Leadership behaviors among deans of students at public research universities in the Southeast

MISSING IMAGE

Material Information

Title:
Leadership behaviors among deans of students at public research universities in the Southeast
Physical Description:
viii, 115 leaves : ill. ; 29 cm.
Language:
English
Creator:
Barth, Richard A
Publication Date:

Subjects

Subjects / Keywords:
Educational Leadership, Policy,and Foundations thesis, Ph. D
Dissertations, Academic -- Educational Leadership, Policy, and Foundations -- UF
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 2004.
Bibliography:
Includes bibliographical references.
Additional Physical Form:
Also available online.
Statement of Responsibility:
by Richard A. Barth.
General Note:
Printout.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 022435035
oclc - 847495555
System ID:
AA00025714:00001

Full Text











LEADERSHIP BEHAVIORS AMONG DEANS OF STUDENTS AT PUBLIC
RESEARCH UNIVERSITIES IN THE SOUTHEAST

















By

RICHARD A. BARTH


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLM[IENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2004















ACKNOWLEDGEMENTS

I wish to acknowledge my dissertation committee for their assistance during the

dissertation process. Dr. Arthur Sandeen, my advisor, sustained me in this effort,

showing patience and understanding above the call of duty. Dr. Wayne Griffin, my

outside committee member, consistently reminded me of the importance of maintaining

balance in my life while working on the dissertation and provided a tremendous amount

of emotional support throughout the long process. Dr. David Honeyman and Dr. Lamont

Flowers were instrumental in assisting me with deciding on the dissertation topic and

helping me design the study.

I also wish to acknowledge Dr. Mike Rollo, Dr. Julie Athman, and Dr. Mike

Mironack. As colleagues and friends they kept me focused on the dissertation and

provided me with the advice and motivation needed to complete the journey.

A special note of thanks goes to Ms. Evelyn Chiang, who shared her time, talent,

and energy to aid me as I completed the dissertation. I truly appreciate the time she spent

teaching me about statistics and analyzing my data. Most of all, I appreciate her

unwavering support and friendship over the past two years.

Finally, I would like to express my gratitude to my parents, Alvin and Dolores

Barth. They made many sacrifices throughout their lives to allow me to pursue my

educational goals. Through their hard work and commitment to education, they have

provided me with opportunities that they never had themselves. I am very fortunate to

have them as my parents.


ii














TABLE OF CONTENTS
page

ACKNOWLEDGFMIENTS......................................................................... ii

LIST O F TA B LES.................................................................................... v

A B STRA C T .......................... ............................................... .... ............ vii

CHAPTER

I IN TR O D U C TIO N .............................................................................. 1

Statement of the Problem ..................................................................... 6
Purpose of the Study.......................................................................... 8
Theoretical Background ...................................................................... 9
D definition of Term s........................................................................... 13
Delimitations and Assumptions............................................................ 18
Organization of the Study................................................................... 19

2 REVIEW OF THE LITERATURE ........................................................ 20

Definitions of Leadership ................................................................... 21
Theories of Leadership ....................................................................... 22
The Full Range of Leadership Model...................................................... 29
Research on Transformational Leadership................................................ 35
Contingencies and Limitations of the Full Range of Leadership Model............. 36
The Dean of Students........................................................................ 37

3 METHODOLOGY ........................................................................... 42

Research Population.......................................................................... 43
The Instrument ...................... ....................................... .................. 44
Reliability and Validity ..................................................................... 46
D ata C collection .................................. ..................................... ... .... 49
D ata A analysis ......................................... ................. ....................... 50
H um an Subjects ......................................... ............ ...... .................... 51

4 ANALYSIS AND PRESENTATION OF DATA....................................... 52

Survey R esponses............................................................................ 53








Response Rates.............................................. ... ........ ...................... 54
Demographic Information................................................................... 54
Research Question 1 ......................................................................... 55
R research Q question 2.......................................................................... 57
Research Question 3.......................................................................... 61
R research Q question 4.......................................................................... 65
R research Q question 5.......................................................................... 70
Sum m ary......................................... ............... ................. ............... 72

5 SUMMARY AND DISCUSSION......................................................... 74

Summary and Discussion of Findings...................................................... 75
Leadership Behaviors Exhibited by Deans of Students................................. 77
Leader Effectiveness as Measured by the
Multifactor Leadership Questionnaire................................................... 79
Satisfaction with the Leader as Measured by
the Multifactor Leadership Questionnaire............................................. 81
Willingness to Exert Extra Effort as Measured by
the Multifactor Leadership Questionnaire............................................ 84
Gender and Perception of Leadership Style............................................... 86
Implications for Student Affairs............................................................ 87
Recommendations for Future Research................................................... 89

APPENDIX

A INSTITUTIONAL REVIEW BOARD APROVAL.................................... 91

B MULTIFACTOR LEADERSHIP QUESTIONNAIRE PERMISSION ............. 93

C INFORMATION LETTER FOR DEANS OF STUDENTS........................... 95

D INFORMATION LETTER FOR PARTICIPANTS..................................... 97

E INFORMED CONSENT FORM........................................................... 99

F DEMOGRAPHIC QUESTIONNAIRE FOR PARTICIPANTS..................... 101

G MULTIFACTOR LEADERSHIP QUESTIONNAIRE SAMPLE ITEMS......... 103

REFERENCE LIST............................................................................... 105

BIOGRAPHICAL SKETCH..................................................................... 115














LIST OF TABLES


Table page

3-1 Organizational Samples Used in Validation and Cross
Validation Analysis of the Multifactor Leadership Questionnaire..................... 47

3-2 Means, Standard Deviations, and Intercorrelations Among
M LQ Factor Scores........................................................................... 48

3-3 Comparison of Overall Fit Measures Among Several Factors Model................. 49

4-1 Summary of Respondents and Deans' Age Distribution................................ 55

4-2 Summary of Educational Level for Deans and Respondents........................... 55

4-3 Deans of Students: Leadership Behaviors................................................. 57

4-4 Summary of Regression Analysis for Variables Predicting
Leader Effectiveness......................................................................... 59

4-5 Summary of Full and Reduced Models Regression for Leader
Effectiveness Adding Transformational Leadership Behaviors........................ 61

4-6 Best Predictor Model by Number of Behaviors for Leader Effectiveness............ 61

4-7 Summary of Regression Analysis for Variables Predicting
Satisfaction with the Leader................................................................. 63

4-8 Summary of Full and Reduced Models Regression for Satisfaction
with the Leader Adding Transformational Leadership Behaviors..................... 65

4-9 Best Predictor Model by Number of Behaviors for Satisfaction
w ith the Leader...................................... ......... ...... ...... ..................... 65

4-10 Summary of Regression Analysis for Variables Predicting
W willingness to Exert Extra Effort........................................................... 68

4-11 Summary of Full and Reduced Models Regression for Willingness
to Exert Extra Effort Adding Transformational Leadership Behaviors............... 69








4-12 Best Predictor Model by Number of Behaviors for Willingness to
Exert Extra Effort............................................................................. 69














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Doctor of Philosophy

LEADERSHIP BEHAVIORS AMONG DEANS OF STUDENTS AT PUBLIC
RESEARCH UNIVERSITIES IN THE SOUTHEAST

By

Richard A. Barth

May 2004

Chair: C. Arthur Sandeen
Major Department: Educational Leadership, Policy, and Foundations

This study examined the leadership behaviors of deans of students at public

research universities in the southeast using Bass and Avolio's Full Range of Leadeship

Model. A sample (n = 96) of student affairs professional staff members working within

dean of students offices at 31 public research universities in the southeast completed the

Multifactor Leadership Questionnaire Rater Form (5x-Short) to examine the relationship

between transformational, transactional, and laissez-faire leadership behaviors of deans

of students and the outcome variables of satisfaction with the leader, perception of leader

effectiveness, and followers' willingness to exert extra effort.

SPSS and SAS statistical software programs were used to run multiple linear

regression analyses on the data. Deans of students exhibited transformational leadership

behaviors more frequently than they exhibited transactional behaviors, which they

exhibited more frequently than laissez-faire behavior. The transformational behavior of

idealized influence-attributed, the transactional leadership behavior of contingent reward,








and laissez-faire behavior were significant predictors at the p < .05 level for the outcome

variable of leader effectiveness. The transformational behaviors of idealized influence-

attributed and idealized influence-behavior, the transactional behavior of contingent

reward, and laissez-faire behavior were significant predictors at the p < .05 level for the

outcome variable of satisfaction with the leader. The transformational behaviors of

idealized influence-attributed, idealized influence-behavior, and individual consideration,

along with the transactional behavior of contingent reward and laissez-faire behavior,

were significant predictors at the p < .05 level for the outcome variable of willingness to

exert extra effort. Transformational leadership behaviors accounted for unique variance

in professional staff members' ratings of the outcome variables above that accounted for

by transactional and laissez-faire leadership. The findings support the theoretical

prediction of the Full Range of Leadership model that leaders who are more

transformational and less transactional are more effective as leaders and more satisfying

to their followers. There was no significant difference in how male and female deans of

students were rated overall by their professional staff members and there was no

significant difference in the way male and female professional staff members rated their

deans.














CHAPTER 1
INTRODUCTION

The federal government and the states began showing an interest in distinguishing

between public and private colleges soon after the ratification of the U.S. Constitution

(Rudolph, 1990; Brubacher & Rudy, 1997). Several states established

nondenominational institutions between 1782 and 1820, beginning with Georgia, North

Carolina, Tennessee, and Vermont establishing state-chartered and state-supported

institutions before 1800 (Rudolph, 1990). The early enthusiasm for establishing state

institutions of higher education developed from the public's need for more democratic

and secular institutions that could be held accountable for fulfilling the needs and

objectives of the state (Rudolph, 1990). These initiatives indicated that higher education

was viewed as being essential to the public good and that state governments were

concerned about religiously governed private colleges dictating the national educational

agenda (Rudolph, 1990; Brubacher & Rudy, 1997).

This early push toward development of public higher education lost momentum in

the aftermath of the famous U.S. Supreme Court case, Trustees of Dartmouth College v.

Woodward, which gave privately incorporated colleges control over their own policies

and activities (Rudolph, 1990). Private colleges were created throughout the United

States after the Dartmouth decision and enjoyed unprecedented autonomy (Rudolph,

1990). Rudolph (1990) stated that the Dartmouth decision "discouraged the friends of

strong state-supported and state-controlled institutions;... by encouraging [private]









college funding and by discouraging public support for higher education, [Dartmouth]

probably helped to check the development of state universities for half a ceniun (p.

211).

While the attempts at establishing state institutions of higher education were

premature in terms of public acceptance and ready implementation at the beginning of the

19h century, the last half of the 19th century was a time when the country's industrialized

society was facing increasingly complex problems and deficiencies that would eventually

lead to the widespread development of public higher education (Rudolph, 1990;

Brubacher & Rudy, 1997). There was an increasing need for highly trained professionals

in areas such as engineering, agriculture, public health, forestry, and nursing, but the

professional schools of the modem university did not exist (Bonnen, 1998). There was

also a growing frustration with the perceived unresponsiveness of colleges, mostly

private, that were providing a classical education and were unwilling to address society's

changing needs (Bonnen, 1998). At the same time, a fear arose that the "American

dream" of unlimited opportunities was being threatened by industrialization and the

growing economic inequality it was causing. The lack of access to the skills and practical

education necessary for a better life was viewed as a serious threat to the survival of the

middle class (Brubacher & Rudy, 1997; Bonnen, 1998).

Part of the response to these concerns was the land-grant idea, which was

eventually expressed in the 1862 Morrill Land Grant Act and the second land-grant act of

1890 (Bonnen, 1998). The land-grant idea was to provide federal and state support for

the development of institutions of higher education devoted to science and education in

the service of society by (a) educating and training professionals for the increasingly








urban and industrial society; (b) providing broad access to education regardless of wealth

or social status; and (c) working to improve the welfare and social status of the farmers

and industrial workers (Bonnen, 1998).

The impact of the land-grant legislation was not felt immediately. At the

beginning of the 20'h century public higher education remained largely undeveloped

(Thelin, 1996). However, this began to change after 1900 when state universities

increasingly became a symbol of state pride. State legislators began recognizing that

universities could be of service to the state; therefore, they started supporting them

financially (Lucas, 1994; Thelin, 1996). Thelin (1996) observed that "applied research, a

utilitarian and comprehensive curriculum, not to mention the public appeal of spectator

sports and the availability of federal funds for such fields as agriculture and engineering

led to the growth and maturation of the state university" (p. 12).

During the period between World War I and World War II, the promise of the

Morrill Act began to be seen in the state universities of the West and Mid-West with

enrollments climbing to between fifteen and twenty-five thousand at some institutions

(Thelin, 1996). However, many of the current large state research universities were still

relatively small during this period and their curricular offerings were limited. At the

beginning of World War II, several state universities had enrollments of fewer than five

thousand students and graduate and doctoral programs were limited (Thelin, 1996;

Brubacher & Rudy, 1997).

After World War II, the convergence of the Servicemen's Readjustment Act and

the tremendous increase in government and foundation research grants available to

universities provided the driving force behind the incredible expansion that took place in









higher education at every level (Rudolph, 1990; Brubacher & Rudy, 1997; Caple, 1998).

The period of growth from 1945 to 1970 in enrollments, university influence in society,

graduate and professional programs, and in construction of new state institutions has been

called higher education's "golden age" and a time that ushered in the modem state

research university (Thelin, 1996; Lucas, 1994; Rhodes, 2001). However, the 21st

century state research university, as well as the society it serves, has changed profoundly

from higher education's golden age.

The state university today as compared to 40 years ago is much larger, more

complex, and offers a much wider range of opportunity for disciplinary, or

interdisciplinary specialization (Keller, 1990; Altbach, 2001). Its faculty and student

body are more characterized by involvement in graduate work, research, upper division

and professional education (Balderston, 1995; Rhodes, 2001). Large state research

universities have become national and international in their teaching, research, and some

public service areas (Brubacher & Rudy, 1997; Rhodes, 2001). They serve as the

foundation for the public college and university system that enrolls 78 percent of all

students and 81 percent of undergraduate students (Neimark, 1999).

The modern state research university finds itself in what Altbach (2001)

categorized as a "curious paradox" (p. 11). Along with its private counterpart, the state

research university is part of a system of higher education that is considered the best in

the world (Altbach, 2001). In writing about both the private and the public American

university, Rhodes (2001) stated:

It has been the foundation of growing national economic prosperity and
manufacturing success, vast improvements in the products of agriculture and
industry, and undreamed-of access to new means of communication;... [the
American university] has provided successive generations the opportunity for









meaningful careers, for service in a free society, and for access to the riches of
human experience, aspiration, and achievement;... it has trained the workforce,
enriched the individual experience, . enlightened public life,... quickened the
social conscience and empowered and inspired each rising generation. (p. 1)

Kerr (1991), in writing specifically about the strengths of the American higher education

system that emerged during the 1980s after a twenty-year period of major transformation,

stated:

Higher education met the test of action from 1960 to 1980 overall quite well, and
emerged from this period clearly larger and mostly better. In particular, it was
providing more services to more people in the American society than ever before.
It had, in many ways, been transformed, and, in the process, it had become a more
central aspect of the life of the nation and was, consequently in turn, a greater
potential source of transformation for the nation. (p. 376)

But despite the strengths of the American university and the overwhelming benefits it has

produced for society, it is facing unprecedented criticism (Altbach, 2001).

The public higher education system has been the target of harsh criticism for

being too expensive, inefficient, poorly managed, and for lacking performance criteria

(Neimark, 1999). Specifically, the research university has been consistently criticized for

failing to engage its undergraduate students in the teaching and learning process

(Blimling & Whitt, 1999: National Association of State Universities and Land-Grant

Colleges [NASULGC], 2001). By encountering this criticism at a time when the

landscape of higher education is changing rapidly, the state research university faces a

tremendous challenge in forging a new path to regain public confidence. Strong and

effective leadership at every level of the university is a critical element of meeting this

challenge (Lucas, 2000; Blimling & Whitt, 1999; Boudreau, 1998).









Statement of the Problem

Institutions of higher education currently face a landscape that is changing at an

unprecedented rate (Lucas, 2000; Blimling & Whitt. 1999). Along with the challenge of

this constant change, public institutions find themselves confronting both a decrease in

public confidence and an increase of external criticism over their perceived failure to

actively engage students in the teaching and learning process (Blimling & Whitt, 1999;

NASULGC, 2001). In addressing the criticism questioning the responsiveness and

relevance of public institutions, reports such as Reinventing Undergraduate Education: A

Blueprint for America's Research Universities (The Boyer Commission on Educating

Undergraduates in the Research University [Boyer Commission], 1998) and Returning to

Our Roots: The Student Experience, by the Kellogg Commission (NASULGC, 1997)

emphasize the need for institutions to change the way they engage their undergraduate

student population by making undergraduates and their learning a higher priority.

In calling research universities' record of educating undergraduates one of failure,

the Boyer Commission (1998) stated:

In a context of increasing stress declining governmental support, increased
costs, mounting outside criticism, and growing consumerism from students and
their families universities too often continue to behave with complacency,
indifference, or forgetfulness toward that constituency whose support is vital to
the academic enterprise. Baccalaureate students are the second-class citizens who
are allowed to pay taxes but are barred from voting, the guests at the banquet who
pay their share of the tab but are given leftovers. (p. 37)

This criticism is not new, as scholars and commentators, such as Ernest Boyer and Page

Smith, have called for reform in undergraduate education for many years (Boyer, 1990;

Smith, 1990; Boudreau, 1998). But the criticism has intensified as higher education has









been slow to change and public trust continues to erode (Boudreau, 1998; NASULGC,

1997).

Observers of American higher education have written extensively on the role that

leaders in academic affairs must play in addressing the challenges higher education faces

and implementing the necessary changes. However, these writers address the issues

facing higher education while giving little or no attention to the role student affairs

leaders can or should play in assisting an institution with making undergraduate

education the first priority (e.g., Lucas, 2000; Balderston, 1995; Peterson, Dill, & Mets,

1997). Boudreau (1998) in his book, Universitas: The Social Restructuring ofAmerican

Undergraduate Education, fails to mention student affairs and the role it plays on campus

even when addressing the issue of students' drug and alcohol use impacting the

classroom experience. Surprisingly, the Kellogg Commission on the Future of State and

Land-Grant Universities in its Returning to Our Roots series of reports fails to

specifically or clearly address the critical role student affairs may play in the lives of

students (NASULGC, 1997, 1998, 1999, 2000, 2001).

While some commentators and reform reports have failed to stress the important

role student affairs must play in addressing the changing environment of higher

education, others have clearly recognized this role. Boyer (1987) devotes an entire

section of his book, College: The Undergraduate Experience in America, to life outside

of the classroom and states that the "college of quality remains a place where the

curricular and cocurricular are viewed as having a relationship to each other" (p. 195).

Schroeder (1999) stresses that in responding to the pressure for improved undergraduate








education, academic personnel and student affairs personnel must work together to create

effective learning environments.

For public research universities that employ the dean of students title, the dean of

students is in a key leadership position to assist the institution in creating a seamless

learning environment for undergraduate students and making undergraduate education a

top priority. The dean of students often oversees several of the common functions found

in a student affairs division, holds the "primary educational role within student affairs,"

and "has assumed the rather undefined but significant role of 'conscience of the campus'"

(Sandeen, 1996, p. 444).

A review of the literature in student affairs, including a search of published

dissertations, revealed no empirical studies conducted on the leadership behavior of deans

of students at public research universities. Overall, the contemporary dean of students

has received minimal scholarly attention in the literature (Robillard, 2000). As a result of

the lack of a research base, little is known about the leadership behavior of deans of

students and its relationship to the professional staff members' perceptions of the

effectiveness of this behavior. This presents a significant gap in student affairs research

since the dean of students plays a major role in the student life program (Ambler, 1993)

and is responsible for many of the common student affairs functions (Sandeen, 1996).

Purpose of the Study

The purpose of this study was to examine the leadership behaviors of deans of

students at public research universities in the southeast using Bass and Avolio's (1997)

Full Range of Leadership model. The study investigated the relationship between

transformational, transactional, and laissez-fair leadership behaviors of the deans and the








outcome variables of subordinate satisfaction with the leader, subordinate perception of

leadership effectiveness, and subordinate willingness to exert extra effort.

Specifically, this study addressed the following research questions:

1. To what extent do deans of students exhibit transformational, transactional, and
laissez-fair leadership behaviors as rated by their subordinates using the
Multifactor Leadership Questionnaire Rater Form (5X-Short)?

2. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' perception of leader
effectiveness as measured by the Multifactor Leadership Questionnaire Rater
Form (5X-Short)?

3. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' satisfaction with
their leader as measured by the Multifactor Leadership Questionnaire Rater Form
(5X-Short)?

4. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' willingness to exert
extra effort as measured by the Multifactor Leadership Questionnaire Rater Form
(5X-Short)?

5. Is there a relationship between gender and the transformational leadership
behavior, the transactional leadership behavior, or the laissez-faire leadership
behavior of deans of students as rated by their subordinates using the Multifactor
Leadership Questionnaire Rater Form (5X-Short)?

Theoretical Background

Leaders in public institutions face the challenge of how to lead at a time when

conditions are changing, public confidence is low, resources are tight, and options are

limited (NASULGC, 1997). Blimling and Whitt (1999) state that doing things the way

they have always been done is not an appropriate response for student affairs during this

period of reform in higher education. In facing the current challenges, there is a need for

visionary leadership within student affairs (Rogers, 1996). The authors of the report,








Returning to Our Roots: The Student Experience, state, "We live in an age of

transformational not technical change. Our leadership, like our institutions, must become

transformational as well" (NASULGC, 1997, p. 21). Therefore, transformational

leadership theory is particularly applicable to the contemporary dean of students' role

within the institution.

Transformational leadership was first distinguished from transactional leadership

by Dowton (1973). However, it was the work of Bums (1978) that first drew major

attention to the ideas associated with transformational leadership (Leithwood, Tomlinson,

& Genge, 1996). Burns (1978) conceptualized two factors, transactional and

transformational, to differentiate ordinary from extraordinary leadership. Transactional

(ordinary) leadership is based on an exchange relationship in which follower compliance

(effort. productivity, and loyalty) is exchanged for expected rewards. Transformational

(extraordinary) leaders raise followers' consciousness levels about the importance and

value of designated outcomes and ways of achieving them. They also motivate followers

to transcend their own immediate self-interest for the sake of the mission and vision of

the organization. Followers' confidence levels are raised and their needs broadened by

the leader to support their development to higher potential. Such total engagement

(emotional, intellectual and moral) encourages followers to develop and perform beyond

expectations (Bums, 1978; Bass, 1985; Sergiovanni, 1991).

Bass (1985) operationalized the work of Bums (1978) by developing a model of

transformational and transactional leadership that he later revised with Bruce Avolio and

that is now referred to as the Full Range of Leadership model (Bass & Avolio, 1994).

The model identifies four distinct transformational leadership behavior constructs:








idealized influence, inspirational motivation, intellectual stimulation, and individualized

consideration. Three behavioral constructs identify transactional leadership: contingent

reward, management-by-exception active, and management-by-exception passive. The

model also includes a leadership behavior referred to as laissez-faire leadership, the most

inactive form of leadership where a leader chooses not to guide performance when the

situation would warrant guidance (Bass & Avolio, 1994). Bass is credited as being the

first researcher to operationalize a transformational leadership model into a measurement

instrument by his development of the Multifactor Leadership Questionnaire (Chemers,

1997; Conger. 1989).

Bass' (1985) conception of transformational leadership and transactional

leadership contrasts with that of Burns (1978) who considered transformational and

transactional leadership practices as opposite ends of a continuum. Bass (1985) contends

that most leaders display transformational and transactional leadership in varying

degrees. Transformational leadership augments transactional leadership. Transactional

practices on their own do little to bring about the enhanced commitment and extra effort

required for the positive change that will occur when the members of an organization

experience transformational leadership (Leithwood et al., 1996).

Bass and Avolio (1997), in establishing the reliability and validity of the

Multifactor Leadership Questionnaire as an instrument that can measure transformational,

transactional, and laissez-faire leadership behaviors as defined by their Full Range of

Leadership model, utilized a validation samples set and a cross-validation samples set.

The validation samples were collected from several different types of organizational

settings including military, business, political, non-profit, educational, and public service








organizations (Bass & Avolio, 1997). The only connection any of the samples in the sets

had to higher education was the use of undergraduate students and the evaluation of

leaders in nursing schools. The samples did not include any studies using student affairs

practitioners at public research universities.

While research has shown transformational leadership behaviors to be

significantly and positively related to outcomes of willingness of followers to exert extra

effort, a perception that the leader's leadership behavior is effective, and an overall sense

of satisfaction with the leader on the part of the followers (Bass, 1985; Seltzer & Bass,

1990; Bass & Avolio, 1997; Bass 1998), the population sample for the current study may

differ from the validation samples in ways that may weaken or enhance the relationships

between leader behaviors and subordinate criterion variables. Researchers have noted

that variables related to subordinate, task, and organizational characteristics can serve to

weaken, neutralize or enhance the relationships between particular leader behaviors and

subordinate criterion variables (Kerr & Jermier, 1978; Howell, Dorfman, & Kerr, 1986).

Therefore, in evaluating the leadership behaviors of deans of students through the

extension of Bass and Avolio's (1997) Full Range of Leadership model to student affair

practitioners, the issue of what evidence exists to suggest the theory is applicable to the

current study's population arises. This issue is important since the current study is

extending the theory to a new population as opposed to testing the theory itself. The

evidence supporting the extension of the theory to student affairs practitioners comes

from Bass' (1998) review of the research on transformational leadership over a wide

range of organizational types and settings.









Bass (1998), in recognizing that variables related to subordinate, task, and

organizational characteristics can affect the relationships between particular leader

behaviors and subordinate criterion variables, stated that situational contingencies do

make a difference. However, Bass (1998) noted that over fifteen years of research

indicates that situational contingencies do not override the general finding that

transformational leadership behaviors are significantly and positively related to outcomes

of willingness of followers to exert extra effort, a perception that the leader's leadership

behavior is effective, and an overall sense of satisfaction with the leader on the part of the

followers. Bass (1998) argues that research has indicated that transformational leadership

is more effective than constructive transactions, which are more effective than corrective

transactions, regardless of situational contingencies.

This study investigated whether deans of students at public research universities

in the southeast exhibit transformational leadership behaviors, and if so, whether this

leadership style enhanced employee perceptions of extra effort, leadership effectiveness,

and follower satisfaction with leaders' methods.

Definition of Terms

Specific terms used in this study are defined below.

Contingent Reward

Contingent reward is a transactional leadership behavior that rewards followers

for attaining specific performance levels. The leader utilizes primarily extrinsic

motivators to reward followers contingent upon effort and performance level achieved.








Dean of Students

The dean of students is a full-time student affairs professional who performs

supervisory and managerial activities within the division of student affairs and who is not

the chief student affairs officer. The dean of students is responsible for several of the

student affairs functions found on university campuses (Sandeen, 1996). Deans of

students generally report directly to the chief student affairs officer with the title vice

president or vice chancellor (Ambler, 1993). Other titles used for individuals having the

responsibilities of the dean of students are director of student life and dean of student life.

For the purpose of this study, the title dean of students will be used to represent those

persons holding the position of dean of students, director of student life, or dean of

student life.

Doctoral/Research Universities-Extensive

Doctoral/Research Universities-Extensive are institutions that typically offer a

wide range of baccalaureate programs and are committed to graduate education through

the doctorate. They award 50 or more doctoral degrees per year across at least 15

disciplines (Carnegie Foundation, 2000).

Doctoral/Research Universities-Intensive

Doctoral/Research Universities-Intensive are institutions that typically offer a

wide range of baccalaureate programs and are committed to graduate education through

the doctorate. They award at least ten doctoral degrees per year across three or more

disciplines, or at least 20 doctoral degrees per year overall (Carnegie Foundation, 2000).








Effectiveness

Effectiveness refers to a leader's ability to meet the job-related needs of the

followers and promote productivity within the department. This capacity also includes

the leader's ability to make contributions to the entire organization while representing the

follower's interests to the senior leadership (Bass & Avolio, 1995).

Full Range of Leadership Model

The Full Range of Leadership model is a leadership model proposed by Bass and

Avolio (1997) developed from Bass' (1985) transformational leadership theory. It

includes elements of transformational leadership, transactional leadership, and laissez-

faire or non-leadership behaviors.

Idealized Influence

Idealized influence is a leadership behavior that result in leaders as role models.

These leaders are seen as courageous, visionary, value driven and as change agents.

They are admired, respected and trusted. Here the leader is viewed as having high moral

standards and uses power only when necessary. This leader provides consistency and is

seen as a risk taker (Bass, 1985; Bass & Avolio, 1994).

Individualized Consideration

Individualized consideration is a leadership behavior that significantly contributes

to the subordinates achieving their fullest potential (Bass, 1998). Leaders that exhibit this

behavior develop subordinates through coaching, mentoring, and providing feedback

(Bass & Avolio, 1994).









Inspirational Motivation

Inspirational motivation is leadership that excites, arouses and inspires

subordinates in ways that increase optimism and pride (Bass, 1985, 1998). Inspirational

motivation provides meaning and challenge in the follower's work. Followers are

involved in the creation of new futures through a shared vision. Expectations are clearly

communicated in such a way that followers are committed to jointly developed goals

(Bass & Avolio. 1994).

Intellectual Stimulation

Intellectual stimulation is a leadership behavior that encourages followers to

analyze problems and seek out innovative solutions. The leader that utilizes intellectual

stimulation provides subordinates with challenging new ideas and stimulates thinking in

new ways (Bass. 1985).

Laissez-Faire Leadership

Laissez-faire leadership is the most extreme form of passive leadership,

considered to be non-leadership (Bass & Avolio, 1994). The leader avoids making

decisions and is inactive rather than reactive or proactive. The leader evades getting

involved when important issues arise and fails to provide assistance when requested. The

leader is not motivated or adequately skilled to perform duties (Bass, 1998).

Management-by-Exception (Active)

Management-by-exception (active) is a contingent reinforcement behavior in

which the leader actively seeks deviations from standards and takes actions when

irregularities occur (Bass & Avolio, 1994). The leader shuns giving directions if old









ways work and the followers continue to work in familiar patterns as long as performance

goals are met (Hater & Bass, 1988).

Management-by-Exception (Passive)

Management-by-exception (passive) is a leadership behavior in which the leader

only takes actions after deviations and irregularities are evident (Hartog, van Muijen, &

Koopman, 1997). The leader waits for problems to materialize prior to any intervention

(Hater & Bass, 1988).

Multifactor Leadership Questionnaire

The Multifactor Leadership Questionnaire (MLQ) is a measurement instrument

developed by Bass and associates to identify and measure (a) the framework of

leadership factors included in Bass and Avolio's Full Range of Leadership Development

model, and (b) a set of three leadership outcomes (follower extra effort, leader

effectiveness, follower satisfaction with the leader's methods) that occur as a result of

leader behavior (Bass & Avolio, 1997).

Professional Staff

Professional staff are fulltime student affairs practitioners who have responsibility

for one or more outside-the-classroom services or programs at a post-secondary

institution. Professional staff typically have at least a master's degree in student affairs,

counseling, or higher education administration and are a member of a professional

association related to student affairs (Winston & Miller, 1991).

Transactional Leadership

Transactional leadership is leadership based on the exchange between leader and

follower (Burns, 1978). It is implemented through a series of implicit bargains in which









the leader offers incentives and rewards in exchange for satisfaction of lower order needs

(Bass, 1985).

Transformational Leadership

Transformational leadership is leadership based on mutual stimulation and shared

vision, going beyond self-interest exchanges (Bass, 1985, 1998). Transformational

leaders broaden and elevate the interest of followers and have a transforming effect.

They motivate their followers and seek to fulfill their higher order needs (Bass, 1985).

Delimitations and Assumptions

For the purpose of this study, the following delimitations, limitations, and

assumptions apply:

Delimitations

1. This study is delimited to deans of students at public research universities that
(a) are classified as either Doctoral/Research Universities-Extensive or
Doctoral/Research Universities-Intensive by the 2000 Carnegie Classification
(Carnegie Foundation. 2000), (b) employ the dean of students title to
recognize a student affairs professional staff member that is not the chief
student affairs officer, and (c) are located in the Southeastern states of
Alabama, Arkansas, Florida, Georgia, Louisiana, Kentucky, Mississippi,
North Carolina, South Carolina, Tennessee, and Virginia. Therefore, it is not
intended to be reflective of the leadership profiles of deans of students at
large.

2. The study is delimited to the leadership factors developed by Bass (1998) of
Idealized Influence, Inspirational Motivation, Intellectual Stimulation,
Individualized Consideration, Contingent Reward, Management-by-
Exception, and Laissez-Faire.

3. This study will examine the perceptions of subordinates of deans of students
regarding transformational, transactional, and laissez-faire leadership
behaviors. It will not examine the perceptions of deans of students' peers or
supervisors.








Limitations

1. The sample composite of deans of students from public research universities
might not be representative of deans of students as a whole.

2. The subordinates who participate in this study might respond to the MLQ as
they believe they should and not answer truthfully.

3. The study will utilize only one measurement of leadership style, the MLQ
Short Form 5x.

Assumptions

1. Professional staff members within a dean of students office are able to identify
leadership qualities based on their perceptions of the dean's effectiveness.

2. Professional staff members within a dean of students office are able to
evaluate their satisfaction with the dean.

3. Professional staff members within a dean of students office are able to
evaluate their willingness to exert extra effort.

4. All subjects responded truthfully and accurately.

Organization of the Study

This study comprises five chapters. Chapter 1 introduces the study's justification,

its purpose, the problem that it addressed, and the research questions that were tested. In

chapter 2, pertinent literature is reviewed with a focus on leadership theory,

transformational leadership, and information on the development and use of the

Multifactor Leadership Questionnaire. This chapter also contains a review of the history

of the student affairs dean. Chapter 3 describes the method that was used for answering

the research questions. Chapter 4 presents the results of the statistical analyses that were

used to answer the research questions. Chapter 5 provides the overall findings of the

study, conclusions drawn from the statistical analyses, implications of the results, and

recommendations for future research.














CHAPTER 2
REVIEW OF THE LITERATURE

The purpose of this study was to examine the leadership behaviors of deans of

students at public research universities in the southeast using Bass and Avolio's (1997)

Full Range of Leadership model. The study will investigate the relationship between

transformational, transactional, and laissez-fair leadership behaviors of the deans and the

outcome variables of subordinate satisfaction with the leader, subordinate perception of

leadership effectiveness, and subordinate willingness to exert extra effort.

Specifically, this study addressed the following research questions:

1. To what extent do deans of students exhibit transformational, transactional, and
laissez-fair leadership behaviors as rated by their subordinates using the
Multifactor Leadership Questionnaire Rater Form (5X-Short)?

2. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' perception of leader
effectiveness as measured by the Multifactor Leadership Questionnaire Rater
Form (5X-Short)?

3. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' satisfaction with
their leader as measured by the Multifactor Leadership Questionnaire Rater Form
(5X-Short)?

4. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' willingness to exert
extra effort as measured by the Multifactor Leadership Questionnaire Rater Form
(5X-Short)?

5. Is there a relationship between gender and the transformational leadership
behavior, the transactional leadership behavior, or the laissez-faire leadership








behavior of deans of students as rated by their subordinates using the Multifactor
Leadership Questionnaire Rater Form (5X-Short)?

Definitions of Leadership

The perceived importance of leadership is evidenced by the volumes of

publications published on the topic. While there are numerous definitions of leadership,

influencing others is a common theme of the definitions. Hilgert and Haimann (1991)

defined leadership simply as "the ability to guide and influence the opinions, attitudes,

and behavior of others" (pp. 16-17). Gulley (1960) proposed that leadership is

"influencing others within a particular situation and social context in a way that induces

them to follow, be modified, or to be directed" (p. 174).

Other definitions explicitly state that leadership is goal directed. Kreitner and

Kinicki (1995) stated that leadership is "influencing employees to voluntarily pursue

organizational goals" (p. 428). Stogdill (1974) defined leadership as "the process of

influencing the activities of an organized group in its efforts toward goal setting and goal

achievement" (p. 57). Nahavandi's (1997) and Dessler's (1995) definitions of a leader

strongly support the idea that leadership is goal directed. Nahavandi (1997) defined a

leader "as any person who influences individuals and groups within an organization,

helps them in the establishment of goals, and guides them toward achievement of those

goals, thereby allowing them to be effective" (p. 4). Dessler (1995) stated "leadership

occurs whenever one person influences another to work toward some predetermined

objective" (p. 364).

Jago (1982) defined leadership in terms of both process and property:

The process of leadership is the use of non-coercive influence to direct and
coordinate the activities of the members of an organized group toward the
accomplishment of group objectives. As a property, leadership is the set








of qualities or characteristics attributed to those who are perceived to
successfully employ such influence. (p. 315)

These definitions imply that anyone who is able to influence others toward

objectives can be considered a leader. However, formal leadership is tied to a

hierarchical position. Yukl (1994) used the term leader "to refer to people who occupy

positions in which they are expected to exert leadership" (p. 5). This was supported by

Nahavandi (1997) stating "the presence of leaders often assumes some form of hierarchy

within a group" (p. 4).

Theories of Leadership

Although leadership has been the subject of debate, examination, and

investigation for thousands of years, it has only been a topic of continuous formal

analysis by scholars for the last 100 years with several of the leadership theories being

developed in the past 50 years. The leadership theories and research can be classified as

trait, behavioral, situational, and transformational approaches. The evolution of

leadership theories and leadership research can be seen by reviewing these major

categories.

Trait Theories

Many of the earliest leadership investigations centered on identifying and

measuring the specific personal characteristics of leaders based on the assumption that

great leaders are born, not made (Megginson, Mosley, & Peitri, 1989; Kreitner & Kinicki,

1995). This approach is commonly referred to as the trait theory of leadership and it

dominated the study of leadership during the first half of the twentieth century. Studies

employing the trait approach attempted to identify distinctive physical or psychological

characteristics related to leadership behavior. The majority of these studies compared








leaders with non-leaders to identify differences that existed with respect to their physical

characteristics, personalities, and abilities (Yukl, 1989).

Prior to World War II, hundreds of leadership trait studies were conducted

identifying dozens of leadership traits (Kreitner & Kinicki, 1995). Stogdill (1948)

reviewed and synthesized the results of over 120 of these studies and came to the

conclusion that no specific traits or personal characteristics stood out as certain, or even

strong, indicators of leadership. Stogdill's findings brought criticism to the trait theories

and initiated a shift from focusing on traits to focusing on the behavior of leaders

(Kreitner & Kinicki, 1995).

Behavioral Theories

During World War II, as both a reaction to the criticism of trait research and the

burgeoning human relations movement, behavioral theories of leadership began to

emerge (Kreitner & Kinicki, 1995). The concept behind behavioral leadership theory is

that group effectiveness is directly affected by leader behavior. Studies in this area focus

on identifying patterns of behavior often referred to as leadership styles that enable

leaders to effectively influence others (Kreitner & Kinicki, 1995).

The studies conducted by Lewin and his associates (Lewin, Lippit, & White,

1939) in the 1930s are considered the precursor to the behavioral approach (Daft, 1999).

Lewin et al. (1939) identified the styles of leadership as autocratic, democratic, and

laissez-faire. According to Daft (1999), "an autocratic leader is one who tends to

centralize authority and derive power from position, control of rewards, and coercion. A

democratic leader delegates authority to others, encourages participation, relies on

subordinates' knowledge for completion of tasks, and depends on subordinate respect for








influence" (p. 69). A laissez-faire leader is permissive and allows followers to do what

they want with minimum direction or discipline (Megginson et al., 1989).

Lewin et al. (1939) concluded that the democratic leadership style was the most

productive of the three. Work continued in the democratic environment when the leader

was not present implying group cohesiveness and motivation. The lowest productivity

was found with the laissez-faire environment in which worker frustration was high.

Work proceeded intensely in the autocratic environment as long as the leader was

present. However, work stopped when the leader was not present and worker aggression

was prevalent in this environment.

Two of the better-known behavioral leadership studies are the Ohio State Studies

and the Michigan Studies. These studies, like most of the behavioral studies, focused on

identifying the leader's orientation toward the employee, the task to be completed, or a

combination of the two (Megginson et al, 1989).

In studies conducted mostly in factories, researchers at Ohio State University

identified two types, or two dimensions, of behavior on the part of supervisors: "initiating

structure" and "consideration" (Daft, 1999). Consideration is an employee relation

oriented type that is identified by characteristics such as being friendly, considerate,

supportive, open and consultative. Leader behavior focuses on a concern for employees'

needs and the leader strives to create an environment of mutual respect and trust (Daft,

1999). Initiating structure types are task oriented and are prone to be directive, to

coordinate, to plan and to problem solve. Leader behavior focuses on defining and

organizing what employees should be doing to maximize output (Kreitner & Kinicki,

1995). The Ohio State researchers found that the best results were obtained when leaders








engaged in high levels of both task-focused and relationship-centered behavior (Daft,

1999).

The University of Michigan studies compared the behavior of effective leaders

with ineffective leaders. The Michigan researchers developed two types of leadership

behavior termed employee-centered and job-centered (Daft, 1999). The employee-

centered leader focuses on the needs of the followers and stresses interaction and support.

The job-centered leader directs activities toward efficiency by focusing on reaching task

goals and facilitating the structure of the work tasks (Daft, 1999). The employee-

centered and job-centered styles of leadership roughly correspond to the Ohio State

Studies' concepts of consideration and initiating structure respectively. However, unlike

the Ohio State studies, the researchers at Michigan considered the two leadership styles to

be distinct, with a leader being one or the other, but not both (Daft, 1999). The Michigan

researchers findings indicated that employee-centered leaders were more productive than

job-centered leaders (Megginson et al., 1989).

The findings of behavioral studies such as the Ohio State and Michigan studies

have been questioned and criticized by other researchers (Daft, 1999; Kreitner & Kinicki,

1995). The criticism has included references to later research indicating that styles other

than the ones considered optimal by the studies can be effective (Kreitner & Kinicki,

1995; Daft, 1999). Other critics have pointed out that while "it is relatively easy to call

certain behaviors of leaders effective once the desired outcomes have been observed, it is

much more difficult to stipulate in advance the behavior of leaders that result in the

desired outcomes" (Bensimon, Neuman, & Birnbaum 1989, p. 14).









Situational Theories

The limitations of trait and behavioral theories led researchers to explore a new

direction in leadership study. The new focus was on the situation in which leadership

occurred (Hersey & Blanchard, 1969). The concept behind situational or contingency

leadership theories is that leader effectiveness does not depend on who the leader is or on

the leader exhibiting a high degree of certain behaviors. Instead, effective leadership is

based on engaging in different combinations of task and relationship behavior in different

situations. The appropriate style of leadership to be used depends on the situation, the

people, the organization, or other environmental factors (Megginson et al, 1989).

Research conducted by Fiedler (1974), McGregor (1960), Mannheim, Rim and

Grinberg (1967), and Hunt and Liebscher (1973) concluded that work settings that vary

in task structure and climate foster differential leader behavior. Vroom and Yetton (1973)

proposed that it is the leader's decision making behavior that affects group performance.

According to their approach the effectiveness of a decision making procedure depends on

aspects of the situation such as the likelihood that followers will cooperate if allowed to

participate in the decision making process.

One of the most widely cited situational approaches is Hersey and Blanchard's

situational leadership theory. Hersey and Blanchard (1977, 1988) postulated a model that

identifies the readiness level of the followers and links it to the willingness and ability of

the followers to achieve the goals of the organization. Situational leadership theory takes

into consideration the followers' developmental level in order to determine the leader's

approach to accomplishing tasks. There are four categories of leader task and

relationship behavior for this model: (a) high task/low relationship, in which actions are








initiated and decisions made by the leader; (b) high task/high relationship, wherein the

leader provides a considerable amount of direction but also listens to input from

followers; (c) low task/high relationship, which incorporates a shift in problem-solving

from the leader to the followers; and (d) low relationship/low task which results in almost

total delegation of decision making to followers. The appropriate category of leader

behavior is based upon the follower's readiness level as it relates to the task to be

accomplished (Hersey & Blanchard, 1988). Therefore, this model requires a high degree

of discernment on the part of leaders.

Situational leadership theory has been heavily criticized. The major criticism is

that the model lacks a sound theoretical foundation for the hypothesized relationships

among variables (Graeff, 1997). Other researchers have criticized the model by stating

that leader use of supportive behavior is an important contribution to effective leadership

at all levels of subordinate readiness (Goodson, McGee, & Cashman, 1989).

Transformational Leadership

Burns (1978) conducted a comprehensive study of leadership and concluded that

all leadership could be classified as either transactional or transformational. He stated

that a leader-follower interaction that is transactional in nature has the leader offering a

reward for the expected valued response of the follower. Therefore, in transactional

leadership, motivation is achieved when the leader is able to appeal to the self-interest of

the followers. Incentives and rewards are used for influencing motivation. Beyond the

achievement of their related goals, both leader and follower experience no enduring

relationship (Burns, 1978). By contrast, transforming leadership moves to a level of

morality in that both leaders and followers so engage with one another that they raise









each other to a greater sense of purpose and to aspirations that are noble and transcending

(Burns, 1978). Burns' work led to the development of several new approaches to the

study of what is referred to as transformational leadership (Daft, 1999). The term is used

to contrast this new leadership with the older, transactional leadership approach.

Burns (1978) defined the transforming leader as one who "recognizes and exploits

an existing need or demand of a potential follower. But, beyond that, the transforming

leader looks for potential motives in followers, seeks to satisfy higher needs, and engages

the full person of the follower" (Burns, 1978, p. 4). Burns (1978) integrated Maslow's

(1954) theory of human needs and Kohlberg's (1981) theory of moral development to

build his definition of transforming leadership and to examine moral leadership, which he

views as "going beyond simply satisfying the follower's wants or desires to being

actually instrumental in producing the social change that will satisfy both the followers'

and leaders authentic needs" (p. 4).

Motivated by Burn's development of transformational leadership, Bass (1985)

sought to investigate what type of action or strategies leaders use in transforming

followers toward achieving organizational goals. He views the constructs of transactional

and transformational leadership as complementary. Therefore, transformational

leadership behaviors are likely to be ineffective in the absence of a transactional

relationship between leader and follower (Bass, Avolio, & Goodheim, 1987). According

to Bass (1985), transformational leadership augments transactional management to

achieve higher levels of follower performance with the primary difference residing in the

process by which the leader motivates followers and in the types of goals set. The ability








of transformational leaders to obtain performance beyond basic expectations of followers

has been labeled the "augmentation hypothesis" (Waldman, Bass, & Yammarino, 1990).

Bass (1985) viewed the transactional leader as one who operates within the

existing system, tends to avoid risk, focuses on time constraints and efficiency, and

prefers process over substance for maintaining control. The transactional leader fulfills

the needs of followers in exchange for performance that meets basic expectations. A

transactional leader is most likely to be effective in a predictable and stable environment

where measuring current performance against prior performance is the most successful

strategy (Bass, 1985, 1998).

Bass (1985) characterized the transformational leader as one who seeks new ways

of working, seeks opportunities in the face of risk, prefers effective answers to efficient

answers, and is less likely to support the status quo. The transformational leader attempts

to shape and create environmental circumstances as opposed to merely reacting to them

(Avolio & Bass, 1988). He or she will use transactional strategies when appropriate, but

will also motivate by appealing to followers' ideals and moral values and challenge them

to think about problems in new ways. The transformational leader raises the level of

intellectual awareness of the followers about the importance of valued outcomes and

motivates followers beyond their own self-interest for the sake of the organization (Bass,

1985).

The Full Range of Leadership Model

Bass (1985) operationalized the concept of transformational and transactional

leadership by developing a leadership behavior model that he later refined with Bruce

Avolio and is referred to as the Full Range of Leadership Model (Bass & Avolio, 1994;









Bass & Avolio, 1997; Bass, 1998). The Full Range of Leadership Model contains three

classifications of leadership processes: (a) transformational, (b) transactional, and (c)

laissez-faire or non-leadership behavior (Bass & Avolio, 1994; Bass, 1998). The model

predicts that leaders who are more transformational and less transactional are more

effective as leaders and more satisfying to their followers (Bass & Avolio, 1997).

Components of Transformational Leadership

The first set of leadership behaviors in the full range of leadership model

identifies four distinct transformational leadership behaviors, called the "Four I's":

idealized influence, inspirational motivation, intellectual stimulation, and individualized

consideration (Bass & Avolio, 1994, p. 3). According to Bass and Avolio (1994)

transformation leaders employ one or more of the "Four I's" to achieve better results than

leaders that only exhibit transactional behavior.

Idealized influence is a transformational leadership behavior that results in leaders

being role models for the individuals they are leading. It is characterized by the leader

putting the followers' needs above the leader's own personal needs, consistently

demonstrating high ethical standards, and using power only when necessary (Bass &

Avolio, 1994). Bass and Avolio (1997) divided idealized influence into idealized

influence-attributed and idealized influence-behavior. Idealized influence-attributed is

characterized by a leader who is risk-taker, makes followers feel good to be with him or

her, creates a sense of belongingness to the common cause, and cares about the interests

of the followers (Bass & Avolio, 1997). Idealized influence-behavior is characterized by

a leader who displays a high ethical and moral code, is a risk-taker, and has a strong

sense of mission (Bass, 1998).








Inspirational motivation is characterized by behaviors that provide meaning,

challenging goals, a sense of vision and mission, and belief that followers can reach goals

they may have originally thought too difficult to achieve. Optimism and enthusiasm are

expressed by the leader in getting followers to become engaged in envisioning attractive

future states (Bass, 1985; Bass & Avolio, 1994).

Intellectual stimulation is characterized by leader behavior that questions

underlying assumptions, reframes problems, and finds creative solutions to difficult

problems. This behavior develops the potential for followers to solve problems in the

future and encourages creative thought (Bass, 1998; Bass & Avolio, 1997).

Individual consideration is a transformational behavior that focuses on the growth

and development of each follower, providing them with new opportunities to learn, and

giving them personalized attention. The leader coaches, mentors, and teaches in an

attempt to help followers reach the established goals (Bass, 1998; Bass & Avolio, 1997).

Components of Transactional Leadership

Transactional leadership behavior is expressed by the rewarding or disciplining of

the follower depending on the adequacy of the follower's performance (Bass, 1998). The

model breaks transactional leadership into the two components of contingent reward and

management-by-exception.

Contingent reward is characterized by the leader stressing an exchange where the

-leader assigns or gets agreement on what needs to be done and promises rewards or

actually rewards others in exchange for satisfactorily carrying out the assignment" (Bass,

1998, p. 6). Reward is contingent upon the effort expended by the follower and

performance level achieved (Bass, 1998).








Management-by-exception is defined by Bass (1998) as a "corrective transaction"

and occurs when a leader intervenes to make a correction only when something has gone

wrong or a mistake has been made (p. 7). Management-by-exception can either be active

or passive. Management-by-exception active is characterized by the leader actively

watching for deviations from the norm, and taking action when irregularities occur (Bass

& Avolio, 1997). Management-by-exception passive is characterized by the leader

intervening only after a correction is needed. There is no active monitoring for deviations

from the norm (Bass & Avolio, 1997).

Laissez-Faire Leadership

Laissez-faire leadership is the third classification of leadership in the Full Range

of Leadership Model and was added to address behaviors that indicate a non-transaction

of leadership (Bass & Avolio, 1997). Laissez-faire leadership is the most inactive form

of leadership and is characterized by the leader avoiding decisions and not using his or

her authority. Bass (1998) states that Laissez-Faire behavior is "the avoidance or absence

of leadership and is, by definition, most inactive, as well as the most ineffective

according to almost all research on style" (p. 7).

The Augmentation Effect of Transformational Leadership

The Full Range of Leadership Model predicts that transformational leadership

will add to the effectiveness of transactional leadership (Bass & Avolio, 1997; Bass,

1998). Although effective leaders are both transformational and transactional, Bass

(1985, 1998) states that transformational leadership augments transactional leadership in

predicting leadership behavior effects on the outcome variables of subordinate

willingness to exert extra effort, perception of leader effectiveness, and satisfaction with








the leader. Specifically, the augmentation effect states that transformational leadership

behaviors should account for unique variance in followers' ratings of the outcome

variables over and above that accounted for by transactional leadership (Bass, 1998).

The Optimal Leader Profile

According to the Full Range of Leadership Model, every leader displays

transformational, transactional, and laissez-faire leadership behaviors to some degree

(Bass, 1998). However, the leader with an optimal profile infrequently displays laissez-

faire leadership and displays successively higher frequencies of the transactional

behaviors of management-by-exception passive, management-by-exception active, and

contingent reward. The optimal leader profile displays the five transformational

leadership behaviors of idealized influence-attributed, idealized influence-behavior,

inspirational motivation, intellectual stimulation, and individualized consideration the

most frequently (Bass. 1998).

The Multifactor Leadership Questionnaire

According to Chemers (1997), the research of Bass and his associates provided

the support that was needed for applying transformational leadership concepts to

complex, formal organizations. Both Chemers (1997) and Conger (1989) give Bass

credit for being the first researcher to operationalize a transformational leadership model

into a measurement instrument by his development of the Multifactor Leadership

Questionnaire (MLQ).

Bass (1985) developed the MLQ to assess the leadership constructs of

transformational and transactional leadership explicated by his theory. The MLQ was

initially generated by exploratory methods and then tested in the field using factor








analysis (Bass, 1985). The MLQ has undergone several modifications to answer

criticisms about its validity and to be a better gauge of the full range of leadership

(Avolio, Bass, & Jung, 1995). The current form of the MLQ measures five

transformational leadership constructs, three transactional leadership constructs, and one

nonleadership construct. The nine scales are (a) idealized influence-attributed, (b)

idealized influence-behavior, (c) inspirational motivation, (d) intellectual stimulation, (e)

individualized consideration, (f) contingent reward, (g) management-by-exception

(active), (h) management-by-exception (passive), and (i) laissez-faire leadership (Avolio

et al, 1995). The first five scales refer to transformational leadership, the next three to

transactional leadership, and the last scale to nonleadership. The MLQ also measures

three outcomes of leadership: (a) extra effort of followers, (b) effectiveness of the leader,

and (c) follower satisfaction with the leader (Avolio et al, 1995).

The MLQ Hierarchy of Correlations

The Full Range of Leadership model predicts a hierarchy of correlations of the

MLQ components with the outcome variables of leader effectiveness, satisfaction with

the leader, and willingness to exert extra effort. The predicted hierarchy of correlations

states transformational behaviors will have higher correlations with the outcome variables

than contingent reward. Contingent reward will have higher correlations than

management-by-exception active, which will have higher correlations than management-

by-exception passive. Laissez-faire leadership will have the lowest correlations scores

(Bass & Avolio, 1997).








Research on Transformational Leadership

A review of the literature on transformational leadership indicates that it has a

consistent, reliable, and positive relationship to effectiveness measures, whether

organizationally based or subjectively determined as predicted by Bass (1985) in the

development of his theory. The empirical work on transformational leadership covers a

wide area, and applies the concepts in a number of different disciplines and settings.

Transformational leadership has been found to have a substantive and significant

relationship on organizational and group effectiveness (Avolio, Howell, & Sosik, 1999;

Avolio, Waldman, & Einstein, 1988; Barling, Weber, & Kelloway 1996; Geyer &

Steyrer, 1998; Howell & Avolio, 1993; Jung & Avolio, 1999; Lowe, Kroeck, &

Sivasubramaniam, 1996), and perception of performance of the leader (Hater & Bass,

1988; Yammarino, Spangler, & Bass, 1993). Lowe et al. (1996), in a meta-analysis of 39

studies, found that a strong relationship exists between the transformational scales and

leadership effectiveness measures using either organizationally determined criteria or the

MLQ. Bass, Waldman, Avolio, and Bebb (1987) found transformational leadership to

have a powerful modeling effect on followers and on the organizational culture.

Transformational leadership was found to be predictive of innovation and creativity

(Howell & Higgins, 1990; Keller, 1992; Sosik, 1997), positive work attitude and product

knowledge (Yammarino et al., 1997), and followers feeling empowered (Howell &

Higgins, 1990). Furthermore, transformational leadership is predictive of satisfaction

with the leader (Barling, Weber & Kelloway, 1996; Druskat, 1994; Howell & Frost,

1989; Koh, Steers, & Terborg, 1995; Ross & Offermann, 1997), follower commitment

(Yammarino et al., 1997), organizational commitment (Barling et al., 1996; Koh et al.,








1995), and organizational citizenship (Koh et al., 1995). It has also been found that

individuals can be trained to exhibit transformational leadership behavior (Avolio, 1999;

Bass, 1998; Bass & Avolio, 1997; Shea & Howell, 1999).

The large amount of research studies done on transformational leadership which

appear in a broad range of scholarly publications suggest that it is one of the most

important contemporary leadership topics. Overall, the amount of research on

transformational leadership that indicates it has a consistent, reliable and positive

relationship to effective measures is impressive. The following section analyzes possible

contingencies and limitations.

Contingencies and Limitations of the Full Range of Leadership Model

Transformational leadership theory is moderated by situational variables

including level of the leader, the leader's personality, type of organization, the

organizational environment, characteristics of the followers, and type of criterion used to

determine effectiveness (Bass, 1985; Lowe et al., 1996). Transformational leaders are

more likely to arise in times of crises or of major change, and in organic types of

organizations that are not highly structured with routine tasks and functions (Bass, 1985).

Bass (1998) states that organizational turbulence is a condition that often supports the

emergence of transformational leadership in contrast to transactional leadership, which

"is likely to emerge and be relatively effective when leaders face a stable, predictable

environment" (p. 52).

Bass (1985) speculated that the effectiveness of transformational leadership

behavior may be contingent on the type of tasks to be performed. For example, Bycio,

Hackett, and Allen (1995) found that transactional behaviors are very important in








situations where safety is a major concern. Bass (1998) has also stated that where safety

is a priority, management-by-exception active may play a more prominent role in

determining organizational effectiveness than it does in other situations.

In terms of differences between a leader's gender, Druskat (1994), Bass, Avolio,

and Atwater (1996), Carless (1998), and Bass (1998) noted that women tend to display

transformational behaviors more often than men. This is in contrast to a study by Maher

(1997) finding that there are no differences between men and women leaders in

displaying transformational behaviors. According to Bass (1998) the differences that

were found may be explained by the fact that women are socialized to display more

nurturing, caring and developmental behaviors than men, and these behaviors are

essential elements or transformational leadership. Maher (1997) argued that any potential

differences that may have been found may not be universal and can be attributed to

situational or contextual variables.

The Dean of Students

In 1870, Harvard's President Eliot appointed Professor Ephraim Gurney as the

first college dean (Rentz, 1996; Brubaker & Rudy, 1997; Caple, 1998). Although

Gurney's responsibilities remained primarily academic, he assumed responsibility for

student discipline previously handled by President Eliot. Twenty years later, Harvard

divided the dean's position into two offices essentially creating an academic dean and a

student affairs dean (Caple, 1998). An instructor of English, LeBaron Russell Briggs,

took on the nonacademic responsibilities related to students. For this reason, he is

generally considered the first dean of students (Rentz, 1996; Brubaker & Rudy, 1997;

Rhatigan, 2000). In his book about Dean Briggs, Brown (1926) wrote that Brigg's goals








for being dean included "(a) To help the student disciplined, and not merely to humiliate

him; (b) to make it easy for the faculty to do its work; and (c) to develop a sentiment

among the students which would render discipline less and less necessary" (p. 101). As

his position developed, Dean Briggs took on several responsibilities outside of discipline

including registration, student record keeping, assisting entering students, counseling

students, and monitoring extracurricular activity (Brown, 1926; Rentz, 1996; Brubaker &

Rudy, 1997).

By the early 1900s, the combination of increased student influence on

extracurricular activities and increasing enrollments resulted in institutions adding

personnel to take on the responsibilities related to student life outside of the classroom

(Leonard, 1956). As a result, the positions of dean of men and dean of women became

"mainstays of morality and decorum" (Dressel, 1981, p. 94). Deans of men and women

were responsible for many out-of-class services and activities. Among their functions

were student discipline, housing, counseling, advising, student governance and other

student organizations, career development, health, supervision of facilities and social

events, and parental and public relations (Cohen, 1998; Dinniman, 1977; Rentz, 1996).

Deans of women were expected to give special attention to supervising the female

student's social life, housing, health, and social hygiene (Dinniman, 1977).

Around the World War I period, student personnel professional associations

began to emerge as a result of student personnel deans traveling to neighboring campuses

to meet and discuss the common problems and issues they each faced (Dinniman, 1977;

Rentz, 1996). The deans of women formed the National Association of Deans of Women

in 1916, which later became the National Association for Women in Education. The








deans of men formed the National Association of Deans and Advisors of Men in 1919

(Fenske, 1989; Caple, 1998). The National Association of Deans and Advisors of Men

later became the National Association of Student Personnel Administrators while in 1924

members of the existing gender specific associations founded the American College

Personnel Association (Rentz, 1996). Denniman (1977) noted that the formation of the

professional organizations, publications on student personnel work, and the development

of training programs for deans of students, "were all indications of the deanship's

professionalization and growing influence in higher education" (p. 8).

After World War I, the student personnel movement experienced an expansion

driven by the acceptance and application of mental testing and counseling techniques

developed by the Army during the war (Fenske, 1989; Brubacher & Rudy, 1997; Caple,

1998). As testing and counseling gained credibility as tools to help students, their use

became widespread. Employing counseling on a large scale offered the student personnel

movement a greater degree of professionalism while the development and use of new

pedagogical and psychological theories gave support to the functions of student personnel

work (Garland, 1985; Caple, 1998). The importance of students' non-cognitive needs in

their overall development was becoming increasingly recognized resulting in the

expanding and diversifying of student affairs functions on college and university

campuses (Brubacher & Rudy, 1997; Caple, 1998).

The period proceeding and just following World War II saw an increase in the

emphasis placed on student affairs functions (Caple, 1998; Rhatigan, 2000). The

philosophical basis for the student affairs profession was sharpened during this period by

the publication of the American Council on Education's 1937 and 1949 reports titled The








Student Personnel Point of View (Garland, 1985; Fenske, 1989; Rentz, 1996; Caple,

1998). The reports emphasized the philosophical basis for student personnel work and

provided the foundation and assumptions that many professionals believed to represent

the spirit of the profession (Garland, 1985; Rentz, 1996).

While the period after World War II was a time of expansion for student affairs as

a profession, Schwartz (1997) reports that it was not a good period for deans of women.

He argues that in the rush to return to normalcy and to reward men returning from the

war, the role women had played in the success of the student personnel movement was

largely ignored. While offices of the dean of men were often "expanded to become dean

for student personnel, dean of students, and vice-presidents for student personnel

services" deans of women "were given lesser positions, dismissed, or allowed to retire

quicil)" (Schwartz, 1997, p. 433).

From 1950 to 1972, the title "dean of students" was the most frequently used title

to designate the chief student affairs officer (CSAO). However, it was during this period

that a shift to the designation of "vice president" was emerging. By 1972, 18% of the

National Association of Student Personnel Administrators member institutions were

using the title of vice president to designate their CSAO (Crookston, 1974). The use of

the vice president title continued to increase through the 1970s and 1980s becoming the

preferred designation of the chief student affairs officer at public universities (Ambler,

1993). While the use of the dean of students title as designating the CSAO

fell from favor at public institutions, many public universities retained the title to

designate a major student affairs officer who has responsibility for many of the aspects of

student life and who reports directly to the CSAO (Ambler, 1993).








Sandeen (1996) reports that the contemporary dean of students is often

responsible for several of the traditional student affairs functions, responds to student

crises, enforces the institution's community standards, and "responds to the general

concerns of students, faculty, staff, parents, and community members" (p. 444). A

review of the web sites for National Association of State Universities and Land-Grant

Colleges member institutions in the Southeast that employ the dean of students title

indicate that the common student affairs functions that are supervised by the dean of

students include judicial affairs, Greek life, orientation and first-year programs,

leadership development, disability services, student organizations and activities, and

student government. Other functions that were not as common but were the

responsibility of several of the deans of students include withdrawals, parent programs,

service learning, international student services, housing and residence life, multicultural

affairs, and gender issues.














CHAPTER 3
METHODOLOGY

Kerlinger (1986) believed that a research design must encompass both the research

problem and the plan of investigation necessary to acquire empirical evidence on the

problem. While a design does not explain precisely what to do, it implies the direction of

observation and analysis. The purpose of this chapter is to present the research questions

that guided the study, the research population and sample that was studied, the instrument

that was employed, and the statistical analysis that were conducted.

The purpose of this study was to examine the leadership behaviors of deans of

students at public research universities in the southeast using Bass and Avolio's (1997)

Full Range of Leadership model. The study investigated the relationship between

transformational, transactional, and laissez-fair leadership behaviors of the deans and the

outcome variables of subordinate satisfaction with the leader, subordinate perception of

leadership effectiveness, and subordinate willingness to exert extra effort.

Specifically, this study addressed the following research questions:

1. To what extent do deans of students exhibit transformational, transactional, and
laissez-fair leadership behaviors as rated by their subordinates using the
Multifactor Leadership Questionnaire Rater Form (5X-Short)?

2. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' perception of leader
effectiveness as measured by the Multifactor Leadership Questionnaire Rater
Form (5X-Short)?

3. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans








of students and their subordinate professional staff members' satisfaction with
their leader as measured by the Multifactor Leadership Questionnaire Rater Form
(5X-Short)?

4. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' willingness to exert
extra effort as measured by the Multifactor Leadership Questionnaire Rater Form
(5X-Short)?

5. Is there a relationship between gender and the transformational leadership
behavior, the transactional leadership behavior, or the laissez-faire leadership
behavior of deans of students as rated by their subordinates using the Multifactor
Leadership Questionnaire Rater Form (5X-Short)?

Research Population

The population for this study was student affairs professional staff members

supervised by deans of students at public institutions of higher education classified as

either Doctoral/Research Universities-Extensive or Doctoral/Research Universities-

Intensive in the 2000 Carnegie Classification (Carnegie Foundation, 2000) located in the

southeastern states of Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana,

Mississippi, North Carolina, South Carolina, Tennessee, and Virginia. The total number

of institutions representing this population is 45. Five institutions in this population did

not have an equivalent position to a dean of students for the purpose of this study and two

institutions had the position vacant at the time of the study. Therefore, 38 institutions

were used in the study.

The population for the study was primarily identified through the institutions' web

sites and the National Association of Student Personnel Administrators' membership

directory. Institutions were contacted through e-mail and through telephone calls from

the researcher when information was not available through the institution's web site or

the membership directory.









The Instrument

The research instrument utilized in this study was the Multifactor Leadership

Questionnaire (MLQ) Form 5X (Bass & Avolio, 1997). The MLQ was used to collect

data regarding the independent variables of transformational leadership, transactional

leadership, and laissez-faire leadership and the dependent variables of subordinate

perception of leader effectiveness, satisfaction with the leader, and willingness to exert

extra effort.

The MLQ Form 5X was developed by Bass and Avolio to address the criticisms of

an earlier MLQ survey instrument designed by Bass (Bass & Avolio, 1997). Through the

use of confirmatory factor analysis, Avolio, Bass and Jung (1995) refined the original

MLQ into an instrument that better represented each leadership component within the

Full Range of Leadership model. Their findings from the validation and cross validation

studies resulted in the selection of the items for the MLQ Form 5X.

The MLQ Form 5X is a 45-item questionnaire using a Likert scale to measure leader

behaviors. Thirty-six of the items measure the independent variables of leadership

behaviors and nine items measure the dependent variables of outcome factors (Bass &

Avolio, 1997). The MLQ measures five transformational leadership components, three

transactional leadership components, and one nonleadership component. The

components for transformational leadership are: idealized influence-attributed, idealized

influence-behavior, inspirational motivation, intellectual stimulation, and individualized

consideration. The components for transactional leadership are: contingent reward,

management-by-exception-active and management-by-exception-passive. Laissez-faire

leadership is the nonleadership component (Bass & Avolio, 1997). Each of the








leadership components are measured by four interconnected items that are as low in

correlation as possible with items measuring the other eight components (Bass & Avolio,

1997). The items that measure the individual leadership components and outcome

behaviors are as follows:

Transformational Behaviors:

Idealized Influence-Attributed is measured by items 10, 18, 21, and 25.

Idealized Influence-Behavior is measured by items 6, 14, 23, and 34.

Inspirational Motivation is measured by items 13, 26, 36, and 9.

Intellectual Stimulation is measured by items 2, 8, 30, and 32.

Individual Consideration is measured by items 15. 19, 29, and 31.

Transactional Behaviors:

Contingent Reward is measured by items 1, 11, 16, and 35.

Management-by-Exception (Active) is measured by items 4, 22, 24, and 27.

Management-by-Exception (Passive) is measured by items 3, 12, 17, and 20.

Laissez-Faire (Nonleadership) Behaviors:

Laissez-Faire is measured by items 5, 7, 28, and 33.

Outcome Behaviors:

Willingness to Exert Extra Effort is measured by items 39, 42, and 44.

Leadership Effectiveness is measured by items 37, 40, 43, and 45.

Satisfaction with Leader is measured by items 38 and 41.

The MLQ requires a ninth grade reading ability and takes approximately 15 minutes

to complete. The respondents are asked to rate their supervisor, judging how frequently

each statement in the item fits the supervisor. Numerical values are given for each of the








item responses. The values are: 0 = Not at all, 1 = Once in a while, 2 = Sometimes, 3 =

Fairly often, 4 = Frequently, if not always. A lower score indicates that the leader's

behaviors were perceived to be inconsistent with the description of the leadership

component and a higher score is indicative of the perception of the presence of behaviors

consistent with the leadership component.

Participants were also asked to complete a researcher-developed demographic

information sheet. The demographic information sheet requested information on the

deans of students' gender, age, educational level, and number of years in current

leadership position. It also requested information on respondents' gender, age,

educational level, number of years in student affairs, and number of years working with

their current dean.

Reliability and Validity

In their instrument manual, Full Range Leadership Development: Manual for the

Multifactor Leadership Questionnaire, Bass and Avolio (1997) summarized the results of

the tests employed for examining the MLQ 5X's construct validity and reliability. For

their study, Bass and Avolio (1997) used a validation sample and a cross-validation

sample. The sample studies for both the validation sample and the cross-validation

sample were conducted by independent researchers and were based on data generated by

subordinates who evaluated their supervisors within a broad range of organizations and at

varying levels within the organizations (Bass & Avolio, 1997). Table 3-1 gives a

description of the validation samples set and the cross-validation samples set.

The scale scores for both the validation and cross-validation sets of samples are

provided by Bass and Avolio (1997) and are presented in Table 3-2. Reliabilities for









Table 3-1 Organizational Samples Used in Validation and Cross Validation Analysis of
the Multifactor Leadership Questionnaire
Description of Validation Samples Description of Cross-Validation Samples
Set N Set N
1. Undergraduate Students 254 1. Business Organizations in U.S. 215
(American and Taiwanese)
2. Political Organization in U.S. 428
2. Undergraduate Students (American) 162
3. Business Organizations in U.S. 549
3. Nursing Schools in U.S. 45
4. Fire Department in U.S. 325
4. U.S. Government Research
Organization 66 5. Not-for-Profit Government
Organization 189
5. Business Organizations in U.S. 457

6. Business Organizations in U.S. 320

7. College Educators in Nursing
School in U.S. 475

8. U.S. Army Organization 202

9. Oil Platforms Offshore from
Scotland 99

Note. Reproduced by special permission of the Publisher, MIND GARDEN, Inc., 1690 Woodside Road
#202, Redwood City, CA 90461 USA www.mindgarden.com from the Multifactor Leadership
Questionnaire for Research by Bernard M. Bass and Bruce J Avolio. Copyright 1995, 2000 by Bernard M
Bass and Bruce J Avolio. All rights reserved. Further reproduction is prohibited without Publisher's written
permission.

each of the leadership factors and the outcome scales range from .74 to .94 for the

validation sample and .73 to .93 in the cross validation sample (Bass & Avolio, 1997).

The scales' reliabilities are generally high, exceeding standard cut-offs for internal

consistency recommended in the literature (Alovio, Bass, & Jung, 1995).

Bass and Avolio (1997) used confirmatory factor analysis to test the convergent and

discriminant validity of the MLQ Form 5X scales. Bass and Avolio (1997) state that

the confirmatory factor analysis tests were specifically run to determine whether the data

collected from the validation and cross-validation sample sets confirmed the nine-factor

model proposed by Avolio and Bass (1991). Table 3-3 shows the comparison of the

Goodness of Fit (GFI), the Adjusted Goodness of Fit Index (AGFI), the Root Mean










Square Residuals (RMSR), and the Chi-square test results. The fit measures and the chi-

square tests improved as the model progressed from a one-factor to a nine-factor solution

Table 3-2 Means, Standard Deviations, and Intercorrelations Among MLQ Factor Scores
Variable M SD 1 2 3 4 5 6 7 8 9 10 11 12


1. Attributed 2.56 .84
Charisma 2.69 .90

2. Idealized 2.64 .85
Influence 2.71 .89

3. Inspirational 2.64 .87
Motivation 2.69 .91


.86
.87

.79
.83


.87
.89

.86 .91
.90 .91


4. Intellectual 2.51 .86
Stimulation 2.50 .86

5. Individualized 2.66 .93
Consideration 2.62 .94

6. Contingent 2.20 .89
Reward 2.04 .94

7. Management- 1.75 .77
by-Exception 1.71 .81
(Active)

8. Management- 1.11 .82
by-Exception 1.17 .88
(Passive)

9. Laissez-Faire 0.89 .74
0.99 .88


.84 .85 .90
.84 .85 .88

.82 .87 .84 .90
.86 .88 .84 .90


.68 .69 .73 .70 .75
.51 .58 .62 .60 .62

-.12 -.03 -.10 -.08 -.12
-.10 -.08 -.05 -.05 -.11


-.54 -.54 -.55 -.52 -.54
-.54 -.59 -.50 -.41 -.51


-.53 -.54 -.51 -.47 -.49
-.57 -.50 -.50 -.40 -.50


.87
.86

.03 .74
.21 .73


-.34 .28 .82
-.07 .44 .83


-.29 .18 .74 .74
-.07 .40 .82 .87


10. Extra Effort 2.60 1.16 .68 .69 .73 .69 .74
2.51 1.14 .71 .75 .78 .75 .82

11. Effectiveness 2.62 .72 .51 .44 .46 .41 .44
2.66 .88 .62 .48 .52 .40 .53


.62 .03 -.36 -.34 .91
.63 -.01 -.36 -.35 .86

.32 -.14 -.35 -.41 .45 .91
.26 -.04 -.41 -.45 .48 .87


12. Satisfaction 2.57 1.28 .25 .22 .21 .18 .27 .19 .06 -.21 -.25 .23 .15 .94
2.38 1.28 .35 .18 .22 .08 .24 .11 .18 -.17 -.19 .19 .40 .93
Note. Each factor was rated on the 5-point scale from 0 (not at all) to 4(frequently, if not always). Alpha
coefficients are reported in boldface along the diagonal. First values in each column show correlations from
the validation set of samples (N=1,394 after listwise deletion) and second values in each column show
correlations from the cross-validation set of samples (N=1,490 after listwise deletion). Reproduced by
special permission of the Publisher, MIND GARDEN, Inc., 1690 Woodside Road #202, Redwood City,
CA 90461 USA www.mindgarden.com from the Multifactor Leadership Questionnaire for Research by
Bernard M. Bass and Bruce J Avolio. Copyright 1995, 2000 by Bernard M Bass and Bruce J Avolio. All
rights reserved. Further reproduction is prohibited without Publisher's written permission.









(Bass & Avolio, 1997). Bass and Avolio (1997) stressed that the Goodness of Fit Index

of .91 for the full model exceeded the .90 cut-off criterion recommended in the literature

and that the Root Mean Square Residual of .04 of the full model satisfied the criterion

cut-off of less than .05 recommended by Joreskog and Sorbom (1989).

Table 3-3 Comparison of Overall Fit Measures Among Several Factor Models
Fit Measure One- Two- Three- Five- Nine-Factor
Factor Factor Factor Factor Model
Model Model Model Model (Full Model)

Chi-Square/df 5674/594 5260/593 3529/591 3341/584 2394/558
(0,8 5 W -4i (4258/593) (4229/591) (4126/584) (2967/558)
GFI .75 (.66) .77 (.81) .86 (.81) .86 (.82) .91 (.88)
AGFI .72 (.62) .74 (.79) .84 (.79) .84 (.80) .89 (.86)
RMSR .07 (.08) .08 (.07) .05 (.07) .05 (.07) .04 (.05)
Note. First values based on validation set of samples (N= 1,394). Values in parentheses are based on cross-
validation set of samples (N=1,490). GFI = Goodness of Fit Index; AGFI = Adjusted GFI; RMSR = Root
Mean Square Residual. Reproduced by special permission of the Publisher, MIND GARDEN, Inc., 1690
Woodside Road .2O2. Redwood City, CA 90461 USA www.mindgarden.com from the Multifactor
Leadership Questionnaire for Research by Bernard M. Bass and Bruce J Avolio. Copyright 1995, 2000 by
Bernard M Bass and Bruce J Avolio. All rights reserved. Further reproduction is prohibited without
Publisher's written permission.

One of the most comprehensive reviews of the MLQ to date was a meta-analysis

conducted by Lowe, Kroeck, and Sivasubramaniam (1996). The researchers looked at

approximately 40 studies from a variety of countries, institutions, and organizational

levels. They concluded that the MLQ is a valid and reliable measure of transformational,

transactional, and laissez-faire leadership (Lowe, Kroeck, & Sivasubramaniam, 1996).

Data Collection

Each dean of students at the institutions involved in the study was sent a letter of

introduction from the researcher. The letter explained the study, provided a copy of the

MLQ for the dean's review, and provided the dean with the contact information for the

researcher and the researcher's academic advisor.








All data was collected via the self-administered MLQ and the researcher-developed

demographic information sheet that were mailed to each member of the sample as part of

the survey packet. Participants, through the letter of introduction, were assured of

confidentiality and anonymity in the final reporting of results.

Survey packets were mailed to the professional staff members within the dean of

students office or its equivalent at each of the 38 institutions in the study during the late

spring and early summer of 2003. For institutions that had four or fewer professional

staff members in the Dean of Students Office, survey packets were sent to each

professional staff member. For institutions that had over four professional staff members

in their Dean of Students Office, four staff members were randomly selected and the

survey packets sent to those four staff members.

The survey packet included a cover letter, an informed consent form, a copy of

the Multifactor Leadership Questionnaire Rater Form (5x-Short), a demographic

information sheet, and a postage-paid return envelope. Twenty-one days after the initial

mailing a duplicate packet containing a reminder letter was sent to all members of the

sample who had not returned the survey packet materials. Twenty-four days after the

second mailing, a third and final packet was mailed to the remaining non-respondents.

Data Analysis

Research Question 1 examined the degree to which deans of students exhibited

transformational, transactional, and laissez-faire leadership behaviors. Data collected by

MLQ was analyzed using SPSS to compute the mean scores and standard deviations for

the leadership scales of the Multifactor Leadership Questionnaire.








Research questions 2, 3. and 4 employed multiple regression analyses to evaluate

the degree of relationship between each dependent variable (willingness to exert extra

effort, leadership effectiveness, and satisfaction with the leader) and the multiple

independent variables (transformation, transactional, and laissez-faire behaviors). SPSS

and SAS statistical software was utilized to run standard multiple regressions with the

composite independent variables of transformational leadership, transactional leadership,

and laissez-faire leadership. Further analyses focusing on the component behaviors that

comprise the composite independent variables were conducted using SAS multiple

regression for full and reduced models and all-possible subsets.

Research question 5 utilized independent-samples t-tests to examine whether

there was a relationship between gender and the perception of transformational leadership

behavior, transactional leadership behavior, or laissez-faire leadership behavior.

Human Subjects

All respondents of the study were assured of confidentiality in the handling and

reporting of results. No adverse affects were foreseeable by participating or refusing to

participate in the study; therefore, the risk to human subjects was considered to be

negligible. Approval to proceed with this study was secured through the Institutional

Review Board of the University of Florida before participants were contacted.














CHAPTER 4
ANALYSIS AND PRESENTATION OF DATA

The purpose of the study was to examine the leadership behaviors of deans of

students at public research universities in the southeast using Bass and Avolio's (1997)

Full Range of Leadership model. The study investigated the relationship between

transformational, transactional, and laissez-fair leadership behaviors of the deans and the

outcome variables of subordinate satisfaction with the leader, subordinate perception of

leadership effectiveness, and subordinate willingness to exert extra effort.

Specifically, this study addressed the following research questions:

1. To what extent do deans of students exhibit transformational, transactional, and
laissez-fair leadership behaviors as rated by their subordinates using the
Multifactor Leadership Questionnaire Rater Form (5X-Short)?

2. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' perception of leader
effectiveness as measured by the Multifactor Leadership Questionnaire Rater
Form (5X-Short)?

3. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' satisfaction with
their leader as measured by the Multifactor Leadership Questionnaire Rater Form
(5X-Short)?

4. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' willingness to exert
extra effort as measured by the Multifactor Leadership Questionnaire Rater Form
(5X-Short)?

5. Is there a relationship between gender and the transformational leadership
behavior, the transactional leadership behavior, or the laissez-faire leadership








behavior of deans of students as rated by their subordinates using the Multifactor
Leadership Questionnaire Rater Form (5X-Short)?

Survey Responses

The population for this study was student affairs professional staff members

supervised by deans of students at public institutions of higher education classified as

either Doctoral/Research Universities-Extensive or Doctoral/Research Universities-

Intensive in the 2000 Carnegie Classification (Carnegie Foundation, 2000) located in the

southeastern states of Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana,

Mississippi, North Carolina, South Carolina, Tennessee, and Virginia. The total number

of institutions representing this population is 45. Five institutions in this population did

not have an equivalent position to a dean of students for the purpose of this study and two

institutions had the position vacant at the time of the study. Therefore, 38 institutions

were used in the study.

Survey packets were mailed to the professional staff members within the Dean of

Students Office or its equivalent at each of the 38 institutions in the study during the late

spring and early summer of 2003. For institutions that had four or fewer professional

staff members in the Dean of Students Office, survey packets were sent to each

professional staff member. For institutions that had over four professional staff members

in their Dean of Students Office, four staff members were randomly selected and the

survey packets sent to those four staff members.

The survey packet included a cover letter, an informed consent form, a copy of

the Multifactor Leadership Questionnaire Rater Form (5x-Short), a demographic

information sheet, and a postage-paid return envelope. Twenty-one days after the initial

mailing a duplicate packet containing a reminder letter was sent to all members of the








sample who had not returned the survey packet materials. Twenty-four days after the

second mailing, a third and final packet was mailed to the remaining non-respondents.

Response Rates

The Multifactor Leadership Questionnaire Rater Form (5x-Short) was distributed

as a self-administered survey mailed to 137 professional staff members in the dean of

students office at the 38 institutions included in the study. The initial mailing and the two

reminder mailings resulted in 96 surveys being returned for a response rate of 70%.

Included in the calculation of the 70% response rate were the professional staff members

at seven institutions included in the study that did not return surveys for their respective

institutions. Therefore, while there were 38 institutions included in the study, the number

of deans evaluated was 31.

Demographic Information

The demographic information sheet was designed to obtain information from the

respondents on both the dean of students and the respondent in the areas of gender, age,

and education level. Information was also collected on the number of years the dean of

students had been in the dean position, the number of years the respondent had been in

student affairs, and the number of years that the respondent had worked with their current

dean. The demographic information sheet can be found in Appendix F.

The age demographic responses were divided into five categories. Table

4-1 presents the distribution of the respondents and deans with respect to age. The

majority of the respondents were in the 30's age group category and the majority of the

deans were in the 40's age group category. The deans tended to be in their 40's or older

and the respondents tended to be in their 40's or younger.









Table 4-1 Summary of Respondents and Deans' Age Distribution
Age Category 20 30 40 50 60

%of Deansa 0.00 6.45 51.61 35.48 6.45

% of Respondentsb 23.95 40.62 25.00 7.92 3.12
Note.
an =31. bn = 96.

The majority of the respondents were female with 57.3% of the sample being

female and 42.7% being male. Of the deans, 45.2% were female and 54.8% were male.

The respondents and deans' educational levels are presented in Table 4-2. The deans'

educational levels were Baccalaureate Degree only (BS/BA) 0.0%, Masters Degree

19.3%, Juris Doctor (JD) 9.7% and Doctorate (PhD/EdD) 71.0%. The respondents'

educational levels were Baccalaureate Degree (BS/BA) 10.4%, Masters Degree 67.7%,

Juris Doctor (JD) 5.2% and Doctorate (PhD/EdD) 16.6%.

Table 4-2 Summary of Educational Level for Deans and Respondents
BA/BS MS/MA JD PhD/EdD

Deans % 0.0 19.3 9.7 71.0

Respondents % 10.4 67.7 5.2 16.6
Note.
a b
n =31. n = 96.

The average length of time the respondents had worked in student affairs was 7.25

years. The average length of time the respondents had been working under the dean they

evaluated for the study was 3.8 years. The deans that were evaluated for the study had

served in the dean position for an average of 7.6 years.

Research Question 1

Research Question 1 examined the extent to which deans of students exhibit

transformational, transactional, and laissez-faire leadership behaviors as rated by their









subordinates using the Multifactor Leadership Questionnaire Rater Form (5X-Short).

Table 4-3 provides the means and standard deviations calculated for each of the

leadership behaviors in the Full Range of Leadership model. The higher the mean score

for the leadership behavior, the higher the subordinates' perception of the leadership

behavior being present in the leadership style of the deans.

The leadership behavior with the highest mean score was inspirational motivation

(2.94), indicating that it was the leadership behavior that subordinates perceived most

frequently in their dean of students. Among the transformational leadership behaviors,

inspirational motivation was followed by idealized influence-attributed (2.82), idealized

influence- behavior (2.75), individualized consideration (2.66), and intellectual

stimulation (2.61).

The transactional leadership behavior with the highest mean score was contingent

reward (2.64). The 2.64 mean score for contingent reward indicates it was perceived

more often in deans of students than the transformational leadership behavior of

intellectual stimulation, but perceived less often than the transformational leadership

behaviors of inspirational motivation, idealized influence-attributed, idealized influence-

behavior, and individualized consideration. Among the transactional leadership

behaviors, contingent reward was followed by management-by-exception passive (1.24),

and management-by-exception-active (1.15).

The least frequently exhibited behavior by deans of students as perceived by the

subordinates was laissez-faire leadership. Laissez-faire leadership behavior had a mean

score of.91.









Table 4-3 Deans of Students: Leadership Behaviors
Leadership Behavior Type Behavior

Transformational


Inspirational
Motivation

Idealized Influence
Attributed

Idealized Influence
Behavior

Individualized
Consideration

Intellectual
Stimulation


Transactional


Contingent
Reward

Management-by-
Exception
Passive

Management-by-
Exception
Active

Laissez-Faire


Laissez-Faire
Note. n = 96.


2.94 .66 1.00 4.00


2.82 .89 0.00 4.00


2.76 .77 1.00 4.00


2.66 .69 0.50 4.00


2.61

1.68


1.00

0.67


4.00

2.58


2.64 .69 0.75 4.00



1.24 .84 0.00 3.00


1.15

.91


0.00

0.00


3.00

3.50


Research Question 2

Research Question 2 examined whether there was a relationship between the

transformational leadership behavior, the transactional leadership behavior, or the laissez-

faire leadership behavior of deans of students and their subordinate professional staff


M

2.76


Min.

0.80


Max.

4.00








members' perception of leader effectiveness as measured by the Multifactor Leadership

Questionnaire Rater Form (5X-Short).

The initial analysis was performed between the dependent variable (leader

effectiveness) and the composite independent variables of transformational leadership,

transactional leadership, and laissez-faire leadership using standard multiple regression

analysis. The analysis was performed using both SAS Regression and SPSS Regression.

Further analyses were conducted using SAS multiple regression for full and reduced

models and all possible subsets.

The full and reduced model regression and the all-possible subsets regression

focused on the components in the Multifactor Leadership Questionnaire (MLQ) that

compose the composite independent variables of transformational leadership,

transactional leadership, and laissez-faire leadership. The transformational leadership

section of the MLQ is composed of five scales of four items each. The component

behaviors for transformational leadership are idealized influence-attributed (II-A),

idealized influence-behavior (II-B), inspirational motivation (IM), intellectual stimulation

(IS), and individualized consideration (IC). The transactional leadership section of the

MLQ is composed of three scales of four items each. The component behaviors for

transactional leadership are contingent reward (CR), management-by-exception active

(MBE-A), and management-by-exception passive (MBE-P). The laissez-faire (LF)

leadership section of the MLQ is a single scale with four items.

Assumptions were tested by examining normal probability plots of residuals and

scatter diagrams of residuals versus predicted residuals. No violations of normality,








linearity, or homoscedasticity of residuals were detected. In addition, box plots revealed

no evidence of outliers.

For the standard multiple regression analysis, no interaction effects between the

independent variables of transformational leadership, transactional leadership, and

laissez-faire leadership were found. The regression analysis revealed that the model

significantly predicted leader effectiveness, F(3,92) = 137.71, p< .0001. R2 for the model

was .818, and adjusted R2 was .812, indicating that the model accounts for 81.2% of the

variance for leadership effectiveness. Table 4-4 displays the unstandardized regression

coefficients (B), the standardized regression coefficients (P), the observed t value (t), the

significance level (p), and the semipartial correlations (sr) for each variable.

All three of the independent variables were significant at the p < .05 level. For

transformational leadership, 3 = .651, t = 12.22, and p = .0001. For transactional

leadership, P3 = .159, t = 2.99, and p = .004. For laissez-faire leadership, P3 = -.414, t = -

6.85, and p = .0001. Comparing across variables, transformational leadership was the

strongest predictor of leader effectiveness, and there was a significant negative

relationship between laissez-faire behavior and leadership effectiveness.

Under Bass and Avolio's (1997) Full Range of Leadership model, the dependent

variable of leader effectiveness is higher for leaders who are perceived as having higher

Table 4-4 Summary of Regression Analysis for Variables Predicting Leader
Effectiveness
Variable B 13 p sr
Transformational .863 .651 12.22 .0001* .544
Transactional .351 .159 2.99 .004* .133
Laissez-Faire -.441 -.414 -6.85 .0001 -.305
Note. n = 96. R2 = .818.
*p <.05.








scores on the transformational leadership component behaviors. Therefore, a multiple

regression analysis using full and reduced models was calculated to examine the extent to

which the transformational leadership component behaviors added to the predictive

accuracy for the dependent variable of leader effectiveness when they are added to the

transactional component behaviors and laissez-faire behavior.

The transformational leadership components significantly added to the predictive

accuracy for the dependent variable of leadership effectiveness when they were added to

the components for transactional leadership and laissez-faire leadership behavior. The F-

value was calculated at 9.88 with p < .05 and R2 increased by 8% when the

transformational components were added to the reduced model. In addition, it was found

that of the variance in leader effectiveness that was not associated with transactional and

laissez-faire leadership, 57% was associated with the transformational leadership

behaviors. Table 4-5 displays the unstandardized regression coefficients (B), the

standard errors (SE), the observed t values (t), and the significance levels (p) for the

component behaviors in the reduced model and the full model.

The component behaviors were analyzed using an all-possible subsets regression

with a R selection method to determine which groups of behaviors were the best

predictors of leader effectiveness. Table 4-6 displays the R2 and the adjusted R2 for the

best one-behavior model through the model that includes all nine behaviors. The highest

adjusted R2 was .851 for both the six-behavior model (II-A, II-B, IC, CR, MBE-A, LF)

and the seven-behavior model (II-A, II-B, IM, IC, CR, MBE-A, LF). However, the four-

behavior model was close to being as strong as a predictor of leader effectiveness as the

six-behavior and seven-behavior models with an adjusted R2 of .846.









Table 4-5 Summary of Full and Reduced Models Regression for
Adding Transformational Leadership Behaviors


Leader Effectiveness


Behavior B SE t p
Reduced
CR .870 .071 12.05 .0001
MBE-A .070 .060 1.16 .2488
MBE-P .059 .077 0.77 .4441
LF -.487 .077 -6.30 .0001*
Fullb
II-A .276 .077 1.77 .0005*
II-B .178 .091 1.96 .0532
IM -.076 .082 -0.94 .3523
IS .071 .110 0.64 .5214
IC .101 .106 0.95 .3452
CR .419 .085 4.92 .0001*
MBE-A .049 .052 0.94 .3505
MBE-P .054 .066 0.82 .4122
LF -.360 .068 -5.29 .0001*
Note.
n =96. n =96.
aR = .780. 'R2= .863.
*p < .05.

Table 4-6 Best Predictor Model by Number of Behaviors for Leader Effectiveness
Number
in Model R2 Adjusted R2 Behaviors in Model


.696
.776
.841
.853
.857
.860
.862
.862
.863


.693
.771
.835
.846
.850
.851
.851
.850
.849


II-A
II-A, CR
II-A, CR, LF
II-A, II-B, CR, LF
II-A, II-B, IC, CR, LF
II-A, II-B, IC, CR, MBE-A, LF
II-A, II-B, IM, IC, CR, MBE-A, LF
II-A, II-B, IM, IC, CR, MBE-A, MBE-P, LF
II-A, II-B, IM, IS, IC, CR, MBE-A, MBE-P, LF


Note. n =96.

Research Question 3

Research Question 3 examined whether there was a relationship between the

transformational leadership behavior, the transactional leadership behavior, or the laissez-

faire leadership behavior of deans of students and their subordinate professional staff


1
2
3
4
5
6
7
8
9








members' satisfaction with their leader as measured by the Multifactor Leadership

Questionnaire Rater Form (5X-Short).

The initial analysis was performed between the dependent variable (satisfaction

with the leader) and the composite independent variables of transformational leadership,

transactional leadership, and laissez-faire leadership using standard multiple regression

analysis. The analysis was performed using both SAS Regression and SPSS Regression.

Further analyses were conducted using SAS multiple regression for full and reduced

models and all-possible subsets.

The full and reduced model regression and the all-possible subsets regression

focused on the components in the MLQ that compose the composite independent

variables of transformational leadership, transactional leadership, and laissez-faire

leadership. The transformational leadership section of the MLQ is composed of five

scales of four items each. The component behaviors for transformational leadership are

idealized influence-attributed (II-A), idealized influence-behavior (II-B), inspirational

motivation (IM), intellectual stimulation (IS), and individualized consideration (IC). The

transactional leadership section of the MLQ is composed of three scales of four items

each. The component behaviors for transactional leadership are contingent reward (CR),

management-by-exception-active (MBE-A), and management-by-exception passive

(MBE-P). The laissez-faire (LF) leadership section of the MLQ is a single scale with

four items.

Assumptions were tested by examining normal probability plots of residuals and

scatter diagrams of residuals versus predicted residuals. No violations of normality,








linearity, or homoscedasticity of residuals were detected. In addition, box plots revealed

no evidence of outliers.

For the standard multiple regression analysis, no interaction effects between the

independent variables of transformational leadership, transactional leadership, and

laissez-faire leadership were found. The regression analysis revealed that the model

significantly predicted satisfaction with the leader, F(3,92) = 91.757,p< .0001. R2 for the

model was .750, and adjusted R2 was .741, indicating that the model accounts for 74.1%

of the variance for satisfaction with the leader. Table 4-7 displays the unstandardized

regression coefficients (B), the standardized regression coefficients (P3), the observed t

value (t), the significance level (p), and the semipartial correlation (sr) for each variable.

The independent variables of transformational leadership and laissez-faire

leadership were significant at the p < .05 level. For transformational leadership, P3 = .700,

t= 11.22, and p= .001. For laissez-faire, 3 = -.283, t = -3.99, andp = .0001.

Transactional leadership was not significant at the p < .05 level with P3 = .052, 1 = 0.83,

and p = .410. Comparing across variables, transformational leadership was the strongest

predictor of leader effectiveness, and there was a significant negative relationship

between laissez-faire behavior and leadership effectiveness.

Under Bass and Avolio's (1997) Full Range of Leadership model, the dependent

variable of satisfaction with the leader is higher for leaders who are perceived as having

Table 4-7 Summary of Regression Analysis for Variables Predicting Satisfaction with
the Leader
Variable B 13 p sr
Transformational .923 .700 11.22 .0001* .585
Transactional .113 .052 .83 .410 .043
Laissez-Faire -.299 -.283 -3.99 .0001* -.208
Note. n = 96. R2 =.750.
*p < .05.








higher scores on the transformational leadership component behaviors. Therefore, a

multiple regression analysis using full and reduced models was calculated to examine the

extent to which the transformational leadership component behaviors added to the

predictive accuracy for the dependent variable of satisfaction with the leader when they

were added to the transactional component behaviors and laissez-faire behavior.

The transformational leadership components significantly added to the predictive

accuracy for the dependent variable of satisfaction with the leader when they were added

to the components for transactional leadership and laissez-faire leadership behavior. The

F-value was calculated at 13.69 with p < .05 and R2 increased by 15.7% when the

transformational components were added to the reduced model. In addition, it was found

that of the variance in satisfaction with the leader that was not associated with

transactional and laissez-faire leadership, 79.6% was associated with the transformational

leadership behaviors. Table 4-8 displays the unstandardized regression coefficients (B),

the standard errors (SE), the observed t values (t), and the significance levels (p) for the

component behaviors in the reduced model and the full model.

The component behaviors were analyzed using an all-possible subsets regression

with a R2 selection method to determine which groups of behaviors were the best

predictors of satisfaction with the leader. Table 4-9 displays the R2 and the adjusted R2

for the best one-behavior model through the model that includes all nine behaviors. The

highest adjusted R2 was .849 for the full nine-behavior model. The second highest

adjusted R2 was .789 for both the four-behavior (II-A, II-B, CR, LF) and five-behavior

(II-A, II-B, IM, CR, LF) models.









Table 4-8 Summary of Full and Reduced Models Regression for Satisfaction with the
Leader Adding Transformational Leadership Behaviors
Behavior B SE t p
Reduced
CR .760 .089 8.54 .0001*
MBE-A -.022 .076 -0.29 .7713
MBE-P .052 .097 0.53 .5974
LF -.420 .098 -4.31 .0001*
Fullb
II-A .467 .091 5.11 .0001*
II-B .216 .109 1.99 .0496*
IM -.104 .097 -1.07 .2875
IS .107 .131 0.82 .4159
IC -.002 .127 -0.01 .9895
CR .204 .102 2.00 .0486*
MBE-A -.035 .062 -0.56 .5739
MBE-P .034 .079 0.43 .6671
LF -.235 .081 -2.89 .0048*
Note.
n = 96. bn = 96.
`R' = .645. bR2= .802.
*p <.05.

Table 4-9 Best Predictor Model by Number of Behaviors for Satisfaction with the Leader


Adjusted R2
.719
.752
.778
.789
.789
.788
.786
.784
.849


Behaviors in Model


II-A
II-A, LF
II-A, II-B, LF
II-A, II-B, CR, LF
II-A, II-B, IM, CR, LF
II-A, II-B, IM, IS, CR, LF
II-A, II-B, IM, IS, CR, MBE-A. LF
II-A, II-B, IM. IS, CR, MBE-A, MBE-P, LF
II-A, II-B, IM, IS, IC, CR, MBE-A, MBE-P, LF


Research Question 4

Research Question 4 examined whether there was a relationship between the

transformational leadership behavior, the transactional leadership behavior, or the laissez-

faire leadership behavior of deans of students and their subordinate professional staff


Number
in Model
1
2

4
5
6
7
8
9
Note. n = 96.


R2
.722
.757
.785
.798
.800
.801
.802
.802
.863









members' willingness to exert extra effort as measured by the Multifactor Leadership

Questionnaire Rater Form (5X-Short).

The initial analysis was performed between the dependent variable (willingness to

exert extra effort) and the composite independent variables of transformational

leadership, transactional leadership, and laissez-faire leadership using standard multiple

regression analysis. The analysis was performed using both SAS Regression and SPSS

Regression. Further analyses were conducted using SAS multiple regression for full and

reduced models and all possible subsets.

The full and reduced model regression and the all possible subsets regression

focused on the components in the MLQ that compose the composite independent

variables of transformational leadership, transactional leadership, and laissez-faire

leadership. The transformational leadership section of the MLQ is composed of five

scales of four items each. The component behaviors for transformational leadership are

idealized influence-attributed (II-A). idealized influence-behavior (II-B), inspirational

motivation (IM), intellectual stimulation (IS), and individualized consideration (IC). The

transactional leadership section of the MLQ is composed of three scales of four items

each. The component behaviors for transactional leadership are contingent reward (CR),

management-by-exception active (MBE-A), and management-by-exception passive

(MBE-P). The laissez-faire (LF) leadership section of the MLQ is a single scale with

four items.

Assumptions were tested by examining normal probability plots of residuals and

scatter diagrams of residuals versus predicted residuals. No violations of normality,








linearity, or homoscedasticity of residuals were detected. In addition, box plots revealed

no evidence of outliers.

For the standard multiple regression analysis, an interaction effect between the

independent variables transformational leadership and transactional leadership was found

to be significant at thep < .05 level (p = .0114, 1 = -2.58). However, adding the

interaction term to the model only increased predictability of extra effort by 1.7%;

therefore, the interaction term was excluded from the model for the purposes of this

study. No other interaction effects between the independent variables were significant.

The regression analysis revealed that the model significantly predicted

willingness to exert extra effort, F(3,92) = 88.947, p < .0001. R2 for the model was .744,

and adjusted R2 was .735, indicating that the model accounts for 73.5% of the variance

for willingness to exert extra effort. Table 4-10 displays the unstandardized regression

coefficients (B). the standardized regression coefficients (13), the observed t value (t), the

significance level (p), and the semipartial correlation (sr) for each variable.

All three of the independent variables were significant at thep < .05 level. For

transformational leadership, 13 = .699, t = 11.07, and p = .0001. For transactional

leadership, 13 = .158, t = 2.50, andp = .014. For laissez-faire leadership, 3 = -.283, t = -

3.94, andp = .0002. Comparing across variables, transformational leadership was the

strongest predictor of willingness to exert extra effort, and there was a significant

negative relationship between laissez-faire behavior and willingness to exert extra effort.

Under Bass and Avolio's (1997) Full Range of Leadership model, the dependent variable

of willingness to exert extra effort is higher for leaders who are perceived as having

higher scores on the transformational leadership component behaviors. Therefore, a









Table 4-10 Summary of Regression Analysis for Variables Predicting Willingness to
Exert Extra Effort
Variable B (3 t p sr
Transformational 1.029 .699 11.07 .0001* .584
Transactional .387 .158 2.50 .014* .132
Laissez-Faire -.334 -.283 -3.94 .0002* -.208
Note. n = 96. R2 = .744.
*p <.05.

multiple regression analysis using full and reduced models was calculated to examine the

extent to which the transformational leadership component behaviors added to the

predictive accuracy for the dependent variable of willingness to exert extra effort when

they are added to the transactional component behaviors and laissez-faire behavior.

The transformational leadership components significantly added to the predictive

accuracy for the dependent variable of willingness to exert extra effort when they were

added to the components for transactional leadership and laissez-faire leadership

behavior. The F-value was calculated at 13.32 with < .05 and R2 increased by 16.3%

when the transformational scales were added to the reduced model. In addition, it was

found that of the variance in willingness to exert extra effort that was not associated with

transactional and laissez-faire leadership, 77.5% was associated with the transformational

leadership behaviors. Table 4-11 displays the unstandardized regression coefficients (B),

the standard errors (SE), the observed t values (t), and the significance levels (p) for the

component behaviors in the reduced model and the full model.

The component behaviors were analyzed using an all-possible subsets regression

with a R selection method to determine which groups of behaviors were the best

predictors of willingness to exert extra effort. Table 4-12 displays the R2 and the adjusted

R2 for the best one-behavior model through the model that includes all nine behaviors.









Table 4-11 Summary of Full and Reduced Models Regression for Willingness to
Exert Extra Effort Adding Transformational Leadership Behaviors
Behavior B SE t p
Reduced
CR .918 .102 9.01 .0001*
MBE-A .071 .087 0.81 .4201
MBE-P .146 .112 1.30 .1953
LF -.478 .112 -4.28 .0001*
Full'b
II-A .228 .105 2.16 .0334*
II-B .275 .125 2.20 .0305*
IM -.185 .112 -1.65 .1028
IS .086 .152 0.56 .5736
IC .477 .146 3.26 .0016*
CR .240 .117 2.05 .0433*
MBE-A -.006 .072 -0.09 .9297
MBE-P .155 .091 1.71 .0914
LF -.319 .094 -3.41 .0010*
Note.
an =96. bn =96.
OR2= .627. bR2= .790.
*p<.05.

The highest adjusted R2 was .772 for the seven-behavior model (II-A, II-B, IM, IC, CR,

MBE-P, LF). However, the adjusted R2 for the five-behavior model (II-A, II-B, IC, CR,

LF) of .763 and the adjusted R2 for the six-behavior model (II-A, II-B, IM, IC, CR, LF) of

.768 are within 1% of the seven-behavior adjusted R2.

Table 4-12 Best Predictor Model by Number of Behaviors for Willingness to Exert Extra
Effort


Adjusted R2
.650
.721
.745
.757
.763
.768
.772
.770
.768


Behaviors in Model
IC
II-A, IC
II-A, IC, LF
II-A, II-B, IC, LF
II-A, II-B, IC, CR, LF
II-A, II-B, IM, IC, CR, LF
II-A, II-B, IM, IC, CR, MBE-P, LF
II-A, II-B, IM, IS, IC, CR, MBE-P, LF
II-A, II-B, IM, IS, IC, CR, MBE-A, MBE-P, LF


Number
in Model
1
2
3
4
5
6
7
8
9
Note. n = 96.


R2
.654
.727
.753
.768
.776
.782
.789
.790
.790








Research Question 5

Research Question 5 examined whether there was a relationship between gender

and transformational leadership behavior, the transactional leadership behavior, or the

laissez-faire leadership behavior of deans of students as rated by their subordinates using

the Multifactor Leadership Questionnaire Rater Form (5X-Short).

The analysis was completed by evaluating each of the composite leadership

behaviors (transformational, transactional, and laissez-faire) separately. For each

leadership behavior, independent-samples t-tests using SPSS were run: (a) with the test

variable being the overall mean score for the leadership behavior and the grouping

variable being gender of participant; (b) with the test variable being the mean score for

the leadership behavior and the grouping variable being the gender of the dean; (c) with

the test variable being the mean score for male deans only and the grouping variable

being gender of participant; and (d) with the test variable being the mean score for female

deans only and the grouping variable being gender of the participant.

The t-tests did not reveal any significant differences in how deans' leadership

behaviors were perceived depending on either the deans' gender or on the participants'

gender. For transformational leadership, there was no significant difference in the mean

score for all deans as perceived by male participants (M= 2.75, SD = .616) and female

participants (M= 2.76, SD = .716) with t(94) = -0.080, and p = .936. There was no

significant difference between the mean scores for male deans (M = 2.69, SD = .654) and

female deans (M = 2.84, SD = .690) as perceived by all participants with t(94) = -1.087,

and p = .280. There was no significant difference between the mean scores for male

deans (M= 2.61, SD = .726) and female deans (M= 2.92, SD = .684) as perceived by








female participants with t(53) = -1.633, andp = .108. There was no significant difference

between the mean scores for male deans (M= 2.78, SD = .567) and female deans (M

2.71, SD = .703) as perceived by male participants with t(39) = 0.380, and p = .706.

There was no significant difference in the mean scores for female deans as perceived by

male participants (M= 2.71, SD = .703) and female participants (M = 2.92, SD = .684)

with t(42) = -0.969, andp = .338. There was no significant difference in the mean scores

for male deans as perceived by male participants (M= 2.78, SD = .567) and female

participants (M= 2.61, SD = .726) with t(50) = 0.971, and p = .336.

For transactional leadership, there was no significant difference in the mean score

for all deans as perceived by male participants (M= 1.70, SD = .410) and female

participants (M= 1.67, SD =.402) with t(94) = 0.382, andp = .703. There was no

significant difference between the mean scores for male deans (M = 1.74, SD = .334) and

female deans (M = 1.60, SD = .466) as perceived by all participants with t(94) = 1.637,

and p = .106. There was no significant difference between the mean scores for male

deans (M= 1.75, SD = .296) and female deans (M= 1.58, SD = .475) as perceived by

female participants with t(53) = 1.56, and p = .126. There was no significant difference

between the mean scores for male deans (M= 1.73 SD = .377) and female deans (M =

1.64, SD = .464) as perceived by male participants with t(39) =.702, and p = .487. There

was no significant difference in the mean scores for female deans as perceived by male

participants (M= 1.64, SD = .464) and female participants (M= 1.58, SD = .475) with

t(42) = 0.388, andp =.700. There was no significant difference in the mean scores for

male deans as perceived by male participants (M= 1.73, SD = .376) and female

participants (M= 1.75, SD = .296) with t(50) = 0.178 andp =.859.









For laissez-faire leadership, there was no significant difference in the mean score

for all deans as perceived by male participants (M = .884. SD = .806) and female

participants (M= .932. SD =.867) with t(94) = -0.275, and p =.784. There was no

significant difference between the mean scores for male deans (M= .942, SD = .847) and

female deans (M = .875, SD = .834) as perceived by all participants with t(94) = 0.391,

and p = .697. There was no significant difference between the mean scores for male

deans (M= 1.14, SD = .929) and female deans (M= .732, SD = .767) as perceived by

female participants with t(53) = 1.774, and p = .082. There was no significant difference

between the mean scores for male deans (M= .730, SD = .707) and female deans (M=

1.125 SD .913 ) as perceived by male participants with t(39) = -1.557, and p =.128.

There was no significant difference in the mean scores for female deans as perceived by

male participants (M= 1.125, SD = .913) and female participants (M= .732, SD = .766)

with t(42) = 1.525, andp = .135. There was no significant difference in the mean scores

for male deans as perceived by male participants (M = .730, SD = .707) and female

participants (M= 1.14, SD =.929) with t(50) = -1.776, andp = .082.

The findings imply that there is no significant relationship between gender and the

transformational leadership behavior, transactional leadership behavior, or laissez-faire

leadership of deans of students as perceived by their professional staff members.

Summary

A total of 137 surveys were mailed to professional staff members working in a

dean of students office at 38 public institutions of higher education classified as either

Doctoral/Research Universities-Extensive or Doctoral/Research Universities-Intensive in

the 2000 Carnegie Classification (Carnegie Foundation, 2000) located in the southeastern









states of Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, North

Carolina, South Carolina, Tennessee, and Virginia. Of the 137 professional staff

members surveyed, 96 returned completed surveys resulting in a response rate of 70%.

The data provided information on the extent to which deans of students exhibited

transformational, transactional, and laissez-faire leadership behaviors as perceived by the

professional staff members in the deans' office. Multiple regression analyses were used

to examine the relationship between the transformational, transactional, and laissez-faire

leadership behaviors of deans of students and their subordinate professional staff

members' perception of leader effectiveness, satisfaction with the leader, and willingness

to exert extra effort. The data was also evaluated to examine whether there was a

relationship between gender and transformational leadership behavior, the transactional

leadership behavior, or the laissez-faire leadership behavior of deans of students.

Chapter 5 follows with a summary of the study and a discussion of the findings.

Suggestions for future research are also presented.














CHAPTER 5
SUMMARY AND DISCUSSION

This chapter presents a summary and discussion of the findings, implications for

student affairs, and suggestions for future research. The purpose of the study was to

examine the leadership behaviors of deans of students at public research universities in

the southeast using Bass and Avolio's (1997) Full Range of Leadership model. The

study investigated the relationship between transformational, transactional, and laissez-

fair leadership behaviors of the deans and the outcome variables of subordinate

satisfaction with the leader, subordinate perception of leadership effectiveness, and

subordinate willingness to exert extra effort.

Specifically, this study addressed the following research questions:

1. To what extent do deans of students exhibit transformational, transactional, and
laissez-fair leadership behaviors as rated by their subordinates using the
Multifactor Leadership Questionnaire Rater Form (5X-Short)?

2. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' perception of leader
effectiveness as measured by the Multifactor Leadership Questionnaire Rater
Form (5X-Short)?

3. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' satisfaction with
their leader as measured by the Multifactor Leadership Questionnaire Rater Form
(5X-Short)?

4. Is there a relationship between the transformational leadership behavior, the
transactional leadership behavior, or the laissez-faire leadership behavior of deans
of students and their subordinate professional staff members' willingness to exert








extra effort as measured by the Multifactor Leadership Questionnaire Rater Form
(5X-Short)?

5. Is there a relationship between gender and the transformational leadership
behavior, the transactional leadership behavior, or the laissez-faire leadership
behavior of deans of students as rated by their subordinates using the Multifactor
Leadership Questionnaire Rater Form (5X-Short)?

The study examined the leadership behaviors of deans of students at public

institutions of higher education classified as either Doctoral/Research Universities-

Extensive or Doctoral/Research Universities-Intensive in the 2000 Carnegie

Classification (Carnegie Foundation, 2000) located in the southeastern states of Alabama,

Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, South

Carolina, Tennessee, and Virginia. The total number of institutions representing this

population is 45. Five institutions in this population did not have an equivalent position

to a dean of students for the purpose of this study and two institutions had the position

vacant at the time of the study. Therefore, 38 institutions were used in the study.

The study utilized Bass and Avolio's (1995) Multifactor Leadership

Questionnaire Rater Form (5X-Short) to assess transformational, transactional, and

laissez-faire leadership behaviors of deans of students as perceived by their professional

staff member subordinates. A researcher-developed demographic sheet was used to

obtain the age, educational level, and gender of the participants and of the deans that

were evaluated. Statistical analyses were conducted using both SPSS and SAS statistical

software.

Summary and Discussion of Findings

This study assessed the transformational, transactional, and laissez-faire

leadership behaviors of deans of students using Bass and Avolio's (1997) Full Range of









Leadership model. According to the Full Range of Leadership model, every leader

displays each of the leadership behaviors to some degree (Bass, 1998). However, the

leader with an optimal profile infrequently displays laissez-faire leadership and displays

successively higher frequencies of the transactional behaviors of management-by-

exception (passive), management-by-exception (active), and contingent reward. The

optimal leader profile displays the five transformational leadership behaviors of idealized

influence-attributed, idealized influence-behavior, inspirational motivation, intellectual

stimulation, and individualized consideration the most frequently (Bass, 1998).

Under the Full Range of Leadership model, the hierarchy of correlations of the

Multifactor Leadership Questionnaire (MLQ) components with the outcome variables of

leader effectiveness, satisfaction with the leader, and willingness to exert extra effort is

typically transformational behaviors > contingent reward > management-by-exception

active > management-by-exception passive > laissez-faire leadership (Bass & Avolio,

1997). In addition. Bass (1985, 1998) states that transformational leadership augments

transactional leadership in predicting leadership behavior effects on the outcome

variables. Specifically, the augmentation effect states that transformational leadership

behaviors should account for unique variance in subordinates' ratings of the outcome

variables over and above that accounted for by transactional leadership (Bass, 1998).

The results of this study produced a profile of leadership behaviors for deans of

students that was similar to, but did not match exactly, Bass and Avolio's (1997) optimal

leader profile. The results provide evidence of the augmentation effect of

transformational leadership behaviors on subordinate ratings of the outcome variables

and provide no evidence of a relationship between gender and rating of deans of students'









leadership behaviors. Conclusions drawn from the results of this study are discussed in

relation to the research questions that served as the basis for the study.

Leadership Behaviors Exhibited by Deans of Students

The findings suggest that deans of students at public research universities in the

southeast, as a group, exhibited transformational behaviors more frequently than

transactional behaviors, which they displayed more frequently than laissez-faire

leadership behaviors. The mean scores for deans of students' leadership behavior

presented in Table 4-3 indicate that deans of students exhibited transformational

leadership "fairly often" (M= 2.76, SD = .67), transactional leadership "sometimes" (M=

1.68, SD = .40), and laissez-faire leadership "once in a while" (M= 0.91, SD = .84).

Therefore, the composite independent variables followed the Full Range of Leadership

model optimal profile which calls for transformational leadership behaviors to be

displayed more frequently than transactional leadership behaviors, which should be

displayed more frequently than laissez-faire leadership behavior (Bass & Avolio, 1997).

However, the mean scores for the component behaviors of the composite independent

variables did not exactly match the optimal profile.

The mean scores for the component behaviors of the composite independent

variables indicate that one of the transactional behaviors, contingent reward (M= 2.64,

SD = .69), was exhibited more than the transformational behavior of intellectual

stimulation (M= 2.61, SD = .78). While this finding prevents an exact match with the

optimal leadership profile proposed by Bass and Avolio (1997), it is not necessarily a

negative finding. Research on transformational leadership indicates that the frequency

with which transformational and transactional leadership behaviors emerge and are








effective can depend to some extent on the work environment, the organization, the goals

and tasks involved, and the distribution of power between the leaders and followers

(Bass, 1998).

Among the several factors that could have contributed to the relatively high mean

score for contingent reward is the possible stability of the deans' offices due to the deans'

length of time in the dean position. For this study, the average number of years the deans

of students had served in their position was 7.6 years. Therefore, due to their years of

experience in the dean position, the environment faced by the deans was a fairly

predictable and stable environment. Bass (1998) states that transactional leadership is

likely to emerge more frequently and be relatively effective when leaders are engaged in

a stable and predictable environment. What makes the dean of students position unique,

is that while it may operate in a stable and predictable environment, it often deals with

conditions of crisis and uncertainty which creates conditions that research indicates

makes the emergence of transformational leadership behaviors more likely (Bass, 1998).

Therefore, the high mean scores for the transformational leadership behaviors as well as

the transactional leadership behavior of contingent reward may be a reflection of the

environment within which the deans of students serve as leaders.

Additionally, the nature of the leader-subordinate relationship within a dean of

students office may provide some insight to the relatively high contingent reward mean

score. Bass (1998) speculates that in situations where the subordinate has power and

information, transactional leadership will emerge more frequently than in situations

where the leader retains most of the power and information. Sandeen (1996) states that

the dean of students often overseas several of the common functions found in a student









affairs division. These functions include units such as financial aid, services for students

with disabilities, and judicial affairs, which require the professional staff members within

the unit to have a high degree of knowledge and expertise in that particular area.

Therefore, a dean of students at a public research institution is typically supervising

individuals that have power and information within a specific area of responsibility.

Leader Effectiveness as Measured by the Multifactor Leadership Questionnaire

The initial analysis for examining whether there was a relationship between the

transformational leadership behavior, the transactional leadership behavior, or the laissez-

faire leadership behavior of deans of students and their subordinate professional staff

members' perception of leader effectiveness as measured by the Multifactor Leadership

Questionnaire Rater Form (5X-Short) was done using SAS and SPSS standard regression.

The analysis revealed that the model significantly predicted leader effectiveness with the

composite independent variables accounting for 81.2% of the variance for leadership

effectiveness. For leader effectiveness, all three of the variables were significant at the p

< .05 level. However, transformational leadership was the strongest predictor of leader

effectiveness with a standardized regression coefficient of .651 compared to a

standardized regression coefficient for transactional leadership of .159. Therefore, the

composite variable of transformational leadership was a stronger contributor to the model

in predicting leader effectiveness than the composite variable of transactional leadership.

There was a significant negative relationship between laissez-faire behavior and

leadership effectiveness with laissez-faire leadership having a standardized regression

coefficient of-.414 and a semipartial correlation of-.305. This indicates that deans who

frequently exhibit laissez-faire leadership behaviors are more likely to be perceived as









ineffective by their professional staff members than dean's that rarely display such

behavior.

Further analyses were performed on the outcome variable of leader effectiveness

using SAS full and reduced model regression and SAS all possible subsets regression.

These analyses used the component behaviors of transformational and transactional

leadership to examine which component behaviors were having the strongest influence

on predicting leader effectiveness. While the full and reduced model regression did find

that the transformational leadership components significantly added to the predictive

accuracy of leader effectiveness, the only transformational leadership behavior

component that was found to be significant in the full model at the p < .05 level, was

Idealized Influence Attributed. The transactional leadership behavior of contingent

reward was also significant at the p < .05 level, as was laissez-faire leadership.

The significant predictive value of idealized influence-attributed and contingent

reward was also seen in the all-possible subsets regression. For the one-behavior model,

idealized influence-attributed was the best predictor of leader effectiveness with an

adjusted R2 of .693. For the two-behavior model, idealized influence-attributed and

contingent reward were the best predictors of leader effectiveness with an adjusted R2 of

.771.

The findings of the full and reduced model regression and the all-possible subsets

indicate that, for the participants in this study, the transactional behavior of contingent

reward was a stronger predictor of leader effectiveness than several of the

transformational leadership component behaviors. This finding is not altogether

unexpected since Bass (1985, 1998) states that in general the best leaders are both









transformational and transactional, and that contingent reward behavior plays an

important role in effective leadership. However, it is not what the Full Range of

Leadership model would have predicted and it runs counter to much of the research on

the Full Range of Leadership model (Bass, 1998; Waldman, Bass, & Yammarino, 1990;

Howell & Avolio, 1993).

The findings on contingent reward suggest that there is something about the

nature of the work within a dean of students office that supports the transactional

behavior of contingent reward being a fairly strong predictor of leader effectiveness.

Avolio, Bass, and Jung (1999) speculated that contingent reward could be the foundation

through which leaders build trust and structure the professional development expectations

of their followers. Studies that examined factors contributing to attrition in student

affairs work found that the work is often associated with long hours and stressful

conditions and that professional staff members are often dissatisfied with professional

development opportunities (Barr, 1990; Carpenter, 1990; Bender, 1980). Therefore, one

possible reason that contingent reward emerged as a stronger predictor of leader

effectiveness than expected is that it is effective in developing trust in a stressful work

environment and in clarifying professional development opportunities for professional

staff members within a dean of students office.

Satisfaction with the Leader as Measured by the Multifactor Leadership Questionnaire

The initial analysis for examining whether there was a relationship between the

transformational leadership behavior, the transactional leadership behavior, or the laissez-

faire leadership behavior of deans of students and their subordinate professional staff

members' perception of satisfaction with the leader as measured by the Multifactor









Leadership Questionnaire Rater Form (5X-Short) was done using SAS and SPSS

standard regression.

The analysis revealed that the model significantly predicted satisfaction with the

leader with composite independent variables accounting for 74.1% of the variance for

satisfaction with the leader. The independent variables of transformational leadership

and laissez-faire leadership were significant at the p < .05 level. The independent

variable of transactional leadership was not significant at the p < .05 level.

Transformational leadership was the strongest predictor of satisfaction with the leader

with a standardized regression coefficient of .700.

There was a significant negative relationship between laissez-faire behavior and

satisfaction with the leader with laissez-faire leadership having a standardized regression

coefficient of-.283. This indicates that professional staff members who perceive their

dean as frequently exhibiting laissez-faire leadership behaviors are less likely to be

satisfied with their dean as a leader.

Further analyses were performed on the outcome variable of satisfaction with the

leader using SAS full and reduced model regression and SAS all possible subsets

regression. These analyses used the component behaviors of transformational and

transactional leadership to examine which component behaviors were having the

strongest influence on predicting satisfaction with the leader. The full and reduced model

regression found that the transformational leadership components significantly added to

the predictive accuracy of satisfaction with the leader. The transformational leadership

behavior components that were found to be significant in the full model at the p < .05

level, were idealized influence-attributed and idealized influence-behavior. The








transactional leadership behavior of contingent reward was also significant at the p < .05

level, as was laissez-faire leadership.

The significant predictive value of idealized influence-attributed, idealized

influence-behavior, contingent reward, and laissez-faire leadership was seen in the in the

all-possible subsets regression. For the four-behavior model these components were the

best predictor of satisfaction with the leader with an adjusted R2 of .789. The only model

with a higher adjusted R2 was the full nine-behavior model. For the one-behavior model,

idealized influence-attributed was the best predictor of satisfaction with the leader with

an adjusted R2 of .719. For the two-behavior model, idealized influence-attributed and

laissez-faire behavior were the best predictors of satisfaction with the leader with an

adjusted R2 of .752. Idealized influence-attributed, idealized influence-behavior, and

laissez-faire comprised the best three-behavior model with an R2 of .778.

The findings of the full and reduced model regression and the all-possible subsets

regression indicate that for the outcome variable of satisfaction with the leader, the results

were closer to what the Full Range of Leadership model would predict than the findings

for leader effectiveness. The transactional behavior of contingent reward was found to be

a significant predictor, but it was not as strong as predictor as it was for leader

effectiveness. The transformational leadership behaviors of idealized influence-attributed

and idealized influence-behavior, were stronger predictors than contingent reward and for

the standard multiple regression with the composite independent variables, transactional

leadership was not a significant predictor.

The strong predictive value of the idealized influence components of

transformational leadership for satisfaction with the leader is consistent with Avolio,









Bass, and Jung's (1999) findings on the component behaviors of the Full Range of

Leadership model. Idealized influence is a charismatic type of leadership that builds

followers' identification with the leader and the leader's vision. It is characterized by the

leader establishing himself or herself as role model through exhibiting high ethical

standards and it energizes followers as well as providing them with a clear sense of

purpose (Avolio, Bass, & Jung, 1999).

Willingness to Exert Extra Effort as Measured by the Multifactor
Leadership Questionnaire

The initial analysis for examining whether there was a relationship between the

transformational leadership behavior, the transactional leadership behavior, or the laissez-

faire leadership behavior of deans of students and their subordinate professional staff

members' willingness to exert extra effort as measured by the Multifactor Leadership

Questionnaire Rater Form (5X-Short) was done using SAS and SPSS standard regression.

The analysis revealed that the model significantly predicted willingness to exert extra

effort with the composite independent variables accounting for 73.5% of the variance for

willingness to exert extra effort. All three of the independent variables were significant

at the p < .05 level. However, transformational leadership was the strongest predictor of

willingness to exert extra effort with a standardized regression coefficient of .699

compared to a standardized regression coefficient for transactional leadership of .158.

Therefore, the composite variable of transformational leadership is a stronger contributor

to the model in predicting willingness to exert extra effort than the composite variable of

transactional leadership.

There was a significant negative relationship between laissez-faire behavior and

willingness to exert extra effort with laissez-faire leadership having a standardized









regression coefficient of-.283 and a semipartial correlation of-.208. This indicates that

deans who frequently exhibit laissez-faire leadership behaviors are less likely to have

subordinates that are willing to exert extra effort than deans that rarely display such

behavior.

Further analyses were performed on the outcome variable of willingness to exert

extra effort using SAS full and reduced model regression and SAS all possible subsets

regression. These analyses used the component behaviors of transformational and

transactional leadership to examine which component behaviors were having the

strongest influence on predicting subordinate willingness to exert extra effort. The full

and reduced model regression did find that the transformational leadership components

significantly added to the predictive accuracy for willingness to exert extra effort.

Adding the transformational components to the model increased R2 by 16.3%, and of the

variance in willingness to exert extra effort that was not associated with transactional and

laissez-faire leadership, 77.5% was associated with the transformational leadership

behaviors. The transformational leadership behavior components that were found to be

significant in the full model at the p < .05 level, were idealized influence-attributed,

idealized influence-behavior, and individualized consideration. The transactional

leadership behavior of contingent reward was also significant at the p < .05 level, as was

laissez-faire leadership.

The predictive value of individualized consideration, idealized influence-

attributed, idealized influence-behavior, contingent reward, and laissez-faire was seen in

the all-possible subsets regression. For the five-behavior model, these components were

the best predictors of willingness to exert extra effort with an adjusted R2 of .763. The









best one-behavior model was individual consideration with an adjusted R2 of .650. The

best two-behavior model was individual consideration and idealized influence-attributed

with an adjusted R2 of .721. The best three-behavior model was individual consideration,

idealized influence-attributed, and laissez-faire with an adjusted R2 of .745. The best

four-behavior model was individual consideration, idealized influence-attributed, laissez-

faire, and contingent reward with an adjusted R2 of .757.

The full and reduced model regression and the all-possible subsets regression

indicate that individual consideration was the strongest predictor of professional staff

members' willingness to exert extra effort. Bass (1985) discussed individualized

consideration in terms of the leader continuously assisting followers in developing the

followers' full potential and the leader focusing on the needs of each follower as an

individual. Therefore, although individualized consideration was not found as a

significant predictor of the other outcome variables, it is consistent with the Full Range of

Leadership model that it would serve as a significant predictor for willingness to exert

extra effort.

Gender and Perception of Leadership Style

This study utilized independent samples t-test in examining how deans' leadership

behaviors were perceived depending on either the dean's gender or the participant's

gender. The findings indicate that there was no significant difference in how male and

female deans of students were rated overall by their professional staff members and that

there was no significant difference in the way male and female professional staff

members rated their deans. These findings are consistent with those of Maher (1997) that









found there were no differences between men and women leaders in displaying

transformational behaviors.

Other studies have found that women tend to be more transformational than their

male counterparts (Bass, 1999). Bass (1999) speculated that the findings from studies

that found women tend to be more transformational examined organizations dominated

by males. In these situations, women may have had to be better leaders than their male

counterparts to attain the same leadership positions. Bass (1999) stated that more studies

needed to be done to examine what happens when women are in the majority. For this

study, female deans were a slight majority at 53.9%. Therefore, this study may indicate

that that the finding of previous research that females are more transformational does not

hold true when women are in the majority.

Implications for Student Affairs

The results of this study suggest that professional staff members working within a

dean of students office, or its equivalent, are more willing to exert extra effort, have

higher levels of satisfaction with the dean of students, and view the dean's leadership as

more effective when the dean utilizes transformational leadership behaviors more

frequently than transactional leadership behaviors. While the study may not be entirely

conclusive, the results are consistent with the theoretical prediction of Bass and Avolio's

(1997) Full Range of Leadership model. The model predicts that leaders who are more

transformational and less transactional are more effective as leaders and more satisfying

to their followers (Bass & Avolio, 1997). Where the results differ from the theoretical

prediction of the model is on the predictive value of the transactional behavior of

contingent reward. The findings indicate that contingent reward is a stronger predictor of








the outcome variables of leader effectiveness, willingness to exert extra effort, and

satisfaction with the leader than the model would have predicted.

The results have implications for student affairs since using effective leadership

practices improves the work experience of subordinates (Daft, 1999; Fiedler, 1974;

Hersey & Blanchard, 1988; Nahavandi, 1997; Seltzer & Bass, 1990; Stogdill,1974).

Employee satisfaction has been found to be a key indicator of the total quality of an

organization, and low employee moral negatively impacts many areas of an organization

(Tuttle, 1994). Malaney and Osit (1998) indicated that low student affairs staff moral

will spill over into the staffs work with students, producing student dissatisfaction with

the student affairs staff. Since professional student affair staff members within a dean of

students office are often on the front lines, interacting with students on a daily basis, their

satisfaction with their dean is a key variable for providing high quality services to

students.

Deans of students' professional staff members' willingness to exert extra effort is

also a key component of an effective dean of students office. Long hours and stressful

conditions are common characteristics of student affairs work (Barr, 1990; Carpenter,

1990). Professional staff members who are willing to exert extra effort are more likely to

effectively meet the challenges associated with student affairs work and increase the

organizational quality of the office. Having professional staff members willing to exert

extra effort is also important to the productivity of a dean of students office at a time

when state legislatures are demanding that institutions be more cost-effective and rely

less on state support (Arnone, 2004).









Through improving the work experience of their professional staff members and

increasing their professional staff members' willingness to exert extra effort, deans of

students that employ a transformational leadership style can improve the student affairs

program at their institutions. Bass and Avolio (1998) found that transformational

leadership can be developed within individuals through a leadership development

program they refer to as the "Full Range of Leadership Development." It is a

comprehensive training program that works with participants in improving their

leadership behavior profile and with dealing with the obstacles to changing their

leadership behavior (Bass and Avolio, 1998). This type of leadership development

program may assist deans of students in becoming more transformational in their

leadership, which would improve the work experience of their professional staff

members. Therefore, this type of leadership program could be developed for deans of

students at public research institutions.

Recommendations for Future Research

The findings of this study support the existence of the basic theoretical prediction

of Bass and Avolio's (1997) Full Range of Leadership model within the dean of students

offices at public research universities. However, the study is not conclusive and suggests

a number of areas for future research.

1. The current study was limited to deans of students at public research
universities in the southeast. Before the findings can be generalized to the
dean of students population as a whole, it is recommended that the study be
duplicated with institutions in other Carnegie Classifications, with private
institutions, and with institutions in other parts of the country.

2. A key to improving leadership effectiveness is identifying characteristics of
subordinates, organizational environment, and work tasks that neutralize or
enhance leadership behaviors (Kerr & Jermier, 1978). Therefore, it is








recommended that research be done on the characteristics of the student
affairs work that may neutralize or enhance transformational leadership.

3. A comprehensive search of the literature revealed very limited studies on the
leadership behavior of deans of students or on what type of leadership is
effective within a dean of students office. Research to examine these areas
could prove valuable to student affairs practitioners.

4. This study did not attempt to examine how perceptions differ between
transformational and transactional deans of students. Future research could
focus on what transformational deans and transactional deans believe they
ought to be doing in differing circumstances.

5. Future research focusing on the relationship between transformational
leadership and outcomes beyond those measured by the MLQ would be
valuable. Included in this research could be a measure of job satisfaction.

6. Research suggest that transformational leadership can be learned (Bass &
Avolio, 1998). Future research employing a pre-test and post-test design
around a transformational leadership development program for deans of
students could be helpful.














APPENDIX A
INSTITUTIONAL REVIEW BOARD APPROVAL
















@ UNIVERSITY OF

0FLORIDA


Inslilutional Re ievv Board


Mr. Richard A. Barth
155 Tigert Hall
Campus


9XA Psycholugy Bldg.
POBo\ 112250
Gadincsllc. FL 32611-2250
Phone. (352) 392-0433
Fax (352) 392-9234
E-mail !rb2aiull edu
http. "rp il edt'irb ':rb02


\0'


FROM:


C. Michael Levy, PhD, Chair
University of Florida
Institutional Review Board 02


SUBJECT: Approval of Protocol #2003-U-263


TITLE:


Leadership Behaviors Among Deans of Students at Public Research Universities in the
Southeast


SPONSOR: Unfunded


I am pleased to advise you that the University of Florida Institutional Review Board has recommended
approval of this protocol. Based on its review, the UFIRB determined that this research presents no
more than minimal risk to participants. Given your protocol, it is essential that you obtain signed
documentation of informed consent from each participant. Enclosed is the dated, IRB-approved
informed consent to be used when recruiting participants for the research.

It is essential that each of your participants sign a copy of your approved informed I
consent that bears the IRB stamp and expiration date. I

If you wish to make any changes to this protocol, including the need to increase the number of
participants authorized, you must disclose your plans before you implement them so that the Board can
assess their impact on your protocol. In addition, you must report to the Board any unexpected
complications that affect your participants.

If you have not completed this protocol by 10-Mar-2004, please telephone our office (392-0433), and
we will discuss the renewal process with you. It is important that you keep your Department Chair
informed about the status of ,his research protocol.



CML:dl


'quai Opporti ny/A Affnnnat Amlon Ir.nston