance leader.
Self-Rank Motivational Leader Effectiveness: a rank that represents an athlete’s ranking
of oneself relative to the group, of his/her ability as a motivational leader (e. g., ranked
3rd out of 9 athletes).

Self-Rank Performance Leader Effectiveness: a rank that represents an athlete’ 5 ranking
of oneself relative to the group, of his/her perceived effectiveness as a performance leader
(e. g., ranked 3rd out of 9 athletes).

Self-Score Performance Leader Effectiveness: a score that reﬂects an athlete’s rating of
his/her perceived effectiveness as a performance leader.

Self-Score Motivational Leader Effectiveness: a score that reﬂects an athlete’s rating of
his/her perceived effectiveness as a motivational leader.

Self-Evaluation of Performance Leader Ability: a score that reﬂects an athlete’s
evaluation of his/her own consistency of performance leadership in the boat.
Self-Evaluation of Motivational Leader Ability: a score that reﬂects an athlete’s
evaluation of his/her own consistency of motivational leadership in the boat.
Supplemental Terms

Crew Boat: collective members of a speciﬁc boat. This study examined the members of
coxed eight (8+) and coxed four (4+).

Crew Team: the collection of individuals making up the school rowing team.

 

 

 

 

mana
regar
main
leade
ofle
CODU
achic
Whit

C001

lead.
This
follo
53ml
arhle
r’31er

163d<

thou:

Chapter 1
INTRODUCTION
Nature of the Problem

Despite the extensive lines of research on leadership in the social behavior,
management, and sport science domains, 3 lack of consensus exists among them
regarding the conceptual deﬁnition and essential features of effective leadership. The
main reason for these divergent perspectives is due to the inherent complexity of the
leadership process. However, researchers from different viewpoints agree that the process
of leadership involves social inﬂuence between a leader and followers, occurs in the
context of a group, and reﬂects the aspiration of the group to attain collective action and
achievement. Thus, in its purest form, leadership can be deﬁned as a social process in
which one individual inﬂuences a group of individuals toward the accomplishment of a
common goal or objective (Bass, 1990; Chemers, 1997, 2000).

The purpose of this dissertation is to develop and test a conceptual model of
leadership in athletes within the framework of social cognitive theory (Bandura, 1997).
This chapter provides a general overview of research on leadership speciﬁc to athletes,
followed by a brief synopsis of the theoretical tenets of social cognitive theory. This
synthesis is provided as the backdrop from which a conceptual model of leadership in
athletes (see Figure 1 on p. 2) was proposed in order to test the hypothesized, theoretical
relationships between the determinants, self-regulatory mechanisms, and effects of
leadership in athletes.

Drawing from the sport psychology literature, leadership has been traditionally

thought of as a behavioral process that inﬂuences individuals and groups toward

cart}. 5 7:22. a. CI:_:<C n. 6.7.2.53 LSCQE C. Teena—.23.: _.: >::c.c.v.

 

 

 

bmnmwaaaaaa ammiﬁed. Emnreammg #33 ER hug meSE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

52:833.?
wen—£35...
D wee... mun—512.8
D banana—nu
aux—312.8
D was...» 3.5 5333.. mg
0232. we? , U Hansen—wean?
U >mo==n 532:5: f
D 02:55.». U 3378339.-
ﬁmaeai 38:35:...»— Fauna—dz:
Gee—012.322. D wamu.anQ.OeE D 32:58:: All
D Hear 55.5.»:3 U mannaatoaoum
D Mme U Zena—.36 D Mgr—nae:
U roman—.3? dean—8
Havana—:1
Manes~
H25.
3379:39— QEES Hem—m U 03.296
D 32.3 M328
hmuzezezi D male—“3.8 $3.03an a AL
0 .33an: 1

 

 

 

 

 

EMER N. moot: 00mins“ 09888»— Zoai ow romaoaau 5 >333

 

 

emabhs
focusin
in spon
proces:
posnio
\Vhﬂe;
& C he]

hasrec

design;
to be a:
relative
reasons
examir
team P
Were a
the tEa
COmpa
imerac

DOSitio

established goals (Chelladurai, 1984; Chelladurai & Riemer, 1998). Although by
focusing only on the behavioral mechanisms of leadership, the commonly used deﬁnition
in sport has neglected to consider the social and cognitive dimensions of the leadership
process. Furthermore, the inherent hierarchical power structure of a sport team has
positioned the coach in the central leadership role exceedingly above his/her athletes.
While ample research has been conducted on the leadership role of the coach (see Reimer
& Chelladurai, 1995, for a review), the less clearly deﬁned leadership role of the athlete
has received limited empirical scrutiny.

Previous research on leadership in athletes has identiﬁed characteristics of
designated leaders or team captains. Initially, leadership speciﬁc to athletes was thought
to be associated with the centrality of position, or the actual location of the athlete
relative to his/her teammates (Grusky, 1963). Subsequent work identiﬁed additional
reasons why athletes were assigned a leadership position. Tropp and Landers (1979)
examined interpersonal attraction and leadership as a function of spatial positioning (i.e.,
team position and structural system) in female collegiate ﬁeld hockey players. Athletes
were asked to rate each team member on the extent to which the athlete was a leader on
the team. The authors compared athletes who received higher peer ratings on leadership
ability with the athletes who were scored lower on leadership ability by their peers. This
comparison revealed that athletes who performed in ﬁeld positions requiring lower
interaction with teammates received higher leader scores by their peers than athletes in
positions requiring moderate to high interaction. In particular, athletes playing the goalie
position were rated higher in leadership ability than any other position. Differences

between designated team captains and non-captains were also considered and revealed

 

 

 

that te
experi
The a‘

amou

indiv
form
(198
SOCi<
Part
athle
cons
locu
perf
tear
Wer
th

leve

res;
eac]

is sl

that team captains who scored higher on peer-rated leadership had more years of
experience at the varsity level, and were scored higher by their peers on attraction ratings.
The authors concluded that the nature of the task required by the athlete, rather than the
amount of interaction with teammates, explained leader emergence.

Opposing the concept of designated leaders was the notion of emergent leaders, or
individuals perceived as leaders. These leaders did not necessarily receive the title of a
formal leader, such as “team captain.” With regard to informal leadership, Yukelson et al.
(1983) identiﬁed certain characteristics of individuals classiﬁed as either high or low on
sociometn'c prestige and status among intercollegiate baseball and soccer teams.
Participants completed a nomination instrument that asked them to list the top ﬁve
athletes they considered leaders on the ﬁeld, in addition to the top ﬁve athletes they
considered friends off the ﬁeld. Athletes also completed a measure on intemal/external
locus of control. Coaches provided a score for each athlete on the overall level of
performance during the season. The authors discovered that for baseball and soccer
teams, athletes who scored higher on peer-rated leadership status were better performers,
were upperclassmen, and reported an internal locus of control. In contrast, individuals
who scored on the lower end of the peer-rated leadership distribution displayed lower
levels of performance, were lowerclassman, and reported an external locus of control.

Unfortunately, initial attempts to operationalize leadership speciﬁc to the athlete’s
responsibilities were marginal at best. Tropp and Landers (1979) asked athletes to rate
each teammate on leadership by responding to the question, “How much of a team leader
is she?” Yukelson et al. (1983) assessed leadership with the question, “Who do you look

to with admiration or for leadership within the group during practice or competitive

 

 

situ
the
lead

char

mea
athle

me]

and c
meas
the t<

harm

peGrs

situations?” Furthermore, early work on the athletic leadership role neglected to consider
the athlete’s self-perception of leadership ability and the coach’s perception of effective
leadership. While both of these studies discovered that leaders share common
characteristics (i.e., upperclassmen or years of experience), these qualities were
descriptive at best and failed to disclose how coaches and athletes constructed the
meaning of the athletic leadership role. Namely, the roles and responsibilities of the
athlete leader were not clearly deﬁned or associated with a theory that would examine the
mechanisms associated with leadership.

Rees and Segal (1984) were among the ﬁrst to provide an operational deﬁnition
and contrast the different dimensions of the athletic leadership role. Speciﬁcally, they
measured the instrumental leadership role (i.e., athlete(s) considered the best player(s) on
the team), and the expressive leadership role (i.e., player(s) who contribute to group
harmony). By examining more than one dimension of leadership, the authors were able to
assess leader role differentiation. Thus, they determined the extent to which different
types of leadership emerged, and the degree to which these leader roles were
differentiated (e. g., different athletes designated for each type of leadership) or integrated
(e.g., same athlete is identiﬁed as a leader in both dimensions). Each athlete was asked to
exclude himself and rank the top ﬁve instrumental leaders and the top ﬁve expressive
leaders on the team. Results demonstrated that ﬁrst string starters were designated as
instrumental leaders, while both ﬁrst and second string starters were identiﬁed as
expressive leaders. In addition to playing status, athletes designated as leaders by their

peers were more likely to have higher class standing and greater degrees of interpersonal

‘4-

 

 

 

dill

des

lea:
pers
earl
athl
app;
to IE
desi
“cor
amb
role
3558;
COlTr
mecl

leadc‘

ICadE

leade

fOHm

slimet

attraction. The authors also found a high degree of integration, with many athletes
designated as both instrumental and expressive leaders.

While Rees and Segal (1984) distinguished instrumental from expressive
leadership, they considered the perception of effective leadership only from the
perspective of followers. As a result, the exclusion of self-rankings and coach rankings in
earlier research depicted an asymmetrical representation of effective leadership in
athletes (Rees & Segal, 1984; Tropp & Landers, 1979; Yukelson et al., 1983). This
approach failed to provide an understanding of athletes’ perceptions of their own abilities
to lead a group, and how they viewed their personal leadership abilities relative to their
designated leaders. Moreover, the inﬂuence of “being the best athlete on the team” and
“contributing to group harmony” as responsibilities of the athlete leader provided a rather
ambiguous account of what athlete leaders were expected to do. Finally, the notion of
role differentiation was not based on theory. In essence, this atheoretical approach
assessing the followers’ perspectives provided an important account of the descriptive
correlates of leadership. However, Rees and Segal (1984) did not consider the
mechanisms by which conceptions of leadership were constructed and translated into
leader conduct and the normative expectations regarding leader responsibilities.

Glenn and Horn (1993) provided a comprehensive assessment of emergent team
leadership by having high school female soccer athletes rate their personal abilities to
lead the team, in addition to peer-rated and coach-rated assessments for each athlete’s
leadership ability. Their results demonstrated the disparity between coach, leader, and
follower expectations of effective leadership in athletes. Speciﬁcally, they found that

athletes who scored high on self-reported leadership ability also reported higher scores on

 

lﬁaS

rece

“HS
IUOI

Per :

con
und
con
5UP]

COn.

have

masculinity, femininity, and perceived soccer competence. In contrast, athletes who
received a higher peer score on leader ability reported higher scores on masculinity,
perceived soccer competence, and competitive trait anxiety, whereas, the coaches’ ratings
of leader ability were associated with only the athlete’s skill ranking.

While Glenn and Horn (1993) provided an assessment of leader identiﬁcation that
was more comprehensive (i.e., self, peer, and coach ratings), they did not determine the
motivational mechanisms reﬂecting the reasons for engaging in leadership behaviors.
Perceptions of competence in sport and soccer were measured, but competence speciﬁc
to leadership ability was not. Furthermore, the authors employed a leader behavior
inventory that provided one word descriptors of leader behaviors and characteristics (e. g,
determined, impressionable, and honest). While this construct provided additional
descriptive information on leader characteristics, the manner in which these skills were
constructed and used by athletes to lead a group was not measured. In order to assess and
understand individual differences in leadership, researchers must determine the
conception and expectations of athlete leadership (i.e., “What is an athlete leader
supposed to do to lead the group?”) and the motivational mechanisms behind leader
conduct (“How conﬁdent is the athlete leader in leading the group?”). More importantly,
the hypothesized relationships between meaning and construction with the mechanisms
of inﬂuence should be derived from theory that allows for researchers to test proposed
determinants and regulatory mechanisms of athlete leadership.

Statement of the Problem
Based on the summary of ﬁndings on leadership in athletes, researchers in sport

have yet to address both the conceptual and operational underpinnings of leadership that

4 IN“

 

 

are spe
somec
becau
adﬂet
leade
hassi
ofaﬂ
regul
front

ﬁani

cont
lead
it is
anal

5:8

are speciﬁc to the athlete. Coaches have often been overheard saying, “We just need
someone to step up and be the leader,” or, “Our team performed poorly this season
because there was no leadership.” Yet, what exactly has “leadership” meant to the
athlete? How have conceptions regarding leadership been rendered and translated into
leader conduct? In summary, the main limitations of early work on leadership in athletes
has shown the need for future research to (a) construct an adequate operational deﬁnition
of athlete leaders’ roles and responsibilities, (b) measure athletes’ motivational and self-
regulatory mechanisms regarding leadership, (c) triangulate conceptions of leadership
from the leader, follower, and coach perspectives, and (d) develop a conceptual
framework based on theory.

An additional issue regarding the limitations of past work on leadership in athletes
concerns the assertion of the level or levels of analysis at which the perceptions of
leadership are expected to occur. Given the inﬂuence of group composition and structure,
it is possible for a team effect to emerge. Recent research on leadership has implemented
analyses and methodologies that account for individual and team level effects (Cogliser
& Schriesheim, 2000; Grifﬁn & Mathieu, 1997; Kozlowski & Doherty, 1989). In
particular, the examinations of group-level consensus have provided additional insight
regarding the potential impact of group-level perceptions (James, 1982). Analyzing data
at only the level of the athlete neglects considering the social-cognitive property of
psychologically belonging to a group, may overestimate the ﬁndings, and may lead one
to assume that the same relationship will also hold at the team level (Moritz & Watson,
1998). Therefore, it is not only important to examine whether certain variables represent

a team-level construct, but also to determine whether, once aggregated to the group level,

it has any predictive power or implications for the leadership process. Research in sport
has yet to explore correlates of leadership in athlete leadership as a team-level construct,
however, consideration of this concept has both theoretical and practical signiﬁcance
given that the adoption of leadership beliefs may occur at both the individual and team
levels.

Regardless of how it is designated, or how it emerges, leadership is an integral
part of athletic performance. A leaderless group, with no guidance or direction, will often
become a ﬁ'ustrated group. Additional empirical research on the leadership role of the
athlete is necessary to understand the pertinent psychological, social, and regulatory
correlates effective of peer leadership at both individual and group levels. Beyond the
implications of improved performance, researchers must discern the positive
psychological beneﬁts of leadership, so that parents, coaches, and practitioners may
continue to foster positive psychological and behavioral development (Larson, 2000).
Overview of Social-Cognitive Correlates of Leadership

In order to obtain a better understanding of athletes’ conceptions regarding
leadership, researchers must ascertain their perceptions of competence, interest, value-
orientations, and conduct speciﬁc to the leadership role of the athlete. To do so requires a
comprehensive theory, which deﬁnes the determinants and mechanisms through which
the leadership role and responsibilities are discernible. Social cognitive theory is one
theory that takes into account the motivational and self-regulatory mechanisms of
leadership, as well as the levels of individual and collective beliefs and behaviors among
sport teams. By distinguishing the reciprocal inﬂuence between cognitive/personal

factors, social inﬂuences, and behaviors to explain human functioning and motivation,

the social cognitive theoretical model can address limitations with previous research on
leadership in athletes (Ames, 1992a; Bandura, 1997, 2001b; Nicholls, 1984).

Central to Bandura’s (1997) notion of social cognitive theory is human agency, or
the intentional thoughts, beliefs, and behaviors which ultimately reﬂect an individual’s
ability to generate actions relative to a speciﬁc goal or objective. Thus, one component
of human agency is leadership, in which an individual may choose to lead a group, put
forth a certain amount of effort to guide and direct a group, and persist to a certain degree
in order to attain collective action. By adopting a social cognitive framework, sport
psychology research can examine the interactions between the motivational components
of choice, effort, and persistence in athlete leaders.

A primary mechanism believed to develop and maintain agency beliefs and
behaviors, such as leadership, is self-efficacy. Self-efficacy can be deﬁned as “the belief
in one’s capabilities to organize and execute courses of action required to produce given
attainments” (Bandura, 1997, pg. 3). Research on self-efﬁcacy in sport has predominately
focused on the examination of efﬁcacy speciﬁc to individual performance (Bandura,
1990, 1997; Feltz & Lirgg, 2001). However, recent research in sport has begun to
examine the social cognitive properties of psychologically belonging to a group (i.e.,
collective efficacy).

Chemers and his colleagues have deﬁned efficacy speciﬁc to the context of
leadership as the conﬁdence one has to lead a group of individuals (Chemers, 2000;
Chemers, Watson, & May, 2000; Watson, Chemers, & Preiser, 2001). Watson et al.
(2001) examined athletes’ conﬁdence about their general leadership capabilities (e. g., “ I

know what it takes to make a group accomplish its task”). Results indicated that teams

10

 

who he
mhkne
taunt
resear
teani
eﬂice
auue
to th

capa

othe
tnec
this

to v

the
”101
ind-
ll'ar
tur,
Var:

0ft

who had conﬁdent leaders reported stronger collective efﬁcacy beliefs. Moreover,
athletes who reported their leader as effective also reported greater conﬁdence in their
team than did athletes who perceived their leaders as less effective. However, this
research is limited, as it is difficult to understand exactly what the leader must do for the
team to accomplish a task. Furthermore, the social-cognitive correlates of leadership
efﬁcacy in any setting have yet to be examined. Thus, additional research is warranted on
athletes’ perceptions of their capabilities to perform leadership ﬁrnctions that are speciﬁc
to their sport, and the determinants that lead to enhanced efﬁcacy beliefs about their
capabilities to lead the team toward the objective or goal.

In addition to one’s efficacy to lead, more recent trends in leadership research in
other settings (i.e., military and organizational) have focused on auxiliary self-regulatory
mechanisms such as leadership skills (Mumford et al., 2000). Research stemming from
this approach has centered on the development of skills (e. g., collaboration, or the ability
to work well with others, proﬁcient listening skills, empathy, effective confrontation
techniques, etc.) that inﬂuence and regulate the leadership process rather than emphasize
the effectiveness of leader behavior or outcome. The implication of the skills-based
model is that this development may occur in a progressive systematic fashion, and that
individual differences in leadership skills may reﬂect meaninng developmental
transitions in the acquisition and use of sport-speciﬁc leadership skills, which may, in
turn, reveal individual differences in leader emergence and effectiveness. For example,
varsity leaders in sport may display different leadership characteristics than junior varsity

or novice athletes with less experience. Therefore, further exploration of the viable skills

ll

that are acquired and developed through sport experiences may serve as a possible
regulating function of leadership.

In addition to the proposed regulatory mechanisms of leader efﬁcacy beliefs and
skills, social cognitive theory hypothesizes that certain contextual, personal, and
situational factors may serve as determinants of achievement-related outcomes, such as
leadership. While leader skills and efficacy beliefs may be considered as more proximal
in nature (i.e., account for more variance in leadership and directly regulate leadership
conduct), personal attributes may be considered as more distal predictors of leadership
identiﬁcation and effectiveness. Two personal attributes that have consistently emerged
in leadership research are gender and achievement. One social cognitive view that has
emphasized society’s structural inﬂuence on gender differentiation is Eagly’s social role
theory (Eagly, 1987). The theoretical tenets of social role theory have postulated that
gender differences in social behavior are a product of the social roles that regulate
behavior in adult life. Eagly (1987) classiﬁed the traditional notion of masculinity as
“agentic” and femininity as “communal.” Men have been expected to possess
predominately agentic attributes (i.e., independent, masterful, assertive, and competent),
while women have been classiﬁed as having high levels of communal qualities (i.e.,
friendly to others, unselﬁsh, concerned for others, and emotionally expressive). As a
result of these roles, Eagly proposed that men and women bring differing dispositional
skills and beliefs to leadership situations (Eagly & Johannesen-Schmidt, 2001; Eagly,
Makhijani & Klonsky, 1992).

One way to continue to understand the simultaneous occupation of gender role

and leader role is to examine this complex interaction in the traditionally masculine

12

conn
phat
beha
adﬁe
high:
abﬂh
sanu
than
regai
conc
prox'
Wom
“(NU

0fth

COgn
adue
iﬁChc
deﬁn.
the p,
&:L€g

Oﬁent

context of sport. For example, Glenn and Horn (1993) hypothesized that female soccer
players who were more androgenous (i.e., able to exhibit both masculine and feminine
behaviors) were more likely to emerge as team leaders. However, they found that female
athletes who rated themselves as both masculine and feminine also rated themselves
higher on leadership ability, whereas female athletes who were rated high in leadership
ability by their peers exhibited high levels of masculinity. Therefore, for this particular
sample of female athletes, masculinity was perceived by peers as a more favorable
characteristic of the female leader. This ﬁnding reﬂects the emerging positive perspective
regarding women’s movement into sport as a viable way to challenge traditional
conceptions of gender differences. Klenke (1996) has argued that the setting of sport
provides a provocative context from which to examine the development of leadership in
women, because women’s position in sports has come to reﬂect the changing role of
women, not only in this particular context, but also in society in general. Further scrutiny
of the relationship between gender and leadership is necessary in order to have a more
useful understanding of the inﬂuence of gender roles on leadership roles in sport.

In addition to personal determinants such as gender-role orientations, social
cognitive theory proposes that value-laden beliefs that reﬂect the meaning behind
achievement activities may also serve to inﬂuence agency-related beliefs (Bandura, 1997;
Nicholls, 1992). Speciﬁcally, achievement goal orientations exemplify how individuals
deﬁne personal ability, successful experiences, and task difﬁculty, which in turn, reﬂect
the purpose for engaging in activities (Ames, 1992a; Bandura, 1997; Duda, 2001; Dweck
& Leggett, 1988; Nicholls, 1984, 1989). Two achievement goals, task- and ego-

orientation, have been extensively researched in athletic settings. With task orientation,

l3

 

the MM
and perce
state of eg
position 1

ability (3

proposet

but also
correlat
\Vherea
psycho
the fun
associa
Temp]
Orient;
DTEdoj
chem
means
3850C:
Such;
inyeg

Fe80).

the individual used self-referenced strategies to deﬁne perceived ability, task difficulty
and perceptions of success. A more differentiated conception of ability characterized a
state of ego involvement in which the individual was more concerned about one’s
position relative to others, ultimately relying on normative standards to demonstrate
ability (Nicholls, 1984,1989).

Nicholls (1992), whose work has focused on youth and emerging adults, has
proposed that achievement goals not only reﬂect different perspectives of achievement,
but also different views of the world. Speciﬁcally, he proposed that ego orientation may
correlate with a relative lack of concern for social issues regarding fairness and justice.
Whereas, task orientation may emphasize values such as fairness and cooperation. Sport
psychology research on athletes’ goal orientations, beliefs about success, and views about
the functions of sport, has indicated that athletes’ dispositional tendencies may be
associated with variations in sport social behaviors (Duda, 2001; Duda, Olson, &
Templin, 1991). In particular, the empirical ﬁndings on the relationship between goal
orientations and sportspersonship attitudes have revealed that athletes with a
predominance in ego orientation are more likely to endorse unsportsmanlike play and
cheating behaviors, and report the use of aggressive acts in sport as a more legitimate
means to succeed (Duda et al., 1991). However, research in sport has yet to explore the
association of task and ego orientations with other dimensions of sport social behaviors,
such as leadership.

Research in sport has begun to broaden the conception of achievement-goals by
investigating the notion of multiple goals (i.e., self-enhancing ego and self-defeating

ego). In addition to multiple goals related to achievement, achievement goal theorists

l4

have also operationalized social goals, or beliefs about the causes of success in
relationships with others (Jarvien & Nicholls, 1996). Exploration of the ways in which
social goals in sport interact with task and ego dimensions to inﬂuence leadership beliefs
and behaviors in sport has yet to be conducted (Roberts, 2001). Similar to the academic
setting, sport psychology research should consider the manner in which students’ beliefs
about the social reasons for sport achievement (i.e., social responsibility) affect their
achievement-related beliefs and behaviors in athletic settings (Urdan & Maehr, 1995). It
is possible that athletes may have achievement goals related not only to conceptions of
ability to perform in sport, but also to conceptions of ability to lead their team. Therefore,
certain social goals may inﬂuence performance and additional achievement outcomes
(i.e., leadership) that are hypothesized to occur in sport. By examining the independent
effects of social goals, as well as their interactive effects with task and ego goals,
researchers might come to a more complete understanding of athlete motivation and
achievement in sport pertaining to performance and social behaviors.

In addition to one’s dispositional tendencies, the tenets of social cognitive theory
have also proposed that achievement-related beliefs (i.e., leadership) might be inﬂuenced
by perceptions of situational elements, or motivational climate. Although general
discussions of motivation have typically focused on the person as the causal source,
Maehr and Braskamp (1986) identiﬁed the situational inﬂuences on achievement
patterns. In particular, the role of group membership and expectations of others have been
emphasized simply because, in achievement settings, individuals do not act in isolation
from the social groups in which they hold membership. The authors proposed that, to a

certain degree, the normative and role-related expectations of the group may determine

15

 

th:

f0:

CO

the value of engaging in certain behaviors. For example, the set of expectations that exists
for an individual who is identiﬁed as the leader of the group may be inﬂuenced by the
coach and/or teammates (Maehr & Braskamp, 1984; Vroom & Yetton, 1973).

The manner in which these expectations are communicated to group members
may be reﬂected through the situational goal structures that are established and most
salient to athletes. Goals that reﬂect individuals’ perceptions of the achievement-setting
climate have been referred to as mastery and performance goals (Newton & Duda, 1999;
Newton, Duda, & Yin, 2000). The mastery goal climate emphasizes improving personal
competence, learning, and mastering skills based on a set of internalized standards. The
performance goal climate, on the other hand, emphasizes outperforming others,
demonstrating superior ability with little effort, and valuing high ability.

Consistent with the theoretical tenets of achievement goal theory, individual
differences in leadership have been expected to emerge given the motivational climate
emphasized by the coach. For example, in a mastery climate, athletes are encouraged to
collaborate and master skills relevant to the activity (e. g, working together to solve a
performance-related problem). In contrast, in a performance climate, athletes are
encouraged to outperform others and engage in intra-team rivalry. Thus, how the climate
is perceived may explain individual differences in regulatory mechanisms and effects of
leadership. Furthermore, current research in sport has explored perceptions of
motivational climate as a group level construct (Gano-Overway et al., 2001; Magyar &
Feltz, & Simpson, 2002). Therefore, climate may operate as individual- and group-level

predictors of leadership’s effects in sport.

l6

To date, few studies have attempted to connect the constructs of structural goals
with leadership behaviors (Duda & Balaguer, 1999; Kozlowski & Doherty,]989). Similar
to the integration of social goals into personal achievement goals, ﬁrture research in sport
psychology may beneﬁt from examining the social dimensions (i.e., leadership
development) of the perceived motivational climate. The inﬂuence of the motivational
climate, as established by the coach, and its subsequent effect on the development of
leadership in athletes has yet to be examined.

Contextual Factors in the Current Study

The sport chosen for the current investigation was collegiate rowing. Rowing has
various boat classes in which multiple combinations of individuals will row in the boat
(e.g., double, four, eight). The current study examined athletes who rowed in the eight
(8+), in which eight rowers are seated in seats one (bow) through eight (stroke) with the
cox positioned in seat nine, and the four (4+) in which four rowers are facing the cox who
is positioned in seat ﬁve. The coxswain is the only person facing the ﬁnish line while the
rowers are facing the coxswain with their backs to the ﬁnish line. This arrangement
places the coxswain in a very inﬂuential position with responsibility to steer the boat and
call the race (e. g., letting the rowers know how far they have to go to the ﬁnish line, or
how far they are from an opponent).

Rowing was selected for this research because it is a team sport that is considered
to be a closed-motor skill, or performed in a stable environment in which the basic
actions involved essentially remain unchanged throughout the activity. Rowing is unique
compared to other team sports such as basketball, which may be considered more of an

open-motor skill, in which the team performs in a constantly changing environment,

17

directly impacting the performance of the skill. For example, basketball players
constantly interact with one another and with their opponents during a game, which
creates multiple opportunities for variability. Rowing, on the other hand, is a repetitive
skill in which teammates are highly dependent on one another and exercise a controlled
form of interaction between crew members while performing the skill. Therefore, the
examination of performance in crew allows researchers to control other extraneous
variables which could conceivably inﬂuence the observation of leadership (i.e.,
interaction with opponents). According to Steiner’s (1972) task typology, rowing may be
classiﬁed as a unitary additive task, meaning that all rowers perform the same aspects of
the task at the same time, and the sum of each individual rower’s output determines the
group output. The sport of rowing was also chosen because with the advent of Title IX
there has been an increase in women’s crew teams at the collegiate level. As a result,
females have been afforded the opportunity to participate in athletics at the collegiate
level.
Conceptual Model of Leadership in Athletes

In conclusion, current research on leadership in sport has failed to test a conceptual
model of athlete leadership that integrates the social cognitive properties that may
inﬂuence the leadership role of the athlete (see Figure l on p2). Stemming from social
cognitive theory, the current model proposed three principal categories: personal and
situational determinants, regulatory mechanisms, and team and leader effects. The
hypothesized personal determinants were demographic variables (previous experience in
sport, and previous experience with leadership and rowing ability), gender role

perspective (agentic and communal), and achievement goal orientations (task, ego, and

18

lea:

ma

leader). Situational determinants were perceptions of the motivational climate (i.e.,
mastery, performance, and leadership dimensions). The proposed regulatory mechanisms
of leadership were leadership skills, leadership efﬁcacy, and task self-efficacy. A total of
12 dependent variables that were speciﬁc to leadership effects were examined (self score
of performance leadership, self rank of performance leadership, peer score of
performance leadership, peer rank of performance leadership, evaluation of #1
performance leader, evaluation of #1 motivational leader, self evaluation of performance
leadership ability, and self evaluation of motivational leadership). Collective efficacy was
also assessed as a team effect variable.

Stemming from social cognitive theory, the personal determinants (demographic,
gender roles, and achievement goal orientations) were hypothesized to directly inﬂuence
self-regulatory mechanisms (leadership skills and efﬁcacy beliefs). Personal determinants
were also hypothesized to serve as more distal predictors of effects related to leadership
(performance and motivational effectiveness and evaluation) and team achievement
beliefs (collective efficacy). Similar to the personal determinants, the situational
determinants (perceived motivational climate) were hypothesized to inﬂuence self-
regulatory mechanisms of leadership, and leadership and team effects. Overall, the
personal and situational determinants were hypothesized to serve as proximal
determinants (i.e., account for more variance) of self-regulatory mechanisms, and more
distal determinants (i.e., account for less variance) of leadership and team effects. The
self-regulatory mechanisms were hypothesized to be proximal predictors (i.e., account

for more variance) of leadership and team effects.

19

Rt’.‘

J

Research Questions

1.

Of the variables included in the conceptual model, what are the strongest individual
and team level social-cognitive determinants and self-regulatory mechanisms of the
leadership role in athletes?

How do the personal determinants that were measured, such as perceptions of gender
roles and achievement goals, inﬂuence leadership in athletes?

How do the situational/contextual elements that were measured, such as individual
and team perceptions of the motivational climate, inﬂuence leadership in athletes?
How do the regulatory mechanisms that were measured, such as leadership skills and
efficacy beliefs, inﬂuence leadership in athletes?

What are the differences in motivation between athletes identiﬁed as leaders and
those identiﬁed as followers?

How do personal background characteristics, such as previous experience in sport,
and previous experience with leadership and rowing ability, inﬂuence leadership in
athletes?

What are the cross-sectional differences between varsity and non-varsity rowers in
leadership regulatory mechanisms and effects?

How do self-reported motivational and regulatory mechanisms of leadership

inﬂuence peer and coach ratings of leader conduct?

Conceptual Model Hypotheses

1.

Gender role perspectives will differentially predict self-regulatory mechanisms of

leadership and leadership’s effects.

20

(a) Agentic attributes will positively predict performance-execution skills, negative
tactics leader skills, leadership efﬁcacy, and task self-efficacy.

(b) Communal attributes will positively predict motivational-interpersonal skills, respect-
communication skills, leadership efﬁcacy, and task self-efﬁcacy.

(c) Agentic attributes will positively predict the evaluation of the #1 ranked performance
leader, in addition to self and peer leader effectiveness scores.

((1) Communal attributes will positively predict the evaluation of the #1 motivational
leader, self and peer scores of motivational-leader effectiveness.

2. Goal orientations will differentially predict self-regulatory mechanisms of leadership
and leadership’s effects.

(a) Task orientation will positively predict performance-execution skills, motivational-
interpersonal skills, respect-communication skills, leadership efficacy, and task self-
efficacy.

(b) Ego orientation will positively predict performance-execution skills, negative tactics
skills, leadership efficacy, and task self-efficacy, and negatively predict perceived
importance of motivational-interpersonal skills, respect-communication skills, and
performance and motivational leader evaluations.

(c) Leader orientation will positively predict the perceived importance of performance-
execution skills, motivational-interpersonal skills, respect-communication skills,
leadership efficacy, task self-efﬁcacy, and negatively predict perceived importance of
negative tactics leader skills.

((1) Task orientation will positively predict collective efﬁcacy, the evaluation of the #1

performance-leader, the evaluation of the #1 motivational-leader, self scores of

21

performance-leadership effectiveness, self scores of motivational-leader
effectiveness, peer scores of performance-leader effectiveness, and peer scores of
motivational leader effectiveness. Leader orientation will positively predict self and
peer scores for performance-leader and motivational-leader effectiveness. Ego
orientation will positively predict self and peer scores of performance leadership.

3. Perceptions of motivational climate will differentially predict self-regulatory
mechanisms of leadership, and leadership effects.

(a) Individual perceptions of mastery dimensions and leadership dimensions of
motivational climate will positively predict performance-execution leadership skills,
motivational-interpersonal leadership skills, respect-communication leadership skills,
leadership efficacy, and task self-efficacy.

(b) Individual perceptions of performance motivational climate will positively predict
performance-execution skills and negatively predict motivational-interpersonal skills,
respect-communication skills, leadership efficacy, and task self-efficacy.

(c) Perceptions of mastery climate will positively predict collective efﬁcacy, the
evaluation of the #1 performance leader, the evaluation of the #1 motivational leader,
self and peer scores of performance-leader effectiveness, and self and peer scores of
motivational-leader effectiveness. Performance climate will positively predict self-
score and rank of performance-leader effectiveness.

4. Leadership skills will differentially predict leader and team effects.

(a) Performance-execution leadership skills will positively predict collective efficacy,
evaluation of the #1 performance leader, and self and peer scores and ranks of

performance-leadership.

22

(b) Motivational-interpersonal leadership skills will positively predict collective efﬁcacy,
evaluation of the #1 motivational leader, and self and peer scores and ranks of
motivational-leadership.

(c) Respect-Communication leadership skills will positively predict collective efﬁcacy,
evaluation of the #1 performance leader, and #1 motivational leader, and self and peer
scores and ranks of performance— and motivational-leadership.

(d) Negative tactics leadership skills will negatively predict collective efﬁcacy, and self
and peer score and rank of performance-and motivational-leadership.

5. Leadership efficacy and task self-efﬁcacy beliefs will positively predict team (e.g.,
collective efficacy) and leader effects.

6. Personal and situational determinants will serve as distal predictors of self and peer
ratings of performance and motivational leadership, while self-regulatory
mechanisms will serve as more proximal predictors of leadership effects. Self-
regulatory mechanisms will mediate the direct inﬂuence of personal determinants on
team and leader effects.

Additional Hypotheses

7. There will be cross-sectional differences in self-regulatory mechanisms and
leadership effects based on leadership role experience (yes or no), and academic class
standing (Freshman, Sophomore, Junior and Senior).

(a) Athletes who report previous experience with a leadership role (i.e., student council,
youth group, 4-H, etc), and are upperclasswomen (juniors and seniors) will report
higher mean scores on all leadership skills, leadership efﬁcacy, and task self-efﬁcacy

than athletes who have no experience with leadership and are lowerclasswomen.

23

(b) Athletes who have previous experience with a leadership role and are
upperclasswomen will report higher self-scores and will be scored higher by their
peers on performance and motivational leadership effectiveness.

8. There will be group differences between athletes designated as the #1 leader of the
boat versus athletes who are not given a #1 ranking.

(a) Athletes designated as #1 performance leaders will report higher mean scores on
agentic attributes, leadership orientation, performance-execution skills, respect-
communication skills, negative tactics leader skills, leadership efficacy and task self-
efﬁcacy than athletes not designated a #1 ranking.

(b) Athletes assigned as #1 motivational leaders will report higher mean scores on
communal attributes, motivational-interpersonal skills, respect-communication skills,
leadership efﬁcacy, task self-efﬁcacy, and collective efﬁcacy than athletes not
designated a #1 ranking.

Basic Assumptions

The primary assumption for this study was based on the notion that athlete
leadership exists in sport. Furthermore, it was assumed that having a group of athletes
complete questionnaires that are designed to assess the different dimensions of leadership
would actually provide information regarding the perceived interaction between the

leader and followers on a rowing team. In addition to the assumption of the existence of a

conceptual deﬁnition of leadership, it was assumed that there were speciﬁc personal and

situational determinants and regulatory mechanisms of athlete leadership that can be

measured and demonstrated to predict the leadership constructs in the sport of rowing.

24

Delimitations

The generalizability of the results of this research will be limited to intercollegiate
rowers. It is recognized that the results from the current investigation should be
interpreted with caution for competitive junior, club and elite rowers. Due to the unequal
representation of gender in the current sample population, generalizability of the results
to both males and females should also be considered with caution. Furthermore, the
leadership constructs developed in this study may not be suitable for team sports other

than unitary additive sports.

25

Chapter 2
REVIEW OF LITERATURE
From the “Great Man ” to the “Contingent Fit ” of Leadership

Despite the substantial research conducted on leadership in experimental,
organizational, and military settings, sport researchers have made little effort to apply
these ﬁndings to the study of leadership in athletes. In order to conduct a theoretical study
on leadership in athletes, this chapter provides a general overview of the conceptual and
theoretical underpinnings of leadership that have been researched outside of the sport
context. For the purpose of this paper a selection of research was reviewed that
emphasized the trait/personality, behavioral, situational, contingency, and
transactional/transformational perspectives. For extensive review, see Bass, 1990;
Chemers, 1997; Northouse, 2001. These synthesized ﬁndings are provided to supplement
the atheoretical approach to research on leadership in athletes that has been conducted to
date. This synthesis is followed by a summary of Bandura’s (1986, 1997) social cognitive
theory and the proposed motivational and self-regulatory mechanisms of athlete
leadership.
Trait/Personality Perspective

Early work on leadership emphasized a person-centered approach by focusing on
the role of the leader from the leader’s perspective. The inception of leadership research
began with the “great man” theories, which identiﬁed the supposed innate qualities and
characteristics of successful social, political, and military leaders of the 20th century
(Chemers, 1997). Traditionally, it was believed that people were born with these traits

and that only individuals who possessed them became great leaders.

26

Stogdill (1948) was the ﬁrst to challenge the great man concept of leadership with
his synthesis of leadership research that was conducted between 1904 and 1947.

After reviewing 124 studies he found that the individual in a leadership role, on average,
displayed the traits of intelligence, alertness, insight, responsibility, initiative, persistence,
self-conﬁdence, and sociability. Although he identiﬁed certain traits that were
characteristic of leaders, he concluded that an individual does not become a leader solely
because he/she possesses certain traits. Rather he emphasized that leadership was
primarily determined by situational factors, not personality traits, and that the traits of the
leader must be relevant to the situation in which the leader is functioning.

Stogdill (1974) then conducted a further review in which he examined 163 studies
that were conducted between 1948 and 1970. In this review he identiﬁed traits such as
self-conﬁdence, a sense of personal identity, vigor, and persistence as ﬁrndamental
characteristics of emergent leadership. In his second review he validated the original idea
that trait characteristics function as leadership correlates, and also expressed that a pattern
of characteristics should be emphasized to distinguish leaders, rather than just one single
trait to distinguish leaders. Based on his ﬁndings, potential antecedents of leadership
include the understanding that traits may vary given the leader, and that the situation may
vary in which the leader is ﬁrnctioning (i.e., both personality and situational factors).

Research stemming ﬁom the trait/personality perspective has proposed that traits
and characteristics such as knowledge, self-monitoring, extroversion, dominance, and
achievement-motivation continually relate to leadership. Researchers from this
perspective have reasoned that due to measurement and methodological limitations,

previous research on trait leadership consistently failed to identify a uniform set of traits

27

among leaders (Kenny & Zaccaro, 1983; Zaccaro, Foti, & Kenny, 1991). Zaccaro and his
colleagues postulated that ﬁndings from both experimental and ﬁeld-based settings
revealed that a large portion of leadership variance (i.e., 59% in Zaccarro et al., 1991)
was explained by trait attributes (i.e., self-monitoring), and that this signiﬁcant amount of
variance indicated that individuals who were perceived as leaders in one group also
emerged as leaders in subsequent groups across different group situations. Therefore,
Zaccaro et al. (1991) suggested that leadership does not involve the possession of one
speciﬁc trait or characteristic, but rather a compilation of attributes, and that it is within
the ability of the leader to perceive the needs of the group and modify one’s approach in
order to emerge consistently as the leader.
Behavioral Perspective

Disillusioned by the seemingly failed attempt to understand leadership from the
trait/personality approach, researchers turned to investigate behavioral outcomes of
leaders as a means to delineate effective leadership. During the 1950’s, a series of studies
emerged ﬁ'om research laboratories in psychology of two prominent universities (Ohio
State University and University of Michigan), and from these research programs
originated two higher-order dimensions of leadership behaviors. The ﬁrst reﬂected an
emphasis on task performance, or the emphasis on organizing group members, clarifying
goals, and directing members toward task accomplishment. The second dimension
focused on the importance of the socioemotional well-being and interpersonal
relationships among group members. The most inﬂuential research on leadership during
this time was the development of the Leader Behavior Description Questionnaire

(LBDQ; Hemphill & Coons, 1957). This construct examined the degree to which the

28

leader emphasized a task/performance dimension labeled as initiation of structure (i.e.,
the extent to which the leader deﬁnes the responsibilities of the individual or group), and
interpersonal dimension referred to as consideration (i.e., how much the leader engages in
two-way communication). The Ohio State researchers considered these dimensions as
orthogonal, and the degree to which a leader emphasized one dimension was not
dependent on the other dimension. Thus, leaders could display variations on both
dimensions of initiation of structure and consideration leadership behaviors (e. g.,
high/high). Although this construct had been linked to various group outcomes (e. g.,
satisfaction, cohesiveness, etc), the ﬁndings remain equivocal. In some situations, high
initiation of structure was perceived as most effective, while other situations called for a
strong emphasis in both behaviors.

Critics attributed the lack of consistency in ﬁndings to the fact that the ratings of
behavior reﬂected the perceptions of the rater and not the actual behavior, and that one
style of behavior may not be perceived as appropriate in every situation (Chemers, 1997,
2001). Although research from the person-centered approach provided some empirical
support for the role of the leader, the situational elements of the leadership process must
also be considered.

Situational Perspective

Despite Stodgill’s (1948, 1974) earlier recommendation to consider both the
person and the situation/context, researchers continued to neglect this interaction until the
emergence of the situational approach to leadership. Contrasting with the trait approach,
situational leadership emphasized situational/contextual determinants of leadership and

postulated that different situations required different kinds of leadership (Bass, 1990;

29

Hughes, Ginnet, & Curphy, 1996). In particular, four styles of leadership, varying in
degree of supportive and directive behavior, labeled ‘directing’, ‘coaching’, ‘supporting’,
and ‘delegating’, have been deﬁned within this perspective. Directing leadership style
reﬂects the leader’s emphasis on communication of goal achievement, allocation of less
time to interpersonal needs, and supportive behaviors. The coaching leadership style
emphasizes both goal achievement and maintenance of subordinates’ socioemotional
needs. In contrast, the supporting leadership style tends to de-emphasize task instruction
and develops employees’ skills relative to the task to be accomplished through supportive
behavior. Finally, delegating leadership style involves less task input and social support
in reference to the task; subordinates are allowed to take responsibility for the task as
they see ﬁt.

In addition to the leadership style, the situational leadership perspective
highlighted the developmental elements of leadership. Subordinates were classiﬁed along
a developmental continuum based on their interest and ability to perform the required
task. Individuals who were interested and conﬁdent in their work and knew how to
perform the task were ranked high on this continuum. In contrast, people considered to be
lower on the continuum have limited skill capacity to accomplish the task at hand but feel
as if they have the motivation or conﬁdence to get the job done. Leaders are expected to
diagnose the developmental level of the subordinate and then adapt their leadership style
accordingly. For example, an individual classiﬁed developmentally as high will receive
mostly a delegating style ﬁom the leader, which provides minimal support and direction

and allows the follower to take on more responsibility. In contrast, an individual

30

class

1116 1:

suite
deci
expé
gite
deci

deci

SllU

sut
int.

Co

bet

bel

classiﬁed developmentally as low will receive mostly directing behavior style, in which
the leader provides much more direction to the follower.

An additional situational perspective that considered the perspective of the
subordinates was the application of Vroom and Yetton’s (1973) normative model of
decision-making to the study of leadership. This model considered both the normative
expectations of the group and the type of leadership decisions leaders should employ
given these expectations. While this model portrayed a complex methodology to
decision-making, this process of decision-making considered all parties involved with
decisions (i.e., both leaders and followers).

Overall, the situational perspective provided one of the ﬁrst accounts regarding
the developmental status of subordinates and the characteristics of subordinates. This
view of leadership highlighted the importance of match between the leader and followers
and the ability of the leader to be ﬂexible to accommodate the needs of the subordinates.
However, because limited empirical research exists to support the claims of the
situational perspective, the ﬁndings have been inconsistent and equivocal (Chemers,
1997). Without the basic empirical support of the proposed mechanisms, there is concern
regarding the validity of matching leaders relative to the developmental status of
subordinates and applying this approach to group settings as opposed to one-on-one
interactions.

Contingency Perspective

Contingency theory provided another perspective that examined the interaction

between leaders and followers based on the proposition that leaders have dominant

behavioral tendencies and that leadership is contingent on situations which are more

31

conducive to the-leaders’ behavioral dispositions. One of the most widely used
contingency perspectives was Fiedler’s ( 1967) contingency model of leadership which
suggested that the leader’s effectiveness was contingent on how well the leader’s style ﬁt
the context. Therefore, while the situational leadership theory emphasized the ﬂexibility
of the leader, the contingency model maintained that leaders were more consistent and
less ﬂexible in their behavior. Three main components are featured in the contingency
model of leadership. First is the leader-member relationship, which refers to the group
atmosphere andthe degree of conﬁdence, loyalty, and attraction group members feel for
their leader. Second is task structure, referring to the degree to which the requirements of
a task are clearly spelled out. Finally, position power which refers to the degree of
authority the leader has to reward or punish followers.

To empirically validate his model, F iedler (1967) developed the Least Preferred
Coworker (LPC) scale, which requested that leaders rate the worst co-worker with whom
they had ever worked and the one person who most interfered with a successful
accomplishment of the task. These three facets of the contingency approach to leadership
reﬂected the degree to which the leadership situation provided the leader with a sense of
predictability, certainty, and control. Situations that were favorable were those in which
the leader had a good leader-member relationship, task clarity, and sufﬁcient leader
position power. Least favorable situations represented the opposite extreme, with poor
leader-member relations, unstructured tasks, and weak position power. Finally,
moderately favorable situations were postulated to fall in between these two dimensions.
Therefore, F iedler (1967) hypothesized that certain styles would be more effective in

certain situations, and classiﬁed all possible combinations of LPC score with group

32

performance and situational favorability to represent the complexity of effective
leadership. For example, task motivated leaders (low LPC) were hypothesized to be
effective in both favorable and unfavorable positions. In contrast, relationship-oriented
leaders (high LPC) were hypothesized to be more effective in moderately favorable
situations.

Chemers (1997) noted the LPC was originally used as an indicator of
psychological distance, and that later research based on the LPC resulted in discrepant
ﬁndings (e. g., high LPC more effective in moderately favorable situations). To resolve
these inconsistent ﬁndings, researchers used the LPC to measure the leader’s
motivational orientation (e.g., leader’s motivation toward task accomplishment),
motivational hierarchy (e. g., primary and secondary motivations of the leader), cognitive
complexity, and value-attitude (e. g., judgments of group and task accomplishments).

While this model examined the collective combination of leadership styles and
situational variables, the contingency perspective has been criticized for not explaining
why individuals with certain leadership styles have been more effective in some
situations than others (Chemers, 2001). Furthermore, the application of the LPC measure
of leadership has been questioned for lacking face validity, taking too much time to
complete, and failing to directly analyze the process by which leaders’ motivational
orientation affects group processes and outcomes.

Although the construct validity and empirical support of the contingency model
have been questioned, F iedler and his colleagues have been commended for providing the
foundation for the development of additional contingency theories (Bass, 1990; Chemers,

1997). Among them, path-goal theory provided a more integrative framework from

33

which to study leadership, and expanded upon the trait/situational and contingency
theories of leadership (House, 1971; House & Shamir, 1993). Path-goal theory of
leadership examined how leaders motivate subordinates to accomplish designated goals.
The theoretical assumptions stated that subordinates would be motivated if they thought
they were capable of performing a task, if they believed their efforts would result in a
certain outcome, and if they believed the payoff for doing their work were worthwhile
(House, 1996).

House and Mitchell (1974) submitted that leader behavior, characteristics of the
subordinates, task characteristics, and motivation all interacted to embody the major
components of path-goal theory. They proposed that leaders adopt four types of
leadership behaviors: directive, supportive, participative, and achievement-oriented.
Directive leadership characterizes a leader who gives subordinates instructions about the
task to be performed, expectations related to performance, the amount of time to perform
the task, and the timeline in which it should be completed. Supportive leadership
describes a leader who is friendly and approachable and attends to the well-being of
subordinates. Participative leadership refers to a leader who encourages followers to
participate in collaborative decision-making. The fourth leader behavior, achievement-
orientation depicts a leader who challenges subordinates to perform at the highest level
possible.

The subordinates’ characteristics take into account factors which may inﬂuence
their perceptions of the leader’s behavior (i.e., need for affiliation, preference for
structure, desire for control, and perceived ability). Task characteristics expand beyond

the situational components of the followers’ perspectives and include the design of the

34

task to be performed, the formal authority of the organizational structure, and the primary
work of the subordinates. Finally, path-goal theory acknowledged the contingent
relationship that exists between the leader and subordinates, and proposed that leaders
must choose a leadership style that best ﬁts the needs of the subordinates and the type of
task they are performing.

Some of the methodological limitations noted were linked to the potential
confounds of assessing all measures of leadership (e. g., leader behavior, subordinate
satisfaction) from the same person. While House (1996) has contended that researchers
have inaccurately applied the path-goal theory to the study of leadership, issues remain
related to the complexity of the path-goal model and its failure to adequately explain the
relationship between the leader and worker motivation.

T ransactional/ Transformational Perspective

The transactional approach to leadership emerged as a way to describe the
leadership process from the followers perspective. Transactional theories depicted social
interactions from a cost-beneﬁt standpoint. The premise of a cost-beneﬁt analysis of
human relationships was that people who viewed interactions as rewarding would
continue the relationship, while costly interactions would be terminated. Researchers
from this perspective claimed that people sought to maximize rewards and minimize cost,
thus, the result of an interaction reﬂected a combination of cost and rewards yielding
either a proﬁt or a loss.

The process of fair exchange has been linked to perceptions of justice (Chemers,
1997 ; Philips, Douthitt, and Hyland, 2001). A just exchange is when an individual feels

the investment in the interaction is proportional to the perceived cost. When individuals

35

receive less than what they feel they deserve, they feel angry, and when they receive
more than they deserve, they feel guilty. Phillips and colleagues (2001) explored the role
of justice in team member satisfaction with the leader and attachment to the team. Results
indicated that perceptions of justice mediated the relationship between leader
consideration and satisfaction with the leader. In contrast, justice perceptions only
partially mediated the relationship between team decision accuracy and decision
inﬂuence with perceived satisfaction of leadership. Therefore, subordinate perceptions of
justice heavily inﬂuenced subsequent perceptions of leader satisfaction with leader
behavior.

An additional transactional approach is Graen’s vertical dyad linkage (VDL)
model, which examined leadership at the dyadic level. A dyad consists of the relationship
between the leader and one subordinate. This approach provided a very different
perspective of leadership because previous research on leadership assumed leaders
treated followers in a collective way, as a group, using an average leadership style. With
the VDL model, a leader’s relationship to the work unit as a whole is viewed as a series
of vertical dyads. This dyad perspective emphasizes two types of linkages, those that are
based on expanded and negotiated role (extra-roles) responsibilities, called the in-group,
and those that are based on the formal employment contract (deﬁned roles), or the out-
group. Subordinates are expected to become part of the in-group or out-group based on
how well they work with the leader and how well the leader works with them. Also
becoming part of one group or another is based on how well subordinates involve
themselves in expanding their role responsibilities with the leader. Early research from

the dyadic perspective focused on the quality of the leader/member relationship and

36

exam
work

of th

and ‘.
crea‘
trans
( 197
char
conl
W611
acce

incr

Whi
She
pos
stir

Ila

WI
on

leg

examined the nature of the differences between in-groups and out-groups. More recent
work has examined the relationship between vertical dyads and the organizational climate
of the work group (Kozlowski & Doherty, 1989).

While transactional leadership focused on the mere transaction between leaders
and followers, transformational leadership concerned the process whereby a leader
created a connection with the followers that raised the level of motivation and
transformed the values of followers from self-interest to collective achievement. House
(1976) was the ﬁrst to publish on the theory of charismatic leadership. He proposed that
charismatic leaders encompassed speciﬁc personality characteristics of dominance,
conﬁdence, strong values, and a desire to inﬂuence others. These characteristics, in turn,
were hypothesized to establish the followers’ trust in the leader’s ideology, unquestioning
acceptance and affection toward the leader, obedience, emotional involvement, and
increased conﬁdence.

Recent examination of transformational leadership has shown the process by
which a leader transforms the group of followers (Bass, 1990; Chemers, 2000; House &
Shamir, 1993). Bass (1990) has described transformational leaders as individuals who
possess internal values such as idealized inﬂuence, inspirational motivation, intellectual
stimulation, and individualized consideration. This notion of “charismatic” leadership has
transcended mainstream psychology into the corporate business setting, where effective
leaders are deﬁned as individuals with vision and the capability to empower subordinates.
While this theory expanded the notion of leader effectiveness beyond performance
outcome and considered some of the motivational mechanisms of collective achievement,

research on transformational leadership has been criticized for lacking conceptual clarity,

37

of i
has

the

em
16$

sitt

de

his

being anti-democratic (i.e., lacking in collaborative efforts), and having weak empirical
support (House & Shamir, 1993; Northouse, 2001).
Conclusions

Overall, this small selection of ﬁndings on leadership alludes to the great amount
of interest in this topic, and the potential for future research. While previous investigators
have attempted to incorporate all of the different elements proposed to be involved with
the leadership process (i.e., type of task, position of power, subordinate characteristics),
this increase in model complexity has made it difﬁcult to demonstrate a consistent
empirical validation of these leadership models. However, in light of the progression that
research on leadership in other settings has experienced (i.e., from the person, to the
situation, to the interaction between the two), a much more comprehensive understanding
of leadership has evolved. Thus, this summary of ﬁndings on the leadership process
provides sport researchers with certain issues to consider for ﬁrture research on leadership
in athletes.

To date, however, the only connection that has been made between leadership
ﬁndings from other settings and leadership in sport is speciﬁc to the leadership in coaches
(Chelladurai & Reimer, 1998; V03 Strache, 1979). Chelladurai developed his
multidimensional model of leadership in coaches in part from the Leadership Behavior
Description Questionnaire (LBDQ) and Vroom and Yetton’s (1973) theory of normative
decision-making. From this model, be constructed the Leadership Scale for Sports (LS S)
which assessed ﬁve different dimensions of leadership: training and instruction,
democratic behavior, autocratic behavior, social support, and positive feedback. While

his research ﬁndings have provided a valuable contribution to the understanding of

38

pre
du
rol

fol

lee
the
id:
en

re

athlete preferences of coach leadership, his notion of congruence between athletes’
preferred leadership style and coaches’ leadership behavior does not directly apply to the
dynamic of peer leadership. The leadership style of the athlete leader will vary given the
roles and responsibilities that are typically deﬁned by the coach and accepted by peer
followers. Therefore, the athlete leader is in a very different leadership position than that
of the coach, and thus, ﬁrture research should construct measures of athlete leadership to
reﬂect these differences.

One of the common themes to emerge from this collection of work is that
leadership is not an inherent trait in the leader, but rather a pattern of attributes and skills
that have the potential to be developed. While various styles and behaviors were also
identiﬁed by this work, two prevailing dimensions of leadership behavior consistently
emerged across the studies. These two predominant dimensions of leadership were
related to the task (i.e., initiation of structure, instrumental) and interpersonal (i.e.,
consideration, expressive) elements. Accordingly, these dimensions may also be pertinent
to the role of the athlete leader. For example, certain athletes may provide leadership that
is speciﬁc to the performance of the sport task and may set an example for peer athletes
through the demonstration of hard work, dedication, and consistently performing at one’s
best. In contrast, certain athletes may provide leadership that is relevant to the
motivational aspects of sport by providing inspiration and encouragement to teammates,
or fostering cohesion and team unity. Future research should examine whether the
dimensions of performance and motivational leadership are a function of the athlete
leader role, and if so, should identify the determinants and regulatory processes of these

leadership roles.

39

the

let

er
bet

C0

id:

1111

hll

me

Another theme that emerged from this summary of research from other domains is
related to developmental and motivational issues with regard to leadership and the
importance of leader ﬂexibility. Older athletes who are inﬂexible and take advantage of
the power associated with being an athlete-leader may not be perceived as an authentic
leader by his or her teammates as compared with athlete-leaders who are more ﬂexible
and considerate of the developmental needs of teammates. For example, incoming
freshman becoming acclimated to the workings of the team may ﬁnd athletes with more
experience inspiring if these seasoned athletes provide the social support necessary to
become adjusted. The mere transition into class status (i.e., junior and senior) may not
correlate with leadership, but rather, athletes with more experience who are willing to
share their experiences and teach younger athletes about commitment and the positive
aspects of the sport may be more likely to emerge as leaders. Future exploration of the
identiﬁed leader’s motivational mechanisms will provide a better understanding of the
underlying dimensions of the leadership process.

This summary of research, in particular, portrays the complexity of the leadership
process and the need to consider, among other things, the leader/member match, the type
of situation, the type of task being performed, and the amount of power being allocated to
the leader. For example, is the same individual designated as the leader by the coach also
seen as the leader by the athletes? Does this athlete leader take advantage of power
allocated to him or her? Is the same individual who is seen as inspiring at practice also
considered a leader in competition and under pressure situations? Due to the inherent
hierarchy of sport, the possibility of athlete leadership and the success of this leadership

may be partially determined by the coach. Therefore, future work should examine how

40

C01”

3th

reg

151

int

hu

”C?

athlete leaders were selected, the criteria for this selection, and whether the coach
consistently endorses the role of the athlete leader on his or her team.

Due to the complexity of the leadership process, future research in sport needs to
adopt a theoretical approach to the study of leadership that will test the antecedents and
regulatory mechanisms of leadership. The selected theory should explain how leadership
is constructed and given meaning by and maintained among athletes, and at which point

in the process the different layers may occur (e. g., individual versus team).

Social Cognitive Theory

In order to address the limitations found in leadership conducted on athletes,
social cognitive theory was chosen as the theoretical framework to study the reciprocal
inﬂuence between cognitive/personal factors, social inﬂuences, and behaviors to explain
human ﬁrnctioning and motivation. Speciﬁcally, the social-cognitive perspectives on
achievement goals, gender roles, and self-efficacy beliefs were used to delineate athletes’
construction of meaning and motivational mechanisms of leadership (Ames, 1992a,
1992b; Bandura, 1997; Nicholls, 1989). Central to Bandura’s (1986, 1997) social
cognitive theory is the notion of human agency, or the intentional thoughts and behaviors
which ultimately reﬂect an individual’s desire to generate actions relative to a speciﬁc
goal or objective.

Human agency may be inﬂuenced by personal beliefs regarding intention,
forethought (i.e., direction and coherence), self-reaction (i.e., monitoring and correction)
and self-reﬂection (i.e., meaning) of action (Bandura, 2001a, 2001b). Agency ﬁrst begins
with intentions or plans to take action. Individuals can create cognitive representations of

past, present, and future experiences to anticipate the possible consequences of ﬁrture

41

 

attic

pets

thei

mo:

011C

the

C01

eng

the

be

far

hu

11.

d1

actions. By using symbols or internal representations of behavior, individuals create a
personal standard, one they wish to achieve, which serves as a motivator and regulator for
their behavior. Individuals may then use self-reaction and self-reﬂection to evaluate their
motivation, values, and meaning of their actions. Bandura (1997; 2001a) has asserted that
once these agency-related beliefs are formed, they will inﬂuence whether an individual
chooses to perform a certain activity, the amount of effort one will expend, and the
degree to which one will persist on a task. For example, an athlete may have a symbolic
conception of what it means to be a leader in sport, and based on this standard, will
engage in leader conduct and monitor his/her actions accordingly.

Embedded within social cognitive theory is the notion of triadic reciprocality, or
the interrelationship between personal, behavioral, and environmental factors that are
believed to inﬂuence agency beliefs. Although Bandura (1997) hypothesized that these
factors inﬂuenced each other bi-directionally, he also proposed that they did not inﬂuence
human agency equally, but rather that each component prompted cognition and behavior
with varying levels of strength and degree. It is this interplay of self-generated (previous
behavior, desired standards) and external sources of information that helps a person
deﬁne his/her self-capabilities and inﬂuences future actions.

Therefore, one component of human agency is leadership, in which an individual
may actively choose to lead a group, the amount of effort to put forth to guide and direct
a group, and the amount of persistence required to move the group to the successful
collective action. Rather than completely concentrating on the traits of the leader,
Bandura (1997) claimed that the application of social cognitive theory to the study of

leadership more adequately illustrated the reciprocal interaction between the leader and

42

followers. For example, he described how different leaders might have very different
styles and approaches to leadership, but are similar in their efficacy to motivate their
followers regardless of the situation. This same principle may be applied to the study of
leadership in athletes. By adopting a social cognitive framework, sport psychology
research can begin to examine the interactions between the motivational components of
choice, effort, and persistence in athlete leaders.

Social Cognitive Yheory of Leadership

Early research on leadership stemming from a social cognitive perspective mainly
focused on the followers’ cognitive processing of leader effectiveness. This research
represented a prototype or “implicit theories” of leadership that reﬂected followers’
leader preferences (Smith & Foti, 1991). Murphy (2002) recently proposed that
leadership should examine the social cognitive view of the self in the role of the leader.
Speciﬁcally, she argued that research should explore the structure of the self, taking into
consideration personal attributes, social roles, past experiences, and future goals that
represent a self-schema, in addition to the function of the self, or how individuals manage
to enact behavior.

Bandura and Wood were the ﬁrst to test self-regulative functions in a leadership
situation. Managerial decision-making was experimentally manipulated by controlling
perceptions of ability (Wood & Bandura, 1989) and controlling the environment
(Bandura & Wood, 1989). In essence, these investigations discovered that when people
perceived their ability as a ﬁxed capacity, were more likely to suffer from self-doubt, a
weak sense of efficacy, and a disinterest in the task to be performed. In contrast,

individuals who viewed ability as an acquirable skill were more likely to adopt positive

43

beliefs and self-reactions to performance attainments, persist when faced with difﬁculty,
and set challenging goals. This model of leader self-regulation shows that, how the leader
cognitively appraises sources of information has an important impact on the ﬁnal
decisions that are made. Therefore, an individual who holds the belief that he/she can
lead the group through adversity and times of change is more likely to persevere when
faced with obstacles. This type of perseverance provides an essential function of the
leadership process.

Recent reviews on leadership research also discussed the need to examine
leadership as a group level process (Chemers, 2000; Hogg, 2001). Hogg (2001) cogently
stated that leadership was a relational property within groups (i.e., leaders exist because
of followers, and followers exist because of leaders) and argued for the importance of
examining leadership through understanding the social cognitive properties of
psychologically belonging to a group. On the other hand, while Chemers’ (2000)
integrative theory of leadership depicted leadership as part of a group dynamic, with the
primary ﬁsnction of leadership being collective achievement. Therefore, Chemers argued
that future research should explore the role of leadership within a social cognitive
perspective and consider individuals’ leadership efficacy relative to task self-efficacy and
collective efﬁcacy of successful mastery.

Leadership Effects

Previous research on leadership in athletes was problematic due to the various
deﬁnitions and measurements used to assess leadership outcomes. Leadership research
from other domains has identiﬁed different leadership criteria such as leader

identiﬁcation, effectiveness, and evaluation. These primary indicators of leadership are

44

distinct from one another and may be assessed from the leader’s perspective using self-
report measures or the followers’ perspectives using peer-ratings. For example, leader
identiﬁcation has been examined both in experimental and ﬁeld settings (Bass, 1990;
Biernat et al., 1998; Zaccaro, Foti, & Kenny, 1991). One method used to identify the
leader is to have each group member provide the leader with a score and/or rank order
their preference for the leader of the group (i.e., top 3 leaders in the group). The
individual with the highest score or rank is then considered the leader of the group as
identiﬁed by his/her peers.

In addition to the identiﬁcation of the leader, the perceived effectiveness of the
group leader is also considered an important dimension of the leadership process. Leader
effectiveness is typically determined relative to a successful performance outcome, such
as the number of correct responses or successful completion of the task (Bass, 1990).
This begs the question, “Is an athlete-leader effective only if the team wins?” Watson,
Chemers, and Prieser (2001) found a strong relationship between teammates’ evaluations
of leader quality and collective efﬁcacy perceptions in basketball teams. However,
perceived leader effectiveness did not signiﬁcantly predict team offensive (average points
per game) and defensive performance (average points allowed per game) or overall
success (team rank). Both individual and collective efficacy beliefs accounted for a
signiﬁcant portion of the variance in these performance variables. Based on the ﬁndings
of Watson et al., (2001) it is possible that teams with low to marginal performance
outcomes may still evaluate the group leader as effective, and that the relationship
between perceived leader effectiveness and performance may be mediated by the team’s

collective efficacy. Leadership effectiveness may also be measured by evaluating the

45

frequency and consistency of leader behaviors from both the leader and follower
perspectives.
Self-Regulatory Mechanisms of Leadership

Until recently, limited attention has been given to how leaders cognitively
represent themselves as leaders and how their ability to engage in appropriate leader
conduct is relative to this cognitive standard. Bandura’s (1997) social cognitive theory
identiﬁed important motivational and self-regulatory mechanisms rooted in cognitive
activity that may regulate leadership development and functioning. One of the primary
mechanisms proposed to regulate leadership thoughts and behavior was self-efﬁcacy
(Chan & Drasgow, 2001; Chemers, 2000; Chemers, Watson, & May, 2000; Murphy,
2002; Paglis & Green, 2002; Watson, Chemers, & Preiser, 2001). Speciﬁc to leadership,
efﬁcacy beliefs reﬂect an individual’s conﬁdence in his/her ability to lead a group toward
the accomplishment of goal to a certain degree or level.

In the military setting, Chemers et al. (2000) used a construct that assessed military
offrcers’ conﬁdence in their generalized ability to lead others and to engage in speciﬁc
leadership skills. Their results demonstrated that individuals who reported higher
leadership efficacy were rated more strongly on objective performance evaluations in
leadership simulations and demonstrated higher leadership ratings from peers and
superiors. Furthermore, leadership efficacy emerged as the most signiﬁcant predictor of
leader performance evaluations, and accounted for more variance than self-esteem and
perceived optimism. More recently, Chan and Drasgow (2001) examined efficacy beliefs
relative to military and collegiate students’ motivation to lead (MTL). Individual

perceptions of leadership efficacy were included in their conceptual model as both an

46

antecedent and mediating mechanism. Leadership efficacy beliefs independently
predicted the motivation to lead and mediated the relationship between personality and
motivation to lead.

Recently, Paglis and Green (2002) developed a multi-dimensional construct of
leadership efficacy in managers. Speciﬁcally, they identiﬁed three dimensions of
leadership efﬁcacy related to direction-setting, gaining-commitment, and overcoming
obstacles. The ﬁndings revealed that a participant’s self-esteem, internal locus of control,
perceptions of job autonomy, and subordinate level of ability signiﬁcantly predicted the
direction-setting and overcoming-obstacles dimensions of leadership self-efﬁcacy.
Leadership efﬁcacy was also related to subordinate’s perceptions of manager’s
motivation for leading change and found that the direction-setting and gaining-
commitment dimensions were signiﬁcantly correlated with subordinate ratings. This
study suggests that leadership efficacy is multi-dimensional and parallels the different
dimensions of leadership. However, it should be noted that the total or unidimensional
subscale was more predictive of group outcomes than the multidimensional approach.

To date, Watson et al., (2001) has provided the only account of leader conﬁdence
in athletes, and this perceived conﬁdence in leadership, in addition to the evaluation of
leader quality, was examined relative to collective efficacy. The authors found a strong
relationship between teammates’ evaluations of leader quality and collective efficacy
perceptions in basketball teams. In essence, teams that perceived greater effectiveness in
their leader’s capabilities reported greater conﬁdence in their teams than did individuals
who perceived less effective leadership on their teams. According to Chemers’ (1997)

integrative theory of leadership, efﬁcacy to lead others should be measured in relation to

47

one’s efficacy to perform the skill individually (i.e., task self-efficacy) and as a group
(i.e., perceived collective efficacy).

However, Watson et al., (2001) has utilized a measure that assesses general
leadership effectiveness (i.e., “ my captain’s behavior is very motivating to me”), which
makes it difﬁcult to ascertain what it is exactly that leaders do. Therefore, future studies
in sport should expand beyond the mere identiﬁcation of athlete leaders and leadership
behavior in order to clearly operationalize leadership that is speciﬁc to athletes’ roles and
responsibilities. For example, athletes should rank order their preferred leader, and
evaluate the leader relative to speciﬁc skills that are relevant to the type of leadership
being provided (i.e., performance versus motivational leadership). More importantly,
additional research is needed to establish the determinants and regulatory mechanisms
that predict leadership effects and distinguish the preferred leaders from the non leaders.

An additional mechanism proposed to inﬂuence leader performance and
effectiveness was the acquisition of leadership skills (Mumford et al., 2000). Recent
trends in leadership research in the organizational/industrial and military settings
attempted to assess the process of leadership by adopting a skill-based model of
leadership development. Research within this perspective examined the learning and
development of skills that inﬂuence leadership rather than emphasize the effectiveness of
leadership behavior or outcome. First, leadership skills may be partially distinguished
from personality traits in that they are not always inherent attributes of the leader, but
rather can be learned and developed over time (i.e., complex problem solving skills,
creative thinking, etc.). Second, leadership skills are not entirely dispositional in nature as

the use of these skills ﬂuctuates given certain conditions. Finally, by examining the skills

48

of leadership, one can understand the developmental processes that may be involved in
teaching people how to become leaders.

Mumford and his colleagues have examined leadership skills in the military
setting (Mumford et al., 2000a, 2000b). While his work on leadership skills has been
ﬁeld-based and speciﬁc to the military, the organizational structure of the military is
hierarchical, and may provide useful ideas for future research on leadership development
in athletes. For example, Mumford et al., (2000a, 200b) examined the developmental
differences in the multi-dimensionality of leadership patterns and skill acquisition among
military officers as a function of experience and timing of leadership opportunities. They
examined the development of the speciﬁc leadership skills: complex problem solving,
solution construction, social judgment, creative thinking, and leadership expertise. As
expected, the senior-level leadership positions, by nature, required multi-dimensional
leadership patterns and skills. Speciﬁcally, senior level ofﬁcers were better at translating
new ideas into actions, and spent more time appraising the implications of a novel
situation, which ultimately resulted in a higher quality solution to the problem. The
implication of the skills-based model was that not only was there a subsequent
development in skills relative to the rank of the individual, but also that this development
occurred in a progressive systematic fashion. The authors also found that there were
potential differences between the impact of assignments and the training on leadership
skill development, and recommended that potential interventions consider the nature of
the skills that are relevant for the leader given current rank and thus developmental level.

Similar to the military setting, athletes may also experience developmental

transitions in leadership skill development. For example, varsity leaders in sport may

49

display different leadership characteristics than junior varsity or novice athletes with less
experience in sport. In the same manner in which skills were identiﬁed as leadership
skills in the military setting, future research on leadership in athletes should identify skills
that would be pertinent to leadership and collective achievement in sport. Individual
differences in leadership skills may reﬂect meaningful developmental transitions in the
acquisition and use of sport-speciﬁc leadership skills, which may in turn reveal individual
differences in leader identiﬁcation and effectiveness. Further exploration of the viable
skills that are acquired and developed through sport experiences may serve as a possible
regulating function of leadership.
Social Cognitive Determinants of Leadership

Efficacy beliefs and regulatory skills were hypothesized to regulate human
functioning through four mediating processes: cognitive, motivational, affective and
selective processes. The cognitive appraisal of potential efﬁcacy determinants was
proposed to include two distinct processes that enabled an individual to attend to and use
certain cues as potential indicators of efficacy, and then to integrate and weigh these
sources of information to form efficacy judgments (Bandura, 1986, 1997). Bandura has
explained the importance of the selection process people undergo in the shaping of the
environment in which they exist. These selection and appraisal processes may occur at
both the individual level regarding personal efficacy and at the group level, regarding
one’s conception of the group’s capability, or collective efficacy (Bandura, 2001a; Feltz
& Lirgg, 1998). People who believe in their ability to lead others will actively choose
leadership responsibilities and appraise information regarding their leadership ability

accordingly.

50

While social cognitive theory has identiﬁed many important factors used to assess
efficacy judgments (e. g., perceived importance of the skill, requisite knowledge to
perform the skill, task difﬁculty, perceived ability, etc), two belief systems have been
found to demonstrate a signiﬁcant inﬂuence in the cognitive appraisal of efficacy
information- conception of ability and perceived controllability of the environment.
Therefore, construction of knowledge and meaning will inﬂuence the appraisal of
information surrounding people regarding their ability to achieve a certain task. This
construction occurs through personal orientations and perspectives (i.e., gender and
achievement) and through the perceptions of one’s environment and context (i.e.,
motivational climate).

Gender Roles

Understanding human motivation and behavior on the basis of gender deserves
considerable attention due to the subtle yet unwavering inﬂuence of gender on our
everyday lives GBredemeier, 1992; Eagly, 1987; Gill, 1994). While sex differences
typically refer to the biological differences between women and men, the term “gender”
is used to reﬂect the social and psychological attributes and behaviors of women and
men. Various interrelated theories have been proposed to explain psychological,
biological, and sociostructural processes by which conceptions of gender were developed
and gender behavior was produced. Early viewpoints of gender differences in psychology
and mainstream society depicted the masculine and feminine dimensions as a single
continuum in which individuals were classiﬁed as either masculine or feminine. This
restricted dichotomy did not allow for individuals to encompass both characteristics, or

what was considered an androgynous attribute. Furthermore, social scientists became

51

concerned about the distinct comparisons being made between women and men and
noted the importance of the similarities and considerable overlap between the two sexes
(Bredemeier, 1992).

In an effort to address these limitations, Bem (1974) proposed that our gender
schema, or what she deﬁned as our network of associations and experiences embodying
the culture’s conception of sex roles, was used as a lens to guide our behavior and our
interpretation of the environment. Speciﬁcally, she emphasized that the dimensions of
masculinity and femininity were not on a single continuum that dichotomized individuals,
but rather were independent constructs that allowed for individuals to vary, in fact, in
both masculinity and femininity. To empirically test this variation, Bem (1974)
constructed the Bern Sex-Role Inventory (BSRI), a 60-item questionnaire with 20 items
representing the stereotypical feminine attributes (e. g., affectionate, gentle), 20 items for
masculine characteristics (e. g., ambitions, assertive), and 20 neutral or ﬁller items (e. g.,
truthful, happy). The inventory was constructed based on the belief that people
internalized society’s sex-typed standards of desirable behavior for men and women. As
such, the items were generated based on the sex-typed social desirability and not on the
differential endorsement by males’ and females’ cultural deﬁnitions of sex-typed
behavior. The BSRI deﬁned masculinity and femininity as orthogonal constructs, thus
individuals could be classiﬁed as high on both dimensions (androgyny), low on both
dimensions (undifferentiated), or high on one dimension and low on the other (either
feminine or masculine). Bem later provided a 30-item reduced version of the BSRI (30
items) with the items selected in order to maximize both the internal consistency and

orthogonality of the scales (Bem, 1974; Covey & Feltz, 1991).

52

Although Bem’s research provided an important contribution to the understanding
of individual differences in the cognitive components of gender schematic processing,
critics of gender schema theory argued that people do not categorize themselves as either
male, female, or androgynous, and act in accordance with this conception invariantly
across various situations. More importantly, certain researchers claimed that
sociostructural components of gender must also be considered rather than just the
cognitive or psychological mechanisms of gender roles. Within a sociological framework,
the sources of gender differentiation were based more in social and institutional practices
than ﬁxed properties of the individual.

One predominant theory that has emphasized society’s structural inﬂuence on
gender differentiation is Eagly’s social role theory (Eagly, 1987). The theoretical tenets
of social role theory postulated that sex differences in social behavior were a result of the
social roles that regulated behavior in adult life. Within the social role theoretical
approach, Eagly adopted a slightly different deﬁnition of sex and gender than most
gender theorists. She conﬁned the use of sex differences to the observed differences
between male and female behavior and gender to the meanings that society and
individuals assign to female and male categories. Social roles were hypothesized to
emanate from shared expectations or stereotypes that applied to people of a particular
category or social position. Gender roles were deﬁned as the shared expectations about
appropriate behavior and dispositional qualities that were consistent with their socially
identiﬁed gender. Therefore, the social norms regarding appropriate attributes and

behavior speciﬁc to women established the female gender role, while the shared account

53

of acceptable characteristics and behavior for men constituted the male gender role
(Eagly, 1987) .

Social role theory was developed partly in response to limitations found in earlier
theoretical accounts of gender (i.e., gender schema theory), and the psychological
research on sex differences. Speciﬁcally, Eagly’s position has been that understanding
childhood socialization and development does not necessarily shed light on the factors
that maintain sex differences in adults. Social role theory has placed an emphasis on the
structural and cultural inﬂuences on adult social roles and has provided a more proximal
predictor of adult social behavior. According to social role theory, the set of skills,
competencies, and beliefs that people acquire when occupying a certain social role
function as primary determinants of sex differences (i.e., differences between men and
women). Therefore, as an individual occupies a certain social role, say that of a female
athlete, she would acquire a set of skills and competencies relevant to canying out
behaviors speciﬁc to this role. Further, the female athlete may acquire a variety of
attitudes and beliefs as she collects information, interpreting events and interactions from
the lens of her respective role.

Social role theory postulated that there were differential expectations of men’s
and women’s attributes based the societal division of labor between the sexes and the
socially constructed expectations associated with the gender roles that men and women
generally hold (Eagly, 1987; Carli & Eagly, 2001). Eagly (1987) has classiﬁed the
traditional notion of masculinity as agentic and communal for femininity. Men were
expected to possess predominately agentic attributes (i.e., independent, masterﬁrl,

assertive, and competent), while women were classiﬁed as having high levels of

54

communal qualities (i.e., friendly to others, unselﬁsh, concerned for others, and
emotionally expressive). Therefore, it was assumed that men would attempt to control
their environment and obtain tangible outcomes, such as task completion and high
performance. In contrast, women were expected to engage in more socially-oriented
behaviors and show more concern for interpersonal relations.
Gender and Leadership

Research exploring the relationship between gender and leadership has produced
equivocal ﬁndings (Chemers, 1997; Eagly & Johannesen-Schmidt, 2001). In an effort to
better understand the role of gender in the organizational context, Eagly and her
colleagues have synthesized the research on gender and leader emergence, style and
evaluation (Eagly & Johnson, 1990; Eagly & Karau, 1991; Eagly, Makhijani, & Klonsky,
1992). Speciﬁcally, Eagly wanted to determine if sex differences in leaders could be
explained by the theoretical tenets of social role theory. Beginning with leadership style,
Eagly and Johnson (1990) analyzed two predominant styles, task and interpersonal style,
which had been identiﬁed in previous research. With task style the leader directed the
group speciﬁc to the accomplishment of task characteristics and performance. In contrast,
with interpersonal style, the emphasis shifted from the task and performance to the
maintenance of interpersonal relationships, in which the primary focus of the leader is
looking out for the well-being of subordinates and establish friendships with group
members. The authors also examined the leader’s tendency to espouse a democratic
versus autocratic style of delivery. Consistent with the theoretical perspective of social

role theory, women were expected to adopt an interpersonal leadership style that was

55

more democratic in nature, while men were expected to exhibit task leadership with more
of an autocratic approach.

However, the authors proposed that sex differences in these leadership styles
would be less pronounced in organizational settings (i.e., ﬁeld studies) that compared
occupants of the same managerial role than assessment and laboratory research (i.e.,
experimental settings). Their position was that individuals who occupied organizational
leader roles were typically socialized into these roles in which the expectations and
guidelines regarding effective style were explicit and clear to the occupant. In contrast,
people in experimental settings interacted with strangers on a short-term basis, and the
protocol concerning acceptable leadership behavior was more ambiguous and, therefore,
should result in more gender stereotypic behavior. Their synthesized results supported the
proposed hypothesis by failing to detect a signiﬁcant difference between female and male
leadership styles among the collection of studies that were conducted in the
organizational ﬁeld setting. Whereas, gender stereotypic leadership styles were present in
assessment and laboratory research settings. Furthermore, gender stereotypic differences
in autocratic and democratic styles emerged in all three categories of research
(organizational, assessment, and laboratory).

Following the synthesis of research on leadership style, Eagly and Karau (1991)
conducted a second meta-analysis on the relationship between gender and leader
emergence. Speciﬁcally, the authors explored the relationship between the gender-type of
task and the subsequent stereotypic classiﬁcation of behavior involved in leadership.
Task activities were deﬁned as a member’s direct contribution to the group’s task, and

were measured by the number of attempted answers to problem-solving tasks. In contrast,

56

social activity was deﬁned by the member’s attempt to maintain satisfactory morale and
interpersonal relationships among group members, and was measured through the
assessment of positive reactions, or the individual’s attempt to show solidarity and
agreement with group members. In line with the theoretical tenets of social role theory,
the authors hypothesized that there would be sex differences in leader emergence among
initially leaderless groups relative to the gender typing of the task to be performed. Men
were hypothesized to engage in more task activity contributions while women were
hypothesized to contribute more to group social activity. Results demonstrated that men
were more likely to emerge as leaders in short-term groups and in groups that were
performing predominately task-oriented activities (i.e., did not involve complex social
interactions). Women, on the other hand, were found to emerge as social leaders slightly
more than men.

Related to the evaluation of leaders, Eagly et al., (1992) examined the subordinate
evaluation of men and women holding leader roles, controlling all characteristics except
sex of the participant. Overall, their synthesis revealed a small tendency for subordinates
to evaluate male leaders more favorably than female leaders. However, this evaluation
was contingent upon leadership style and subordinate sex. Speciﬁcally, female leaders
were devalued when leadership was carried out in a stereotypic masculine-agentic and
autocratic fashion, and when the subordinates were men. Eagly and Johannesen-Schmidt
(2001) emphasized the importance of gender roles because the roles of men and women
had different content, and more importantly, because of the need for women to negotiate
the contradiction that existed between the female gender role and the leader role. In

particular, females adopt roles that were perceived by others to encompass predominately

57

communal qualities, which opposed the predominately agentic qualities that were
believed to be required to succeed as leaders.

While Eagly’s work has provided some valuable contributions to the
comprehension of gender and leadership, information regarding leadership in same-sex
groups (i.e., female leaders and female subordinates) is less understood. This lack of
understanding may be due to the fact that there are few same-sex groups in the
work/organizational setting. Plus, the premise of social role theory is the division of labor
among the sexes and women’s struggle to gain entry into the workforce. Although cross-
gendered groups are less likely to exist in the sport setting, social role theory still
provides a provocative account of the intersection between gender and leadership in
adults. Furthermore, Eagly’s approach has allowed for the examination of perceived
gender roles relative to the skills that are believed to be important for athlete leaders.
Therefore, sport provides a valuable context to apply the social cognitive perspective of
social role theory in order to understand the relationship between gender and skills and
perceived effectiveness and evaluation of female leaders.

Gender Roles in Sport

Early research on gender and sport mainly focused on issues related to role
conﬂict and the classiﬁcation of sex roles among male and female athletes. Similar to the
opposing qualities in the gender role versus the leader role, sport researchers were
interested in the simultaneous occupation of gender and athlete roles and how female
athletes negotiated the extreme contradictions between athleticism and femininity. More
current research has discovered that female athletes do not perceive a conﬂict between

being an athlete and being feminine. In fact, current female conceptions of the athlete role

58

not only challenge the importance of being communal (feminine), but rather endorse the
characteristics of being agentic (masculine)(Krane, 2001).

However, the relationship between gender and leadership in sport could beneﬁt by
further study. Due to the emphasis on leadership in coaches, there is yet a limited
understanding of the simultaneous occupation of the athlete, gender, and leader roles in
sport. Therefore, more research is needed to understand what exactly “leadership” means
to the female athlete in the context of sport. Furthermore, a problem exists with common
deﬁnitions of leadership because they are presented as though leadership is gender
neutral, when, in fact, normative conceptions regarding leadership reﬂect the emphasis of
agentic attributes. As a result, women are perceived to lack the typical leadership
characteristics, such as aggression, competitiveness, and dominance (Eagly & Karau,
1991). However, some researchers have argued that women do not lack the prerequisites
for leadership, but rather lack the opportunities for exercising leadership and lack the role
models they admire and wish to emulate (Klenke, 1996; Pemberton, 1997).

This predicament of limited role models and opportunities to develop leadership
is clearly an issue for women in sport as women have struggled to maintain consistent
leadership (i.e., coaching and administration) (Pemberton, 1997). There is an emerging
positive perspective regarding women’s increased movement into sport as a viable way to
challenge traditional conceptions of gender differences. Klenke (1996) has argued that
the setting of sport provides a provocative context from which to examine the
development of leadership in women because women’s position in sports is a reﬂection
of the changing role of women, not only in this particular context, but also in society in

general.

59

To date, Glenn and Horn (1992) have provided the only account of gender and
leadership in female athletes. Their results demonstrated that both agentic (masculine)
and communal (feminine) dimensions predicted self-reported leadership ability among
high school female soccer players. While agentic attributes emerged as the only predictor
of the two that explained peer ratings of leadership, neither agentic nor communal
attributes predicted coach ratings of leadership. This research provided evidence that the
female athlete role does not necessarily parallel that of the female gender role. Thus, in
this context it is possible that agentic attributes are viewed as more relevant to the roles
and responsibilities of the female athlete (Krane, 2001).

Bussey and Bandura (1999) have postulated that gender schema and sociological
theories focus separately on gender conceptions and the sociocultural inﬂuences on these
conceptions however, they make little effort to integrate the two perspectives. Previous
research on gender has predominately emphasized the normative expectations of gender
and their inﬂuence on gender behavior, but has failed to examine the motivational and
self-regulatory mechanisms behind role occupation. The authors have proposed that
merely knowing a stereotype does not necessarily mean that one strives to behave in
accordance with it, and the have recommend social cognitive theory as a solid theoretical
framework ﬁ'om which to examine the integration of psychological and sociostructural
components of gender roles, and to test the motivational and self-regulatory conceptions
of gender that coincide with behavioral conduct.

Achievement Goals
Previous research on leadership in the work setting has integrated principles of

leadership with the Atkinson’s (1964) expectancy theory of achievement motivation

60

(McClelland & Boyatzis, 1982; Sorrentino, 1973; Sorrentino & Field, 1986). The
achievement motivation perspective, hypothesized that individuals who were highly
motivated would attempt to lead in problem-solving situations. McClelland and Boyatzis
(1982) identiﬁed the Leadership Motive Pattern, or the manager’s high need for power,
low need for affiliation, and high activity inhibition, which consistently identiﬁed
managers who advanced within corporations.

1n controlled settings, researchers examined small groups working on problem-
solving tasks and discovered that achievement motivation predicted individual
differences in leadership. Sorrentino and his colleagues discovered that individuals
scored higher on leadership measures if they scored high on need for achievement and
affiliation compared with individuals who scored lower on achievement and affiliation
(Sorrentino, 1973; Sorrentino and Field, 1986). However, the exact mechanism that
linked achievement with leadership has remained ambiguous given the deﬁnition of
achievement behavior from the expectancy theory perspective. Nicholls (1984) has
argued that his deﬁnition of achievement behavior as “the behavior in which the goal is
to develop or demonstrate competence to oneself or to others (i.e., demonstrate high
ability or avoid demonstration of low ability)” is more exact and easier to measure than
the expectancy theory deﬁnition of achievement behavior (i.e., obtain a standard of
excellence). Furthermore, the Leader Motive Pattern, or the high need for power and low
need for achievement, may not exert the same predictive utility in the sport setting as
observed in the work setting. Therefore, by adopting the conceptual and operational

deﬁnition of achievement goals from an achievement goal perspective (Ames, 1992a;

61

Nicholls, 1984, 1989), future research may clearly delineate individual differences in the
motivational and self-regulatory mechanisms of leadership.

The social cognitive framework of the achievement goal perspective describes
how achievement goals are dispositional tendencies, which exemplify how individuals
deﬁne personal ability, successful experiences, and task difﬁculty (Ames, 1992a, 1992b;
Bandura, 1997; Duda, 2001; Dweck & Leggett, 1988; Nicholls, 1984, 1989). Two
primary achievement goals, task and ego orientation, have been extensively researched in
educational and athletic settings. With task orientation, the objective is to acquire skills
and knowledge, exhibit effort, and experience optimal challenges and personal
improvement. When one is ego-involved, however, there is a preoccupation with personal
ability, or more importantly, the desire to demonstrate superior ability relative to others.

Achievement goal theory provides a framework to allow for the understanding of
how people think about themselves, the tasks they are performing, the evaluation of this
performance, and the purpose and meaning behind their behaviors. Accordingly, the
emphasis of self-mastery and the development of ability versus the use of normative
standards and the demonstration of ability are the two deﬁning standards of perceived
success. When perceived competence is high, both task and ego perspectives are expected
to correspond with the display of adaptive achievement behaviors, such as choosing a
task that is challenging, exerting maximal effort, and persisting when faced with
difficulty. Conversely, an emphasis on ego-involved goals coupled with low perceived
competence will lead to the exhibition of maladaptive behaviors, such as choosing tasks
that are too easy or too hard, exerting minimal effort, and showing a lack of persistence

(Duda, 2001; Dweck & Leggett, 1988; Nicholls, 1992).

62

The
individual d
1992). [Sid
orientations
interest. anw

Newton &

 

proposed tl
ju5tice, an
achieveme
conventio
while Eng
011 nor-mat
with fairn
(task Ofie
reSPEC! fo
pOSIUlatio
Oﬁentatio
aligned ta

PUrpOSe O

 

 

associate.

 

 

The Task and Ego Orientation in Sport Questionnaire was developed to assess
individual differences in goal orientations in the sport setting (TEOSQ; Duda & Nicholls,
1992). Using this construct, researchers have established an empirical link between goal
orientations and achievement-related beliefs such as sources of conﬁdence, enjoyment,
interest, and satisfaction in sport (Duda & Whitehead, 1998; Magyar & Feltz, in press;
Newton & Duda, 1999; Treasure & Roberts, 1998). Nicholls (1989, 1992), however, has
proposed that theories of achievement also reﬂect conceptions of knowledge, fairness,
justice, and general views about the world. Therefore, through the assessment of
achievement goals, researchers may explore how people construct meaning of social
conventions in sport and how these conceptions are regulated by one’s goal perspective
while engaged in a sport task. For example, Nicholls (1989) postulated that an emphasis
on normative comparison (ego orientation) is likely to correspond with a lack of concern
with fairness and justice. Whereas, the perceived importance of self-referenced criteria
(task orientation) will likely coincide with the value of effort, collective engagement,
respect for rules, and proper conduct. Duda (1989) provided empirical evidence for this
postulation in the sport setting. Speciﬁcally, she examined the relationship between goal
orientations and perceived purposes in sport and discovered that high school athletes
aligned task orientation with intrinsic, pro-social, and cooperative views about the
purpose of sport (e. g., master a skill, work well with others). In contrast, ego orientation
was associated with more extrinsic and self-serving motives (e. g., outperform others to
gain recognition).

Achievement goals and views about the functions of sport have also been

associated with variations in sport social behaviors (Duda, Olson, & Templin, 1991;

63

Gano-Overway et al., 2001). In particular, the empirical ﬁndings on the relationship
between goal orientations and sportspersonship attitudes have revealed that athletes with
a predominance in ego orientation are more likely to endorse unsportsmanlike play and
cheating behaviors, and report the use of aggressive acts in sport as a more legitimate
means to succeed. In contrast, the predominance of task orientation with low ego
orientation has been positively associated with sportspersonship attitudes (e. g., respect
for social conventions and rules of sport).

To date, research in sport psychology has yet to link goal orientations to other
dimensions of social skills and behavior such as leadership. For example, leadership
requires skills that represent the underlying dimensions of both task and ego orientation.
If the situation calls for collaboration, such as solving interpersonal conﬂict among
members of the team, one would expect task orientation to emerge as a stronger
correlate. On the other hand, a situation that calls for exerting power over others to make
a ﬁnal decision, would have a stronger association with ego orientation. Therefore, both
task and ego orientation may be related to leadership in some capacity.

Current thinking about achievement goals would beneﬁt ﬁ'om the examination of
additional goal orientations that may function in achievement contexts. Research in
educational psychology has broadened the conception of achievement goals and
considered the ways in which students’ beliefs about the social reasons for academic
achievement (i.e., social responsibility) affect their achievement-related beliefs and
behaviors in the educational setting (Blumenfeld, 1992; Urdan & Maehr, 1995). For
example, Jarvien and Nicholls (1996) examined views about the social world of

adolescents by identifying the various dimensions of social goals and beliefs about the

64

causes of success in peer relationships. Social goals were deﬁned as the kind of social
outcomes individuals preferred, as opposed to outcomes they wanted to avoid in peer
social relations. These goals were assessed with the Social Goal Questionnaire, in which
individuals were asked, “When I’m with people my own age, I like it when. . .,” and
responded to items that reﬂected the dimensions of dominance, intimacy, nurturance,
leadership, popularity and avoidance. One of the dimensions identiﬁed was leadership
(i.e., “I like it when I organize what they do.”).

Based on the work of Jarvien & Nicholls (1996), a similar conception of a social
goal about leadership may function in the sport setting. Commensurate with the
achievement goals that reﬂect the purpose of sport, athletes may also adopt social goals
that reﬂect their beliefs about the social reasons for trying to succeed (e. g, make ﬁiends,
learn responsibility). Therefore, certain social goals may inﬂuence performance,
additional achievement, and social outcomes (i.e., leadership) that are hypothesized to
occur in sport. While current work on goal orientations in sport has examined the notion
of multiple-achievement goals (i.e., self-enhancing ego and self-defeating ego),
exploration of the ways in which social goals in sport interact with task and ego
dimensions to inﬂuence social beliefs and behaviors has yet to be conducted (Roberts,
2001). By examining the independent effects of social goals as well as their interactive
effects with task and ego goals, researchers might come to a more complete
understanding of athlete motivation and achievement in sport pertaining to performance

and social behaviors.

65

Structural Goals

Achievement goal theory also provides a pragmatic approach to the study of
leadership because of the importance placed on the situational goal structure created by
signiﬁcant others (e. g., coaches) in the sport environment. The theory contends that the
degree of task and/or ego involvement is a ﬁinction not only of one’s dispositional
tendencies, but also of the motivational atmosphere created by signiﬁcant others, or what
has been labeled as “perceived motivational climate” (Ames, 1984,1992a, 1992b, 1992c;
Bandura, 1997; Duda, 2001; Maehr & Braskamp, 1986 Nicholls, 1984, 1992).

Ames and her colleagues were among the ﬁrst to examine the goal structures of
achievement settings that were adult-deﬁned and adult-imposed in the educational setting
(Ames, 1984, 1992a, 1992b; Ames & Ames, 1981; Ames & Archer, 1988). Her line of
work began with the investigation of individual differences in attributions, affect, and
learning strategies based on the experimental manipulation of the climate goal structure
as either competitive or individualistic. In the competitive goal structure, participants
were encouraged to evaluate personal performance relative to socially-based criteria (i.e.,
“Let’s see which one of you can solve the most puzzles. . .who will be the winner?”).
Whereas, in the individualistic goal structure, the emphasis was shifted from competition
and outperforming others to deﬁning success relative to one’s personal standards (i.e.,
“Try to solve as many puzzles as you can”).

Findings from this experimental research demonstrated that competitive outcomes
decreased the salience of personal control based on previous performance experience and
effort, and emphasized the focus of performance outcome. The individualistic structure,

however, facilitated a mastery orientation, in which the child attributed success/failure

66

relative to one’s effort, used past performance as a source of evaluation, and also tried
self-leaming strategies, thus fostering a sense of personal control.

Following the series of investigations that experimentally manipulated the goal
structure, Ames and Archer (1988) conducted a ﬁeld correlational study to measure
students’ classroom goals and how they related to choice, effort, and persistence in the
classroom. Similar to the individual goal perspective, perceptions of the situational goals
were also hypothesized to be independent from one another, depicting individual
differences in perceptions of the climate. Thus, individuals may have heightened
perceptions of both mastery and performance goals in the classroom (e. g., teacher
emphasizes both self-referenced and performance outcome strategies in the classroom),
perceive low mastery and low performance, or perceive high in one and low on the other
(i.e., high mastery/high performance). Therefore, participants were classiﬁed into one of
the four groups based on their perceptions of mastery and performance goals in the
classroom setting.

Results demonstrated that perceptions of mastery classroom goals were highly
correlated with the use of learning strategies, a positive attitude toward the class, and with
seeking optimal challenges. Children who perceived mastery class goals were also more
likely to attribute success to effort, and strategy, and to credit the teacher for their
success; whereas, mastery goals were negatively correlated with teacher inﬂuence for the
cause of failure. Perceptions of performance goals in the classroom setting were
negatively associated with a positive attitude toward class, and feelings of self-
competence. Related to attributions for success, performance goals were positively

correlated with effort and strategy attributions for success, and ability, strategy, and task

67

difficulty attributions for causes of failure. Furthermore, it was found that students in the
high mastery/high performance, and high mastery/low performance groups reported using
more learning strategies, had a higher positive attitude toward the class, and sought out
more of a task challenge than students in the high performance/low mastery and low
performance/low mastery groups.

The ﬁndings from this line of research were the ﬁrst to highlight empirically the
importance of a high mastery climate in the classroom. The goal structures identiﬁed
represented the underlying dimensions of the classroom climate related to the basis and
type of evaluation, amount of social comparison, and nature and source of rewards being
emphasized by the teacher. In addition, students’ prior experiences and their expectations
about the manner in which they should interact, work together and regard each other
inﬂuenced perceptions of the situational goal structure that was emphasized in the
classroom.

In an effort to measure whether similar goal dimensions existed in the sport
setting, Duda and her colleagues developed the Perceived Motivational Climate in Sport
Questionnaire (PMCSQ), which assessed athletes’ perceptions of the structural goals
created by the coach (Newton & Duda, 1999; Newton, Duda, & Yin, 2000; Seifriz, Duda,
& Chi, 1992; Walling, Duda, & Chi, 1993). Consistent with the work of Ames, two
primary goal structures emerged that reﬂected mastery-(task-involving) and performance-
(ego-involving) climate dimensions. The mastery situational goal structure was
characterized by the perception that an emphasis on effort and mastery of the skill was
rewarded, mistakes were considered a part of the learning process, and collaboration

among team members was encouraged, with every member having an important role on

68

theteani
punnhed

ability di

concepn
perceive
ofconﬁ
Duda l
perceitn
success
satisfac
Climate
sansra,

attribu-

depent
Partici
Slate-[
SlYOng
eValue
age an

per80r

be fun

the team. In contrast, the performance climate reﬂected the perception that athletes were
punished for mistakes, and that the coach recognized and treated players with higher
ability differently than those with lower ability, and encouraged intra-team rivalry.

Previous research on structural goals in sport predominately focused on the
conceptual link between situational goal structures and motivational indices, such as
perceived competence, intrinsic interest, beliefs about the causes of success, and sources
of conﬁdence and satisfaction in sport (Duda, 2001; Magyar & Feltz, in press; Newton &
Duda, 1999; Seifriz, Duda, & Chi, 1992; Treasure & Roberts, 1998). Athletes who
perceived a mastery sport climate reported greater intrinsic interest in the sport, attributed
success to effort, employed self-referenced sources of conﬁdence, and experienced
satisfaction from mastery experiences. Conversely, athletes who perceived the sport
climate to be performance-based had low intrinsic interest in the sport, derived
satisfaction from outperforming others, relied on normative sources of conﬁdence, and
attributed success to ability.

Whether personal achievement goals or structural goals are more predictive
depends upon the nature of the dependent variable and the developmental level of the
participant (Duda, 2001; Dweck & Leggett, 1988). For dependent variables that are more
state-like in nature (e. g., efficacy beliefs) the motivational climate may emerge as the
stronger predictor, whereas, with variables that are more dispositional (e. g., self-
evaluation of leadership) goal orientations should emerge as the primary predictor. The
age and developmental level of the participant may also inﬂuence the predictive utility of
personal and structural achievement goals. Younger children’s goal orientations may not

be ﬁrlly developed, and thus, the climate may override dispositions and predict

69

achievement-related outcomes. However, researchers also have proposed that the
interaction between personal and structural goals be examined to determine if the
interaction between the two prevail over independent contributions.

Similar to the proposed relationship between personal achievement goals and
perceived purposes of sport, research has established a link between structural goals and
views about the purposes of sport (Ommundsen & Roberts, 1999). Perceptions of a task-
involved climate were associated with the view that sport serves the purpose of the
development of lifetime skills and social skills, whereas perceptions of an ego-involved
climate were associated with the perception that sport served to enhance one’s social
status.

While previous research has established the conceptual link between perceived
structural goals and achievement-related outcomes, limited research has examined the
inﬂuence of the situational goal structure established by the coach and social behaviors in
sport. Preliminary support for the theoretical relationship between motivational climates
and sportspersonship has been provided (Kavussanu & Roberts, 2001). Duda (2001),
however, has noted the potential limitation with the current statistical method used to
examine perceptions of motivational climate. Speciﬁcally, she has questioned whether
individual athletes on a given team share perceptions of the motivational climate because
previous research has overlooked this potential for a team effect of the motivational
climate. In her most recent work, Duda and colleagues (as cited in Duda, 2001) have
attempted to account for the potential team effect. These ﬁndings provided evidence for a
team effect with teams displaying a stronger agreement concerning group perceptions of

the mastery climate, and less agreement regarding the performance climate. Duda has

70

contended that it is incorrect to assume all athletes within the total sample of participants
will perceive the same dimension, but rather data collected on teams should display group
and individual level effects.

Related to predictors of efﬁcacy, Magyar, F eltz, & Simpson (2002) examined
motivational climate as group level predictors of aggregated collective efficacy beliefs in
junior rowers. Results demonstrated that crews that perceived a predominant mastery
climate were more likely to report cogent perceptions of collective efficacy. Performance
climate, on the other hand, did not emerge as a predictor or collective efficacy, thus,
ﬁrrther demonstrating the importance of establishing a mastery climate. Bandura and
Wood (1989) have discussed that when groups perceive a collective endeavor as one that
is unmanageable, the group may perceive a weakened sense of efficacy making group
accomplishments difﬁcult to envision. This ﬁnding lends support to a major hypothesis
of social cognitive theory (Ames, 1992a, 1992c; Bandura, 1997; Duda, 2001) that the
perceived situational goal structure of the sport setting can not only have an impact on
individual achievement-related beliefs, but also impact collective perceptions of
achievement as well.

Recently, Gano-Overway et al., (2001) examined the inﬂuence of motivational
climate perceptions on sportspersonship at both the individual and group levels. Results
found a three-way interaction between achievement goals, mastery climate, and
sportspersonship at the level of the individual athlete. Speciﬁcally, under a mastery
climate, athletes with higher levels of ego orientation had a signiﬁcant positive
relationship between task orientation and sportspersonship. When athletes’ perceptions of

the task-involving climate were low and levels of ego orientation were low, a signiﬁcant

71

positive relationship emerged between task orientation and sportspersonship. When
aggregated up to the team level, the perceptions of a performance climate emerged as the
signiﬁcant predictor, and negatively predicted sportspersonship beliefs among the team.
Therefore, it is possible that perceptions of the motivational climate may inﬂuence social
behaviors at both the individual and team levels.

Maehr and Braskamp (1986) have discussed the inﬂuence of group membership
and the expectations of others on achievement-related beliefs. The authors proposed that,
to a certain degree, the normative and role-related expectations of the group may
determine the value of engaging in certain behaviors. For example, the set of expectations
that exist for an individual who is identiﬁed as the leader of the group may be inﬂuenced
by both the coach and/or teammates. The manner in which these expectations are
communicated to group members may be reﬂected through the situational goal structures
that are established and made most salient to athletes. The authors also discussed
perceptions of justice at the group level. In the work setting, perceptions of equity in the
group have been found to correlate with the desire to belong, participate, and commit to
the group goals. When workers feel the rewards that are being distributed are not
compensatory with the amount of effort invested, issues of equity may become
problematic. Therefore, structural goals may explain variation in group level perceptions
of leadership.

To date, few studies have attempted to connect the constructs of stmctural goals
with leadership behaviors. Duda and Balaguer (1999) have highlighted the need to
integrate achievement goal theory in the examination of leadership in sport. Speciﬁcally,

they have examined athletes’ goal orientations, and perceptions of the motivational

72

climate in relation to their perceptions of their coach’s leadership style. They found that
the climate created by the coach was signiﬁcantly related to athletes’ preferred and
perceived facets of leadership behavior. When the players perceived that their coach
created a mastery climate, they were more likely to perceive their coach as demonstrating
positive leadership behaviors. Whereas, athletes who perceived a predominately
performance climate were less likely to view their coaches as leaders; These ﬁndings
further reinforce the relevance of the perceived climate reﬂectingthe opinions that
athletes may adopt about their coach, and, in particular, the signiﬁcance of the
motivational climate in the display of leadership.

Similar to the integration of social goals and personal achievement goals, future
research in sport psychology may beneﬁt ﬁ'om the examination of thesocial dimensions
(i.e., leadership development) of the perceived motivational climate: The inﬂuence of the
motivational climate as established by the coach and the subsequenteffect of the climate
on the development of leadership in athletes has yet to be examined:

Conclusion
Current research on leadership in sport has failed to test a conceptual model of athlete
leadership that underscores its deﬁnition and purpose in a particulae sport: Based on the
current summary of hypothesized determinants and regulatory mechanisms of leadership,
this dissertation research developed and tested a social cognitive conceptual model of

leadership in athletes.

73

Chapter 3

METHOD

Participants

Data were collected on a total of 528 male and female intercollegiate rowers, ages
18-37 years (M= 19.75, SD= 1.52) and representing 18 collegiate rowing programs from
the Western, Midwestern, and Eastern regions of the United States. Initial screening of
the dichotomous variable, gender, indicated an uneven split with a resultant 20% male (n
= 108) and 80% female (n = 420) athletes in the current sample. This uneven split in the
gender distribution has the potential to produce outliers, deﬂate correlations with other
variables, and result in differential inﬂuence that may be biased (Tabachnik & Fidell,
1996). Therefore, all male rowers in addition to nine female coxswains of male teams
were removed from subsequent analysis and examined separately. This resulted in a
sample of 411 female rowers. Initial data screening based on a priori exclusion criteria
was conducted to detect athletes with signiﬁcant missing data and revealed that 38
athletes (9%) were missing portions of the questionnaire packet. Further screening
revealed that 18 of these athletes were missing one or more pages of the questionnaire
(e. g., skipped back of page, missed Time 2 data collection) and thus, were removed from
the study (n = 393). Two athletes who were injured and did not have the necessary
information to complete the packet were also removed (n = 391). Twenty-three
participants had missing data speciﬁc to the scoring and ranking of boat members.
Further exploration of this missing data revealed a distinct pattern with four athletes

failing to complete the leader ratings or rankings for both performance leadership and

74

motivational leadership. Two athletes chose not to complete the ratings or rankings on
boat members for only motivational leadership, ﬁve athletes provided a score for each
athlete but I'BfUSCd to rank their teammates on the leadership dimensions, and nine
athletes refused to distinguish members of the boat and provided tied-rankings. Finally,
three athletes chose to rank only the top ﬁve members in the boat. Of the 23 athletes,
seven athletes provided satisfactory information (e. g, ranked half of the boat) to replace
the missing data. Speciﬁcally, any athlete who received a score but did not receive a rank
was given a tie-rank with the remaining members of the boat (e. g, after top ﬁve athletes
were ranked, everyone else in the boat received a rank of 6). This resulted in a sample of
375, however, eight spare athletes who did not have any missing data, but who did not
receive peer scores from members in their respective boats, had to be removed.

After screening the data relative to the exclusion criteria, 367 female athletes
remained in the sample. A total of 17 schools and 54 boats were represented in this
sample, the majority of these boats were the coxed eight 8+ (n = 47) and the remainder of
the boats represented the coxed four 4+ (n = 7). The majority of the boats included in the
sample contained at least 50% representation of the boat (e. g, 5 members out of nine),
however 5 boats contained less than 50% representation. Therefore, a second
exclusionary criteria was that boat members of these underrepresented boats had to
demonstrate an acceptable level of consensus (e. g, r..,g greater than .50) on the group
level variables collective efficacy, mastery and performance motivational climate. This
group of underrepresented boats demonstrated consensus on collective efficacy (.51 to
.93, M = .76), performance climate (.86 to .99, M = .96), and mastery climate (.88 to .99,

M = .95), and, thus, were retained in the current sample.

75

Measures

Demographic. All participants completed a demographic background
questionnaire (see Appendix A). This questionnaire packet included a series of items that
assessed athletes’ years of experience in rowing (e. g., “How long have you rowed in a
competitive program?”), best 2K erg score, experience with injury during the season
(e.g., “Have you experienced an illness or injury within the last 6 months?”), experience
with boat selection methods (e. g., “Have you switched seat positions since the beginning
of the competitive season?”), and experience as leaders (i.e., “Have you ever held a
leadership role of any kind?”).

Gender role perspective. The degree of masculinity and femininity, which was
labeled as agentic and communal in the current study, was assessed using the Bem Sex-
Role Inventory (R-BSRI; Bern, 19; Covey & Feltz, 1991). This is a 30-item scale that
asks participants to indicate on a 5-point Likert-type scale how well each attribute
describes them (see Appendix B). Ten items assessed the degree of perceived agentic
characteristics (e. g., “independent”), 10 items assessed communal attributes (e. g.,
“sympathetic”), and 10 items served as ﬁller items (e. g., “moody”). Both agentic and
communal subscales demonstrated acceptable levels of internal consistency with
Cronbach alphas of .83 and .92 respectively.

Goal orientations in sport. A modiﬁed version of the Task and Ego Orientation in
Sport Questionnaire (M-TEOSQ; Duda & Nicholls, 1992) was used to assess individual
differences in goal orientation within the sport context (see Appendix C). Certain items
within the task and ego dimensions were modiﬁed to reﬂect the nature of the sport of

rowing. For example, the notion of learning the skill or a new skill is a rather restricted

76

notion at the collegiate level. Therefore, “I learn a new skill and it makes me want to
practice more” was changed to “I make improvements and it makes me want to practice
more.” Also, “I learn a new skill by trying hard” was changed to “I make improvements
by trying hard.” An additional item, “I work together with my teammates,” was also
included in the task dimension. For ego orientation, the concept of outperforming others
is also rather restricted given the nature of the sport. For example, “I’m the only one who
can do the play or skill” does not apply; everyone on the team knows how to perform the
skill of rowing. Therefore, this item was changed to “I outperform everyone else on my
team.” A second item, “I score the most points/goals/hits, etc,” was changed to an
equivalent concept in the sport of rowing- “I outperform others on the erg.” A leadership
goal orientation subscale taken from the Social Goal Questionnaire (Jarvien & Nicholls,
1996) was added to assess the athlete’s achievement goal related to the role of the leader
(e.g., “I organize what my teammates do”). Similar to Urdan and Maehr’s (1995)
conception of social goals, we have deﬁned social goals as representing the perceived
social purposes of trying to achieve in sport. Therefore, athletes were asked to respond to
the stem “I feel most successful in rowing when...” and respond to each item on a 5-point
Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). The ﬁnal M-TEOSQ
was a 20-item measure with eight items for the task dimension (or = .80) seven items for
ego (or = .81), and ﬁve items for leadership orientation (CL = .78).

Perceived motivational climate. The motivational climate of each team was
measured using a modiﬁed version of the Perceived Motivational Climate in Sport
Questionnaire-2 (M-PMCSQ-Z). The original 33-item inventory developed by Newton

and Duda (1999) that assesses the extent to which athletes perceive the situationally

77

emphasi
items ( sr
employ:
items re
and sew
rowers '
the 5110:
Coach t
Likert s
agrees t
reliabil

reach a

in Mag
task se
the red
ICSpon
ranging
items

CrCrib;
4 “Can

\k

C9Tltri

emphasized goals on their current team as mastery or performance was reduced to 12
items (see Appendix D). This reduced version was based on previous research that
employed the PMCSQ-2 in a study of rowers (Magyar, Feltz, & Simpson, 2002). Five
items represented mastery climate (i.e., “each rower contributes in some important way”),
and seven items assessed perceptions of performance climate (i.e., “coach favors some
rowers more than others”). Five additional items regarding the athletes’ perceptions of
the situational goals that emphasized leadership opportunities were constructed (i.e., “
Coach encourages us to become leaders”). Athletes were asked to respond on a 5-point
Likert scale and indicate the degree to which he/she strongly disagrees (1) to strongly
agrees (5) with each item. Both performance and mastery subscales demonstrated
reliability with Cronbach alphas of .70 and .78, however, the leadership subscale failed to
reach an acceptable level of reliability with an alpha of .46.

Task self-eﬁicacy. A reduced version of the task self-efﬁcacy measure employed
in Magyar et al., (2002) was used to measure task self-efficacy (see Appendix E). The
task self-efficacy construct comprised seven items that were conceptually consistent with
the reduced version of the crew collective efﬁcacy measure. Athletes were asked to
respond to the item “How conﬁdent are you that you can. . . .” on an 11-point Likert scale
ranging from 0 (not at all conﬁdent) to 10 (completely conﬁdent) for each of the seven
items. The task self-efficacy measure demonstrated acceptable level of reliability with a
Cronbach alpha of .92. However, examination of inter-item correlation revealed that Item
4 “can make a signiﬁcant contribution to your boat winning a race” and Item 7

“contribute to your boat’s ability to row it’s fastest time” were correlated at .86, and thus,

78

Item 4 was removed. The ﬁnal six-item subscale achieved acceptable levels of internally
consistency with a coefficient alpha of .90.

Leadership efﬁcacy. A leadership measure was developed based on the
recommendations from Bandura (1997) and his guide to constructing efficacy measures,
in addition to input from rowing experts consisting of two former national team coaches
and two former national team members. The original measure was piloted on 36
intercollegiate rowers and consisted of 22 items assessing both general leadership ability
and task-speciﬁc leadership ability (e.g., getting teammates to rig the boat). Athletes were
asked to respond to the stem “How conﬁdent are you that you can...” on an ll-point
Likert scale ranging from 0 (not at all conﬁdence) to 10 (completely conﬁdent). The
leadership efficacy scale was found to be internally consistent with a Cronbach alpha of
.97. In order to develop a more parsimonious measure, corrected item total correlations
were examined to determine which items maximized internal consistency. This resulted
in the removal of seven items. Initial testing of the 15-item leadership efficacy measure
(see Appendix F) that was used in the current investigation was found to be internally
consistent with a Cronbach alpha of .91. However, Item 1 “Organize boat members to
successfully prepare boat for travel (e. g., place boats of trailer)” and Item 2 “Organize
boat members to complete boat preparation (i.e., clean and rig the boat)” demonstrated a
correlation of .80. In order to control for multicollnearity, item one was removed. Also,
Item 15 “Encourage teammates to ‘win at all cost,’ (e. g., jump the start or cheat)”
demonstrated a low item mean of 1.19 and was removed from the ﬁnal subscale. This
resulted in a 13-item subscale that was used in the current investigation with a Cronbach

alpha of .92.

79

Leadership skills. The Athlete Leadership Skills in Sport Questionnaire (A-LSSQ)
measure was developed for the purpose of this study. This measure assessed the
perceived importance of skills used by athlete leaders and was constructed in line with
Bandura’s (1997) social cognitive theory and Eagly’s (1987) social role theory. Thus,
items represented the cognitive, behavioral and affective leadership skills within the
agentic and communal dimensions of social role theory.

The A—LSSQ was a 52-item questionnaire, which was designed to reﬂect eight
distinct subscales. Speciﬁcally, two subscales reﬂected social role theory: namely agentic
(12 items) and communal ( 12 items). Four subscales reﬂected the theoretical tenets of
social cognitive theory: cognitive (ﬁve items), metacognitive (four items) behavioral
(four items), and affective (ﬁve items). Two additional subscales were created to reﬂect
the positive and negative dimensions of leadership, speciﬁcally, ethical-respectful
approach (ﬁve items), and negative tactics approach (ﬁve items). This measure was
piloted on 317 undergraduate students enrolled in either activity or lecture classes in the
Department of Kinesiology (M age = 20.11 years, SD = 2.04). The sample included 136
female and 174 male students (seven failed to identify their gender) with athletic
experience in a variety of sports (M years = 12.44, SD = 4.03). Participants were asked to
think about the importance of leadership and identify the skills that were important for
athlete leaders to use and respond to the stem “It is important for an athlete leader to. .
on a 5-point Likert scale (not at all important = 1 to very important = 5).

To determine the factor structure of the A-LSSQ, an exploratory principal
component factor analysis with varimax rotation was conducted. This analysis revealed

13 factors with eigenvalues greater than 1.0. However, only six factors were interpretable

80

and reliable and were kept in the ﬁnal solution. Items with .40 or higher were retained on
any given factor. Using this criterion, 19 items did not load on the appropriate factors and
were removed from the ﬁnal solution. In order to generate a more parsimonious measure,
corrected item-total correlations were examined to determine which of the remaining
items maximized internal consistency. The ﬁnal measure represented a 26—item version of
the A-LSSQ (see Appendix G) with the following four subscales: Performance-Execution
(ten items), Motivational-Interpersonal (seven items), Respect-Communication (ﬁve
items), and Negative Tactics (four items). All subscales except the negative reached
acceptable levels of reliability (or = .72, .73, .73, and .60 respectively).

Self and Peer Performance Leader Eﬂectiveness Score and Rank. Following the
procedures outlined by Biernat et al., 1998, participants were asked to rate and rank each
member of their boat including themselves with regard to their overall effectiveness as
“performance” leaders (see Appendix H). Performance leadership was operationalized as
someone who is considered to be the “go to” person, is competent, masterﬁrl, assertive,
conﬁdent, and may lead a boat toward a successful performance outcome. Successful
performance outcome was operationlized as an improved race time or winning a race.
Participants ﬁrst listed the ﬁrst name and/or initials of each boat member according to
seat (e.g., seat 1 through seat 9). After each member was listed in order, the participant
rated each member in the boat on performance leadership using a scale of 1 (not at all
effective) to 5 (entirely effective).

After rating each member, participants were then asked to rank-order each boat
member on performance leadership giving a rank of 1 to their top choice for the

performance leader, 2 for their second choice, and continued to rank each boat member

81

until all nine members were ranked. Participants were instructed to include themselves in
the ranking, and distinguish each member with a different ranking (i.e., two athletes with
the same performance leadership score had to be distinguished with a different ranking).
Any discrepancy in the list of names was corrected using a master list provided by the
coach, and/or consensus among the boat members. Scores from this measure generated
four of the 12 leadership dependent variables tested in the conceptual model: self-score
performance leader effectiveness, self rank performance leader effectiveness, peer score
performance leader effectiveness, and peer rank performance leader effectiveness. Peer
scores and ranks were calculated for performance and motivational leader effectiveness

for each of the 367 athletes using the following equations:

 

 

Peer Score = (seat score sum) — (self-score) (1)
n-l

Peer Rank = (seat rank sum) - (self-rank) (2)
n-l

By removing self-score from the total score assigned to the participant’s seat, the peer
scores controlled for potential self-bias (Biernat et al, 1998; Zacarro et al., 1991). The
control for self-bias was also conducted for peer-rank by removing the self-rank from the
total rank assigned to the participant’s seat number.

Self and Peer Motivational Leader Effectiveness Score and Rank. Participants
were asked to rate and rank each member of their boat including themselves with regard
to their overall effectiveness as a “motivational” leader (see Appendix 1). Participants
were instructed to think of the person who has the capability to motivate the crew, and

were informed that this type of leadership is not limited to performance and may include

82

interactions both in and out of the boat. Motivational leadership was operationalized as
someone who encourages teammates to “stay tough” and “work through the pain” (i.e.,
on the erg or during a race), resolves conﬂict between members of the boat, acts
unselﬁshly, shows concern for others, or helps teammates calm their nerves before testing
and competitions. Similar to performance leadership, participants listed each boat
member according to seat and rated everyone in the boat on motivational leadership using
a scale of 1 (not at all effective) to 5 (entirely effective). After rating each member,
participants were asked to rank-order each boat member on motivational leadership
giving a rank of 1 to their top choice for the motivational leader, 2 for their second
choice, and continued to rank each boat member until all nine members were ranked.
Responses to this measure yielded four of the 12 leadership dependent variables in the
study: self-score motivational leader effectiveness, self-rank motivational leader
effectiveness, peer-score motivational leader effectiveness, and peer-rank motivational
leader effectiveness. Peer score and peer rank for each participant was calculated using
equations one and two.

Evaluation of # 1 Performance Leader. After ranking individual boat members,
athletes were asked to identify the person ranked #1 as a performance leader and respond
to four questions regarding the leader’s frequency and consistency of leadership
behaviors (see Appendix J). Responses were given on a 5-point Likert scale ranging from
1 (not at all) to 5 (all of the time). This measure was constructed to assess the followers’
perceptions in the consistency and effectiveness of performance leadership (e.g., “How

often does this individual contribute to successﬁil performances?”). An acceptable level

83

of internal consistency was observed (or = .75). Scale scores from this measure
represented one of the 12 leadership dependent variables in the current investigation.

Evaluation of # 1 Motivational Leader. Participants were asked to complete a
similar four-item measure (see Appendix K) for the person ranked #1 motivational leader
(e. g., “How often does this individual motivate you to do your best?”). The Motivational
Leader Evaluation Subscale exhibited lower internal consistency with an alpha of .67.
Scale scores from this measure produced one of the 12 leadership dependent variables in
the study.

Self-evaluation of Performance and Motivational Leader Ability. This measure
provided an additional dependent variable that assessed self-perceptions of leadership
relative to the evaluation of others. Participants also completed a seven-item measure (see
Appendix L) assessing the frequency of their own leadership behaviors (e. g., “ How often
do you contribute to successful performances?”).This measure was found to be internally
consistent with a Cronbach alpha of .75, and the total scale score represented one of the
12 leadership dependent variables.

Collective Eﬂicacy. A reduced version of the crew collective efficacy measure
(Magyar et al., 2002) was used to help keep the time demands on the athletes to a
minimum. This scale assessed athletes’ beliefs about their boat’s ability to perform
rowing skills successﬁrlly (see Appendix M). Athletes were asked to think about their
conﬁdence in their boat’s ability to row successfully and responded to the stem “How
conﬁdent are you that your boat can...” for each of the seven items. Items were anchored
on an 11-point Likert scale ranging from 0 (“not at all conﬁdent”) to 10 (“completely

conﬁdent”). All items were generated speciﬁc to the sport of crew (i.e., “successfully

84

execute the race plan”). The measure of crew collective efﬁcacy reached an acceptable
level of internal consistency with a Cronbach alpha of .94.

Coach ’s packet. Coaches were also requested to complete a survey that provided
demographic information, boat selection method, and selection method of athlete leaders
(Please see Appendix N). Coaches were also asked to rank two boats, which also served
as the master roster of athletes.

Procedures

Following approval from the institutional review board, head coaches of men’s
and women’s collegiate rowing programs scheduled to compete at a major regatta in the
Western region of the United States were contacted via e-mail to explain the purpose of
the research and inquire if they would be interested in letting their athletes participate.
Coaches who demonstrated an initial interest in the project were then contacted by phone
to discuss the project protocol in more detail. Speciﬁcally, coaches were informed it
would take the athletes approximately 30 min to complete the questionnaire and that the
questionnaires should be administered by a team manager or assistant coach just prior to
the regatta (e. g., during travel or team meeting). As an incentive to participate, coaches
were also informed that upon receipt of the completed athlete and coach questionnaires,
they would receive group-based feedback on the leadership rankings in the boat
compared with their rankings. Also, if the coach contributed two complete boats with no
missing data and the coach questionnaire, then his/her name would be entered into a
lottery in which the winner would receive an award of $100. Two separate lottery
drawings were offered for the women’s teams and the men’s teams. Coaches who agreed

to have their athletes participate were sent an electronic copy of the questionnaire to

85

guarantee that they would have the time to participate in the project. Once ﬁnal approval
was granted, questionnaire packets were mailed to head coaches, who then distributed
them to assistant coaches and/or research assistants to administer the questionnaires to the
athletes. Participants were administered the questionnaire packet in one of two ways.
Method 1 was a one-time assessment with the administration of the entire packet of
questionnaires, and Method 2 was a two-time assessment with the ﬁrst half of the
questionnaires administered at the beginning of the week (Time 1) and the second half of
the questionnaires administered later in the week (Time 2). The athletes were provided
with an envelope and were asked to place the completed questionnaire(s) in the envelope
and seal it closed before returning it to the research assistant/coach. Participants in the
Method 1 group took approximately 30 min. to complete the questionnaire, while
participants in the Method 2 group took approximately 15 min. for each session. In order
to examine for potential method and order effects, a series of one-way ANOVA’s were
conducted to determine if there were any differences in athletes’ scores who were
surveyed at one time versus athletes who were surveyed at two different times.
Treatment of the Data

Preceding statistical analysis, the raw data were inspected to ensure reliable data
entry and testing of assumptions related to normality, linearity, and homogeneity.
Assumptions regarding univariate and multivariate linearity, normality, and homogeneity
were tested using SPSS Explore and PRELIS 2.51 (Scientiﬁc Software International Inc,
Chicago IL). Following the recommendations of Tabachnik and Fiddell (1996) univariate
distributions were examined using histograms, skewness and kurtosis statistics, whereas

outliers were detected using boxplots and standardized scores (e. g., scores exceeding plus

86

or minus 3 standard deviations from the mean were considered outliers). Multivariate
distribution was examined using Mardia’s coefficient, while multivariate outliers were
detected using Mahalanobis distance. Levene’s test of homogeneity of variance was used
to test the assumption that each group (i.e., boat class) had the same variance across
levels of the dependent variable(s).

Following data screening, maximum likelihood estimation with LISREL 8.52
(Scientiﬁc Software International, Inc. Chicago, IL) was used in the conﬁrmatory factor
analyses of all measures that were constructed or modiﬁed for the purpose of this
research. Following the recommendations of Anderson and Gerbing (1988), a two-step
approach was used in which an acceptable CFA model were established followed by the
testing of the structural model. If these two steps were successfully completed, then the
difference between the conﬁrmatory and structural model were examined. The following
steps were taken to examine the ﬁt of the hypothesized model and, if necessary, to revise
the hypothesized model to improve overall ﬁt.

Scores for each participant were calculated as the mean score for the number of
items within each subscale dimension (i.e., mean of the eight items for task orientation).
Descriptive statistics such as the mean, standard deviation, Pearson Product Moment, and
bivariate correlations were calculated for independent and dependent variables.
Hypotheses 1 through 6, which were speciﬁc to the testing of the conceptual model were
tested with path analysis using maximum likelihood estimation method. Hypotheses 7
and 8, which involved the examination of group differences, were tested with between-

subjects Multiple Analysis of Variance (MANOVA).

87

Chapter 4

RESULTS
Demographic Information

The ﬁnal sample of female rowers (N = 367) had an average of 2.3 years’

experience rowing in college (SD = 1.8) and viewed the sport of rowing to be very
important (M = 8.21, SD = 1.36). The majority of this sample perceived themselves to be
somewhat good at rowing (M = 6.78, SD = 1.31), and had an average 2K erg score of
7:65 min (Min = 6:57, Max = 9:03). Thirty-six athletes (9.8%) received ﬁill-ride
scholarships to row in college, 100 athletes (27.2%) received partial athletic funding, and
237 (62.9%) did not receive an athletic scholarships. The majority of the sample (n =
211) had some type of injury or illness during the season, and of this group, 93 athletes
were removed from practice or competition for an average of 25 days (SD = 31).

A total of 199 athletes (54.2%) rowed varsity, 43 (1 1.7%) rowed junior varsity,
and 125 (34.1%) rowed novice. The majority rowed in the coxed eight 8+ (n = 342) while
ﬁve boats in the sample represented the coxed four 4+ (n = 25). Seat representation was
evenly distributed among this sample of athletes: 45 (12.3%) rowed bow seat, 46 (12.5%)
rowed 2 seat, 36 (9.8%) rowed 3 seat, 40 (10.9%) rowed 4 seat, 40 (10.9%) rowed 5 seat,
42 (11.4%) rowed 6 seat, 33 (8.9%) rowed 7 seat, 41 (11.2%) rowed 8 seat, and 44 (12%)
were coxswains.

The majority of athletes (n = 233, 63.5 %) reported that they had switched seats
and rowed in a different seat for more than three consecutive days during fall
training/racing. Among the athletes who switched seats, 83 athletes (22.6%) reported

switching only one seat, 81 athletes (22.1 %) reported switching among two additional

88

seats, 40 athletes (10.9%) reported switching among three seats, 17 athletes (4.6%)
reported switching among four seats, and 14 athletes (3.8%) reported switching among
multiple seats.

Two hundred eighty-six athletes (77.9%) reported having previous leadership
experiences. Among this group who occupied leadership roles, 128 (34.9%) reported
having leadership experience in sport (e. g., team captain in high school), 38 (10.4%)
reported occupying a leadership role in school (e. g., member of student council), 91
(24.8%) reported having leadership experiences in both sport and school, and 31 (8.4%)
reported leadership experiences in other areas (e. g., youth group leader). The majority of
the entire sample considered themselves to be leaders in crew (n = 241, 65.7%) while
only 26 athletes (7.1%) were current team captains. On average, this sample of athletes
considered leadership to be fairly important (M = 7.45, SD = 2.0).

Descriptive Statistics

The mean scores for all personal and situational determinants and self-regulatory
mechanisms of leadership that were assessed in the current investigation demonstrate that
this sample of athletes reported similar mean scores for agentic and communal attributes
(see Table O-l). For personal and situational achievement goals, athletes reported higher
scores for task orientation, followed by leader orientation, then ego orientation, and also
reported higher scores for mastery climate than performance climate.

Athletes’ perceptions associated with the regulatory mechanisms of leadership
reﬂected the perceived importance of performance-execution skills, motivational-
interpersonal skills, and respect-communication skills in the context of rowing. Overall,

three efficacy beliefs were assessed in this study; self-efficacy was rated the highest,

89

followed by collective efficacy, then ﬁnally leadership efficacy. Therefore, this sample of
athletes reported stronger conﬁdence in their individual abilities to perform successfully
than in their conﬁdence to lead fellow teammates and their conﬁdence that the crew
would perform successfully.

Three different perspectives of leader effects were measured in this investigation.
The ﬁrst set of leader effect scores reﬂected self-perceptions of leadership effectiveness
and ability (e. g., “How often are you a performance leader ?”). The second set of leader
effect scores reﬂected peer scores, or how this group of participants was scored and
ranked by their peers in their respective boats. Finally, the third set of leader effect scores
reﬂected how leaders were evaluated by this sample of athletes (e. g., “How often is this
person (#1 ranked athlete) a performance leader?”).

Speciﬁc to team and leader effect scores, the self scores for performance-leader
and motivational-leader effectiveness were slightly higher than the average of peer
scores, whereas average peer rankings of leader effectiveness were also slightly higher
than self rankings (see Table O-2). Related to evaluation scores, athletes rated their #1
designated leaders as more consistent in leadership ability than themselves.

Leader identiﬁcation and evaluation patterns. To determine the pattern of leader
scores and rankings, Pearson Product Moment correlations between average peer score,
average peer rank, self score and self rank were conducted for both performance and
motivational leadership dimensions (see Table O-3). The intercorrelation between peer-
score_performance-leader effectiveness and peer-score motivational-leader effectiveness
was .84, which suggests that participants reported similar scores for their boatmates’

performance-leadership and motivational-leadership effectiveness. Average peer rank for

90

per
for
sec
roll
bet
we

the

sel
per
mc
scc
(r

rnc
the

C0

91

am

the

performance-leadership effectiveness was also highly correlated with average peer rank
for motivational-leadership effectiveness. Therefore, the intercorrelations between peer
scores and rank suggest that performance-leadership roles and motivational-leadership
roles were highly integrated. With leadership role differentiation the intercorrelation
between the two leader scores should be smaller in magnitude, suggesting that people
were scored differently for performance leadership versus motivational leadership, but
the athletes in the current sample were not scored differently with such a high correlation.

The relationship between self-score performance-leadership effectiveness and
self-score motivational-leadership effectiveness was .53, indicating that athletes
perceived their performance-leadership effectiveness as somewhat comparable to their
motivational-leadership effectiveness. Regarding leadership evaluation scores, athletes
scored their #1 performance leaders slightly differently than their #1 motivational leaders
(r = .48). Self-performance leader evaluation was also moderately correlated with self-
motivational leader evaluation scores. However, because certain athletes ranked
themselves as the #1 leader in the boat (n = 173), there may be some dependency when
correlating self-leader evaluation scores with peer-leader evaluation scores.

Related to performance-leader ranking, 231 athletes did not receive a #1 ranking,
91 athletes were ranked #1 by their peers, and 45 athletes were self-ranked as the #1
leader. Of the 45 athletes who were self-ranked, eight athletes ranked themselves as the
#1 leader, but did not receive a peer rank of #1, and thus, were moved to the non-ranked
group. The distributions of performance-leader rankings for each seat by boat (8 boats
and 4 boats) are presented in Table O-4. For the total sample, 106 (28%) athletes ranked

the stroke seat as the #1 leader (101 from the 8-boat, ﬁve from the 4-boat). The majority

91

of the athletes (n = 140, 37%) voted the coxswain as the #1 leader for performance-leader
effectiveness (126 from the 8-boat, 14 from the 4-boat).

Regarding motivational leadership, a total of 227 athletes were not given a #1
ranking for motivational leadership, while 104 athletes were ranked #1 by their peers, and
36 athletes ranked themselves as #1. Of the 36 self-ranked athletes, 28 were also given a
peer rank of #1. Therefore, the eight athletes that ranked themselves as #1 did not receive
a peer rank of #1 and were included in the non-leader group. This resulted in 132 athletes
identiﬁed as motivational leaders and 235 athletes not identiﬁed as motivational leaders
in the boats. The rankings by boat for motivational leadership are presented in Table O-5.
Relative to seat position, the coxswain was voted as the #1 leader by the majority of the
sample (n = 163, 43%). A total of 55 athletes (52 from the 8-boat, three ﬁom the 4-boat)
voted the stroke seat as the #1 motivational leader.

Examination of the distributions of the stroke and coxswain leader rankings by
roWing status (varsity versus non-varsity and junior varsity/novice rowers) demonstrated
that varsity rowers (n = 202) ranked the stroke and coxswain fairly evenly for
performance leadership (70 and 68 votes respectively). The junior varsity/novice rowers
(n = 173), on the other hand, ranked the coxswain higher on performance leadership (n =
72, 41.6%) compared to the stroke seat (n = 36, 20.8%). For motivational leadership,
however, both varsity (n = 88, 43.6%) and non-varsity rowers (n = 75, 43%) identiﬁed
with the coxswain as the leader compared to the stroke seat (varsity n = 36, 17.85%;
non-varsity n = 19, 11%). Overall, a total of 87 athletes were voted as the #1

performance leader and #1 motivational leader in the boat.

92

Demographic/Background Inﬂuences on Perceptions of Leadership. The
intercorrelations between personal demographic information and leadership effectiveness
and evaluation are presented in Table O-6. Years of experience demonstrated a
signiﬁcant positive relationship with peer scores of performance-leader effectiveness, self
scores for both performance-and motivational-leader effectiveness, and self evaluations
of motivational leadership. Athletes who reported having less experience received a
lower peer-evaluation score on motivational leadership. An inverse relationship emerged
between season best 2K time and peer-and self-leadership scores, suggesting that athletes
with faster erg times were more likely to be scored higher by their peers on both
performance and motivational leadership. Previous experience with leadership also
emerged as a signiﬁcant correlate of leader scores. Speciﬁcally, athletes who reported
having participated in a leadership role received higher peer scores and self-scores on
leader effectiveness and self-evaluation of leader ability.

C onﬁrmatory Factor Analyses (CFA)

The purpose of this investigation was to develop and test a conceptual model
hypothesized to explain determinants and self-regulatory mechanisms of leadership in
athletes. As described in Chapter 3, a two-step modeling approach was explored based on
the recommendations of Anderson and Gerbing (1988), in which an acceptable CFA
model was established, followed by the testing of the structural model. However, due to
the excessive number of indicators that were multidimensional and displayed large
patterns of standardized residuals, the series of attempts at conﬁrmation of the entire
measurement model failed, and thus, separate CFA’s were performed on each measure

that was constructed or modiﬁed for the purpose of this study (see Figure 2).

93

iterr
Scrt
of r
amc
lﬂIF
pro
vs‘hi
Res
bee
iter
sen
resr
maj
dex-
fac

que

Spe
fac
Kli
We]

nut

Step I Data Screening: Preliminary inspection of the data was conducted at the
item level for each measure that was constructed or modiﬁed for this investigation.
Screening was conducted with regard to univariate and multivariate normality, restriction
of range in participant responses, and potential for multicollinearity and dependency
among items (Kline, 1998). Univariate normality was inspected using skewness and
kurtosis estimates, with observed values in excess of i 2.00 identiﬁed as potentially
problematic. Inspection of multivariate normality was conducted using PRELIS 2.52,
which provides Mardia’s coefficient values for multivariate skewness and kurtosis.
Researchers recommend reporting Mardia’s coefficient values for descriptive purposes
because large sample sizes are known to inﬂate these values (Bollen, 1989). Related to
item distributions, large values of skewness and kurtosis may denote that the item did not
serve as a reliable discriminator among the participants (e.g., the majority of the sample
responded similarly). Likewise, items with small standard deviations (e. g., less than .80)
may suggest problems with restriction of range, while unexpectedly large standard
deviations may indicate the presence of an outlier. While most consider confirmatory
factor analysis robust to violations of normal distributions, identifying potential
questionable items may assist with respeciﬁcation.

Step 2 Identiﬁcation: Items were analyzed using the covariance matrix and were
speciﬁed to load on only one latent factor. The loading of the ﬁrst item on each latent
factor was ﬁxed to 1.0 to establish a metric for the factor (Joreskog & Sorbom, 1996;
Kline, 1998; Maruyama, 1998). The number of ﬁ'ee parameters speciﬁed in the models
were less than or equal to the number of observations (i.e., v(v+1)/2, where v is the

number of observed variables).

94

 

 

 

 

 

 

    
 
    
 

Latent Factor 1

 

 

 

@ ® @
r

 

 

_~.

 

m
.a.
I
5
3
.rs

 

<

 

 

m
01
5
3

or

 

 

3 ®
t
a
can

 

 

 

Latent Factor 2

 

 

[TI
\1
5"
3

\1

 

 

@ C

 

 

 

Figure 2. An example 2-factor conﬁrmatory factor analysis measurement model.

95

Step 3 Estimation: Item uniqueness and disturbances were estimated using
standardized residuals, t-value parameter estimates, and squared multiple correlations.
The pattern of correlated residuals provide information regarding the ﬁt of the item,
Individual items with large standardized residuals (greater than 2.59) and modiﬁcation
indices above 5.00 were examined for correlated uniqueness, and any item considered
problematic (i.e., excessive cross-loading) was removed. In an effort to maintain
unidimensional measurement, correlated disturbances were not permitted (Anderson &
Gerbing, 1988; Kline, 1998; Stout, 1987).

Multiple ﬁt indices were considered to establish the criterion of acceptable ﬁt.
The chi-square statistic compares the absolute ﬁt of the hypothesized model with the
population (model-implied) matrix, however, because this statistic is sensitive to sample
size, the difference in X2 between the hypothesized and respeciﬁed models was used to
assess improvement in model ﬁt. Also, to reduce the sensitivity of the X2 test to sample
size, the X2/df was calculated with a ratio of less than 3 considered to be acceptable
(Kline, 1998). Absolute ﬁt was assessed using the goodness of ﬁt index (GFI), which is
based on the proportion of observed covariance matrix explained by the model-implied
covariance matrix. The adjusted goodness of ﬁt index (AGFI) adjusts the GFI downward
for model complexity and takes into account that models with more estimated parameters
tend to ﬁt the data better than simpler models. A discrepancy between these two indices
suggests that trivial or non-signiﬁcant parameters were included in the model. Indices
such as the normed ﬁt index (NF 1), which indicates the proportion in the improvement of
the overall ﬁt over a baseline independent model, and the non-normed ﬁt index (NNFI),

which corrects for model complexity, were used to examine incremental ﬁt. Similar to

96

the NFI, the comparative ﬁt index (CFI) assesses the improvement in the hypothesized
model over an independent model, but is less sensitive to sample size. Researchers
recommend that values demonstrating acceptable ﬁt should be .90 or higher (Kline,
1998). The root mean square error of approximation reﬂects the amount of unﬁtted
residuals between the observed and model-generated covariance matrices. Values less
than .10 are considered adequate, while values less than .05 are considered very good.
Models that were identiﬁed as more complex than the hypothesized model underwent
respeciﬁcation that was not based entirely on statistical consideration, but also consistent
with the theoretical tenets on which the measurement model was based (i.e., self-efficacy
theory).

Evaluation of # 1 Performance and Motivational Leader. In order to evaluate the
consistency of designated leaders, athletes responded to four questions regarding the #1
ranked performance leader and three questions regarding the #1 ranked motivational
leader. Descriptive statistics, model ﬁt indices, and information regarding the parameter
estimates are presented in Table O-7. Results from the conﬁrmatory factor analysis
demonstrated acceptable ﬁt with the data (X2 (13) = 22.36 RMSEA=.06, GFI = .97, NFI
= .91, NNFI = .90, CFI = .97).

Eﬁicacy Measures. Three separate efficacy measures were constructed to measure
the athlete’s conﬁdence in her ability to row successfully, conﬁdence in her boat’s ability
to row successﬁilly, and conﬁdence in her own ability to lead her teammates. All three
measures were expected to correlate and, as a result, were simultaneously subjected to
conﬁrmatory factor analysis to determine the factor structure (i.e., 29 items). The

hypothesized model failed to ﬁt the data (X2 (3 74) = 1470.72, RMSEA = .09,GFI = .79,

97

.\"Fl =
demor
memb
equip
prepa
was (i
did n1
(“enc
and z
self-r
demr
abili
a rat
mod

GFl

iIErr
that
1mm
ilenj
t0 t1
Valu

best

NFI = .81, NNFI = .84, CF] = .85). Initial examination of the conﬁrmatory results
demonstrated an excessive standardized residual between Item 1 “Organize boat
members to successfully prepare for boat travel (e. g., place boats, oars, and other
equipment on trailer)” with Item 2 (“ Organize boat members to complete boat
preparation (i.e., clean and rig the boat)”). Inter-item correlations were examined where it
was discovered that these two items had a correlation of .79, suggesting that the athletes
did not distinguish between the two items, and thus, Item 1 was removed. Item 15

2”

(“encourage teammates to ‘win at all cost ) demonstrated an orthogonal loading of .10
and also was removed ﬁ'om the leadership efficacy measure. In addition, Item 4 from task
self-efficacy (“can make a signiﬁcant contribution to your boat winning a race”)
demonstrated an excessive standardized residual with Item 7 (“contribute to your boat’s
ability to row it’s fastest time”). Thus, athletes were not distinguishing between winning
a race and rowing the fastest time, and so Item 4 was also removed. The respeciﬁed
model demonstrated an acceptable level of ﬁt with a X2(272) = 826.22, RMSEA = .07,
GFI = .85, NFI = .87, NNFI = .90, CFI = .91 (see Table O-8).

Athlete-Leadership Skills in Sport Questionnaire. Descriptive statistics for each
item are presented in Table O-9. Inspection of skewness and kurtosis values demonstrate
that four items demonstrated skewness values just above 2.0, while eight items displayed
kurtosis values above 2.0. Inspection of standard deviations indicated that the majority of
items had values of less than .80, which demonstrated restricted response variability. Due
to the low average ratings, small standard deviations, and large skewness and kurtosis

values, Item 23 (“call someone out who is not giving 100%”), Item 13 (“be one of the

best athletes on the team”), and Item 22 (“help the team by rowing the fastest”) were

98

ide

der

are
the
.75
me
mc
an
(X

ICE

hy

hat
co.
Val
ad:
the
res
Der

Ilei

identiﬁed as problematic and removed from the conﬁrmatory analysis. The revised model
demonstrated marginal ﬁt (X2 (224) = 661.32, RMSEA = .07, GP] = .87, NFI = .73,
NNFI = .78, CFI = .81).

Modiﬁed- Task and Ego Orientation in Sport. Descriptive statistics and ﬁt indices
are presented in Table O-lO. The hypothesized 3-factor structure failed to adequately ﬁt
the data (X2 (167) = 778.87, RMSEA = .10, GFI = .83, NFI = .73, NNFI = .75, CFI =
.78). Inspection of modiﬁcation indices indicated that Items 4 (“my teammates look up to
me”) and 11 (“I lead by example”) from leader orientation demonstrated high
modiﬁcations indexes indicating that the two items could have been assigned to both task
and ego orientation. After removing these two items, the revised model was improved
(X2 (132) = 529.65, RMSEA=.10, GFI = .85, NF I = .75, NNFI = .77, CFI = .80) and
reached a marginal level of ﬁt.

Modiﬁed Perceived Motivational Climate in Sport Questionnaire. The
hypothesized model contained the mastery, performance, and leader climate scales. Initial
inspection of this model revealed that Items 5, 12, and 13 from the leader climate scale
had large modiﬁcation indices ranging from 12.05 to 76.16. After removing Item 12, “the
coach makes it clear who he/she thinks are the leaders,” which demonstrated the largest
values of 76. 1 5 and 67.75, the remaining items for the leader climate no longer
adequately speciﬁed the leader climate latent factor (all t-values less than 1.96). Due to
the instability of this factor, the leader climate was removed and the M-PMCSQ-Z was
respeciﬁed with the remaining 12 items designated to load on the mastery and
performance dimensions. Examination of the revised 2-factor structure indicated that

Item 6 (“rowers are dropped from the boat if they continually make mistakes”)

99

demonstrated a low squared multiple correlation of .03 and standardized estimate of .16,
suggesting that this item was not adequately loading on the mastery latent factor. Given
the orthogonality of this item, it was removed which adequately improved model ﬁt X2
(53) = 163.42, RMSEA = .07, GFI = .93, NF I = .87, NNFI = .88, CFI = .90 (see Table
O-l 1).
Path Analysis

Although the measurement model for the efficacy measures, Athlete Leadership
Skills Questionnaire (A-LSSQ), and Modiﬁed Task and Ego Orientation in Sport
Questionnaire (M-TEOSQ) failed to reach consistent ﬁt indices of .90, Stout (1987) has
highlighted the importance of establishing essential unidimensionality rather than relying
entirely on ﬁt indices of .90 or greater to substantiate model ﬁt. Unidimensionality refers
to the assumption that a single latent factor is sufficient to explain the common variance
among items. When the condition of unidimensionality is satisﬁed, then items that are
hypothesized to load on a latent factor are the unambiguous indicators of a single
construct (e.g., no cross-loading or correlated error variances). However, researchers are
taking notice of the difficulty in establishing strict unidimensional measurement models
with “real-world” data, and propose the notion of “essential” unidimensionality
(Embretson & Reise, 2000; Stout, 1987). Stout (1987) claims that a one-factor
measurement model will demonstrate essential unidimensionality when the average
between item residual covariances approaches zero as the length of the measure
increases. In essence, if an underlying latent dimension is strong enough, it will emerge
from the measure and should be considered above the smaller speciﬁc factors (i.e., noise

or method variance) and inﬂuences that may also emerge and adversely affect the ﬁt of

100

the model. Therefore, given Stout’s recommendation, in addition to the sufficient factor
loadings of each item, low standard errors, and observed internal consistency, the three
measures that failed to ﬁt the data based on certain ﬁt indices derived from the
covariance modeling, actually provide evidence of measuring the underlying latent
dimensions they are hypothesized to measure (e. g., 7 items hypothesized to measure
collective efficacy).

To test the ﬁt and conceptual relationships of the hypothesized model of
leadership, a recursive path analysis on the estimated covariances of directly observed
variables (e.g., subscale scores) was conducted (see Figure 3). Due to low reliability, the
negative tactics skills subscale and the evaluation of the #1 motivational leader subscale
were not included in the model. In an effort to improve internal consistency and control
for multicollinearity, self score for performance-leader effectiveness was averaged with
self score for motivational-leader effectiveness. The new self-score leader effectiveness
variable achieved an acceptable level of reliability with a coefficient alpha of .70. Peer
score performance-leader effectiveness was averaged with peer score motivational-leader
effectiveness, and the new peer-score leader effectiveness variable was internally
consistent with a coefficient alpha of .95. Due to the number of athletes who provided a
self-rank of #1 (n = 173), the self-evaluation scores were not included in the ﬁnal model
to control for potential dependency in the data.

The reduction of variables resulted in a hypothesized model with seven theoretical
causal variables: agentic attributes, communal attributes, task orientation, ego orientation,

leader orientation, mastery, and performance climate. Variables related to demographic/

lOl

Determinants Regulatory
Mechanisms
Pe rfo rm 3 n ce
8 kil ls

//

 

Motivational
Skills

 

a'r
Communal \r-u,
Attributes V" ' '

A I." ‘m
as $3,!!!” I ."I
Mimic," ll"ili’l'.\ V.. 3?; .

//‘\eil!flh‘¢.§§’%1

ll!
1

Leader *1,”
Orientation \‘t'l
I

ll

  

§
Respect-Com.
Skills

gt

\

\

\\

OrieEngétion '. HI.“ .‘ ‘ .-.r
1135‘s“
’1‘

r l\
with." \\\\\s

Leader
‘ - Evaluation

\o

\

Figure 3. Illustration of Full Hypothesized Model.

102

Eﬂects

Peer
Scores

 

Self
Scores

In!" I -
it", ,...'.,,, ,
I“? /

or if ’\\\“\‘

o.‘

/’o“‘\‘

 
   
  

{01h
skil
effi

SC 0

me
inf
rel;

ant

NC
arr.

EV;

eff
has
wh
for

Pat

Par.

background inﬂuences on leadership were not included and examined separately. The
following nine variables were tested as endogeneous variables: performance-execution
skills, motivational-interpersonal skills, respect-communication skills, leadership
efficacy, task self-efficacy, collective efficacy, evaluation of #1 performance leader, self-
score leader effectiveness, and peer-score leader effectiveness.

The direct inﬂuence of personal, situational determinants, and regulating
mechanisms on team and leader effects was tested, in addition to the potential mediating
inﬂuence of leadership skills, leadership efficacy, and task-self efficacy on the
relationship between determinants and effect variables. The correlation matrix, means,
and standard deviations for all variables included in the path model are presented in Table
1. The overall hypothesized model demonstrated an acceptable ﬁt to the data (X2 (35) =
99.96, Root Mean Square Error of Approximation = .07, Comparative Fit Index = .95,
Normed Fit Index = .94, Goodness of Fit Index = .97). Furthermore, examination of the
amount of variance explained in the endogeneous variables ranged ﬁ'om 16% in the
evaluation of the performance leader measure to 40% in the leadership efficacy measure.

Path coefﬁcients (i.e., standardized regression coefficient) demonstrate the direct
effect of an exogenous variable on an endogenous variable in the model. When the model
has two or more causal variables, the path coefficients are partial regression coefficients,
which measures the direct effect of one variable on another in the path model controlling
for other prior variables. Examination of the path t-value was conducted to test individual
path coefficients. A t-value associated with a parameter is calculated by dividing its
unstandardized estimate by its standard error with values greater than 1.96 indicating the

parameter is signiﬁcantly different than zero.

103

4008 H

>805. $88653 b83885. Si 03830.05. 3.303% 000$ 300$ $3.033

 

 

 

_ n w A u a 3 m o E Z S 3 3 G a
_. >moaao ---
u. 0033::E -.3. ..o
u. Ham.» .3... .3... ..o
a. mmo .3... wow .3: _.o
u. r8003 .3... ._ .. .3... .3... _.o
a. Z. 0:308 ..ou .3... .Nw... now .3 _.o
3. w. 0:308 .5. -.ON ..2. ._ _. .3 1.8... ..o
m. w. 90:. .nm... .3: .3... .3... .3... .3: .00 mo
0. E. mrEw .8 .AN... .3... .03 ..m... .3... .oo .00... _.o
.o. w. 20:... be .3... .3... ..o .3... .3... ..cm .3... .3... ..o
2. rm .Am... .8... .Nu... .3... .3... .5: -.8 .3... .8... .3: _.o
3. Hum—Tmm .3... .8: .8... .3... be... .3 ..og .5... .00 .N... .3... ..o
3. 0m .ow .~ _ ... .3... .8 .8 .3... L0. be .00 .5. he... .3... _.o
3. F 98—. .08 .3. .8... .0... .o. .3... ...n. .3... ..m: ..o .00 .3. .NA... _.o
3. 8.0808 .3... .3... .nw... .3... .3... .8 ..om .3... .3... .nn: .3... .3... .3... .00 mo
3. $2.808 .No... .3: .3... .o... :3... ._ _. ..oo .5 .8 .3: .3... .3... .3 b0 .3... ..o
3 Amy ~00 mum bum Nah 93 0.00 NS Poo 0.8 A00 3.3 m. z 3.8 0.3 98 who
A08 A35 6va 038 0qu A. 08 033 TB A. 3V 83v 3.3V 2 .N3 C .ud 05.3 63.3 636

 

208. >838 u >828 0:10:8mw 00335:: " 003855— 0zacc8mw Hmmw n Haw 0180830? mmo u mmo 0:008:03 00002 n
H8002 0180830? Z. 0:308 u 5383~ 0:805 v. 0:308 N 31.03588 0:308“ v. mﬁzm u wagoaBm—soamxoocmoa BE?

3. mem H 303$:mono—Laoémqmoaa 9.5% w. mem n Womcoor0083§8§0= meM rm u ramaaarmn mar—80$ How—Ammo. u Hum—a
man—mammomoﬁ 0m u 00.8038 mmmomoww b. m8: n r0008. “$0.530? mammooao M 85808 r0000. mmwoaésammw #0020008 n
$90008 8009. mmmoséaomm.

.68.. :32. $682

104

[ﬁpodu
mechan
(3)Age
eﬁicaq
(b)C0r
cmnmu
(cyAgc
leader,
(d)Th<
peﬂbrl
dhnen
anﬁbu
leader

a130 e

Speci
leade
SDECi
§gni
respI
task
Perf

genc

Hypothesis 1: Gender role perspectives will differentially inﬂuence self-regulatory
mechanisms of leadership and leadership ’s eﬂects.
(a) Agentic attributes will positively predict performance-execution skills, leadership
efficacy, and task self-efﬁcacy.
(b) Communal attributes will positively predict motivational-interpersonal skills, respect-
communication skills, leadership efﬁcacy, and task self-efﬁcacy.
(c) Agentic attributes will positively predict the evaluation of the #1 ranked performance
leader, in addition to self and peer leader effectiveness scores.
(d) The initial hypothesis regarding leader effects differentiated agentic attributes with
performance-leadership dimension and communal attributes with motivational-leadership
dimensions. However, because of the averaged leadership effect scores, communal
attributes were hypothesized to positively predict evaluation of the #1 performance
leader, and self and peer scores of leadership effectiveness. Communal attributes were
also expected to signiﬁcantly inﬂuence collective efficacy.

Three of the four paths for Hypothesis 1(a) emerged as signiﬁcant (see Table 2).

Speciﬁcally, agentic attributes inﬂuenced performance-execution skills (B = .20, p<.01),
leadership efficacy (B = .44, p<.01), and task self-efﬁcacy (B =28, p<.01). All four paths
speciﬁed to test Hypothesis 1(b) emerged as signiﬁcant with communal attributes as a
signiﬁcant and positive inﬂuence on motivational-interpersonal skills (B =.39, p<.01),
respect-communication skills (B = .17, p<.01), leadership efﬁcacy (B = .24, p <01), and
task self-efﬁcacy (B = .16, p<.01). Communal attributes also signiﬁcantly predicted
performance-execution skills (B =.11,p<.05). Thus, perceptions of attributes related to

gender differentially inﬂuenced the perceived importance of leadership skills, with

105

commu

skills. \

Table 2

Direct

Perfc
Moti
RCSp
Lead
Task
Eval
Peer
Self-
Coll-
‘p<.05,

to col]
anﬁbt
and (h
agent}

illugtr

full [T

communal attributes positively predicting the perceived importance of all leadership

skills, while agentic attributes predicted only performance-execution skills.

Table 2

Direct Eﬂects for Hypothesis 1

 

 

Agentic Communal

Attributes Attributes
Performance-Execution Skills .20** .11*
Motivational-Interpersonal Skills .03 39*"
Respect-Communication Skills .05 .17*
Leadership Eﬁ‘icacy .44*** .24M
Task Self-efﬁcacy .28** .16*
Evaluation of #1 Leader -.04 .08
Peer-Scores Leader Effectiveness .02 .03
Self-Scores Leader Effectiveness .07 .04
Collective Efﬁcacy -.06 .13 *

 

Note. Standardized path coefficients are taken from full model.
*p<.05, **p<.01, ***p<.001

Related to sub-hypotheses 1(c) and 1(d), the direct path from communal attributes
to collective efficacy (B =.14, p<.01) was signiﬁcant. Regarding the inﬂuence of gender
attributes on leader effects, none of the presumed relationships emerged as signiﬁcant,
and thus, failed to provide direct support for sub-hypotheses 1(c) and 1(d). Therefore,
agentic and communal attributes did not directly inﬂuence leadership effects. An
illustration of the standardized path coefﬁcients speciﬁc to Hypothesis 1 taken from the

full model results is provided in Figure 4.

106

2
.8 .09

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Performance
Skills
.20”
.77
.11'
.02
Motivational
.03 Skills
.39" 32
Respect-Com
l, Skills
.17‘
-.04 .
Agentic Leader
Attributes Evalratlon
.08
.02
Peer
“3 Scores
-.05
.07
Self
04 Scores
Commmal '93
Attributes
.13‘ Collective
Efﬁcacy
.44'"
.24'
Leader
.28“ ﬁlmy .so
.16‘
Task
Self-efﬁcacy

Figure 4. Illustration of Direct

 

.81 .51

Effects for Hypothesis 1.

107

.72

.68

Orie
€er

Orje

Hypothesis 2: Goal orientations will diﬂerentially inﬂuence self-regulatory mechanisms
of leadership and leadership ’s eﬂects.
(a) Task orientation was hypothesized to positively predict performance-execution skills,
motivational-interpersonal skills, respect-communication skills, leadership efficacy, and
task self-efficacy.
(b) Ego orientation was hypothesized to positively predict performance-execution skills,
leadership efﬁcacy, and task self-efficacy.
(c) Leader orientation was hypothesized to positively predict performance-execution
skills, motivational-interpersonal skills, respect-communication skills, leadership
efﬁcacy, and task self-efﬁcacy.
(d) Speciﬁc to leader and team effects, both task and leader orientation were
hypothesized to demonstrate a positively predict collective efficacy, the evaluation of the
#1 performance leader, and self-and peer-scores of leader effectiveness. Ego orientation
was hypothesized to positively predict self scores of leadership effectiveness and a
negatively predict peer scores of leader effectiveness.

Three of the ﬁve hypothesized paths emerged as signiﬁcant for task orientation
(see Table 3). Speciﬁcally, task orientation signiﬁcantly predicted the emphasis of
performance-execution skills (B = .14, p<.01), respect-communication skills(B = .19,
p<.01), and task self-efficacy (B = .10, p<.05). Therefore, with increases in task
orientation there is also a subsequent increase in perceived importance of performance-
execution skills, respect-communication skill, and stronger efficacy beliefs. Ego
orientation was hypothesized to predict three paths, but none of the direct paths emerged

as signiﬁcant. Leadership orientation on the other hand, failed to signiﬁcantly predict any

108

of the three leadership skill paths as hypothesized, but signiﬁcantly inﬂuenced leadership
efficacy (B=.30, p<.01), and task self-efﬁcacy (B=.12,p<.01). Thus, a stronger emphasis
on leader goal-orientation directly inﬂuences stronger perceptions of one’s conﬁdence to

lead others, and perform successfully.

Table 3

Direct Eﬂects for Hypothesis 2

 

 

Task Ego Leader

Orientation Orientation Orientation
Performance-Execution Skills .14* .11 .09
Motivational-Interpersonal Skills .08 .01 .09
Respect-Communication Skills .19* .04 .06
Leadership Efﬁcacy .03 -.04 .30“
Task Self-efficacy .10* .09 .12*
Evaluation of #1 Leader .17* .07 -. 13*
Peer-Scores Leader Effectiveness .01 -.16 .34"
Self-Scores Leader Effectiveness .07 -.06 .29M
Collective Efficacy . 12* .03 -. 18*

 

Note. Standardized path coefficients are taken from full model.
*p<.05, **p<.01, ***p<.001

Related to sub-hypothesis 2(d), task orientation signiﬁcantly inﬂuenced collective
efﬁcacy (B = .12, p<.01), and evaluation of the # 1 performance leader (B = .17, p<.01).
Ego orientation was inversely related to peer scores for leader effectiveness (B = -.16,
p<.01). Finally, leader orientation demonstrated a negative inﬂuence on collective
efficacy (B = -. 18, p<.01), and the evaluation of the #1 performance leader (B = -. 13,
p<.05), and positively inﬂuenced self-scores of leader effectiveness (B :29, p<.01), and
peer-scores of leader effectiveness (B =.34, p<.01). Figure 5 presents the path coefficients

taken from the full hypothesized full model speciﬁc to Hypothesis 2.

109

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.09
Perfomranoe
Skills
.14‘ _77 .02
.11 .09
'°° sells?“
.02
.01
.19'
Respect-Com
.04 Skills
.17'
4" Leader
Task 97 .
orientaion Evaluaron
..13'
.01
51.. P”
. E90. .34' . Scores
onentatron .07
--03 Self
1,. Scores
Leader
onentaron .12,
1 .03
.,1 Collective
Emma
.30‘ .03 .09 .1
.12'
Leadersz Task
Efﬁcacy Self-efﬁcacy
.60 .81

.51

Figure 5. Illustration of Direct Effects for Hypothesis 2.

110

Hypothesis 3: Perceptions of motivational climate will diﬂerentially predict self-
regulatory mechanisms of leadership and leadership ’s effects.
(a) Perceptions of mastery climate were hypothesized to positively predict performance-
execution leadership skills, motivational-interpersonal leadership skills, respect-
communication leadership skills, leadership efficacy, and task self-efﬁcacy.
(b) Perceptions of performance climate were hypothesized to positively predict
performance-execution skills and negatively predict motivational-interpersonal skills,
respect-communication skills, leadership efﬁcacy, and task self-efﬁcacy.
(c) Perceptions of mastery climate were also hypothesized to positively predict collective
efﬁcacy, the evaluation of #1 performance leaders, and self and peer scores of leadership.
Performance climate was hypothesized to positively predict self scores of leadership
effectiveness, and negatively predict collective efficacy.

Only three of the ﬁve hypothesized pathways for sub-hypothesis 3(a) emerged as

signiﬁcant with mastery climate predicting the importance of all leadership skills (see

Table 4).
Table 4

Direct Effects for Hypothesis 3

 

 

Mastery Performance

Climate Climate
Performance-Execution Skills .16"‘ .11*
Motivational-Interpersonal Skills .13* .07
Respect-Communication Skills .22* .04
Leadership Efﬁcacy .05 -.05
Task Self-efﬁcacy -.01 -.07
Evaluation of #1 Leader .09 -.08
Peer-Scores Leader Effectiveness .00 -.06
Self-Scores Leader Effectiveness -.05 -.08
Collective Efficacy .18* -.04

 

Note. Standardized path coefﬁcients are taken from full model.
*p<.05, **p<.01, "*p<.001

lll

Speciﬁcally, individual perceptions of mastery climate exhibited a signiﬁcant
inﬂuence on performance-execution skills (B = .16, p<.05), motivational-interpersonal
skills (B = .13, p<.05), respect-communication skills (B = .22, p<.01). Performance
climate was related only to performance-executions skills (B = .11, p<.01). Related to
leadership, individual perceptions of mastery climate signiﬁcantly inﬂuenced collective
efﬁcacy (B = .18, p<.01). Neither mastery climate nor performance climate demonstrated
a signiﬁcant inﬂuence on perceptions of leader effects. The path coefﬁcients for

Hypothesis 3 are presented in Figure 6. .82 m

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.02
.82
l '3‘
.04 a
' D
Mastery Leader
Clinﬂe _m Enml 72
19 ~°°
Peer
Performance ..08 300'” 0
Climde '
«06
._m Self
Scores
.68
.18"
~01 Collective ‘
05 as Eﬁﬂcy
-.07 _ _ ..1
Leadership Task
efﬁcacy Self-efﬁcacy
.00 .81

.51

Figure 6. Illustration of Direct Effects for Hypothesis 3.

112

Hypothesis 4: Leadership skills will diﬂerentially inﬂuence team (e. g., collective eﬁcacy)
and leader eﬂects.
(a) Performance-execution leadership skills will positively predict evaluation of the #1
performance leader, and self and peer scores of leader effectiveness.
(b) Motivational-interpersonal leadership skills will positively predict collective efficacy,
and self and peer scores of leader effectiveness.
(c) Respect-communication leadership skills will inﬂuence evaluation of the #1
performance leader, and self and peer scores of leader effectiveness.

The direct effects of leadership skills on leader and team effects are presented in
Table 5. One of the three paths representing Hypothesis 4(a) emerged as signiﬁcant with
performance-execution skills positively inﬂuencing the evaluation of the #1 performance
leader (B :25, p<.01).
Table 5

Direct effects for Hypothesis 4

 

Performance- Motivational- Respect-

 

Execution Interpersonal Communication
Evaluation of #1 Leader .24* .02 -. 13
Peer-Scores Leader Effectiveness -.09 -.06 .12
Self-Scores Leader Effectiveness -.O9 .05 .07
Collective Efficacy .05 -.08 .07

 

Note. Standardized path coefficients are taken from full model.
*p<.05, **p<.01, ***p<.001

Partial support was also obtained for Hypothesis 4(c) with respect-communication

skills demonstrating an inverse relation with the evaluation of the #1 performance leader
(B = -.13, p<.05), and positively inﬂuencing peer scores of leader effectiveness (B =.11,

p<.05). Thus, athletes who placed greater importance on the use of respect-

113

communication skills for leadership were more likely to scored as effective leaders by
their peers. Related to the evaluation of the #1 performance leader, performance-
execution skills positively inﬂuenced these evaluations while respect-communication
skills were inversely related to these evaluations. Thus, respect-communication skills are

seen as less important than performance-execution skills in performance leadership.

 

 

 

 

    
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.84
‘2‘.
Leader
Evaluation
.02
Perfon'nance
Skills ,.13. .72
~09
‘03 Peer
Scores
.12
'07 Motivational
Skills 09 .64
.05 Self
Scores
.07
aspect-Com.
Skills l
i 05 68
~08
~07 Collective
Efﬁcacy

 

 

Figure 7. Illustration of Direct Effects for Hypothesis 4.

114

Hypothesis 5: Eﬂicacy beliefs will positively predict team (e. g., collective eﬂicacy) and

leader eﬂects.

(a) Leadership efficacy will positively inﬂuence collective efficacy, the evaluation of the

#1 performance leader, and self- and peer- scores of leader effectiveness.

(b) Task self-efﬁcacy was hypothesized to positively predict collective efficacy,

evaluation of the #1 performance leader and self and peer scores of leader effectiveness.
Leadership efficacy predicted self scores of leader effectiveness (B :22, p<.01),

and peer scores of leader effectiveness (B =.28, p<.01). Self-efficacy beliefs predicted

self-scores of leader effectiveness (B =. 18, p<.01), and collective efficacy (B =.42,

p<.01).

Table 6

Direct effects for Hypothesis 5

 

 

Leadership Task Self-

Eﬂicacy efﬁcacy
Evaluation of #1 Leader -.07 .ll
Peer-Scores Leader Effectiveness 28* .01
Self-Scores Leader Effectiveness 22* . 18*
Collective Efficacy .10 .42*

 

Note. Standardized path coefficients are taken ﬁom full model.
*p<.05, **p<.01, ***p<.001

Thus, athlete who reported stronger leadership efficacy beliefs were more likely
to score themselves as effective leaders, and were more likely to be considered as

effective leaders by their peers as well.

115

 

Leadership
Efﬁcacy

 

 

Task
Self-efﬁcacy

 

 

 

 

 

 

 

 

(:9

D

72

 

._07 Leader
Evaluation ‘
.1 1
.28‘
Peer
Scores ‘

 

 

 

 

Self

C

 

Scores

 

 

 

Collective
Efﬁcacy ‘

 

 

Figure 8. Illustration of Direct Effects for Hypothesis 5.

116

®

0

Hypothesis 6: Personal and situational determinants will serve as distal predictors of
leadership eﬂects, while self-regulatory mechanisms will serve as more proximal
predictors of leadership effects and mediate the relationship between personal and
situational determinants with leadership ejfects.

Hypothesis 1 through 5 illustrated the hypothesized direct effects, or standardized
path coefficients that were estimated in the ﬁill hypothesized model. With path analysis,
the standardized effects may be decomposed to represent indirect and total effects.
Indirect effects measure the effect of mediating or intervening variables, while total
effects represent the sum of all direct and indirect effects. The decomposition of the
standardized effects for the ﬁill hypothesized model are presented in Table 7.

When a variable has a non-signiﬁcant direct effect, but a signiﬁcant indirect and
total effect on an endogeneous variable, than ﬁill mediation has occurred. In the current
conceptual model, there are a total of four relationships between personal determinants
and leader effects which have been ﬁilly mediated by the regulatory mechanisms of
leadership skills and efficacy beliefs. For example, agentic attributes do not signiﬁcantly
predict peer-scores for leader effectiveness. However, the indirect effect of .11 is
signifcant in addition to the total effect of . 13. Therefore, part of the inﬂuence of agentic
attributes on peer-scores for leader effectiveness is being transmitted via leadership skills
and efficacy beliefs. In order to determine where the indirect inﬂuence is derived from,
the product of each path that inﬂuences peer scores of leader effectiveness is considered.
Starting with agentic attributes on performance-execution skills (.20), multiplied by
performance-execution skills on peer scores leader effectiveness(-.09) is equal to .01,

agentic attributes to motivational-interpersonal skills (.03 ), multiplied by motivational

117

.320 q

$.32. Eaten... «Em SSE m§mna QESJSBN Si 9.252.633 Dealigxa o: 53.: EB. 53%.. Mbsmna

 

 

mas—€36: cm 90 $87998 r032 mammooam roams 06:83.6
3 banana mmooaéaomm mmonaéaomm mmmomov.

summing 539.2%

982 mmooﬁm no» .8 .3 -.oa

5983 manna .oa .2... .3... .5...

H08. manna .8 ._w.. .3... .3...
0633.52 >332an

983 mmaoﬁm .3 .8 be .3...

5989 mmooa .ON .oa .5... .om

H08— mmwoa ..o .3 . .3... .3...
Ha» Glmz‘maez

933 manna .3... .2 .3 .5...

5332 mmana .8 .ON .8 .8

4.08— mmwonm .3... .8 .2... .3...
Mme 61835.6:

989 mamas .3 -. 5... -.oo .8

5932 manna .8 -.o. .oo .8

4.08. mmoonm .3 -. 3... -.om .om
Emma? Gluiamoz

983 mmwnnm -. 5... ..ﬁ... .8... -. :w...

=5.an mmwoa .2 .3... .oo... .9?

H08. mmwoa -. S .3... mm... -. S...
3386. @3328

989 mmwoa .oo .oo -.om ..m...

5989 made? .3 .8 .8 now

.38— mmwoa .5 .ON ..8 .3...
Eugeniaanw 0:358

983 manna ._ _... -.om -.om -.os

5982 mmwoa .oo ..S ..8 - on

as». mags .. . -..... -.. . A...

118

interpersonal skills on peer scores (-.06) is equal to -.002, agentic attributes to respect-
communication skills (.05), multiplied by respect-communication on peer scores (. 12) is
equal to .006, agentic attributes to leadership efﬁcacy (.44), multiplied by leadership
efficacy on peer scores (.28) is equal to .12, and agentic attributes to task self-efﬁcacy
(.28), multiplied by self-efficacy to peer scores (.01) is equal to .003. When the combined
effects (e. g., -.02, .002, .006, .12, and .003) are summed together, the ﬁnal indirect effect
is equal to .11. The indirect effect (.11) is then added to the direct effect (.02) which
yields a total effect of .13. Therefore, agentic attributes did not emerge as a signiﬁcant
direct causal variable on peer scores of leader effectiveness, however, the effects of
agentic attributes were transmitted onto peer scores via leadership skills, leadership

efficacy, and task self-efficacy. In particular, the most inﬂuential indirect path was

 

leadership efficacy (see Figure 9). Y

  

Performance
Skills

 

---—...

 

 

Motivational
L Skills

 

 

 

/ . Respect-Com.
Skills

 

 

 

 

 

- .02 Peer
A’t‘éli'li'és ’ Scores

 

 

 

 

 

 

 

Leadership
13' Efﬁcacy

 

 

 

 

Task
Self-efﬁcacy

 

 

 

Figure 9. Direct and Indirect Pathways from Agentic Attributes to Peer
Scores of Leader Effectiveness.

119

A similar pattern can be seen for the indirect inﬂuence of agentic attributes on
self-scores of leader effectiveness. Agentic attributes did not directly inﬂuence self-score
of leader effectiveness (.07), yet examination of the decomposed effects establishes that
partial effects are being transmitted from all potential paths onto self-scores via all
leadership skills, leadership efﬁcacy, and task self-efficacy, however, a larger portion of

the indirect effect being transmitted from leadership efﬁcacy (see Figure 10).

 

Performance
Skills

 

P.-

 

Motivational
Skills

 

 

 

-°3 Respect-Com.
; Skills

 

 

 

 

Ag entic .07 Self
Attributes , Scores

 

 

 

 

 

 

 

 

Leadership
18* Efficacy

 

 

 

Task
Self-efﬁcacy

 

 

 

Figure 10. Direct and Indirect Pathways from Agentic Attributes to Self
Scores of Leader Effectiveness

120

Agentic attributes did not directly inﬂuence collective efficacy beliefs -.06 but
both the indirect (.16) and total (. 10) effects were signiﬁcant. The relationship between
agentic attributes and collective efficacy was primarily mediated by task self—efﬁcacy

(see Figure 11).

 

4 Performance

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

/ i Skills
;
Motivational
Skills
.20'
/.03
Respect-Com.
/ : Skills
.05 l
Collective
Agentic ~02
. Efﬁcacy
Attnbutes
.44'
. Leadership
~23 Efﬁcacy
r
Task
Self-efﬁcacy

 

 

 

Figure 11. Direct and Indirect Pathways from Agentic Attributes
to Collective Efficacy.

121

The relationship between communal attributes and self-scores of leadership
effectiveness was fully mediated by leadership skills, leadership efficacy, and task self-
efficacy (see Figure 12). The direct inﬂuence of communal attributes on self-scores (.04)

was not signiﬁcant, while the indirect (. 10) and total (.14) emerged as signiﬁcant.

 

Performance
Skills

 

 

 

Motivational
Skills

 

.11' L

 

 

Respect-Com.
’ Skills
.17‘

Communal -°‘
Attributes

 

 

 

 

 

 

Self
Scores

 

 

 

 

 

 

.24'

.22'

 

' Leadership
.16' Efﬁcacy .18"

 

 

Task
Self-efﬁcacy

 

 

Figure 12. Direct and Indirect Pathways from Communal
Attributes to Self Scores of Leader Effectiveness.

122

Although the inﬂuence of leader orientation was not fully mediated by leadership skills,
leadership efficacy, and task self-efficacy, the signiﬁcant direct effects increased after
accounting for the indirect inﬂuences of the regulatory mechanisms. This ﬁnding
suggests that a strong emphasis on leader orientation signiﬁcantly inﬂuenced how
athletes scored themselves and were scored by others (peers) on leader effectiveness, and
that this positive inﬂuence was strengthened even more so through the athlete’s
conﬁdence to lead others and perform successfully. The direct negative inﬂuence of
leader orientation on collective efﬁcacy (-. 18) was reduced through leadership efficacy
and task self-efficacy.

Overall, there were seven signiﬁcant indirect pathways in the full hypothesized
model. Examination of the direct effects in the full hypothesized model suggested that
leadership skills and efficacy beliefs served as more proximal predictors of leader effects,
over the hypothesized distal predictors of gender role perspectives. In particular,
leadership efficacy emerged as a more inﬂuential mediating mechanisms when examined
in relation to leadership skills and task self-efficacy beliefs. These ﬁndings provide
strong support for Bandura’s (1997) self-efﬁcacy theory which asserts that efficacy
beliefs are the primary mediating mechanism that would explain individual differences in
leader conduct.

Additional Hypotheses
Hypothesis 7: There will be cross-sectional differences in self-regulatory mechanisms
and leadership eﬂects based on leadership role experience Ores or no) and academic

class standing (freshman, sophomore, junior and senior).

123

In order to determine if there were group differences in self-regulatory
mechanisms and leader effects based on previous experience with leadership and class
standing, a 4 x 2 between-subjects multivariate analysis was performed on the following
leadership skills, efﬁcacy beliefs, and leader effect scores: performance-execution skills,
motivationalginterpersonal skills, respect-communication skills, leadership efficacy, task
self-efficacy, peer score leader effectiveness, and self score leader effectiveness.
Independent variables were class standing (freshman, sophomore, junior, and senior) and
previous experience with a leadership role (yes and no). The means and standard
deviations for each of the dependent variables are presented in Table 8.

Results regarding assumptions of homogeneity of variance-covariance matrices
were met with a non-signiﬁcant Box’s M test (F = 1.20, p =06), while the assumption
that each dependent variable had similar variance, using Levene’s test statistic, was met
for all dependent variables except for motivational-interpersonal skills, F (7,359) = 2.62,
p = .01. UsingWilks’ Lambda criterion, the combined dependent variables were
signiﬁcantly affected by the main effect of previous experience with leadership, F (9,350)
= 4.93, p = .000, n = .11. The interaction between previous experience with leadership
and class standing was also signiﬁcant, F (27,1025) = 1.64, p = .001, n = .04. Although
the magnitude of this interaction was small.

Follow-up univariate ANOVAs were conducted to identify which factors
maximized differences among athletes with previous leadership experience compared to
athletes with no leadership experience. The univariate ANOVAs indicated a signiﬁcant
effect for the following regulatory mechanisms: performance-execution leadership skills,

F (1,359) = 8.12, p = .005, n = .02, respect-communication leadership skills, F (l, 359) =

124

Hugo m

bmunxﬁog 9.85:an 5:53 8. Show WE‘RE” $5. masses.“ because}? Swings

 

 

 

 

 

 

 

 

 

 

 

 

Hosanna—.5 .mmwoloano _ 2o Hosanna—.5 Manolo—.8

mgr... 9:12. mew—3:33 5.94.5»: mgr... ...—ES. mouse—=86 mien—.5»:

?— Am—vw Ems ?— AMS 2— $3 3 as a $3 ?— Ams ?— 63
wonmonawooo A. _o :3 N«.oo 6va Woo 6A3 “co :3 wow 38 wow :3 Sow $3 9% :8
mxoocaon
392302: A. 5 no: ...—m 6va so; a: wow 38 boo $3 woo 38 mom :8 who 33
5889.82:
W839 Sou :8 Adm C8 Ao— TZV Poo :5 AN Coo Ago a: so: 38 +3 :8
0033:3830:
romaaarmo Soc 60$ ob. 2.8V ooo 2N8 9.3. 2N3 98 God Mow 2.43 93 ENE moo Cumv
maoﬁ
as... 3:38 «.3 2.5 3.253 .2253 $218 $253 w: 2.: .5. EE
moﬁbmmomom
was 9% 33 To 23 a: ..me 3.. 3.: New 38 N... 38 .....N 33 25 Go
woﬁoqauooo
mooa
ma: Poo $3 Woo 28 3.1;.va ....3 33 wow 6.23 woo God wk: 33 93 Add
woldqamaoa

moon“

 

 

 

 

 

 

 

 

 

 

125

11.13,p = .001, n = .03, leadership efficacy, F(1, 359) = 27.77, p = .000, n =.07, and
task self-efficacy, F (1, 359) = 8.86, p = .003, n :02. Speciﬁcally, athletes with
leadership experience reported greater importance on the leadership skills and also self-
reported higher efficacy beliefs.

Related to leadership effects, there were signiﬁcant group differences on peer-
score performance leadership, F (l, 359) = 24.48, p = .000, 1] =06, self-score
performance leadership, F (1, 359) = 12.26, p = .001, n = .03, peer-score motivational
leadership, F (1, 359) = 28.34, p = .000, n = .07, and self-score motivational leadership,
F (l, 359) = 8.05, p = .001, n = .02. Univariate ANOVAs detected group differences by
class rank for leadership efﬁcacy, F (3, 359) = 2.63, p = .05, n = .02, and task self-
efficacy, F (3, 359) = 2.75, p = .04, n = .02. Athletes with previous experience with
leadership were scored higher by their peers as effective leaders than their teammates in
the same class standing who did not have experiences with leadership. Similarly, athletes
with previous leadership experiences scored themselves higher on leadership
effectiveness than their teammates in the same class standing. Examination of the means
also demonstrate potential cross-sectional differences with senior athletes with previous
leadership experiences reporting the higher mean scores on self-regulatory mechanisms
and effects of leadership. For previous experience with leadership by class standing, there
was one significant interaction for peer scores on performance-leader effectiveness, F (3,
359) = 3.27, p = .02, n= .03. Speciﬁcally, seniors with previous leader experiences (M =
3.68) were given higher scores than all other athletes, and seniors without leadership

experiences (M = 2.58) were scored the lowest out of all groups by their peers. Although,

126

the magnitude of differences were small, the general trends of these ﬁndings provide
some interesting information regarding the role of class rank and previous experience
with leadership in sport.

Hypothesis 8: Athletes assigned a # 1 rank in the boat will report higher mean scores on
leadership skills, eﬁicacy beliefs, and leader scores than athletes who were not identified
as a leader.

For performance leadership, 128 athletes were identiﬁed as the leader in the boat,
while the remaining 239 athletes did not receive a #1 ranking for performance leadership.
Descriptive statistics are presented in Table 9. To determine if these 2 groups differed, a
between-subjects MANOVA was conducted on the following dependent variables: task
orientation, ego orientation, leader orientation, agentic attributes, communal attributes,
performance-execution skills, motivational-interpersonal skills, respect-communication
skills, leadership efficacy, task self-efﬁcacy, and self and peer leader scores. The
assumption related to the variance-covariance matrices was not satisﬁed with a
signiﬁcant Box’s M test, (F = 1.27, p = .02), however, Tabachnik and Fiddell (1996)
state that MANOVA is robust to violations of homogeneity of variance and recommend
using a more stringent alpha when interpreting the results. Therefore, an alpha of .01 was
used to determine statistical signiﬁcance. Non-signiﬁcant Levene’s tested emerged for all
of the dependent variables except leadership efficacy, F (1, 365) = 4.57, p = .03, self
scores for performance leadership, F (1, 365) = 9.53, p = .002, and self scores for
motivational leadership, F (1, 365) = 8.96, p = .003. Results demonstrated a signiﬁcant
multivariate effect with a signiﬁcant Wilks’s Lambda F (15, 351) = 11.78, p = .000.

Signiﬁcant follow-up univariate ANOVAs were identiﬁed for leader orientation,

127

Table 9

Descriptive Statistics for Athlete ranked # 1 Performance Leader versus Athletes Not
Ranked # 1 in Performance Leadership

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ranked #1 Not Ranked #1

(N =128) (N=239)

M (SD) M (SD)
Task Orientation 4.39 (.46) 4.36 (.42)
Ego Orientation 2.88 (.76) 2.82 (. 70)
Leader Orientation 3.56 (72) 3.20 (.69)
Agentic 3.77 (.60) 3.64 (.59)
Communal 3.83 (.72) 3.87 (.72)
Performance-Execution Skills 4.01 (43) 4.00 (.46)
Motivational-Interpersonal Skills 4.04 (.52) 4.04 (53)
Respect-Communication Skills 4.59 (.44) 4.60 (.43)
Negative Tactics Skills 1.35 (.64) 1.45 (.76)
Leadership Efficacy 7.00 (1.13) 6.49 (1.35)
Task Self-efﬁcacy 8.33 (1.26) 8.10 (1.33)
Peer Score-Performance Leadership 3.85 (.77) 2.89 (.71)
Self Score- Performance Leadership 402 (.82) 3.36 (.86)
Peer Score- Motivational Leadership 3.89 (.70) 3.19 (.66)
Self Score- Motivational Leadership 4.08 (.78) 3.62 (.85)

 

128

 

F (1, 365) = 22.38, p = .000, n = .06, peer scores for performance-leadership
effectiveness, F (1, 365) = 140.24, p =.000_, n = .28, and motivational-leadership

effectiveness, F (1, 365) = 88.84, p = .000, n = .20. Athletes ranked as the #1 leader had
higher scores on leader orientation, and were scored higher by their peers on leader
effectiveness. The F statistic for agentic attributes approached signiﬁcance but failed to
meet the .01 alpha criterion F (1, 365) = 4.04,p = .05.

Related to motivational leadership, the table of means and standard deviations
between athletes ranked #1 (n = 132) and athletes not ranked #1 (n = 23 5) are presented
in Table 10. Assumptions related to equal variance-covariance matrices between the two
groups was satisﬁed with a non-signiﬁcant Box’s M test, (F = 1.09, p =23). The
assumption regarding equal error variances across the two groups was met for all
dependent variables except respect-communication skills, F (1, 365) = 9.29, p = .002,
negative tactic skills, F (1, 365) = 6.37, p = .01, peer-score performance leadership
effectiveness, F (1, 365) = 5.12, p = .02, self-score performance leadership effectiveness,
F (1, 365) = 13.38, p = .000, and self-score motivational leadership effectiveness,

F(1, 365) = 10.39,p= .001.

The MANOVA test statistic Wilks’s Lambda = .76 was signiﬁcant F (13, 351) =
7.49, p = .000, indicating the two groups differed on the combination of dependent
variables. To probe the signiﬁcant MANOVA ﬁnding, follow-up ANOVAs were
computed. Signiﬁcant overall F values were obtained for 10 of the 15 dependent
variables. Signiﬁcant differences were observed between the two groups for task
orientation, F (1, 365) = 6.06, p = .01, n = .02, leader orientation, F (1, 365) = 18.65, p =

.000, n = .05, motivational-interpersonal skills, F (1, 365) = 3.85, p = .05, n = .01,

129

Table 10

Descriptive Statistics for Athlete ranked # 1 Motivational Leader versus Athletes Not
Ranked # 1 in Motivational Leadership

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ranked #1 Not Ranked #1

(N =l32) (N=235)

M (SD) M (SD)
Task Orientation 4.4ﬂ41) 4.34 L44)
Ego Orientation 2.89 (.76) 2.82 (.70)
Leader Orientation 3.54 Q3) 3.21 (.69)
Agentic 3.77461) 3.64 (.59)
Communal 3.92 (.69) 3.82 (.73)
Performance-Execution Skills 4.05 (.42) 3.98 (.46)
Motivational-Integ)ersonal Skills 4.11 (.51) 4.00 (53)
Regect-Communication Skills 4.67 (.37) 4.55 (.46)
Negative Tactics Skills 1.33 (.66) 1.46 (.75)
Leadership Efficacy 7.04 (1.20) 6.46 (1.31)
Task Self-efficacy 8.39 (1.29) 8.07 (1.31)
Peer Score-Performance Leadership 3.74 (.86) 2.93 (.74)
Self Score- Performance Leadership 3.96 (.81) 3.39 L89)
Peer Score- Motivational Leadership 3.89 (.70) 3.19 (.66)
Self Score- Motivational Leadership 4.11 (.75) 3.60 (.85)

 

130

 

leadership efﬁcacy, F (1, 365) = 17.44, p = .000, n = .02, task self-efﬁcacy, F (1, 365) =
5.22, p = .000, n = .01, and peer scores for motivational leadership, F (1, 365) = 91.91, p
= .000, n = .20.
Additional Analyses

An exploratory analysis was conducted on the correlates of coach, peer, and self-
rankings of performance and motivational leadership. Coaches scores and ranks were
collected for only 259 athletes from the current sample, and thus, were examined
separately (see Table 11). Athletes who scored higher on agentic attributes, leader goal
orientation, leadership efficacy, and task self-efficacy were more likely to receive a better
ranking from their coaches and peers on performance leadership. Related to motivational
leadership, athletes who reported higher scores on communal attributes, leader goal
orientation, and leadership efficacy were more likely to be ranked as a leader by their

coaches and teammates.

131

Table 11

Correlates of Coach, Peer, and Self Rankings of Leadership Effectiveness (n = 25 9)

 

 

Coach- Coach- Peer-Rank Peer-Rank Self- Self-
Rank Rank Perf. Mot. Rank Rank
Perf. Mot. Leader Leader Perf. Mot.
Leader Leader Leader Leader
Agentic -.17** -.09 -.20*** -.15* -.30*** -.25***
Communal -.05 -.15“ .02 -.04 -.05 -. 14*
Task -. ll -.09 -.06 -.12* -.04 -.18**
Ego -.08 —.06 .01 -.03 -.08 -.08
Leader -.24*** -.19** -.31*** -.32*** -.35*** -.35***
M-Climate -.06 -.01 .04 -.07 .06 -.03
P-Climate .01 .00 -.02 .00 -.03 .Ol
P-Skills -.04 .001 .00 -.04 -.10 -.18**
M-Skills -.06 -.02 -.02 -.05 -. l6" -. l8"
R-Skills -.06 -.08 .04 -. 10 -.09 -.11
LE -.28*** -.20*** -.28*** -.26*** -.42*** -.31"*
T—SE .16” -. 13* -.ll -.09 -.24*** -.2l***
CE .06 .04 .13* .06 .09 .06
2K Erg .08 .07 .13" .03 .05 .03
Time

 

Note. Agentic = Agentic attributes; Communal = Communal Attributes; Task = Task Orientation; Ego =
Ego Orientation; Leader = Leader Orientation; P-skills = Performance-Execution Skills; M-skills =
Motivational-Interpersonal Skills; R-skills = Respect-Communication Skills; LE = Leadership efficacy; SE
= Task Self-efficacy CE= Collective Efﬁcacy; Coach rank performance-leader effectiveness; Coach rank
motivational-leader effectiveness; Peer rank performance leader effectiveness; Peer rank motivational
leader effectiveness; Self rank performance leader effectiveness; Self rank motivational leader
effectiveness.

*p<.05, **p<.01. ***p<.001

132

Chapter 5
DISCUSSION

The current ﬁndings provide an important contribution to the understanding of
leadership speciﬁc to athletes in an additive type sport, such as the sport of rowing. The
primary purpose of this study was to develop and test a conceptual model of leadership in
athletes based on Bandura’s (1986, 1997) social cognitive theory. The hypothesized
model demonstrated an acceptable ﬁt to the data.

Results from the path analysis demonstrated support for the integration of
achievement goals with goals that emulated athletes’ beliefs about the social reasons for
their sport involvement (Blumenfeld, 1992; Urdan & Maehr, 1995). Speciﬁcally,
individual differences in goal perspectives reﬂected individual differences in regulatory
mechanisms and effects of leadership in athletes. Among the three goal orientations
hypothesized to inﬂuence leadership skills and efficacy beliefs, task orientation was most
closely aligned with the perceived importance of leadership skills. Task orientation was
signiﬁcantly correlated with all three leadership skill dimensions and directly inﬂuenced
performance-execution and respect-communication skills in the hypothesized model. In
contrast, ego orientation predicted only performance-execution skills, while leader goal
orientation did not predict any of the leadership skills.

The observed differences among goal orientations highlight the contrast between
task and ego orientation that has been established consistently in previous research
(Duda, 2001). Task orientation was aligned with effort, skills, and adaptive strategies

related to leadership, while ego orientation related only to the performance dimension,

133

that being the perceived importance of performance-execution skills. In contrast, the
social goal of leader orientation was not positively related to any of the leadership skills.

Related to team and leader effects, task orientation positively inﬂuenced the
evaluation of the #1 performance leader and collective efficacy beliefs. In contrast, leader
orientation negatively inﬂuenced the evaluation of the #1 performance leader and
collective efﬁcacy. With an emphasis on task orientation, athletes are more likely to
report favorable evaluations of the #1 leader and perceive collective efficacy among their
teammates, whereas this is not the case with an emphasis on leader orientation.

However, examination of the path model revealed that the leader goal orientation
emerged as the strongest predictor of leadership efficacy, self scores of leader
effectiveness, and peer scores of leader effectiveness. Thus, athletes who reported feeling
more successful in sport when given the responsibility to lead their teammates, also noted
greater conﬁdence in leading teammates, possessed higher self-perceptions of leader
effectiveness, and were also scored as more effective leaders by their peers. Leader
orientation emerged as a dispositional tendency that differentiated athletes who were
more likely to perceive themselves as leaders, and be viewed as leaders by their peers
from athletes who were not considered leaders. With an emphasis on leader orientation,
the goal is not self-mastery or to demonstrate superior competence in athletic ability, but
rather to exercise leadership ability. Therefore, leader orientation relates to leadership in a
manner that is different than task and ego orientation; its inclusion allows for a unique
perspective of leadership that would not be possible with the examination of only task

and ego orientations.

134

Self- Regulation of Leadership in Athletes

Consistent with previous research (e. g., Chemers et al., 2000; Watson et al.,
2001), leadership efficacy emerged as both a direct predictor and a regulating mechanism
of leadership. Speciﬁcally, leadership efficacy mediated the inﬂuence of the hypothesized
determinants on team and leader effects. This ﬁnding strongly supports Bandura’s (1986,
1997) social cognitive theory of self-efficacy with efficacy beliefs serving as the primary
mediating function between cognition and leader conduct. Furthermore, athletes who
reported greater levels of conﬁdence in their ability to lead their teammates had higher
self-reported scores of leadership effectiveness, and were reported as being more
effective leaders by their teammates. Higher scores on task self-efficacy, on the other
hand, signiﬁcantly predicted self-reported leader effectiveness scores, but not leader
effectiveness scores assigned by teammates. Similar to previous ﬁndings (Chemers et al.,
2000; Paglis & Green, 2002), leadership efficacy emerged as the stronger predictor for
both self- and peer-reported leadership scores.

Previous research has not examined leadership efficacy relative to perceptions of
task self-efficacy and collective efficacy (Chan & Drasgow, 2001; Chemers et al., 2000;
Paglis & Green, 2002). The current investigation provides a signiﬁcant contribution to
the self-efﬁcacy literature with the simultaneous examination of leadership efficacy
relative to task self-efficacy and collective efficacy beliefs. Leadership efficacy emerged
as the strongest predictor of leader effects (i.e., self and peer leader scores), while task
self-efficacy emerged as the stronger predictor of the team effect, collective efficacy.
This distinction among the three efficacy measures demonstrated discriminant validity

and highlighted the important role of leadership efﬁcacy in the hypothesized conceptual

135

model of leadership in athletes. Although the path analysis ﬁndings demonstrated that
task self-efficacy partially mediated the relationship between determinants and leader
effects, and predicted self -scores of leader effectiveness, it was the perceptions of
leadership efficacy that emerged as a strongest mediator and predictor of leader effects in
the conceptual model. Therefore, the finding that leadership efficacy predicts leadership
above and beyond self-efﬁcacy suggests that the conﬁdence speciﬁc to leading others is a
more inﬂuential characteristic of leadership.

In addition to efﬁcacy beliefs, leadership skills were hypothesized to directly
inﬂuence leader effects and function as a regulating mechanism between determinants
and leader effects. Higher scores on performance-execution skills directly predicted the
evaluation of the #1 performance leader. The only other set of skills to emerge as
signiﬁcant predictors of leader effects were the skills related to respect-communication,
which predicted leader evaluation and peer scores of leader effectiveness. Although
leadership skills did not provide as strong a mediation effect as leadership efficacy on the
relationship between determinants and leader effects, leadership skills reﬂected athletes’
perceptions of what skills were important in the athlete leader role.

One interesting ﬁnding regarding leadership skills was the inability to measure
negative tactics used by leaders to inﬂuence followers. This dimension of leadership
skills failed to emerge with an acceptable level of reliability and was subsequently
removed from the ﬁnal analysis. Future research should consider different approaches
such as qualitative to assess negative tactics used to lead others. Limited research exists
on leadership skills (Mumford et al., 2000a, 2000b) and to date, no research has

examined the construct of leadership skills in the sport setting. By differentiating the

136

types of skills that can be used in leadership situations, the current investigation
distinguished skill/ strategy preferences and provided a better understanding of what types
of leadership skills exist in the sport of rowing. For example, athletes who utilized
performance-execution skills (e. g., have the ability to identify performance related
problems) were more likely to be evaluated as effective leaders by their teammates.
Gender Role and Leadership

As hypothesized, the current ﬁndings established differential patterns of self-
regulatory mechanisms of leadership skills and conﬁdence given one’s gender role
perspective. Speciﬁcally, agentic attributes signiﬁcantly predicted performance-execution
skills, leadership efficacy and task self-efﬁcacy. Consistent with Eagly’s (1987)
perspective of the agentic approach to leadership, athletes with a stronger disposition
toward agentic characteristics were more inclined to report the importance of
performance related skills in leading their teammates, and reported greater conﬁdence to
lead others and perform rowing skills successfully. The performance related dimensions
of leadership in addition to perceptions of efficacy were key components of the agentic
approach to leadership. Similar to the agentic perspective, communal attributes positively
predicted performance-execution skills, leadership efﬁcacy, and task self-efficacy.
However, the strength of the inﬂuence of communal attributes on these regulatory
mechanisms was not as strong as the agentic inﬂuence. Furthermore, communal attributes
positively predicted motivational-interpersonal skills and respect-communication skills.
The inﬂuence of communal attributes on regulatory mechanisms of leadership was

distinguished by the interpersonal dimensions of leadership.

137

Although the direct effects from agentic and communal attributes on leader
effects did not emerge, there were signiﬁcant indirect effects, which represented the
strongest demonstration of a mediator effect. Agentic attributes indirectly inﬂuenced self-
and peer scores of leadership effectiveness through leadership skills, leadership efficacy,
and task self-efficacy, with the strongest of these indirect effects being leadership
efficacy. This signiﬁcant indirect effect also emerged for the relationship between
communal attributes and self-score of leader effectiveness. Thus, gender-role
perspectives inﬂuence leader effects via the skills and conﬁdence beliefs used in
leadership behavior.

Consistent with the ﬁndings from Glenn and Horn (1993), agentic attributes were
positively correlated with coach and peer rankings of performance leadership. However,
when performance leadership was contrasted with motivational leadership, communal
attributes emerged as the stronger correlate. Therefore, for this sample of female athletes,
both gender perspectives were associated with effective leadership. The distinction in
leadership dimensions has not been investigated in previous research on leadership in
athletes. The current study allows for the distinct differentiation between agentic and
communal attributes in the sport context. Recently, Yoder (2001) claimed that in order
for researchers to understand effective leadership, a gender-sensitive model of effective
leadership must be adopted. Speciﬁcally, the potential moderating inﬂuence of the
interaction between the leader gender and the gender congeniality of the context should
be examined relative to the roles and responsibilities of leaders and their effectiveness. In
the traditional masculine context of sport, both agentic and communal approaches to

leadership were valued with the current sample of athletes. Thus, the current investigation

138

provided additional information regarding the simultaneous occupation of the athlete.
leader, and female roles.
Structural Achievement Goals and Leadership

A signiﬁcant pattern of regulatory mechanisms of leadership also emerged relative to
individual differences in the subjective appraisal of the motivational climate.
Achievement goal theory suggests (Ames, 1992a; Nicholls, 1992) that the way in which
adults organize their environment will increase the likelihood that a particular
achievement behavior or goals will be adopted. In other words, the environment can be
structured to increase the probability that athletes will choose a more adaptive approach
to leadership. The current ﬁndings suggest that one way coaches can develop leadership
in their athletes is through the situational goal structure of their sport team. Speciﬁcally,
mastery motivational climate demonstrated a consistent and signiﬁcant relationship with
all leadership skills, whereas performance climate signiﬁcantly inﬂuenced only the
perceived importance of performance-execution skills. The emphasis of a mastery
climate will encourage manageable leadership strategies that foster collaboration,
learning, and effort, all of which are in control of the athlete at both the individual and
group levels. However, perceptions of favoritism and an emphasis on normative
comparison will not only decrease the salience of leadership skills but may also render a
perception of uncontrollability both for the individual athlete, and the collective group
working toward a common goal. Although perceived motivational climate did not
directly predict leader effects as hypothesized, the perceived climate in which the athlete
was performing in did inﬂuence the types of skills and strategies that were used in

leadership situations.

139

Implications for Coaches
The emergence of leader-goal orientation as a signiﬁcant predictor of leadership

efﬁcacy and leader effectiveness scores provides an explanation as to why certain athletes
were identiﬁed as leaders. Individual differences in the desire to, and the importance of
leading others signiﬁcantly inﬂuenced their conﬁdence to take on leadership
responsibilities and their effectiveness in doing so. Thus, coaches who seek to identify
athletes as potential leaders should consider the dispositional tendency, or leader
orientation of their athletes. Perhaps more importantly, coaches have the ability to
inﬂuence leadership potential by enhancing leadership efficacy in their athletes. By
providing athletes with mastery experiences speciﬁc to leadership situations, athletes will
feel more conﬁdent in performing subsequent leadership activities and responsibilities.

A second important ﬁnding for coaches pertains to the association between
mastery motivational climate and leadership skills. With an emphasis on a mastery goal
structure, coaches may make the use of leadership skills more obvious and acCessible to
the athletes. Coaches can initiate and manage the development of leadership skills by
mentoring athletes during the process of acquiring and reﬁning the various skills that
foster leadership behavior. Therefore, coaches may groom future leaders by establishing
a mastery motivational climate making the use of leadership skills available to athletes,
by providing leadership opportunities that develop mastery experiences and enhance
efﬁcacy beliefs in leadership.

The current ﬁndings not only ﬁll a signiﬁcant void in the sport psychology
literature, but also contribute to better understanding of how athletes perceive leadership

in sport, what characteristics they value in a leader, their conﬁdence to lead others, and

140

whether they identify with peer leaders. This information can be used by coaches,
parents, and sport psychology practitioners to improve the teaching and development of
leadership through experiences in sport.

The development of effective leadership demands time and involvement from the
coach to teach athletes to think for themselves and develop a sense of independence while
at the same time, learning to work with each other. By using team sport participation to
emphasize the development of the fundamental skills that result in leadership, coaches
can provide athletes with the opportunity to acquire interpersonal and relation-
management skills. Coaches who make an effort not only to develop the physical
competence but also to foster social competence may provide their athletes with long
lasting skills that could beneﬁcially impact their lives beyond sport.

Implications for Future Research

The purpose of this research was to provide an understanding of the leadership
process and guidance for future research in this area by validating and testing a
conceptual model of leadership in athletes. The learning, development, and exercise of
leadership represent a dynamic process, which is difﬁcult to measure and operationalize.
As with any complex and dynamic intra-team interaction, it is important to acknowledge
there are other variables that may inﬂuence the leadership process that will not be
included in the current investigation. Furthermore, with a one-time assessment the
dynamic process of leadership over time was not taken into account. The current
investigation assessed beliefs regarding leadership at the beginning of a racing season,
therefore, it is important to note that a lot of changes may have occurred over the course

of the season that may have had a signiﬁcant impact on the leadership process. In order to

141

explore the dynamic changes in leadership, future research should examine leadership
over the entire course of the rowing season in order to assess the emergent patterns that
may occur.
Conclusions

To date, a limited amount of research on leadership in athletes exists. Beyond the
testing of a conceptual model based on theory, research in sport needed additional
research that identiﬁed skills and characteristics that would differentiate potential leaders
from non-leaders. The current investigation identiﬁed the personal and situational
characteristics of leader goal orientation and mastery motivational climate as
distinguishing qualities of the leadership process in athletes. Individual differences in the
regulatory mechanism of leadership efficacy also further distinguished athletes who were

more likely to be identiﬁed as leaders of their teammates.

142

APPENDICES

143

APPENDIX A
Demographic Background Questionnaire

 

 

Initials: Birth Date: / / Today’s Date:
Age : Year in School Gender
CI Freshman [:1 Male
[:1 Sophomore Cl Female
Cl Junior
Cl Senior
Current Crew Program: Head Coach:

 

Coach who spends the most time with you:

 

 

 

How long have you rowed in a competitive program?

 

(e.g., 4 years and 3 months)

How long have you rowed in college?

 

Are you on an athletic [:1 Full Scholarship Cl Partial Scholarship
scholarship?

How important is rowing to you?

0 1 2 3 4 5 6 7 8 9
Not at all Somewhat
Important Important

How good are you at rowing?

O l 2 3 4 5 6 7 8 9

 

Not at all Somewhat

Good Good
What is your best 2K erg time this year ?
What is your all-time personal best 2K ?

 

Please circle the percentage of time you compete in a priority boat:

0 10 20 30 40 50 60 70 80
Never Sometimes

144

D None

1 0
Extremely
Important

1 0
Extremely
Good

90 100
Always

 

Have you experienced an illness or injury within the last 6 months? YES NO

If yes, please describe

 

Were you allowed to practice or compete with this illness or injury? YES NO

If no, how long were you removed from practice or competition?

 

What boat are you currently rowing in?

 

(i.e., Varsity “A” 8+, JV “B” 4+ etc.)

What seat are you currently rowing in?

 

(i.e., 2 seat in V8+, 3 seat in JV4+ etc.)

Have you switched seat positions since the beginning of
this competitive season (say Feb)? YES N 0

If yes, what seats have you rowed in for more than 3 consecutive days?

 

 

Have you ever held a leadership role of any kind? YES NO
(i.e., team captain, student council)

If yes, what kind of leadership experiences have you had?

 

 

Are you a team captain this year? YES NO

Do you consider yourself to be a leader in crew? YES NO

How important is being a leader to you?
0 l 2 3 4 5 6 7 8 9 10

Not at all Somewhat Extremely
Important Important Important

145

APPENDIX B
Gender Role Orientations

Directions: Please indicate how each of these characteristics describe your true personality.

 

Not at all Somewhat Totally
like me like me like me

 

. Defend my own beliefs 1 5

 

. Have leadershii abilities

 

. Affectionate

 

. Eager to soothe hurt feelings

 

. Conscientious

 

. Secretive

 

\IQLIIAUJNI—

. Independent

 

8. Willing to take risks

 

9. Sympathetic

 

10. Warm

 

ll. Moody

 

12. Adaptable

 

l3. Assertive

 

14. Dominant

 

15. Sensitive to needs of others

 

16. Tender

 

17. Reliable

 

18. Conceited

 

19. Strong Personality

 

20. Willing to take a stand

 

21. Understanding

 

22. Love children

 

23. Jealous

 

24. Tactful

 

25. Forceful

 

26. Aggressive

 

27. Compassionate

 

28. Gentle

 

29. Truthful

 

p—n—u—nn—nr—nu—np—o—p—nu—u—u—nu—nu—u—u—u—sp—u—nu—r—nu—u—u—nu—nu—u—u—uu—n
NNNNNNNNNNNNNNNNNNNNNNNNNNNNNN
wwwwwwwwwwwwwwwwwwwwwwwwwwwwww
AA-hb-bA-khabA-bhhhb-h-bAbAAAAbhA-bbh-b
LIIMMMLIIMMMMMMU‘MMMU‘U‘MMMMMMMMMMMM

30. Conventional

 

 

 

 

146

APPENDIX C
Goal Orientations

Directions: Please read each of the statements listed below and indicate how much you
personally agree with each statement by circling the appropriate response.

I feel most successful in rowing when...

 

Strongly Disagree Neutral Agree Strongly

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Disagree Agree
1. I outperform everyone else on my team. 1 2 3 4 5
2. I make improvements and it makes me 1 2 3 4 5
want to practice more.
3. I can do better than my friends. 1 2 3 4 5
4. My teammates look up to me. 1 2 3 4 5
5. Other rowers can’t do as well as me. 1 2 3 4 5
6. I learn something that is fun to do. 1 2 3 4 5
7. Other rowers’ mess-up and I don’t. 1 2 3 4 5
8. I’m in charge of my teammates. l 2 3 4 5
9. I make improvements by trying hard. 1 2 3 4 5
10. I work really hard. 1 2 3 4 5
11. I lead by example. 1 2 3 4 5
12. I outperform others on the erg. l 2 3 4 5
13. Something I learn makes me want to 1 2 3 4 5
go and practice more.
14. My teammates consider me as the l 2 3 4 5
leader of our boat.
15. I’m the best. 1 2 3 4 5
16. A skill I learn really feels right. 1 2 3 4 5
l7. 1 do my very best. 1 2 3 4 5
18. I organize what my teammates do. 1 2 3 4 5
19. I’m in a boat that wins races. 1 2 3 4 5
20. I work together with my teammates. l 2 3 4 5

 

 

 

147

APPENDIX D
Motivational Climate

Directions: Please read each of the following statements carefully and respond to each item in
terms of how you view the head coach of the program. Circle the number that best represents
how you feel.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

On this team ...................
Strongly Disagree Not Agree Strongly
Disagree Sure Agree
1. the coach gets mad when a rower
makes a mistake. I 2 3 4 5
2. each rower contributes in some 1 2 3 4 5
important way
3. the coach believes that all of us
have the potential to become 1 2 3 4 5
leaders.
4. the coach praises rowers only when
they outperform each other. 1 2 3 4 5
5. rowers feel good when they are
given leadership responsibilities. 1 2 3 4 5
6. rowers are dropped from the boat if
they continually make mistakes. 1 2 3 4 5
7. rowers at all skill levels have an
important role on the team. 1 2 3 4 5
8. the coach makes sure rowers
improve on skills they’re not good 1 2 3 4 5
at.
9. only the rowers with the best skills 1 2 3 4 5
get praise.
10. rowers are not concerned with
developing leadership skills. 1 2 3 4 5
11. the coach encourages athletes to
help each other learn. 1 2 3 4 5
12. the coach makes it clear who he/she
thinks are the leaders. 1 2 3 4 5
13. rowers understand who the
designated leaders are for each boat. 1 2 3 4 5
14. the coach emphasizes always trying
your best. 1 2 3 4 5
15. only the top athletes ‘get noticed’ by
the 1 2 3 4 5
coach.
16. the coach wants each of us to
develop the skills necessary to 1 2 3 4 5
become leaders in our sport.
17. the coach favors some rowers more
than others. 1 2 3 4 5
18. the coach thinks only the best
athletes contribute to the success of 1 2 3 4 5
the team.

 

 

 

148

APPENDIX E
Task Self-efficacy

Directions: Please think about your personal performance in the sport of rowing and how
conﬁdent you are for this upcoming regatta. Read each of the statements listed below and
indicate how conﬁdent you are for each statement.

How confident are you that YOU.....

 

Not at all Somewhat Completely
Confident Conﬁdent Conﬁdent

 

l.cansuccessfullyexecutetheraceplan 0 l 2 3 4 5 6 7 8 9 10

 

2. can maintain efﬁcient technique
throughouttherace 0 l 2 3 4 5 6 7 8 9 10

 

3. have the stamina to successﬁrlly hold
your desired stroke rate/boat speed 0 l 2 3 4 5 6 7 8 9 10

 

4. can make a signiﬁcant contribution to

 

 

 

 

 

 

yourboatwinningarace 0 l 2 3 4 5 6 7 8 9 10
5. can use the physical skills necessary to
besuccessful 0 1 2 3 4 5 6 7 8 9 10
6. can use the mental skills necessary to
besuccessful 0 l 2 3 4 5 6 7 8 9 10
7. contribute to your boat’s ability to row
it’sfastesttime O l 2 3 4 5 6 7 8 9 10
149

APPENDIX F
Leader Efficacy

Directions: Please think about your conﬁdence in your ability to be a leader in your boat and
respond to the questions below.

How confident are you that you can .......................

 

Not at all Somewhat Completely
Conﬁdent Conﬁdent Confident

 

l. Organize boat members to successfully

prepare boat for travel (e. g., place boats,

 

 

 

 

oars. and other equipment on trailer). 0 l 2 3 4 5 6 7 8 9 10
2. Organize boat members to complete boat

preparation (i.e., clean and rig the boat). 0 l 2 3 4 5 6 7 8 9 10
3. Calm boat members’ nerves before testing or

competition 0 l 2 3 4 5 6 7 8 9 10
4. “Take charge” when you have to. 0 1 2 3 4 5 6 7 8 9 10
5. Get teammates to pull harder. 0 1 2 3 4 5 6 7 8 9 10

 

6. Encourage all boat members to maintain an
efficient training regimen (i.e., attend all

 

 

practices, weightlifting sessions. etc). 0 l 2 3 4 5 6 7 8 9 10
7. Perform well as a leader across different

situations. 0 l 2 3 4 5 6 7 8 9 10
8. Provide inspiration for other boat members. 0 l 2 3 4 5 6 7 8 9 10

 

9. Be the “go to” person and lead the boat toa
successfulperforrnance. 0 1 2 3 4 5 6 7 8 9 10

 

10. Encourage all boat members to maintain a
healthy lifestyle (i.e., proper nutrition,

 

 

 

 

 

rest/recovery/ injury prevention, etc). 0 l 2 3 4 5 6 7 8 9 10

ll. Successfully build boat members’confidence. 0 l 2 3 4 5 6 7 8 9 10

12. Set reachable goals for your crew. 0 1 2 3 4 5 6 7 8 9 10

13. Intimidate others into performing well 0 l 2 3 4 5 6 7 8 9 10

14. Discipline other teamrrrates when appropriate. 0 l 2 3 4 5 6 7 8 9 10
15. Encourage teammates to “win at all cost,”

(e. g., jump the start or cheat). 0 l 2 3 4 5 6 7 8 9 10

 

 

 

150

APPENDIX G
Leadership Skills

Directions: Please think about leadership in rowing and rate the importance of each statement
below.

It is important for an athlete leader to.....

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Not at all Somewhat Very
Important Important Important
1. Be the “go to” person in a time
of need 1 2 3 4 5
2. Be sensitive to the needs of
every teammate l 2 3 4 5
3. Set performance goals for the
team 1 2 3 4 5
4. Do anything to win (e.g., jump
the start, or cheat , etc) 1 2 3 4 5
5. Make others perform better
through example 1 2 3 4 5
6. Display 100% effort 1 2 3 4 5
7. Console teammates when they
are frustrated l 2 3 4 5
8. Be respected by teammates l 2 3 4 5
9. Have confidence in his/her
ability to lead others 1 2 3 4 5
10. Keep teammates calm before
competition 1 2 3 4 5
ll. Intimidate others to get the l 2 3 4 5
work done
12. Resolve conﬂict between 1 2 3 4 5
teammates
13. Be one of the best athletes on 1 2 3 4 5
a team
14. Foster “togetherness” or 1 2 3 4 5
cohesion
15. Handle pressure situations 1 2 3 4 5
l6. Communicate effectively with l 2 3 4 5
teammates
17. Be respected by coaches l 2 3 4 5
18. Have the ability to identify
performance related problems 1 2 3 4 5
19. Provide external motivators
(e.g., foster team-bonding
situations, develop team
identgy, etc.) 1 2 3 4 5
20. Encourage teammates to win
at all cost (e.g., jump the start
or cheat) l 2 3 4 5

 

 

 

151

 

 

 

 

 

 

 

21. Serve as the liaison between 2 4
coach and teammates

22. Help the team win by rowing 2 4
the fastest

23. “Call someone out” who is 2 4

not giving 100%

24. Behave responsibly 2 4

25. Delegate responsibility to 2 4
teanunaux;

26. Communicate effectively with 2 4

coaches

 

152

 

APPENDIX H
Performance Leadership Identiﬁcation

PLEASE REFER TO THE BOAT YOU ARE CURRENTLY ROWING IN

Directions: A performance leader is a crew member who is considered to be the “go to” person,
is competent, masterful, assertive, conﬁdent, and may lead a boat toward a successful
performance outcome. A successful performance outcome may include an improved race time or
winning a race. Please think of the person in your boat who is a performance leader.

1. List the name or initials of each person in your boat (i.e., varsity top Eight).

After you have identiﬁed each member, please provide each person with a score
reﬂecting his/her leadership capability using the scale of 1 to 5 listed below.

3. After providing a leadership score, rank each member starting with the one person you
consider to be the leader (i.e., rank #1), followed by the next member you consider to be
the second best leader on the team (i.e., #2) until all members are ranked. Please include
only those members that are in your boat and include yourself. If you feel that there is a
tie between members, please try your best to distinguish them and give each athlete a
diﬁ’erent score.

Be assured that your answers will remain confidential and no one except me will see

your responses.

Boat Performance Leader Scale

 

 

 

 

 

 

 

 

 

 

 

 

 

l i 3 4 5
Not at all Somewhat Entirely
Effective Effective Effective
Name or initials Leadership Score Leadership
(please complete this section ranging from 1 to 5 Rank
FIRST) (please complete this (please
section SECOND) complete this
section
LAST)
example Deb F. or DF 4 #3
example Michelle M. or MM 2 #9
Seat 1
Seat 2
Seat 3
Seat 4
Seat 5
Seat 6
Seat 7
Seat 8
COX

 

 

 

 

 

 

153

APPENDIX 1
Motivational Leader Identiﬁcation

Now we want you to think of the person who has the capability to motivate your crew. This type
of leadership is not limited to performance and may include interactions both in and out of the

boat.

Cl Examples of motivational leadership are: encouraging teammates to stay tough and work
through the pain (i.e., on the erg or during a race), resolving conﬂict between members of the
boat, acting unselﬁshly, showing concern for others, or helping teammates calm their nerves
before testing and competitions.

Please rate each crew member on his/her motivational leadership based on the scale below and

provide

a rank for each member in your boat and include yourself.

Be assured that your answers will remain conﬁdential and no one except me will see
your responses.

Boat Motivation Leader Scale

 

 

 

 

 

 

 

 

 

 

1 2 3 4 5
Not at all Somewhat Entirely
Effective Effective Effective
Name or initials Leadership Score Leadership
(please complete this section ranging from 1 to 5 Rank
FIRST) (please complete this (please complete
section SECOND) this section
LAST)
Seat 1
Seat 2
Seat 3
Seat 4
Seat 5
Seat 6
Seat 7
Seat 8
COX

 

 

 

 

 

 

 

154

 

APPENDIX J
Peer Performance Leader Evaluation

Directions: Please rate the person you consider the performance leader (ranked #1) on the
following characteristics

 

 

 

 

 

”mm ~

 

 

 

Name or initials: Seat #:
Not at all Somewhat All of the
time

1. How often does this individual contribute to l 2 3 4 5
successful performances?

2. How oﬁen is this athlete a performance 1 2 3 4 5
leader?

3. How often does this athlete display a hard 1 2 3 4 5
work ethic?

4. How oﬁen does this athlete perform at his/her l 2 3 4 5
highest level?

 

 

155

APPENDIX K
Peer Motivational Leader Evaluation

Directions: Please rate the person you consider the motivational leader (ranked #1) on the

following characteristics

 

 

 

 

 

 

 

 

Name or initials: Seat#:
Not at all Somewhat All of the
time

1. How often does this individual 1 3 4 5
motivate you to do your best?

2. How often is this athlete a motivational l 3 4 5
leader?

3. How often does this athlete display a l 3 4 5
positive attitude?

4. How often does this athlete perform at 1 3 4 5
his/her highest level?

 

156

 

 

APPENDIX L
Self Performance and Motivational Leader Evaluation

Directions: Please think about your own capabilities as a leader in the sport of rowing and
answer the following questions below about your leadership capabilities.

 

 

 

 

 

 

 

 

 

attitude?

 

Not at all Somewhat All of the
time

1. How often do you contribute l 2 3 4 5
to successful performances?

2. How often are you a performance leader ? l 2 3 4 5

3. How often do you display a hard work 1 2 3 4 5
ethic?

4. How often do you perform at your highest 1 2 3 4 5
level?

5. How often do you motivate your 1 2 3 4 5
teammates to do their best?

6. How often are you a motivational leader? 1 2 3 4 5

7. How often do you display a positive 1 2 3 4 5

 

 

157

M;
‘_

APPENDIX M
Crew Efficacy

Directions: Now think about how conﬁdent you are in your crew’s ability toperform in this
upcoming regatta. Read each of the statements listed below and indicate how conﬁdent you are
for each statement.

How conﬁdent are you that your CREW.....

 

Not at all Somewhat-r Completely
Confident Confident- Confident

 

 

 

 

1. can successfully execute the race plan 0 l 2 3 4 5 6 7 8 9 10
2. can maintain efficient technique 0 l 2 3 4 5 6 7 8 9 10
throughout the race
3. has the stamina to successfully hold its
desired stroke rate/boat speed 0 l 2 3 4 5 6 7 8 9 10
4. can win the race 0 l 2 3 4 5 6 7 8 9 10

 

5.canuse the physical skills necessarytobe 0 l 2 3 4 5 6 7 8 9 10

successful

 

6.canusethementalskillsnecessarytobe 0 1 2 3 4 5 6 7 8 9 10

successful

 

7.canrowit’sfastesttime 0 l 2 3 4 5 6 7 8 9 10

 

 

 

158

APPENDIX N
Coach’s Packet

 

 

 

 

Initials: Birth Date: / / Today’s Date: / /
College Program: Gender: Male Female
Is your program a varsity-funded program or a club sport? Varsity Club

Do you have athletic scholarships available to give to your athletes? YES NO

 

Coaching Experience 1

How many years have you been coaching college rowing ?

 

How many years have you been coaching at your current institution?

 

What boat(s) are you coaching this year?

 

What different levels of rowing have you coached (i.e., high school, college novice,
international)?

 

How important is coaching rowing to you?

0 1 2 3 4 5 6 7 8 9 10
Not at all Somewhat Extremely
Important Important Important

 

How good are you at coaching rowing?

O 1 2 3 4 5 6 7 8 9 10
Not at all Somewhat Extremely
Good Good Good

What are some of the reasons you got into coaching?

 

 

159

Boat Selection Method

How much importance do you place on seat racing when determining boat selection?

0 1 2 3 4 5 6 7 8 9 1 0
Not at all Somewhat Extremely
Important Important Important

How much importance do you place on erg scores when determining boat selection?

0 1 2 3 4 5 6 7 8 9 10
Not at all Somewhat Extremely
Important Important Important

How much importance do you place on athlete body size when determining boat
selection?

0 l 2 3 4 5 6 7 8 9 10
Not at all Somewhat Extremely
Important Important Important

How much importance do you place on athlete physiological measures (e. g., VOZmax)
when determining boat selection?

0 l 2 3 4 5 6 7 8 9 10
Not at all Somewhat Extremely
Important Important Important

How much importance do you place on attitude (e. g., dedicated, outgoing, positive, etc.)
when determining boat selection?

0 1 2 3 4 5 6 7 8 9 10
Not at all Somewhat Extremely
Important Important Important

How much importance do you place on your own expertise when determining boat
selection?

0 1 2 3 4 5 6 7 8 9 10
Not at all Somewhat Extremely
Important Important Important

160

Leadership Experience

What were your experiences with leadership while you were competing as an athlete?

 

 

How important is leadership in athletes to you?

O 1 2 3 4 5 6 7 8 9 10
Not at all Somewhat Extremely
Important Important Important

Would you consider leadership skill development in your athletes to be a priority in
your coaching philosophy?

0 l 2 3 4 5 6 7 8 9 10
Not at all Somewhat Very much
a priority of a priority of a priority

Do you designate leaders (captains) at the beginning of the season? YES NO

What is the method used to select leaders (captains) for the entire team?

 

 

What is the method used to select leaders for each boat?

 

 

What do you look for in a leader of your rowers?

 

 

161

Instructions for Leadership Ranking

You are being asked to rank the leadership abilities of the group of athletes performing in
2 boats (preferably the top Eight and the second Eight). Please make sure that the group
of athletes you have listed in the boat is consistent with the same group of crew members
the athletes will be ranking. For example, all 9 athletes designated to row in the top
varsity Eight for the purpose of this exercise should know who they are and who they
should consider for the leader rankings. There should be no discrepancy between the
group of athletes you are ranking and the group of boatmembers the athletes are ranking.
Your list of athletes will be used as the master roster and will be referred to in order
to resolve any discrepancies among the athletes’ responses regarding boat
membership.

I have enclosed ranking forms for you to rank 2 boats, however, please feel free to
photocopy extra forms to rank additional boats.

Please make sure you list the entire name of all athletes in each boat. Do not include
yourself or assistant coaches in the ranking.

A performance leader is a crew member who is considered the “go to” person and is
capable of leading a boat toward a successful performance outcome. A successful
performance outcome may include an improved race time or winning a race. Please think
of the person in the boat who is a performance leader.

4. List the name of all athletes in the boat (i.e., varsity top Eight) on the next page.

5. After you have listed the name of each athlete, please provide each person with a
score reﬂecting his/her leadership capability using the scale of 1 to 5 listed on
the next page.

6. After providing a leadership score, rank each athlete starting with the one person
you consider to be the leader (i.e., rank #1), followed by the next member you
consider to be the second best leader in the boat (i.e., #2) until all members are
ranked. If you feel that there is a tie between members, please try your best to
distinguish them and give each athlete a different rank.

162

PERFORMANCE LEADERSHIP RANKING-BOAT #1

BOAT #1:

 

(i.e., varsity top Eight)

Cl Examples: athlete who is considered the “go to” person, is competent, masterful

9

assertive, conﬁdent, and contributes to improved race time or winning a race.

Performance Leader Scale

 

 

 

 

 

 

 

 

 

 

 

 

 

1 2f 3 §
Not at all Somewhat Entirely
Effective Effective Effective

Name Leadership Score Leadership Rank
(please complete ranging from 1 to 5 (please complete
this section (please complete this this section LAST)
FIRST) section SECOND)
Example Deb Feltz 4 #3
Example Michelle Magyar 2 #10

Seat 1

Seat 2

Seat 3

Seat 4

Seat 5

Seat 6

Seat 7

Seat 8

COX

 

 

 

 

 

 

163

 

MOTIVATIONAL LEADERSHIP RANKING-BOAT #1

Now we want you to think of the person who has the capability to motivate the boat.

This type of leadership is not limited to performance and may include interactions both in

and out of the boat.

Please rate each athlete on his/her motivational leadership based on the scale below and

provide a rank for each crew member.

El Examples of motivational leadership are: encouraging teammates to stay tough and

work through the pain (i.e., on the erg or during a race), resolving conﬂict between
members of the boat, acting unselﬁshly, showing concern for others, or helping

teammates calm their nerves before testing and competitions.

Boat Motivational Leader Scale

 

 

 

 

 

 

 

 

 

 

 

 

 

1 2 3 4 5
Not at all Somewhat Entirely
Effective Effective Effective
Name Leadership Score Leadership Rank
(please complete this ranging from 1 to 5 (please complete
section FIRST) (please complete this this section LAST)
section SECOND)
Example Deb Feltz 4 #3
Example Michelle Magyar 2 #10
Seat 1
Seat 2
Seat 3
Seat 4
Seat 5
Seat 6
Seat 7
Seat 8
COX

 

 

 

 

 

164

 

TABLES

I65

.350 O-—

bmonzwocm M8528 ex. MmfmmxﬁB‘oQ immoozaﬁa QREEEEE 2 u mo d

 

 

32:. m6 £85.38 £5.85 2:355: 38:55:

$2.25». 2:. $3.52.». 62033358

Woimom mxooaooom Nun Ema o.om ohm o.oo o.oo
”Siam ~Booam=oo «.3 o.oo o.oo ..No o.oo _o.oo
wonooméo >E=Q o.oo ..wN o.oo rum o.oo o.oo
banana ~Boonw=oo #3 NE -oo» o.oo o.oo _o.oo
>mgoo 339:8 o.oo o.oo -o.o~ o.oo Foo moo
003358— ?3653 o.oo o.om o.oo o.oo foo o.oo
How—A Damiano: Pom o.oo o.oo o.om Nd o.oo
mmo 0138.30: Non o.oo o. 5 o.oo #3 Moo
.1032 012.830: o.oo o.ow o.oN o.oo _.~o o.oo
guano—Q 03:58 Poo ouo -obm :3 foo moo
wowmonomooo 0:3me Now o.oo 93 o. 2 o.oo o.oo
wear—anions! 33332.:

omROABmooahxmocao—o mem Poo ohm -o. B -o. 3 who o.oo
303$:meow—#:88283— mem Krow o. mo -o. I -o. oo N. 3 m .oo
womomornoaacaowaoo mem k.oo o.oo -_ .oo ohm woo woo
romomqmao mangoN o.oo :3 -o..: =5 :3 v.3
Hour marmaomow m. E foo Low Pmm o.oo _o.oo
Oo=aoo<o mwmomavN o.oo o.oo L .oo 88 :5 5.8

 

166

 

HmEa Ob

bmunwﬁacm M59382 ghgnmw mﬁqmgu \2 u mad

”5.95 -...E ...

 

 

325 mu 283.8% K548: 73555: 33:55:

Haas. @288

wag moono vandaamsoanmaon mmwomésamm ohm o.oq o.om o.oo _.oo moo
:2 ”war voamoasmaanoqu mﬁooa<a=omm Pow Now o.oA .93 o..: 3.8
mo: macaw wand—4339533 mmwoménamm o.oo o.oo -93 o.oo _.oo o.oo
mo: WE? vaROansoa-romqon mmwoﬁzoaomm o.oo N. _ m o.oo o.oo _.oo o.oo
mama macaw Koméaoa—Famqan mmwﬁzosomm .413 ogm .9: -o. _ m o.oo 9mm
moon Wmsw zoa<mao=m_-ramoaq mmooaésomm Ago ..3 o.oo -ogo o.oo Hooo
ma: mooB Zoa<mao=m7ramaoq mmwomésomm 94m 9mm -o.m_ omq _.oo o.oo
mo: 35w zomémosmrramaaq mmwoméaamm wow o.om o.on o.oo o.oo o.oo
m<mEmao= 0». 3 wmamozsmaoo hogan Pam obq .93 -o. E Nd moo
m<m_:mno= 0», E 392325— romaoa 5.3 ohm o.oo mum Noo moo
mm: m<m_:mao= om woqmondmsom hame 233 A. 5 oho -o.m_ o.oo moo moo
mm: m<mEmno= om 32232.3 Page. 215% o.oo o.oo -ooA o.oo ..oo moo

 

167

 

H35 0.»

Elli-:1...“

E838: wags. 393$: Qc§w~a2§u c\u.m~\8§ 3mm» omega @Wmngmzmuu Macaw-a. mg»... ER mucﬁzaae: M85 «Zn-3 é

 

 

_ N u A m o q m o do 2 5
_. was. woos $01,035.50??qu mags...“ ..o
N. van. mooao 3223057532 mmwoaé f... S
w. m2», M88 woﬂogwsoa-romaoq magma... .... ... .3... ..o
a. ma: macaw Zoaagmoaw—Fowaan mum-«9?». .3... .3... .3... 6o
m. m<w_:m:o: E $01.. Pomona ..o. ..o. .oo .8 re
a. m<wEmao= 3 Zoe. r952 -oo -.8 .8 .3 ...m... S
q. mo_m-m<w_=mao= $1.0:ng 580376 .8... .3... .2... on... .3... 2.. S
m. mo_m-m<m_=~ao= 32230:». Fauna—.36 .oo... .3... we... .3... .No... .3... ...m... _.o
0. $03 g wold—Swan Panama mmwgg -.qm... -oo... -.3... -3... .3... .5. -8... 33... _.o
_o. 32E 322325— ramamq manna...“ -.om... -..E... -.wm... -.um... .3 .3 Lo... -.No... .3... _.o
:. males—n $33558 roman. mmoazo .3... -..5... -2... -.X... .3: .3 -oo... -8... a... .3... ..o
5. maﬁwwa. Zoa<maosa mmooaézgm -8... 1:... is... -8... .3... .8 -2... -2... .3... .3... .8... S

 

*quw. :UA.o_. ...—Xbo—

168

0.00.0 CA

9030220: £00.50. @9333?- 3. .08. a: wm§o§§nm 03%..qu

 

 

 

 

 

a as: m2: — m2: N wax. u m2: A mu». 0 m2: 0 m2: 0 m2: 0 0325::
A: u uANV

W2... 2 NA A0.\.V 3 6.0.5 3. 2.0.00 0 2.0.x; NN 3.0.5 Nu 3.0.5 AN 2N.0V z: 30.5 2N0 2.0.5
WE... N Nm 3.5 A0 G.A.\.v 20 3.0.5 .0 3.0.x; N0 3. 10V 00 A0.A.\.V 0N ANo.0.\.V 0A 30.00 .3 ANNXV
$5. 0 Nu A0.0.\.v ..N 32.5 20 3.0.5 00 20.5 mm 2o.\.v Aw 2N.w.\.v 0_ 2-N.A.\.v mo 2A.u..\.v 0N 2A.0.\.V
55—. A H: 200:; A0 2N.0.\.V NA A0.0.\.v Ac 2 _.A.\.V 00 20.0.\.v 00 20.\.V A_ 250.5 00 20.0%.; No 3.0.5
wmaw 0 mm 2o.0.\.v 0A 3.0.5 N 32.5 AA 2N.0.x.v 0_ 20.A.\.V Am 20.0.00 Aw 2N.u.\.v 3 3.0.5 5 3.0V
55—. 0 mm 20.0.5 A0 2A.\.v 00 20.0.00 00 209x; A0 2N.0.\.V A... 2N.w.\.v .3 3.0.5 .0 3.0.x; E 2.0.x;
55.. q 02 2A.0.\.v A0 2&5 0m 20.0.5 00 20. 2.5 um 2o.0.\.v 0A 3.0.x; Nu 3.0.5 _A AA.\.V I 2.2.x;
55w 0 0m 20.A.\.v A0 2u.A.\.v .3 ANo.\.v 0N 2A.0.\.V .3: 3.0.00 00 20.5 N_ A0.\.v 0 2.A.\.V m 3.0.x;
wmsw 0 0N 2A.0.\.V 00 2m.0.\.V 00 Cm.0.\.v Aw 2N.w.§ 0A 6.0.5 3 Q.0.\.~ _0 2.0.00 0 20:0 NN EQXW
A we: we: — m2: N m2: 0 was. A 0325::

A: .1- N0?

was—A _ N 3.5 A 20.5 _ AA.\.V 0 60.5 _A 80.5

wma. N u 2N.0v N 3.5 A 2 0.x; 5 AAmXV A 20.\.v

55w 0 A 20.\.V S 30.5 A 20.5 0 30.x; _ 3.x;

”8:. A A 2N.\.V 0 ANA.\.V : AAA.\..V o 8.x; A 20.5

Was—A 0 .N AAm.® w 2N.\.V 0 WNoﬁ-o ... 2N.\.V N Am.\.v

 

169

Hugo 0-0

biigzo: o\>\o=.§:.o§~ hmbkﬁ. 533.3%“ 8. Mmﬁ

«w -

’

.21.

‘
.EI

 

 

 

 

a war» m2: _ m2: N m2: u m2: A we: 0 m2: 0 mg: q m2: a 086th
A: u 9.3

55w 2 N0 SAX; No 39.5 3 3.0.5 3 3.0.5 3 3on Nu 8.0.5 00 2o.u..\..v 0N 2Ab..\..v Eu A395
55w N N0 3&5 wm $2.5 MN A095 no $.va N.» 39.x; A0 2~.o..\..v mm 20. 15 3 A8. 15 um ANNon
”war 0 AA 2 29.5 um GAR; 3 CAR; mu 39.x; Ac 2&5 00 2c.u..\..v 0m 2335 00 20.5 3 3&5
”man A A0 2 _.A..\..V Aw 2~.w..\..v N0 3&5 A 3.15 3 GAOE 0A 20.A..\..V S 2A.0..\..v um 290.5 N0 32.x;
Wan—A 0 .2 3.0.5 Aw 2~.w..\..v u» G. _..\..v Au 2w.A..\..V 0_ 2S&x.v A0 2w._.§ 3 8.0.x; 0A 89.5 mm A095
55w 0 3 2 _._..\..v .3 3.on Ac 2&5 A2 2 fax; 0N 2A.o..\..v A0 2~.o..\..v Ac 2 _.A..\..v .3 8.0.5 _0 3.0.x;
was... A A0 20. 2.5 A0 2N.o..\°v 0m 20. :5 00 20.qu Ac 2&5 33,15 NM A095 00 3.0.5 o 3.0.x;
Wmsw m 0o 2A.w..\..v A0 2w.2..\..v 02 2a.&.\..v 0N AZQXV .2 3.0.5 3 3.0.x; ”WA 3.35 _0 3.0.x; m 39.3
55w 0 02 2A.0..\..v 0.» 209.0? .2 690.5 Au 2Nw§v 0. 3.0.5 00 2o.~..\.b NA 3.5 5 pAoAb 3 300$
A as: m2: _ m2: N m2: u mo»: A 0325::

A: u NMV

55w 2 2 3.5 A 20.5 _ 3.x; 0 2&5 .0 8&5

EBA. N 0 ANASV q Amman; A 20.5 0 3&5 m 80E

”war 0 N 3.5 0 3&5 0 Soﬁe o 20$; w 295

Wmaﬂ A 0 CAR; A 20.5 I AAAXV N 3.5 N 20.5

szr 0 3 on§ A 20.5 A 200$ 0 Bock; N 3.x;

 

170

A020 00

 

00020: 30008. 3050:. 0:0. 3.53.0.0 000003005 0x30030033 b030mwaﬁrmo 60.03303... 0:0. 0000.03.35:

 

 

2.007%” 0007mm m0=umn m0_m.mn m<w_. om m<m_. om m<w_. 00 m<w_. 00
001.. 2.09... .001.. 2.09.. 2.01.. 309.. $2.001. m0???”
r0002 r0002 r0009. r0009 r0003 2.00000 r0002 r0002
W0§:m mx00a0:o0 .31... .3 .51... .20.... -.0A -._N.. .OA .5.
3.008 0000601033
W025m 23003500 .31... ._N.. .NN..1. .NN..1. .N_...... 201... .001... .N0......
8 :0. m" 0: 8 .0 <23
w0no0m<00 >053. .NN1... 20...... No.1... No.1... .2... .3 .wN..1. .N01...
8 :0: m: 0.: 8 .o <03;
M0000: w0m. NW -.N01... -201... -.31. .00 ..8 ..0A -.N_..1. -.om
ABSESV
P000013003500 .001... .31... 001... .001... -.S -.oA .Nm...... .001...
A0 :0: m: m: 8 5 <3.
220:0 mag—03:6 -20.... ..00 -. .0... -.00 -.8 .00 -.00 -.00
2um::. Nuvmamr un:0:0v
0.5.0:: M00" .001... .A01... .31... .Nm...... -.00 ..0A .20.. 2A....
2 530m: 3
958: m0»: .51... ._N.. .00 ._ 1. -20.. -.oo .8 .0A
2 u 0.00. N u :8
r0300 9801030 -.Nw..1. -.N01... -. _01. -. G1. .3 .00”. -. $1. -. 3......
2n 0.00. N u :3
.360 00mx0010:00 ..o .51. .3 ._N_.. .8 .8 20.. .31.
2umvoPNumoroorwn009v
0.50:. H08: 0003:: -.0 .1... ..N0...... -.No..1. -. .01. .oo .oo -. 3.... -. 5....
2 u 0.00. N n :3
F0301: 0002 -2001... $01... -.A01... ..31... .oo .00 ..N0...... J21...
2 u «00. N n :8

 

20.0. 0007mm 001.. 2.0000: u :03 M020 00102380000000 mm.0o:<0:0mmw 20007mn 7.209... .0000: n :02 m00~0 on Zo:<:00::_-r0000~ mm000<0:0mm.. M02000
001,. P0002 u M0: m00~0 0010~B§00F0000~ mm000<0:0mm.. M0=nmn 309.. 0800a n m0: m00~0 30330:». r0002 mm.000<0:0mm.. 9.0.. on 001. P0009 u
90—0000: 00 90 3 £5280 0010:3500 2.00000. mag. 0.. $2.003. r0002 n m<m_:0:0: 0», m0: 00100803000009 >059 m<w_. 0». $2250: ...0003 u

020—0030: 0», m0: Zoa<mao:m_-r0000~ >035»

.78. 1.02.. 1.3.8.

171

0.00.0 0-3

 

00000.. 5.00500: 3‘ 3300.000. 0000050500. 000330000 @3305... $98.08. @363 000. 3550.00

 

 

 

 

 

2.8.... x0 .2. x00.» 0... >0... 2:00». 22 25.. 0....

000030.. 00.50.0000

560900.000 MN. 00 E 030 .00 .00 .00 .00 .3

:05 30»: m5 283.0000 W028: 0010050000 F0002. 3005.000»— r0m00n macs—.00 3:510
0020—000:

_ 0.00 0.000 -008 0.08 00.000 8 _.0 .00

N 0.0» 0.000 03.5 0.80 0:33.: .3

... 0.00 0.0.0 L93 0.39 39000.00 .00

A 0.00 0.0.0 -003 0.03 38080.00 .0w

0 0.00 0.0.3 0300 0.30 00.000 8 _.0 .00

0 0.00 0000 00.3 0.000 3050390 .00

3 0.00 0.00m 0.30 -0000 005.300 :2

 

2000. 02 .1. 00000000 00 En >9: n >000300 00000000 00 30 Exam.» n W00" 7000: $030 mica 0»- >00~0§B~0000 23 u
203.00 0: 300.2 223 u 29.-203.00 m: 500% 03 H 00500808 m: 300%

172

szo O-m

 

338$. Nu: $3an. Mﬁznbiﬁmk $9.8:me muNNSRmu. @83ng Mag: 33 $595,.

 

 

 

=2: 33: 3v midis—nun 5.18%... raga—3:: 532.8. mo: H333 02.896 man
H3252

9m _ 9.3 98 .93 9Nm 8838 8596a
9m N Nam 93 -93 9am mom—8 8 9o .NN
9m w .93 9NN .9: 93 .39 Z\9u .3
9m a 93 93 -9Nm 93 NZ. G93 .3
9m u .93 93 .93 9.5 63 598 .3
9m a q. E 93 .93 93 .39 39cm .3
9m q .98 9d 9.3 9uN 33933.3 ..NN
9m x Nae 9mo .9mm 9mN .mN\. 33.3 .3
9m o «Nu N3 .93 .92 .92. $93. .3
9m 5 93 93 .93 93 63.393 ..8
9m Z .93 93 .93 93 .3939; .3
9m _N 8% 9.: .93 93 92.53: .3
9m 5 98 NS 93 .93 9355.8 .3
9m I .93 Nam 93 -98 133993 .3
9m 3 93 93 Now 9.8 9039.8 3338
mm _ 9NN 93 -93 NE meann— 8 9o .3
mm N «.3 9.8 .93 93 933298 .mN
mm w 93 93 .93 9NN .3932? 5 .8
mm a 93 93 -93 93 92:38 8896a
mm m 93 93 -93 99. .3252. _N .3
mm a 93 9o... -93 93 .3\.om\3.3 .3
mm q 93 93 -93 .93 3393\qu .mm
mm _ 9: 9m.» .93 TN mow—8 8 9o .3
nm N SNm 9d .93 93 .3\.om:9.& .3
nm w «.3 9; .98 93 98.3293 .3.
mm a 9d N.Nc -9 G 93 .3339? .3
mm m 93 9.3 -9Na N. : .mm\.om\N9N_ .3
mm o 4.3 93. -95 93 .mo\.om\N9$ .3
0m 4 9NN 99. -93 N.No .mN\.ou:c.om .3

 

>58. 03 n $58393 nomqammwos nonmmomosqmgawa oﬁoSémEﬁ mZOM wagon 3:520 mono—£8:

173

 

 

 

 

 

 

I! flittinp
.388 0.0
A -286 >1 a #388. Mﬁg‘aiﬁma $39333 NESQRM. MRSRQE‘ MxSx. 85‘ SLSSSW
~83 325 mu 261:2. mac—8:. $2.83...qu 78:53 won—Von: 23:76 man
Mann—:8: €892.85.— 685525323 .5828
1812.338
_ #3 Cd .93 -93 wow—on 8 ro .3
u 5.3 ca.» -..oo o.om 0:338 .3.
m #3 cam -..2 Wow .£\._w\m.oq .3
a #3 of -Nuw ado .58: :93 .2
e a?» 0.3 ._._o owe 08.838 .3
5 2: :w 6.2 5.3 wgoswa -..
8 an: cum -_ .om ..om 85.358 .2
5 Wow SS .93 com .AQNQMQQ .N.
3 NS ..8 o. E 8.3 $838 ...
um Wow I: ATE .oo.“ 99.353 .8
32738.;—
» PB ago .93 ATE mom—ma 8 _.o .3
q awe 0.: 5.3 boo .meoRuo .X
8 Wow $8 .93 5.3 .otbﬁQoN .wm
5 mum _.o~ .93 -om: .mecauo :3
I #5 9.3 .o..E o._o .ouxpwabo .I
E #3 o.mo .o.& .93 .o_\.Nw\o.o~ .wm
Nu :3 _.o_ .93 o.oo .w_\.-\n.oo ..o
wan—8n.
a Ans o?» 9.3 Wow mom—8 8 re .3
3 58m ouo -24 on. .mm\._o\m.~q .3
3 5.3 ego -..3 93 .mQ._w\m.oo mo
NA 5.3 0.3 L .8 Now .88.. :93 .3
Ma #3 o.ow ...54 Now 083223 .Nw
28:33
3 TE emu Nww Mam manna 8 _.o .3
2 Now o.om 98 Com .m. 53.3 .:
N: 73 ego N3 93 .mw\._wh.o~ ..S
Mu Nwo row one -93 $2533

 

>88. 03 n mesa—manna 89038: ogaomosqmgnma oﬁoSémEow man" magma 28:8.» mono—£8:

174

Had—a 0-5

 

‘I
V:

3.. ~51th Damnzwacm hﬁaaae. Mﬁgmﬁmm ~633me NESQRQ. .m‘nagi $.33 Si NLSFma

 

 

 

 

HESS 33: m5 981.5% K548: Mme H8: Fax—2. mZ—O
EGO

~ .98 :N -93 -9mo maﬁa 8 _.o .Ao
w Nmm Z: 93 -93 69.8298 .3
m NS _.oo 93 -93 .o:.om:o.mu .3
u fmm 9% 9cm 9%» 09.333 .8
E .93 :o -93 -93 .dxerubw .mo
3 N: ..Nm 9% -93 9:5: 88 .3
E Abm 93 Pow 9.3 5:33.: .3
ﬂaw—A

N 93 93 -..ou 9cm mow—8 8 8o .3
m ace 9% -obo Bow .Aw\.$\o.mo .3
o #3 9mm L .2 93 6Q. 36.: .3.
5 9mm 9mm -Zo 9mg .3:o\9oo so
5 ago o.oo -93 -9: .mc\._N\m.om .uA
a krow 9mm -93 -93 .mwinhwm .3
3 9mm 93 Law NE 6:538 am
no 5.3 93 -foq 93 .mo\.oo\q.om .8
F232.

a .93 93 -93 .98 8838 ---
m Nae :5 93A -99. moan; 8 fo .3
I A. G 9.5 -93 98 8838 ---
E w.~m #2 .93 -93 goisowo .3
3 NS Boo 903 -93. .m_\.oo\9w_ .wm

 

>58. 02 n m§§a§8 nomaomaoa oooaomoszmggq onoaﬁéacow min" mag—.8 3:520 mono—mac:

175

.320 O; _

 

 

>\\$>\\Q.w©-m 3‘ Sﬁnwu. Mﬁzkbiﬁau ~633me @3383. $8835 Miss. 33 3.533

 

 

 

 

 

:82 x” 3. 8m >2. 02 >02 23mm.» 2: a: 3.2: a:
?FEKGMQL
mvﬁoﬁrommuaa 393 EN .5 .3 .3 .oo .3
3258 some 3 393.: me E .3 .2 .oo .8
~83 73:83. wold—.338 Fauna—«Eu man

_ 838 8 Ho pm

N mom—on 8 _.o .3

w 8338 ---

A .mﬁwﬁogo .wm

m 8333 ---

o ”0338 ---

q .omxmdQS .wN

w .mihﬁmgo .3

o .ﬂtwﬁogo ON

3 8338 -..-

Z .oQQdodo mm

5 8305a ---

E 8333 ---

3 0:38.05 .No

3 .qQNQQoo .3

a 8333 ---

3. .oﬁmﬁoxﬁw .wm

E 98.353 No

 

22m. 03 M 002303 om 3: >03 n 3,8333 0083mm 0». mm: Em?» n Woo“ 3am: macaw mica om 3333330? 23 n
2238 m: Fgaxw 222 H 23-23.33 m: Sign OE N 0033335 m: :3me 933338 Homage: 8039338333 333.

<38.

176

REFERENCES

177

 

 

 

Ames, C. (1984). Achievement attributions and self-instructions under competitive and
individualistic goal structures. Journal of Educational Psychology, 76, 478-487.

Ames, C. (1990). Motivation: What teachers need to know. Teachers College Record, 91,
409-421.

Ames, C. (1992-a). Achievement goals, motivational climate, and motivational processes.
In G. Roberts (Ed), Motivation in Sport and Exercise. (pp. 161-176). Champaign
IL: Human Kinetics.

Ames, C. (1992-b). Achievement goals and classroom motivational climate. In J. Meece
& D. Schunk (Eds), Students ’ perceptions in the classroom (pp. 327-348).
Hillsdale, NJ: Erlbaum.

Ames, C. (1992-c). Classrooms: Goals, structures, and student motivation. Journal of
Educational Psychology, 84, 261 -27 1 .

Ames, C., & Ames, R. (1981). Competitive versus individualistic goal structures: The
salience of past performance information for causal attributions and affect.
Journal of Educational Psychology, 73, 411-418.

Ames, C. & Archer, J. (1988). Achievement goals in the classroom: Students’ learning

strategies and motivation processes. Journal of Educational Psychology, 80, 260-
267.

Atkinson, J .W. (1964). An Introduction to Motivation. Princeton: Van Nostrand.

Bandura, A. (1986). Social Foundations of Yhought and Action: A Social Cognitive
Theory. Englewood Cliffs, N.J.: Prentice-Hall.

Bandura, A. (1990). Perceived self-efﬁcacy in the exercise of personal agency. Journal of
Applied Sport Psychology, 2, 128-163.

Bandura, A. (1997). Self-eﬂicacy: The exercise of control. New York: W.H. Freeman.

Bandura, A. (2001a). Social cognitive theory: An agentic perspective. Annual Review of
Psychology, 52, 1-26.

Bandura, A. (2001b). Exercise of human agency through collective efﬁcacy. Current
Directions in Psychological Science, 9, 75-78.

Bandura, A., & Wood, RE. (1989). Effect of perceived controllability and performance

standards on self-regulation of complex decision-making. Journal of Personality
and Social Psychology, 56, 805-814.

178

Bass, BM. (1990). Bass and Stogdill ’s handbook of leadership: A survey of theory and
Research (3rd ed.). New York: Free Press.

Bern, S. L. (1974). The measurement of psychological androgyny. Journal of Consulting
and Clinical Psychology, 42, 155-162.

Biernat, M., Crandall, C.S., Young, L.V., Kobrynowicz, D., & Halpin, SM. (1998). All
that you can be: Stereotyping of self and others in a military context. Journal of
Personality and Social Psychology, 75, 301-317.

Blumenfeld, PC. (1992). Classroom learning and motivation: Clarifying and expanding
goal theory. Journal of Educational psychology, 84, 272-281.

Bredemeier, B. (1992). “And ain’t I a woman?” Toward a multicultural approach to
gender and morality. Quest, 44, 179-209.

Bussey, K., & Bandura, A. (1999). Social cognitive theory of gender development and
differentiation. Psychological Review, 106, 676-713.

 

Carli, L.L., & Eagly, A. H. (2001). Gender, hierarchy, and leadership: An Introduction.
Journal of Social Issues, 57, 629-636.

Chan, K.Y., & Drasgow, F. (2001). Toward a theory of individual differences and
leadership: Understanding the motivation to lead. Journal of Applied Psychology,
86, 481-498.

Chelladurai, P. (1984). Discrepancy between preferences and perceptions of leadership
behavior and satisfaction of athletes in varying sports. Journal of Sport and
Exercise Psychology, 6, 27-41.

Chelladurai, P., & Reimer, HA. (1998). Measurement of leadership in sport. In J.L. Duda
(Ed), Advances in Sport and Exercise Psychologz Measurement (pp. 227-256). 1
Morgantown, WV: FIT Press.

Chemers, M.M. (1997). An Integrative Iheory of Leadership. Mahwah, NJ: Lawrence
Erlbaum Associates, Inc., Publishers.

Chemers, M. M. (2000). Leadership research and theory: A functional integration. Group
Dynamics: Theory, Research, and Practice, 4(1), 27-43.

Chemers, M. M., Watson, C.B., & May, ST. (2000). Dispositional affect and leadership
effectiveness: A comparison of self-esteem, optimism, and eﬂicacy. Personality
and Social Psychology Bulletin, 26(3), 267-277.

Cogliser, C.C., & Schriesheim, CA. (2000). Exploring work unit context and leader-
member exchange: A multi-level perspective. Journal of Organizational
Behavior, 21, 487-511.

179

 

Connelly, M. S., Gilbert, J .A., Zaccaro, S.J., Threlfall, K.V., Marks, M., & Mumford,
MD. (2000). Exploring the relationship of leadership skills and knowledge to
leader performance. Leadership Quarterly, 11(1), 65-86.

Duda, J. L. (2001). Goal perspective research in sport: Pushing the boundaries and
clarifying some misunderstandings. In I. G. C. Roberts (Ed), Advances in
motivation in sport and exercise (pp. 129-182). Champaign, E: Human Kinetics.

Duda, J. L., & Balaguer, I. (1999). Toward an integration of models of leadership with a
contemporary theory of motivation (pp. 212-230). In R. Lidor & M. Bar-Eli
(Eds). Sport Psychology: Linking Theory to Practice.

Duda, J.L., & Whitehead, J. (1998) Measurement of goal perspectives in the physical
domain. In J .L. Duda (Ed), Advances in Sport and Exercise Psychology
Measurement (pp. 21-48 ). Morgantown, WV: FIT Press.

Duda, J .L., Olson, L., & Templin, T. (1991). The relationship of task and ego orientation
to sportsmanship attitudes and the perceived legitimacy of injurious acts.
Research Quarterly for Exercise and Sport, 62, 79-87.

Dweck. C.S., & Leggett, E. (1988). A social-cognitive approach to motivation and to
personality. Psychological Review, 95, 256-273.

Eagly, AH. (1987). Sex diﬂerences in social behavior: A social-role interpretation.
Hillsdale, NJ: Earlbaum.

Eagly, A.H., & Johannesen-Schmidt, MC. (2001). The leadership styles of women and
men. Journal of Social Issues, 5 7, 781-797.

Eagly, A.H., & Johnson, ET. (1990). Gender and leadership style: A meta-analysis.
Psychological Bulletin, 90, l-20.

Eagly, A.H., & Karau, SJ. (1991). Gender and the emergence of leaders: A meta-
analysis. Journal of Personality and Social Psychology, 50 (5), 68 5-71 0.

Eagly, A.H., Makhijani, M.G., & Klonsky, B.G. (1992). Gender and the evaluation of
leaders: A meta-analysis. Psychological Bulletin, 111 (1), 3-22.

Embretson, S.E., & Reise, SP. (2000). Item Response Iheory for Psychologists.
Mahwah, NJ: Lawrence Erlbaum Associates, Inc., Publishers.

Feltz, D.L., & Lirgg, CD. (2001). Self-efﬁcacy beliefs of athletes, teams, and coaches. In
R. N. Singer, HA. Hausenblas, & C. M. Janelle (Eds), Handbook of Sport
Psychology (pp. 340-361). New York: John Wiley & Sons.

180

Feidler, FE. (1967). A theory of leader effectiveness. New York: McGraw-Hill.

Gano-Overway, L.A., Guivernau, M., Magyar, T.M., Waldron, J .J ., & Ewing, ME.
(2001). Helping One Another Maintain the Spirit of the Game: Understanding
Females’ Perspectives on Sportspersonship and Developing Strategies for
Encouraging Good Sport Conduct. Research Quarterly for Exercise and Sport,
72(1), S84.

Glenn, S. D., & Horn, T. S. (1993). Psychological and personal predictors of leadership
behavior in female soccer athletes. Journal of Applied Sport Psychology, 5, 17-34.

Gill, D.L. (1994). Psychological perspectives on women in sport and exercise. In Costa
and Guthrie (Eds). Women and Sport (pp. 253-284). Champaign, IL: Human
Kinetics.

Griffin, M.A., & Mathieu, J .E. (1997). Modeling organizational processes across
hierarchical levels: Climate, leadership, and group process in work groups.
Journal of Organizational Behavior, 18, 731-744.

Grusky, O. (1963). The effects of formal structure on managerial recruitment: A study of
baseball organization. Sociometry, 26, 345-353.

Hemphill, J.K. & Coons, AH. (1957). Development of the leader behavior description
questionnaire. In R.M. Stogdill & A.E. Coons (Eds), Leader behavior: Its
description and measurement. Columbus: Ohio State University, Bureau of
Business Research.

Hogg, MA. (2001). A social identity theory of leadership. Personality and Social
Psychology Review, 5, 184-200.

House, R.J. (1971). A path-goal theory of leadership. Administrative Science Quarterly,
16, 321-338.

House, R.J. (1996). Path-goal theory of leadership: Lessons, legacy, and a reformulated
theory. Leadership Quarterly, 7, 323-352.

House, R.J. & Mitchell, TR. (1974). Path-goal theory of leadership. Journal of
Contemporary Business, 3, 81-98.

House, R.J. & Shamir, B. (1993). Integrating transformational, charismatic, and visionary
theories. In M. M. Chemers & R. Ayman (Eds), Leadership theory and research:
Perspectives and directions (88-108). San Diego, CA: Academic Press.

Hughes, R.L., Ginnett, R.C., & Curphey, G]. (1993). Leadership: Enhancing the lessons
of experience. Homewood, IL: Irwin.

181

James, L. R. (1982). Aggregation bias in estimates of perceptual agreement. Journal of
Applied Psychology, 67(2), 219-229.

Jarvinen, D.W., & Nicholls, JG. (1996). Adolescents' social goals, beliefs about the
causes of social success, and satisfaction in peer relations. Developmental
Psychology, 32(3), 43 5-441.

Joreskog, K., & Sorbom, D. (1996). LISREL 8: User’s Reference Guide. Chicago
Scientiﬁc Software Int.

Kavussanu, M., & Roberts, G. C. (2001). Moral functioning in sport: An achievement
goal perspective. Journal of Sport and Exercise Psycholog, 23, 37-54.

Kenny D. A., & Zaccaro, SJ. (1983). An estimate of variance due to traits in leadership.
Journal of Applied Psychology, 68, 678-685.

Kozlowski, S. W., & Doherty, ML. (1989). Integration of climate and leadership
examination of a neglected issue. Journal of Applied Psychology, 74(4), 546-553.

Krane, V. (2001). We can be athletic and feminine, but do we want to? Challenging
Hegemonic femininity in women’s sport. Quest, 53, 115-133.

Larson, R. W. (2000). Toward a psychology of positive youth development. American
Psychologist, 55(1) , 170-183.

Maehr, M.L., & Braskamp, LA. (1986). The motivation factor: A theory of personal
investment. Lexington, Mass: Lexington Books/Heath.

Magyar, T.M., & Feltz, D.L. (in press). The inﬂuence of dispositional and situational
tendencies on adolescent girls’ selection of sport conﬁdence sources. Psychology
of Sport and Exercise.

Magyar, T.M., Feltz, D.L., & Simpson, LP. (2002). Personal and situational determinants

of individual and collective efficacy in adolescent rowers. Research Quarterly for
Exercise and Sport, 73(1), S93.

McClelland, D.C. & Boyatzis, RE. (1982). Leadership motive pattern and long-term
success in management. Journal of Applied Psychology, 67, 73 7-743.

Moritz, S.E., & Watson, CB. (1998). Levels of analysis issues in group psychology:
Using efﬁcacy as an example of a multilevel model. Group Dynamics: Iheory,
Research, and Practice, 2, 285-298.

Mumford, M. D., Zaccaro, S.J., Johnson, J.F., Diana, M., Gilbert, J.A., & Threlfall, K.V.

(2000a). Patterns of leader characteristics: Implications for performance and
development. Leadership Quarterly, 11(1), 115-133.

182

Mumford, M.D., Marks, M.A., Connelly, M.S., Zacarro, S.J., & Reiter-Palmon, R.
(2000b). Development of leadership skills: Experience and timing. Leadership
Quarterly, 11 (1), 87-1 14.

Murphy, SE. (2002). Leader self-regulation: The role of self-efficacy and multiple
intelligences. In Ronald E. Riggio, Susan E. Murphy, & Francis J. Pirozollo
(Eds). Multiple Intelligences and Leadership.

Newton, M. & Duda, J .L. (1999). The interaction of motivational climate, dispositional
goal orientations, and perceived ability in predicting indices of motivation.
International Journal of Sport Psychology, 30, 63-82.

Newton, M., Duda, J .L., & Yin, Z. (2000). Examination of the psychometric properties of
the Perceived Motivational Climate in Sport Questionniare—2 in a sample of
female athletes. Journal of Sport Sciences, 18, 275-290.

Nicholls, J. G. (1984). Achievement motivation: Conceptions of ability, subjective
experience, task choice, and performance. Psychological Review, 91(3), 328-346.

Nicholls, JG. (1989). Ihe competitive ethos and democratic education. Cambridge, MA:
Harvard University Press.

Nicholls, J .G. (1992 ) The general and the specific in the development and expression of
achievement motivation. In G.C. Roberts (Ed), Motivation in sport and exercise.
(pp. 31-55). Champaign IL: Human Kinetics.

Northouse, PG. (2001). Leadership Theory and Practice. Thousand Oaks, CA: Sage
Publications, Inc.

Ommundsen, Y., & Roberts, GO (1999). Effect of motivational climate proﬁles on
motivational indices in team sport. Scandinavian Journal of Medicine and Science
in Sport, 9, 389-397.

Paglis, L.L., & Green, S.G. (2002). Leadership self-efficacy and managers’ motivation
for leading change. Journal of Organizational Behavior, 23, 215-235.

Phillips, J .M. Douthitt, E.A., & Hyland, M. M. (2001). The role of justice in team
member satisfaction with the leader and attachment to the team. Journal of
Applied Psychology, 86, 316-325.

Rees, C.R., & Segal, M.W. (1984). Role differentiation in groups: The relationship
between instrumental and expressive leadership. Small Group Behavior, 15, 109-
123.

Riemer, H. A., & Chelladurai, P. (1995). Leadership and satisfaction in athletics. Journal
of Sport and Exercise Psychology, 1 7, 276-293.

183

Roberts, GO (2001). Understanding the dynamics of motivation in physical activity: The
inﬂuence of achievement goals on motivational processes. In G.C. Roberts (Ed),
Advances in motivation in sport and exercise (pp. 1-49). Champaign, IL: Human
Kinetics.

Roberts, G.C. & Treasure, D.C. ( 1995). Achievement goals, motivational climate and
achievement strategies and behaviors in sport. International Journal of Sport
Psychology, 26, 64-80.

Seifriz, J ., Duda, J .L., & Chi, L. (1992). The relationship of perceived motivational
climate to intrinsic motivation and beliefs about success in basketball. Journal of
Sport and Exercise Psychology, 14, 375-391. ,

Smith, J .A., & Foti, R. J. (1998). A pattern approach to the study of leader emergence.
Leadership Quarterly, 9, 147-160.

 

Smoll, F.R., & Smith, RE. (1989). Leadership behaviors in sport: A theoretical model
and research paradigm. Journal of Applied Social Psychology, 19(18) , 1522-
1551.

Sorrentino, RM. (1973). An extension of theory of achievement motivation to the study
of emergent leadership. Journal of Personality and Social Psychology, 26, 356-
368.

Sorrentino, R.M., & Field, N. (1986). Emergent leadership over time: The functional
value of positive motivation. Journal of Personality and Social Psychology, 50,
1091-1099.

Stodgill, RM. (1948). Personal factors associated with leadership: A survey of the
literature. Journal of Psychology, 25, 35-71.

Stodgill, RM. (1974). Handbook of leadership: A survey of theory and leadership. New
York Free Press.

 

Stout, W. (1987). A nonparametric approach for assessing latent trait unidimensionality.
Psychometrika, 52, 589-617.

Tabachnik, B.G., & Fidell, LS. (1996). Using multivariate statistics (3rd Ed). New
York, NY: HarperCollins Publishers Inc.

Treasure, D.C., & Roberts, GO (1998). Relationship between female adolescents
achievement goal orientations, perceptions of the motivational climate, belief

about success and sources of satisfaction in basketball. International Journal of
Sport Psychology, 29, 211-230.

Tropp, K., & Landers, D. (1979). Team interaction and the emergence of leadership and
interpersonal attraction in ﬁeld hockey. Journal of Sports Psychology, 1, 228-240.

184

Urdan, T.C., & Maehr, ML. (1995). Beyond a two-goal theory of motivation and
achievement: A case for social goals. Review of Educational Research, 65(3),
213-243.

Vos Strache, C. (1979). Players’ perceptions of leadership qualities for coaches. Research
Quarterly, 50, 679-686.

Vroom V.H. & Yetton PW. (1973). Leadership and decision-making. Pittsburgh:
University of Pittsburgh Press.

Walling, M.D., Duda, J .L., & Chi, L. (1993). The perceived motivational climate in sport

questionnaire: Construct and predictive validity. Journal of Sport and Exercise
Psychology, 15, 172-183.

Watson, Chemers, & Preiser. (2001). Collective efﬁcacy: A multilevel analysis.
Personality and Social Psychology Bulletin, 2 7, 1057- 1 068.

Wood, RE, & Bandura, A. (1989). Impact of conceptions of ability on self-regulatory
mechanisms and complex decision-making. Journal of Personality and Social
Psychology, 56, 407-415.

Yoder, J .D. (2001). Making leadership work more effectively for women. Journal of
Social Issues, 5 7, 815-828.

Yukelson, D., Weinberg, R., Richardson, P., & Jackson, A. (1981). Interpersonal
attraction and leadership within collegiate sport teams. Journal of Sport Behavior,
6, 29-36.

Zaccaro, S. J ., Foti, R.J., & Kenny, DA. (1991). Self-monitoring and trait-based variance
in leadership an investigation of leader ﬂexibility across multiple group situations.
Journal of Applied Psychology, 76(2), 308-315.

185