An Ex Post Facto Study on the Relationship Between Self-Reported Peer-to-Peer Mentoring Experiences and Instructor Confidence, Institutional Loyalty, and Student Satisfaction among Part-Time Instructors

Andrews University Andrews University

Digital Commons @ Andrews University Digital Commons @ Andrews University

Dissertations Graduate Research

2012

An Ex Post Facto Study on the Relationship Between Self-An Ex Post Facto Study on the Relationship Between Self-

Reported Peer-to-Peer Mentoring Experiences and Instructor Reported Peer-to-Peer Mentoring Experiences and Instructor

Con;dence, Institutional Loyalty, and Student Satisfaction among Con;dence, Institutional Loyalty, and Student Satisfaction among

Part-Time Instructors Part-Time Instructors

Carolyn A. Watson

Andrews University

, [email protected]

Follow this and additional works at: https://digitalcommons.andrews.edu/dissertations

Part of the Educational Assessment, Evaluation, and Research Commons, and the Higher Education

Commons

Recommended Citation Recommended Citation

Watson, Carolyn A., "An Ex Post Facto Study on the Relationship Between Self-Reported Peer-to-Peer

Mentoring Experiences and Instructor Con;dence, Institutional Loyalty, and Student Satisfaction among

Part-Time Instructors" (2012).

Dissertations

. 1529.

https://digitalcommons.andrews.edu/dissertations/1529

https://dx.doi.org/10.32597/dissertations/1529/

This Dissertation is brought to you for free and open access by the Graduate Research at Digital Commons @

Andrews University. It has been accepted for inclusion in Dissertations by an authorized administrator of Digital

Commons @ Andrews University. For more information, please contact [email protected].

Thank you for your interest in the

Andrews University Digital Library

of Dissertations and Theses.

Please honor the copyright of this document by

not duplicating or distributing additional copies

in any form without the author’s express written

permission. Thanks for your cooperation.

ABSTRACT

AN EX POST FACTO STUDY ON THE RELATIONSHIP BETWEEN

SELF-REPORTED PEER-TO-PEER MENTORING EXPERIENCES

AND INSTRUCTOR CONFIDENCE, INSTITUTIONAL

LOYALTY, AND STUDENT SATISFACTION

AMONG PART-TIME INSTRUCTORS

Carolyn A. Watson

Chair: Erich Baumgartner

ABSTRACT OF GRADUATE STUDENT RESEARCH

Dissertation

Andrews University

School of Education

Title: AN EX POST FACTO STUDY ON THE RELATIONSHIP BETWEEN SELF-

REPORTED PEER-TO-PEER MENTORING EXPERIENCES AND

INSTRUCTOR CONFIDENCE, INSTITUTIONAL LOYALTY, AND

STUDENT SATISFACTION AMONG PART-TIME INSTRUCTORS

Name of researcher: Carolyn A. Watson

Name and degree of faculty chair: Erich Baumgartner, Ph.D.

Date completed: August 2012

Problem

The most recent data from the U.S. Department of Educational Statistics estimate

that part-time faculty now comprises almost half of the faculty labor force; many believe

this statistic has been gravely underestimated. Powerful unions like the American

Association of University Professors (AAUP) and the American Association of Colleges

and Universities (AACU) believe that universities are relying too heavily on part-time

faculty members and that this over-reliance threatens the quality of higher education

today. The debate surrounding the use of part-time faculty seems to focus on issues of

instructor confidence, loyalty, and student satisfaction. Many question whether a part-

time faculty member can deliver quality instruction and contribute to the community of

learners as well as their full-time tenured counterparts. This dissertation explores whether

peer mentoring is an effective means to increase confidence, loyalty to the organization,

and student satisfaction scores among part-time faculty members.

Method

An ex post facto research design was used to explore the quality of a previous

peer-mentoring experience and its relationship to several dependent variables. The

sample was comprised of the eligible 600 part-time faculty who taught in the School of

Business at a large, private, Christian, mid-western university. Data were collected using

an online survey instrument, comprised of four subscales. After the data were collected,

descriptive statistics were generated and a Pearson r was calculated and correlational

matrixes generated to initially determine what, if any, significant relationships existed

between the independent and dependent variables. Linear regression models were

generated to answer the research hypotheses.

Results

One of the study’s major findings was the significant relationship that exists

between mentoring and instructor confidence. Independent variables, such as age and

gender, did not significantly affect these results. However, part-time faculty who taught

for other universities tended to score higher in the measure of instructor confidence than

those with experience teaching only for the University.

While the fidelity or the quality of the mentoring program was not significantly

related to instructor confidence, it was significantly related to institutional loyalty. This

finding was independent of the type of mentor and the other demographic variables,

including teaching at other universities. This was a surprising find, particularly in light of

the way teaching at multiple institutions is portrayed negatively in the literature.

Finally, the research asked whether part-time faculty members who received

mentoring have students with higher means on end-of-course survey forms, which are

used to measure student satisfaction. The data analysis revealed that no significant

relationship exists between mentoring and student satisfaction scores. The research

design and response rate (25%) limit the ability to generalize from these findings.

Conclusions

The research re-affirmed that mentoring is an effective management strategy.

Part-time faculty members who receive mentoring tended to score significantly higher on

measures of instructor confidence. The quality of the peer-mentoring experience did not

appear to be as important as the fact that mentoring, in some form or another, occurred.

In addition, teaching at other universities did not negatively influence the significant

relationship between mentoring, instructor confidence, and institutional loyalty.

Secondly, the quality or fidelity of the mentoring program seems to be important

as it relates to institutional loyalty. While any type of mentoring could result in increased

confidence, if the goal of the University is to develop a sense of institutional loyalty, then

developing a quality mentoring program and fostering quality mentoring relationships

would seem to be important.

This study found no significant relationship between a faculty member’s self-

reported perception of a peer-mentoring experience and the level of student satisfaction.

That no relationship was found in the course of this study does not mean that a

relationship does not exist. It could be that the instrument used to measure student

satisfaction was not valid or that the “halo effect” influenced the student ratings.

Finally, I developed subscales created to measure instructor confidence and the

fidelity of the mentoring program, which have the potential to aid further research and

administrators in the development of both professional development activities and

organizational mentoring programs. Further research can use alternative strategies for

improving the estimates of reliability and validity of these two instruments.

Andrews University

School of Education

AN EX POST FACTO STUDY ON THE RELATIONSHIP BETWEEN

SELF-REPORTED PEER-TO-PEER MENTORING EXPERIENCES

AND INSTRUCTOR CONFIDENCE, INSTITUTIONAL

LOYALTY, AND STUDENT SATISFACTION

AMONG PART-TIME INSTRUCTORS

A Dissertation

Presented in Partial Fulfillment

of the Requirements for the Degree

Doctor of Philosophy

Carolyn A. Watson

August 2012

AN EX POST FACTO STUDY ON THE RELATIONSHIP BETWEEN

SELF-REPORTED PEER-TO-PEER MENTORING EXPERIENCES

AND INSTRUCTOR CONFIDENCE, INSTITUTIONAL

LOYALTY, AND STUDENT SATISFACTION

AMONG PART-TIME INSTRUCTORS

A dissertation

presented in partial fulfillment

of the requirements for the degree

Doctor of Philosophy

Carolyn A. Watson

APPROVAL BY THE COMMITTEE:

________________________________ _____________________________

Chair: Erich Baumgartner Dean, School of Education

James R. Jeffery

________________________________

Member: Isadore Newman

________________________________

Member: Sylvia Gonzalez

________________________________ _____________________________

External: David Penno Date approved

iii

TABLE OF CONTENTS

LIST OF TABLES ..................................................... viii

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

ACKNOWLEDGMENTS ............................................... xii

Chapter

I. INTRODUCTION ..................................................1

Background of the Problem .......................................4

Statement of the Problem .........................................8

Purpose of the Study ............................................9

Assumptions ..................................................9

Research Questions ............................................10

Research Question 1 .......................................10

Research Question 2 .......................................10

Research Question 3 .......................................11

Theoretical Framework .........................................11

Significance of the Study ........................................15

General Research Methodology ..................................17

Limitations ...................................................17

Delimitations .................................................19

Definition of Terms ............................................19

Organization of the Study .......................................22

II. LITERATURE REVIEW ..........................................24

Part-time Faculty ..............................................25

The Rise in Use of Part-time Faculty ..........................25

The G.I. Bill ........................................27

The Money Factor ...................................28

Flexibility in Scheduling ..............................29

Supply and Demand ..................................30

The Debate Related to the Use of Part-time Faculty ...................31

Quality of Instruction ......................................31

The Role of Academic Freedom .............................33

Job Security and Other Financial Considerations ................34

Contributions of Part-time Faculty............................35

Issues Related to Part-time Faculty ................................36

Instructor Confidence ......................................36

Instructor Confidence Defined ..........................38

Role of Inclusion in Instructor Confidence ................39

Institutional Loyalty .......................................41

Transitory Nature of Part-time Appointments ..............41

Institutional Loyalty Defined ...........................42

Types of Institutional Loyalty ..........................44

Student Satisfaction .......................................47

Student Satisfaction Defined ...........................48

Validity of Using Student Satisfaction Surveys .............49

Students and SETs ...................................51

Instructors and SETs ..................................51

SETs as Measure of Instructor Effectiveness ...............52

Summary of Issues Related to Part-time Faculty .................53

Mentoring ...................................................54

Mentoring: A Definition ...................................55

Types of Mentoring Programs ...............................57

Spontaneous Mentoring .........................................57

Systematic Mentoring ................................. 59

Peer Mentoring ......................................62

Developing Systematic Mentoring Programs ...................65

Benefits of Mentoring .....................................69

Conclusion ...................................................71

III. METHODOLOGY ...............................................74

Description of Participants ......................................74

Data Collection Procedures ......................................76

Design of the Study ............................................76

Internal Validity ..........................................77

External Validity .........................................80

Statement of Hypotheses ........................................80

Hypothesis Statements Related to Instructor Confidence ..........81

Hypothesis Statements Related to Mentoring and

Instructor Confidence ..............................81

Hypothesis Statements Related to Mentoring, Fidelity,

and Instructor Confidence ...........................82

Hypothesis Statements Related to Mentoring, Type of

Mentor, and Instructor Confidence ....................83

Hypothesis Statements Related to Institutional Loyalty ...........85

Hypotheses Related to Mentoring and Institutional

Loyalty ..........................................85

Hypotheses Related to Fidelity and Institutional Loyalty .....86

Hypothesis Statements Related to Type of Mentor and

Institutional Loyalty ................................87

Hypothesis Statements Related to Student Satisfaction ............89

Hypotheses Related to Mentoring and Student

Satisfaction ......................................89

Hypotheses Related to Fidelity and Student Satisfaction ......90

Hypothesis Statements Related to Type of Mentor and

Student Satisfaction ................................91

The Variable List ..............................................92

The Instrumentation ............................................95

Online Survey Instrument ..................................96

Part I—Informed Consent .............................96

Part II—Tell Us about You ............................96

Part III—Tell Us How You Feel About the University .......96

Part IV—Tell Us What You Know ......................98

Part V—Tell Us About Your Mentoring Experience ........103

Part VI—Conclusion ................................105

End-of-Course Surveys ...................................106

Estimates of Validity and Reliability ..............................107

Estimates of Validity .....................................107

Field Testing ...........................................108

Estimates of Reliability ...................................110

Fidelity Measures .............................................111

Data Analysis Plan ............................................113

Summary of Methodology ......................................116

IV. SETTING OF THE RESEARCH ...................................118

The University ...............................................118

The University Mentoring Program ...............................120

The Philosophy of the Mentoring Program ....................120

The Mentoring Program Outcomes ..........................121

Qualifications of the Mentor ...............................122

The Mentoring Program Model .............................124

Conclusion ..................................................127

V. RESULTS .....................................................128

Response Rate ...............................................128

Descriptive Statistics ..........................................129

Fidelity Measure .............................................138

Correlation Matrix ............................................141

Multiple Regression Analysis ...................................143

Instructor Confidence Variables ............................143

Hypotheses Related to Mentoring and Instructor

Confidence ......................................145

Results Related to Mentoring and Instructor Confidence ....146

Hypotheses Related to Fidelity and Instructor Confidence ...150

Results Related to Fidelity and Instructor Confidence .......151

Hypothesis Statements Related to Type of Mentor and

Instructor Confidence .............................152

Results Related to Type of Mentor and Instructor

Confidence ......................................153

Institutional Loyalty Variables ..............................158

Hypotheses Related to Mentoring and Institutional

Loyalty .........................................159

Results Related to Mentoring and Institutional Loyalty ......160

Hypotheses Related to Fidelity and Institutional Loyalty ....160

Results Related to Fidelity and Institutional Loyalty ........162

Hypothesis Statements Related to Type of Mentor and

Institutional Loyalty ...............................165

Results Related to Type of Mentor and Institutional

Loyalty .........................................167

Student Satisfaction Variables ..............................167

Hypotheses Related to Mentoring and Student

Satisfaction .....................................168

Results Related to Mentoring and Student Satisfaction ......169

Hypotheses Related to Fidelity and Student Satisfaction .....170

Results Related to Fidelity and Student Satisfaction ........171

Hypothesis Statements Related to Type of Mentor and

Student Satisfaction ...............................172

Results Related to Type of Mentor and Student

Satisfaction .....................................173

Conclusion ..................................................174

VI. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS ..........175

Summary ...................................................175

Purpose ................................................176

Research Methodology....................................177

Limitations .............................................178

Conclusions .................................................179

Conclusions for Research Question 3 ........................182

Discussion ..................................................183

Instructor Confidence .....................................184

Institutional Loyalty ......................................186

Student Satisfaction ......................................188

Theoretical Framework ...................................190

Recommendations for Practice ..................................192

Recommendations for Further Research ...........................200

Summary of the Study .........................................203

vii

Appendix

A. DEFINITION OF VARIABLES ....................................205

B. UNIVERSITY MENTORING MODEL ..............................212

C. LIST OF EXPERT JUDGES .......................................226

D. INSTRUMENTATION ...........................................228

E. INSTITUTIONAL REVIEW BOARD DOCUMENTS ...................240

REFERENCE LIST ....................................................245

VITA ................................................................260

viii

LIST OF TABLES

1. Likert’s System 4 Management Theory ....................................14

2. Descriptive Statistics for Response Rate ..................................129

3. Descriptive Statistics for Mentoring Status ................................130

4. Descriptive Statistics for Type of Mentoring Experiences .....................130

5. Descriptive Statistics for Gender ........................................132

6. Descriptive Statistics for Ethnicity .......................................133

7. Descriptive Statistics for Educational Level ................................135

8. Descriptive Statistics for Age ...........................................136

9. Descriptive Statistics for Years of Service ................................. 136

10. Descriptive Statistics for Number of Modules Taught .......................137

11. Descriptive Statistics for Taught at Other Universities ......................138

12. Descriptive Statistics for How Many Other Universities .....................139

13. Descriptive Statistics for Teaching Level ................................. 140

14. Fidelity Score Frequencies ............................................141

15. Correlation Matrix for Confidence, Loyalty, Student Satisfaction, and

Received Mentoring ...............................................142

16. Correlation Matrix for Confidence, Loyalty, Student Satisfaction, and

Fidelity .........................................................142

17. Correlation Matrix for Confidence, Loyalty, Student Satisfaction, and Type

of Mentor ........................................................144

18. Regression Models for Hypothesis Statements: Mentoring, Instructor

Confidence, and Age ...............................................147

19. Regression Models for Hypothesis Statements: Mentoring, Instructor

Confidence, and Gender ............................................147

20. Regression Models for Hypothesis Statements: Mentoring, Instructor

Confidence, and Ethnicity ...........................................148

21. Regression Models for Hypothesis Statements: Mentoring, Instructor

Confidence, and Length of Service ....................................148

22. Regression Models for Hypothesis Statements: Mentoring, Instructor

Confidence, and Number of Courses Taught ............................148

23. Regression Models for Hypothesis Statements: Mentoring, Instructor

Confidence, and Educational Level ...................................149

24. Regression Models for Hypothesis Statements: Mentoring, Instructor

Confidence, and Teaching Level ......................................149

25. Regression Models for Hypothesis Statements: Mentoring, Instructor

Confidence, and Teaches at Other Universities ..........................149

26. Regression Models for Hypothesis Statements: Type of Mentor, Instructor

Confidence, and Age ...............................................154

27. Regression Models for Hypothesis Statements: Type of Mentor, Instructor

Confidence, and Gender ............................................155

28. Regression Models for Hypothesis Statements: Type of Mentor, Instructor

Confidence, and Ethnicity ...........................................155

29. Regression Models for Hypothesis Statements: Type of Mentor, Instructor

Confidence, and Length of Service ....................................156

30. Regression Models for Hypothesis Statements: Type of Mentor, Instructor

Confidence, and Number of Courses Taught ............................156

31. Regression Models for Hypothesis Statements: Type of Mentor, Instructor

Confidence, and Educational Level ...................................157

32. Regression Models for Hypothesis Statements: Type of Mentor, Instructor

Confidence, and Teaching Level ......................................157

33. Regression Models for Hypothesis Statements: Type of Mentor, Instructor

Confidence, and Teaches at Other Universities ..........................158

34. Regression Models for Hypothesis Statements: Fidelity, Institutional

Loyalty, and Age ..................................................163

35. Regression Models for Hypothesis Statements: Fidelity, Institutional

Loyalty, and Gender ...............................................163

36. Regression Models for Hypothesis Statements: Fidelity, Institutional

Loyalty, and Ethnicity ..............................................163

37. Regression Models for Hypothesis Statements: Fidelity, Institutional

Loyalty, and Length of Service .......................................164

38. Regression Models for Hypothesis Statements: Fidelity, Institutional

Loyalty, and Number of Courses Taught ...............................164

39. Regression Models for Hypothesis Statements: Fidelity, Institutional

Loyalty, and Educational Level ......................................164

40. Regression Models for Hypothesis Statements: Fidelity, Institutional

Loyalty, and Teaching Level .........................................165

41. Regression Models for Hypothesis Statements: Fidelity, Institutional

Loyalty, and Teaches at Other Universities .............................165

42. Definition of Variables ...............................................206

LIST OF FIGURES

1. Peer Mentoring Process and Outcomes ....................................16

2. Mentoring Process and Outcomes Revised................................. 191

xii

ACKNOWLEDGMENTS

It has been said that behind every successful man is a successful woman. A

greater truth might be that behind each success is a myriad of individuals whose

mentoring, assistance, guidance, and encouragement were essential. I would like to take

the opportunity to acknowledge the infinite patience and wisdom of my committee

members, Dr. Erich Baumgartner (chair) and Dr. Sylvia Gonzalez (content specialist).

Their commitment to excellence inspired me to achieve beyond what I could by myself. I

would like to extend a special thank-you to my methodologist, Dr. Isadore Newman. His

ability to communicate complex statistical concepts into terms a fledging Ph.D. student

can understand and enjoy is unparalleled.

I would also like to acknowledge the assistance of Spring Arbor University and

the University in which this research took place. The visionary leadership of both schools

recognized early on the value and contribution of this study and worked diligently to

support this research and remove all obstacles. I would also like to thank the part-time

faculty members in the School of Graduate and Professional Studies at Spring Arbor

University. Working with them for over a decade, I have experienced firsthand their

commitment to excel and their passion for their students. They were the inspiration for

this research.

CHAPTER ONE

INTRODUCTION

Harvard College, founded in 1636, is the oldest institution of higher learning in

America (Presidents and Fellows of Harvard College, 2007). Sixteen years after the

Pilgrims landed on Plymouth Rock, young Harvard tutors, not faculty, stood before

individuals and small groups of students. These young men taught and instructed the next

generation of clergy in order “to advance Learning and perpetuate it to Posterity;

dreading to leave an illiterate Ministry to the Churches” (Presidents and Fellows of

Harvard College, 2007, para. 4). For centuries, the Harvard model of education stood

resolute, and faculty members in colleges and universities continued to be merely tutors,

young men who had recently completed baccalaureate studies.

However, in the 19

century it became increasingly common to hire faculty to

teach in one particular area of expertise or specialization (Carrell, as cited in Schuster &

Finkelstein, 2006). This specialization meant more “formal preparation” and a “graduate

education” (Schuster & Finkelstein, 2006, p. 23). This advanced preparation marked a

shift in the transitory or temporary nature of the college tutor to a lifelong career

commitment to teaching (McCaughey, 1974).

Students also became empowered; educators like Charles Elliot, President of

Harvard College, advocated for an elective system of higher education. He proposed that

each student be allowed to select the course of study he or she would pursue (Schuster &

Finkelstein, 2006). For a hundred years, full-time, professional faculty, who were more

specialized, taught students who were more empowered, until the economic,

demographic, and technological shifts of a post-World War II era interrupted this

seamless fabric of academic life (Schuster & Finkelstein, 2006).

In 1944, Congress enacted the G.I. Bill. This legislation resulted in the infusion of

large numbers of returning veterans into an unprepared university system (Schuster &

Finkelstein, 2006). This influx of students, coupled with a decrease in the availability of

full-time, qualified faculty, resulted in universities and colleges turning to the use of part-

time faculty members in the 1960s (Bowen & Schuster, 1986). According to Bowen and

Schuster (1986), in 1960, 35% of all faculty appointments were for part-time faculty

members. By 2006, this 35% had grown to 46%; if you include other forms of contingent

faculty (graduate student instructors, post-doctoral fellows, and full-time non-tenure-track

faculty) the percentage of contingent faculty rose to 65% (AAUP, 2006). The means that

fully “two-thirds of all faculty employed in 2003” (AAUP, 2006, p. 6) were part-time and

contingent faculty. Schuster and Finkelstein (2006) call this a “seismic shift” (p. 222).

The most recent data from the U.S. Department of Educational Statistics confirm

this and estimate that part-time faculty alone now comprises almost half of the faculty

labor force (Snyder & Dillow, 2012). In fact, many believe that this fact has been

“gravely underestimated” (R. Wilson, 1999, p. A15). Because of the variance in how

researchers define part-time faculty, some believe they are underreported (Lerber, 2006).

Graduate assistants, guest lecturers, visiting professors, and fellowships are just some of

the labels used (Ivey, Weng, & Vahadji, 2005; Reichard, 1998). Researchers seem to

agree that the use of part-time faculty is grossly underestimated, which raises the

question, “Are universities and colleges relying too heavily on part-time faculty?”

The Association of American Colleges and Universities (AACU) believes so.

“Over-reliance on part-time and other ‘contingent’ instructional staff diminishes faculty

involvement in student learning” (Benjamin, 2002, p. 4). In addition, the American

Association of University Professors (AAUP) concurs. In 2006, the AAUP issued the

following statement: “As the faculty collectively grows more contingent, the quality of

higher education itself is threatened” (AAUP, 2006, p. 12).

More and more voices seem to be suggesting that the use of part-time faculty

members jeopardizes the academic integrity of the instructional process (Benjamin, 2002;

Elman, 2002). One author went so far as to say that the use of part-time faculty is the

“major scandal of higher education today” (Arden, 1989, p. A48). Most of those who

deplore the use of part-time faculty cite concerns over the part-time faculty members’

isolation from the larger, educational community (Beem, 2002; Benjamin, 2002; Bradley,

2004). Balch (1999) expands on these concerns and lists several other factors such as

teaching quality, commitment to the college or university, and the time available for part-

time faculty members to spend with students outside of class. Do these “invisible people

of higher education,” as Arden (1989, p. A48) calls them, contribute to the culture of

excellence that American universities and colleges strive for?

The AAUP believes they do not. And the union believes this continued growth

trend is a “problem” (AAUP, 2006, p. 6). This powerful union is careful to note that the

problems are not related to the individuals who are, for the most part, “able teachers and

scholars” (AAUP, 2006, p. 6). Rather the problem lies in the nature of the part-time

faculty member’s employment conditions, the lack of support provided to part-time

faculty, and the lack of academic freedom (AAUP, 2006).

Others have questioned whether this choir of voices has not missed an important

side of the debate. While citing some concerns regarding the uses of part-timers, Balch

(1999) also concluded that part-time faculty members are qualified, highly committed,

and fulfill their duties and responsibilities conscientiously. For others, the use of adjuncts

represents an essential link between the theoretical strongholds of ivory tower

universities and the front lines of the pragmatic professional. The President of Indiana

University concluded that “part-time instructors are necessary to teach specific courses or

to bring specific professional experience to the classroom” (Brand, 2002, p. 21). Arden

(1989) agrees, “Adjuncts are a source of expertise that greatly enriches the educational

experience of students” (p. 17).

It is difficult to know who to believe as the controversy swirls. If the use of part-

time faculty poses such a danger and a threat to the academic integrity of colleges and

universities, why does the trend continue? One needs a clearer understanding of the

issues related to the use of part-time faculty to more fully comprehend the problem and

how best to address it.

Background of the Problem

The numbers do not lie. By conservative estimates, university and college

administrators are giving almost half of post-secondary faculty teaching assignments to

part-time faculty. For better or worse, part-time faculty members are here to stay

(Benjamin, 2002; Snyder & Dillow, 2012). The U.S. Department of Labor predicts that

post-secondary teaching positions will grow faster than the national average and that a

significant number of these positions will be for part-time professors (Bureau of Labor

Statistics, 2004).

Today, the U.S. Department of Education confirms this continuing trend (Snyder

& Dillow, 2012). But tenured faculty, administrators, and accrediting bodies are

wrestling with questions and concerns related to the use of part-time faculty (AAUP,

2006, 2009; Schuster & Finkenstein, 2006). Can a part-time faculty member deliver

quality instruction and contribute to the community of learners as well as their full-time,

tenured counterparts? The debate surrounding the use of full-time versus part-time

faculty members seems to focus on the following three concerns: quality of instruction,

isolation and loyalty, and student interaction.

Some maintain that adjunct faculty members are at a disadvantage. They are

afforded few opportunities for professional growth, and the opportunities that are offered

to them seem to be scheduled during times when work commitments keep adjuncts from

attending (Phillips, 2002). Benjamin (2002) notes that part-time faculty members lack

“professional evaluation . . . supports and, often-collegial involvement enjoyed by the

full-time, tenure-track faculty” (p. 7). When university administrators do not give part-

time faculty the same level of support as full-time, tenured track faculty, it is no wonder

that questions of confidence and quality emerge. According the AAUP (2006), this

“clearly represents a substantial limitation on their functioning as faculty” (p. 9).

This lack of support is evident when, after an adjunct faculty member is assigned

his or her first teaching assignment, the part-timer is set adrift to either sink or swim.

These “add-on” faculty members seldom interact with the full-time faculty and rarely

receive “constructive feedback on the effectiveness of their teaching” (Beem, 2002, p. 2).

In addition to the question of quality instruction, some note that part-timers tend

to be isolated from the larger educational community (AAUP, 2006; Balch, 1999; Beem,

2002; Benjamin, 2002; Bradley, 2004). Balch (1999) notes that a part-time member’s

“commitment to the college or university” is a concern (p. 33). Some adjuncts themselves

note there is a “strong sense of second class citizenship” (Foster & Foster, 1998, p. 11).

Whereas full-time, tenured track faculty members have a “stake in the institution,”

“temporarily employed faculty . . . feel less connected to the institution and less

empowered” (Bradley, 2004, p. 30). Cohen (1999) describes the issue like this:

Part-timers are . . . symbolic of a new class of migrant workers. While they are not

picking grapes, these laborers are wandering around the fields of academia, scraping

together teaching assignments from different institutions while the fruits of full-time

professorship—security, remuneration, stature, and academic freedom—remain out of

reach. (para. 3)

These twin concerns of the quality of instruction provided by part-time faculty

members and the feelings of isolation they experience are major issues that raise the

question, “What impact might this have on a student’s perception and satisfaction with

his or her educational experiences?” Do full-time faculty members truly provide a better

educational experience than part-time instructors do? Elman (2002) believes they do and

notes that the use of part-time faculty seriously hinders the quality of the student’s

educational experience.

Benjamin (2002) also believes that the use of part-time faculty has a negative

impact on student learning. The transitory nature of part-time faculty appointments and

the fact that they rarely have a campus office could lead one to conclude that full-time

faculty members devote more time and energy to their students than part-time faculty

(Benjamin, 2002; Bowen & Schuster, 1986). If this is true, it seems logical to conclude

that students would be more satisfied with the instruction they receive from full-time

faculty. AAUP (2006) notes that part-time faculty are not in a position to develop

relationships with their students, they lack the support to provide students with a quality

education experience, and, without the protection that academic freedom provides, part-

time faculty are less likely to challenge students and hold them to high academic

standards. But evidence that these negative results actually exist is virtually non-existent.

The reality is that research concerning the effectiveness of adjuncts is “scant” (Klein,

Weisman, & Smith, 1996, para. 12).

There is a plethora of literature that records recommendations on how to govern

the use of adjunct faculty. The specific recommendations vary, but common themes are

to provide support, training, and professional development opportunities and to include

adjunct faculty in the greater, scholarly community (AAUP, 2006; Arden, 1989; Balch,

1999; Beem, 2002; Rice, 2004). Many of the concerns and recommendations fall into the

categories of increasing instructor confidence (AAUP, 2003; Meacham, 2002), loyalty to

the organization (AAUP, 2006; Beem, 2002; Bradley, 2004; Cohen, 1999), and student

satisfaction (AAUP, 2006; Balch, 1999; Benjamin, 2002; Elman, 2002).

Part-time faculty members themselves have identified a key strategy to deal with

these issues: mentoring by more experienced faculty. Feldman and Turnley (2001)

concluded that part-time faculty members believe that with “more mentoring from senior

colleagues and greater integration into their larger work groups” instructional quality

could be improved (p. 14). Since a mentor is a guide, someone who has gone before and

can now show the way (Clutterbuck & Megginson, 1999; Cunningham, 1999; Tobin,

1998), it seems logical to assume that an experienced, part-time faculty member may be

uniquely qualified to mentor other part-time faculty members. In an effort to provide

insight in to issue, this dissertation explores the relationship between instructors’ peer-to-

peer mentoring experiences and their confidence as instructors, their loyalty to the

institution, and the satisfaction of students with their teaching.

Statement of the Problem

Since adjuncts themselves have identified mentoring as a means to improve

instruction and minimize the feelings of isolation often associated with a part-time faculty

appointment, this study picks up this line of thought. I found no studies that specifically

explored the ability of peer mentoring to assist in increasing the competence and

confidence of part-time faculty or to decrease the sense of isolation that surrounds them.

As noted earlier, these dual issues of instructor confidence and instructor isolation from

the rest of the faculty give rise to questions about how students perceive the quality of the

instruction they receive. Since universities routinely measure student satisfaction through

end-of-course evaluations this is a factor that can easily be explored.

With almost 50% of all undergraduate courses in the United States being taught

by adjunct faculty (Snyder & Dillow, 2012), it is essential that university leaders address

how best to provide support and inclusion for these isolated members if universities and

colleges are to continue to provide quality education. The focus of this study is the

relationship between part-time faculty members’ peer-mentoring experiences and their

confidence as instructors, their loyalty to the hiring institution, and their students’

satisfaction with their teaching.

Purpose of the Study

The purpose of this study was to investigate what relationship, if any, exists

between a part-time faculty member’s self-reported perceptions of the quality of a peer-

mentoring relationship with instructor confidence, institutional loyalty, and student

satisfaction. Part-time instructors at a large, mid-western, Christian university served as

the participants for this study. The University has developed and implemented an

institutional mentoring program whereby part-time faculty members are mentored by

other part-time faculty. As such, this institution was uniquely situated to provide insight

into these previously stated relationships. More specifically, this study helped to answer

the following question: “Is peer-to-peer mentoring an effective means to support part-

time faculty, bolster confidence, increase institutional loyalty and consequently, produce

students who are satisfied that they received the best education and preparation?”

Assumptions

During the course of designing this investigation, I made several assumptions

regarding mentoring and its benefits as well as the effectiveness of peer-to-peer

mentoring over other forms of mentoring. Literature and studies suggest that mentoring is

a positive intervention that results in both personal and professional growth and

satisfaction among first-year teachers at the elementary and secondary levels (Darling-

Hammond, 2003; Edwards, 2000). This study assumes that the mentoring of part-time

faculty members at the post-secondary level will have similar results.

My review did not find any research that investigated the relationship between

peer mentoring among part-time faculty members. Consequently, this study assumes that

mentoring has not been widely practiced as a means of including part-time faculty

members in the teaching and learning community of post-secondary institutions.

Another key assumption made by this study is the belief that part-time instructors

are keenly interested in personal and professional development. They desire increased

opportunities to be included and to improve their teaching effectiveness. Part-time faculty

members are willing to invest the necessary time and energy to be actively involved in a

mentoring relationship, which could possibly lead to increased instructor confidence,

institutional loyalty, and student satisfaction.

Research Questions

This study sought to determine what relationship exists between a part-time

faculty member’s self-reported perceptions of the quality of a peer-mentoring experience

and several dependent variables: instructor confidence, institutional loyalty, and student

satisfaction. The primary research question was, “What relationship does peer-to-peer

mentoring have to instructor confidence, institutional loyalty, and student satisfaction

among part-time instructors in the University’s School of Business?” From this

overarching question, the following three research questions were identified.

Research Question 1

Is a part-time faculty member’s self-reported perception of the quality of a peer-

mentoring experience related to his or her degree of instructor confidence?

Research Question 2

Is a part-time faculty member’s self-reported perception of the quality of a peer-

mentoring experience related to his or her degree of institutional loyalty?

Research Question 3

Is a part-time faculty member’s self-reported perception of the quality of a peer-

mentoring experience related to his or her students’ satisfaction with the quality of the

learning experience as measured by a student evaluation of teaching/end-of-course form?

Theoretical Framework

Albert Bandura’s social cognitive theory served as the main theoretical

framework for this dissertation. The foundation of social cognitive theory rests on the

premise that an individual’s thoughts, actions, and emotions influence his or her behavior

and that these three constructs are impacted and changed through reciprocal, social

relationships (Bandura, 1989). Mentoring, a primary focus of this study, occurs in the

context of personal relationships and within the social environment. This approach to

personal and professional growth and development makes social learning theory an

appropriate theoretical framework for this dissertation. According to social cognitive

theorists, learning is a “social process” (Zins, Boodworth, Weissberg, & Walberg, 2004,

p. 3).

Bandura’s (1989) social cognitive theory postulates that human beings have a

superior ability to learn vicariously, that is to say, we have the ability to learn from the

successes and mistakes of others. This aspect of social learning theory is critical. Bandura

believes that if we learned only from our own personal experiences, “the process of

cognitive and social development would be greatly retarded, not to mention exceedingly

tedious and hazardous” (p. 21). By observing others, we are able to extend our own skills

and conceptual knowledge. Bandura believes that we draw upon personal experiences as

well as the experiences of others who are “well-informed on the matters of concern” in

order to develop our own competency (p. 13). In fact, social learning theory emphatically

states that the most valuable knowledge is imparted socially. Typically, this vicarious

learning generates “new instances of behavior that go beyond what they have seen or

heard” (p. 25).

In fact, the attainment of these goals and increasing mastery of a particular skill

set results in what social learning theorists call self-efficacy (Bandura & Schunk, 1981).

According to Bandura (1989), “When people aim for and master valued levels of

performance, they experience a sense of satisfaction” (p. 48). Social learning theory

supports mentoring as an important means to convey knowledge and skill to another

person. Bandura claims that “knowledge and reasoning skills are best gleaned from those

who are highly knowledgeable and skilled” (p. 13).

Typically, an individual who is successful within an institution believes that the

work he or she does is important and valuable (Murray, 1991). According to Levinson,

Darrow, Klein, Levinson, and McKee (1978), a mentor is someone who takes a special

interest in helping to ensure that an individual develops into a successful professional.

Mentoring is a trusting, social relationship, which exists between a mentor, who is

experienced and successful, and a protégé, who is less skilled and experienced. The

purpose of a mentoring relationship is to develop confidence and competency. Mentors

can be helpful as they act as a counselor and guide in order to provide direction on how to

become a leader in their chosen vocation (Tobin, 1998).

This dissertation seeks to determine what relationship peer mentoring has with the

mentee’s confidence as an instructor, his or her institutional loyalty, and finally student

satisfaction scores. Bandura (1989) clearly believes that an influential social relationship,

such as mentoring, can result in increased confidence and personal satisfaction. This

study suggests that in addition to these outcomes, a socially influential relationship, such

as mentoring, will also result in increased commitment and loyalty to the organization.

Moreover, this study also examines if the increased confidence and loyalty of faculty

members are also notable in higher levels of satisfaction recorded by students with their

learning experience.

Bandura’s (1989) social learning theory supports the conclusion that an influential

relationship, like mentoring, results in greater confidence and loyalty. But Bandura’s

theory addresses only how one learns. It does not address what ramifications such

learning will have on the organization, its leadership, employees, or consumers. We have

already seen, through the literature, that a social learning strategy, like mentoring, would

likely result in higher levels of confidence and organizational loyalty among employees.

Because part-time instructors work in the organizational context of a university or

college, Bandura’s theory needs to be extended to address the results those twin outcomes

would have on the students of these employees or the organization’s “consumer.” This is

what System 4 Management Theory (Likert, 1961, 1967) does. The System 4

management strategy, which Likert calls participative group, will generally result in high

levels of trust, which in turn leads to higher levels of confidence and loyalty among

employees. This confidence and commitment to group and organizational goals results in

a high performing organization. It seems logical to assume that when organizational goals

are met, not only will the employees experience satisfaction but the recipients of the

organization’s services (i.e., students) will as well.

As the name implies, Likert hypothesized that there are four systems of

management. High functioning and high performing organizations have leaders who

operate using the fourth system, which Likert (1967) refers to as Participative Group.

System 1, which Likert (1961) labeled as exploitive, is characterized by a complete lack

of trust and the use of coercion and fear to achieve goals. The opposite set of behaviors

and skills is where we find System 4 or the participative-group function. Likert (1967)

maintains that high trust levels, mutual respect, high levels of participation, and a

commitment to individual, group, and organizational goals are characteristics of this

fourth system. Table 1 illustrates this continuum.

Table 1

Likert’s System 4 Management Theory

Trust

Motivation

Interaction

System 1:

Exploitive/Authoritative

Distrust

Fear/Punishment

Little interaction

System 2:

Benevolent/Authoritative

Cautious Trust

Reward/Punishment

Little interaction

System 3:

Consultative

Incomplete Trust

Reward/Punishment

and Involvement

Moderate

interaction

System 4:

Participative-Group

Complete Trust

Participation and

Improvements

Extensive

interaction

Note. From A Faster Learner’s Guide to Leadership: Rensis Likert, by D. Richards, n.d.

Retrieved July 30, 2009, from http://www.odportal.com/leadership/fastlearner/likert.htm

The most effective managers and organizations operate using this fourth system.

This fourth generation of management theory emphasizes supportive behavior on the part

of managers and team members. The function of supportive behavior is to increase and

maintain each individual’s sense of worth and value to the organization. Likert (1967)

states that when the organization engages in supportive behavior, “the group is eager to

help each member develop to his full potential” (p. 167). One result of this supportive

culture is a high level of peer loyalty, success in achieving organizational goals, and

effort on the part of team members to coach, train, encourage, and motivate each other

(Likert, 1967).

This dissertation actually modifies Likert’s theory by postulating that when social

learning occurs in the form of peer mentoring, members of the organization experience

increased trust levels and increased opportunities for participation and interaction. This

results in confident and loyal employees. These confident and loyal employees have a

positive impact on the organization’s customers, the students, which, in turn, results in

higher levels of students satisfied with the instruction they have received.

This dissertation tested whether peer mentoring was an effective supportive

behavior strategy and tool that managers could utilize to facilitate the achievement of

System 4 outcomes such as high employee confidence, institutional loyalty, and high

levels of customer (i.e., student) satisfaction. I assumed that a high performing

organization, by default, would have customers who are satisfied with the product and

services they received. Figure 1 illustrates this process and outcome.

Significance of the Study

As one of five regional coordinators of faculty services for Spring Arbor

University, I have spent the past 12 years in the trenches, beside the most gifted and

committed part-time faculty members. Adjuncts desire to grow and develop in their roles

Figure 1. Peer mentoring process and outcomes.

as faculty members; they are willing and able to join the larger, scholarly community.

They only lack the avenue to do so. This study is important because it will assist those in

similar leadership roles to lead this army of part-time laborers more effectively.

This study is important because the results could be used to inform the

University’s leaders, and those of similar schools, how best to support the large number

of adjuncts who teach in their programs. This University relies heavily on part-time

faculty members to fill teaching appointments. As the research suggests, this was an

intentional decision based on two elements—cost effectiveness and the belief that

practitioner professors were uniquely qualified to teach adult learners.

To those who work in the field of higher education, the significance of this study

is apparent. As reliance on part-time faculty appointments increases, so do the questions.

The problems are real, the debate rages; yet I was not able to locate any research

designed to investigate the relationship between peer-to-peer mentoring and instructor

confidence, institutional loyalty, and student satisfaction. It is evident that part-time

• Peer Mentoring provides

support through

• Social Learning

• Training, coaching

• Encouragement

• Motivation

Results in increased

confidence

• in Adult learning theory

• in Effective teaching methods

• in content knowlege

• in knowledge of univesity

polices and procedures

Results in increased

loyalty

• Faculty more committed to

university mission

• Faculty more likely to accept

teaching assignments

• Faculty more experienced

• Less turn over

Results in more

satisfiied students

• Increased satisfaction with

learning experiences

• Increased student satisfaction

scores

High Performing

Organization

faculty appointments are not going away. Evidence is to the contrary; part-time faculty

members are here to stay and our reliance on them is increasing. Consequently, it is

essential that universities and administrators wrestle with the issues in order to answer the

question, “Can a part-time faculty member deliver quality instruction and contribute to

the community of learners as well as their full-time, tenured counterparts?” Moreover, “is

peer-to-peer mentoring an effective means to achieve that outcome?”

General Research Methodology

Since this study sought to collect data “after the fact,” an ex post facto research

design was used to gather data from the part-time faculty members in the University’s

College of Adult and Professional Studies. The group surveyed was a convenient sample

representative of part-time faculty members teaching at this University at the present and

in the future. The survey collected demographic information and data to measure

instructor confidence and institutional loyalty. Moreover, one section of the survey was

designed to elicit the participants’ perceptions of the fidelity or quality of their peer-

mentoring experience. I also reviewed historical data from the University’s End of

Course Survey forms in order to determine the measure for student satisfaction. Once the

survey responses were collected, general linear regression models were generated in

order to test the research questions and to determine whether the findings were

significant.

Limitations

Several factors limit the ability to infer and generalize from the study’s findings.

Since an ex post facto research design was used, random manipulation of the independent

and predictor variables was not possible. Therefore, it is impossible to conclude, with any

certainty, the predictor variable was the cause of the significant relationships rather than

some other spurious variable. Adequate safeguards do not exist to infer causation (Ary,

Jacobs, & Sorenson, 2010). Broadly generalizing causation from this study’s findings

would be inappropriate.

The response rate (25%) also affects the ability to generalize from the study’s

findings. For example, some responses were not included in the analysis because the

participant had not received mentoring or failed to answer all the questions on a particular

subscale. Of the 147 who responded, only 63 (or 43%) had received mentoring. In

addition, some of the hypotheses stated that those with a part-time mentor would do just

as well or better than those mentored by a full-time administrator counterpart. Cases

available to test the hypotheses related to type of mentor were further reduced. For

example, of the 63 who had received mentoring, only 31 (or 49.2%) were mentored by

another part-time faculty member. When you consider those mentored by a full-time

faculty member (n = 23 or 36.5%) or an administrator (n = 8; 12.7%) the cases available

for analysis are even less. While the findings will provide helpful information to the

University’s leadership, broadly generalizing or inferring causation from the results

would be inappropriate.

Finally, while the instrument used to collect data on instructor confidence and the

fidelity of the mentoring program variables was developed using multiple strategies to

increase the estimates of its validity and reliability, the fact remains that these two

subscales lacked extensive testing and re-testing to increase validity and reliability. The

initial estimates of reliability, the ability of the instrument to be consistent, were adequate

(α

=.937 for the Instructor Confidence subscale and α

=.817 for Fidelity of Mentoring

Experience Subscale). And while key strategies, such as field testing, expert judges,

logical and concrete validity were used to increase the estimated validity of these two

subscales, more testing and analysis are needed.

Delimitations

I elected to limit the scope of this study to part-time instructors who teach in the

School of Business at a large, private, mid-western, Christian university. (The School of

Business comprises almost 80% of the student enrollment and represents the largest

sector of part-time faculty.) I selected this University because it is a large, Christian

university that offers adult degree programs, uses part-time faculty members almost

exclusively to staff classroom-teaching assignments, and has an institutional peer-

mentoring program. In addition, this University utilizes end-of-course survey forms as a

means of assessing student satisfaction.

This study explored the relationship of peer-to-peer mentoring and an individual

part-time faculty member’s confidence level, institutional loyalty, and degree of students’

satisfaction with the instructor’s competence. I did not attempt to measure or evaluate the

overall quality of instruction that part-time instructors provide. Conclusions regarding the

institution’s effectiveness as a teaching community were beyond the scope or interest of

this study.

Definition of Terms

I have provided the following definition of terms in order to ensure consistency

and clarity of communication concerning the key constructs involved in the study. The

independent and dependent variables are operationally defined, and definitions are

provided for other important and frequently used terms.

Andragogy: Refers to the teaching philosophy, first espoused by Malcolm

Knowles, which delineated the differences between traditional students and adult

learners. Different from pedagogy (which is teacher driven), andragogy is student driven

and recognizes that adult learners bring the following characteristics into the classroom:

motivation, discipline, life experience, and a desire for practical application that links

theory to practice (Knowles & Associates, 1984).

College of Adult and Professional Studies (CAPS): Refers to the University’s

department that develops and implements programs designed specifically for the working

adult. Such programs are typically based on principles of andragogy and have the

following common characteristics: lock-step cohort design, modules offered in 5- or 6-

week increments, weekly classes meet for 4 hours, accelerated curriculum design and

course learning measured by higher order thinking skills, and the students’ ability to

apply learning to work and personal settings (Bash, 2005).

Fidelity Measure: Refers to the extent to which the peer-mentoring program

adhered to the prescribed protocol (Mowbray, Holter, Teague, & Bybee, 2003).

Considering the fidelity of any intervention or treatment variable is important to ensure

that experiences similar and that any significant differences are documented. In the case

of this study, the prescribed protocol refers to the recommended elements of quality

mentoring programs as delineated in the scholarly literature and not the mentoring

program requirements as outlined by the University.

Institutional Loyalty: A self-reported measure in which a part-time instructor

indicates his or her degree of positive regard toward the organization (Ashforth, Spencer,

& Corley, 2008) as well as how much the part-time instructor believes the University

values his or her contribution and well-being (LaMastro, 2000). Frequently one finds the

term organizational commitment used almost synonymously with institutional loyalty.

Consequently, I use these terms interchangeably. Institutional loyalty was measured using

a previously validated organizational commitment scale (Allen & Meyer, 1990; Meyer &

Allen, 1988).

Instructor Confidence: A self-reported measure in which a part-time instructor

indicates his or her level of confidence with the course content (Donaldson, 1988;

Knowles, Holton, & Swanson, 2005), adult learning methodology (Bash, 2005; Fleming

& Garner, 2009; Galbraith, 2004; Merriam, 2001), and institutional policies and

procedures (Wilson & Elman, 1990). A researcher-developed questionnaire was used to

measure this variable.

Part-time Instructor: The instructional faculty who are not contracted full-time

with the University. Typically, a part-time instructor’s teaching load is less than 50% of

the load carried by full-time, contracted faculty. Contracts for part-time instructors are

issued on an individual teaching assignment basis. Part-time instructors do not have

dedicated office space or benefits such as health insurance, retirement, sick days, etc.

(AAUP, 2006; Barnetson, 2001; Fulton, 2000; Magner, 1999).

Peer-to-Peer Mentoring: Experiences of adjunct faculty who have been mentored,

coached, and guided by other, part-time instructors (Kram & Isabella, 1985; Routman,

2000). The researcher-developed fidelity subscale was used to gather data on this self-

reported measure.

Student Satisfaction: The student self-reported perception of the instructor’s

effectiveness in terms of overall course quality, teaching skill, and availability (Appleton-

Knapp & Krentler, 2006; Kelly, Ponton, & Rovai, 2007; Sproule, 2002). An aggregate of

the University’s end-of-course survey forms was used to measure this variable.

Organization of the Study

This chapter provided the reader with the history and background of the problem.

I offered social learning theory and Likert’s System 4 Management Theory as appropriate

theoretical frameworks for the study. The research problem and general research

questions were given along with operational definitions of key terms and variables.

Chapter 2 begins with a review and analysis of relevant literature. Literature

relevant to the phenomenon of rise in use of part-time instructors is noted. The concerns

and benefits regarding the use of a large number of part-time instructors are investigated.

The literature review also examines the roots of mentoring and its use by post-secondary

institutions to train and orientate new faculty members. The concept of peer-to-peer

mentoring is investigated; this includes a review of studies and other relevant literature

that examine the benefits of using peer mentors as opposed to non-peer mentors.

As universities continue to turn to the use of part-time faculty, the issues of

availability and loyalty arise. New phrases such as “Roads Scholars” and “Freeway

Faculty” have entered the post-secondary institutions’ vocabulary. I examined this trend

and its relationship to institutional loyalty. Elements that contribute to instructor

confidence and its role in teaching effectiveness are shared. Also, student satisfaction as a

measure of instructor effectiveness was investigated.

Chapter 3 provides the reader with detailed information regarding the research

design. This includes information on the sample and selection criteria. This chapter also

outlines the data collection plan and includes information on the survey instruments and

other tools used to gather data related to the study. Chapter 4 describes the setting of the

research and the peer-mentoring program, which comprises the “intervention” used in the

collection and analysis of the data. The data analysis plan will be explained. Chapter 5

provides the results of the study, and Chapter 6 discusses the conclusions and

recommendations, including thoughts for further research.

CHAPTER TWO

LITERATURE REVIEW

The American Association of University Professors (AAUP), the Association of

American Colleges and Universities (AACU), and others have expressed concern that

universities are too dependent on part-time faculty members, and this dependency

threatens the academic integrity of the American academy (AAUP, 2006; Benjamin,

2002). This chapter reviews literature as it relates to the previously mentioned issues.

First, I review the historical perspective, which includes statistics related to the

rising use of part-time faculty members as well as some of the reasons for this continuing

trend. I discuss critical issues surrounding the debate in the rise in use of part-time faculty

members. As outlined in Chapter 1, instructor confidence, institutional loyalty, and

student satisfaction are important issues related to this dependence on part-time faculty

members and will be discussed in section two. Finally, the practice of mentoring, which

includes information related to peer mentoring, will be examined in general with a closer

look at the peer mentoring. This literature review will assist in building a framework to

help answer the research question, “What is the relationship between peer-to-peer

mentoring and a part-time faculty member’s instructor confidence, institutional loyalty,

and student satisfaction scores?”

Part-time Faculty

The Rise in Use of Part-time Faculty

The use of part-time faculty is not the sole purview of the modern post-secondary

institution. As far back as the middle ages through the post-Civil War era, part-time

faculty were used in order to provide expertise that was lacking among full-time faculty

(Jacobs, 1998). Schuster and Finkelstein (2006) believe that “faculty is central to the

well-being of the academy” (p. 3). The powerful union, the American Association of

University Professors (AAUP), could not agree more. The AAUP asserts, “The integrity

of higher education rests on the integrity of the faculty profession” (AAUP, 2003, p. 69).

Bowen and Schuster (1986) state the matter plainly; there can be no question, “The main

duty of every institution of higher education is to place a competent faculty” (p. 3).

This is the central debate over the use of part-time faculty members reduced to its

most fundamental elements. The use of part-time faculty has come a long way since

priests and scholars roamed the European countryside, visiting monasteries and

universities to study and offer expertise (Toutkoushian & Bellas, 2003). These authors

assert that a century of progress has been reversed because the American academy has

come to over-rely on part-time faculty members (Bowen & Schuster, 1986).

Consider that in 1960, according to Bowen and Shuster (1986), 35% of faculty

appointments were for part-time faculty members. Benjamin (2002) asserts that use of

part-time appointments over full-time ones rose 103% between the years of 1975 and

1995. In 1998, the AAUP claimed that the 25 years from 1973 to 1998 saw a substantial

increase in the number of universities relying on part-time appointments.

In 1998, university administrators gave 43% of all faculty appointments to part-

timers. In 2004, this trend and reliance on part-time faculty has remained consistent,

increasing by several percentage points, from 43 to 49.3% (Snyder & Dillow, 2012).

Leatherman (2000a), in an article entitled, “Part-timers Continue to Replace Full-timers

on College Faculties,” observed that adjuncts comprise nearly 50% of the professoriate.

This is consistent with more recent figures from the AAUP that estimates that 48% of

faculty in U.S. institutions are part-time, non-tenure-track faculty (AAUP, 2009). Many

believe these numbers are “gravely underestimated” (Wilson, 1999, p. 15).

In an interview conducted by Rice (2004), Finkelstein and Schuster agreed that

the American professoriate is “to a considerable degree, a part-time profession. . . .

Faculty members in the U.S. are currently split close to 50-50 between part and full-time”

(p. 28). Some universities, particularly the University of Phoenix, use exclusively part-

time faculty members (Feldman & Turnley, 2001).

This reliance grew, not only in 4-year colleges, but in community colleges and

private institutions as well. Between 1972 and 1977, both private and public universities

experienced a sharp decline in their financial support (Ivey et al., 2005). As a result, since

the 1980s, the majority of new hires were part-timers and not tenure-track faculty. In

2003, the AAUP (2006) stated that full-time, tenure-track faculty positions comprised

only 24% of the faculty labor force. Ivey et al. (2005) compared this to 1969 when 96.7%

of new hires were for full-time, tenure-track positions.

When one reviews the literature, which seems replete with dire warnings

regarding the consequences associated with using part-time faculty, one has to wonder,

“Why does the trend continue?” There are four key reasons why universities continue to

depend on part-time faculty: the G.I. Bill, money, flexibility in scheduling, and supply

and demand.

The G.I. Bill

At the end of World War II, almost 30 million American military personnel

employed by the war efforts were thrust back into American life. This number comprised

almost one-quarter of the entire American workforce. Recruiting and drafting efforts took

millions of young people out of school and sent them to war (Mosch, 1975). The

resulting influx of millions, coupled with an educational deficit, set the stage for an

unprecedented domestic crisis (Mosch, 1975).

As a result, the G.I. Bill, signed by Franklin D. Roosevelt, provided educational

benefits, among other provisions, to World War II veterans (Toby, 2010). By the time the

bill’s initial provisions ended in 1956, over 7.8 million of the 16 million World War II

veterans took advantage of these resources; in 1947, almost half of U.S. college students

were veterans (U.S. Department of Veterans Affairs, n.d.). Universities worried about the

educational consequences of admitting large numbers of veterans; they could not expand

facilities or faculty fast enough to accommodate this influx of students (Toby, 2010).

In 1952 and again in 1966, this bill was extended to offer education, grants, and

job-training skills to more veterans (Mosch, 1975). Twenty years later, baby boomers

(who were coming of age) flocked to universities in ever-increasing numbers, which

resulted in another boon and even more increased enrollments (Bowen & Schuster,

1986). This increase in enrollment, along with an increase in the number of qualified,

full-time faculty who were retiring, resulted in an unprepared university system turning to

part-time faculty members (R. Smith, 1980).

The Money Factor

Like any business, educational administrators have had to look for ways to

increase revenues and decrease costs. So it comes as no surprise that, without exception,

most agree that this rise in use of part-time faculty over the past 40 years is primarily due

to Cherchez L’ Argent, or the money motive (Noble, 2000). In 1980, federal and state

governments subsidized 46% of the cost of higher education; by 2003, the percentage had

decreased to 35% (Lerber, 2006). In addition, federal financial aid polices became more

restrictive (German, 1996). Because of this decline in government spending, colleges and

universities had to cut costs to make ends meet (Ochoa, 2011). By not hiring full-time,

tenure-track faculty, universities were able to realize a cost savings. Once a full-time

position was cut, the faculty dollars saved were rarely seen again. Instead, universities

began spending whatever surplus funds were realized on the school’s physical

infrastructure and technology (Ochoa, 2011).

Consequently, the use of part-timers mushroomed into a source of cheap labor

(Magner, 1999). In fact, Noble (2000) estimates that the use of adjunct faculty results in

an almost 42% net gain for the bottom line of universities. Noble (2000) states, “Lower

faculty status is associated with the production of a larger net gain” (p. 94).

Generally, universities pay part-time faculty members less than full-time faculty

and these institutions rarely provide benefits or job advancement opportunities

(Barnetson, 2001; Fulton, 2000; Magner, 1999; Ochoa, 2011). Marcus (1997) observed,

“Because they are so cheap, institutions really come to depend on them” (p. 10). Terry-

Sharp (2001) surveyed 421 anthropology departments on their utilization of part-time

faculty and learned that 60% of the part-time instructors employed in these departments

earned less than $3,000 a course. Terry-Sharp (2001) observes that “this salary is

comparable to salaries received by fast-food workers, baggage porters, or theatre lobby

attendants” (p. 20).

In addition, part-time faculty members are less likely to have access to health

benefits, retirement plans, or life insurance (Fulton, 2000; Marcus, 1997; Ochoa, 2011;

Rhoades, 1996; Terry-Sharp, 2001). In fact, more than 60% of the institutions that

responded to the Terry-Sharp (2001) survey offered part-time faculty no benefits at all.

The compensation paid to adjunct faculty becomes an issue of quality. Shakeshaft (2002)

remarked that it is unrealistic to expect additional time and commitment from a

professional who is “being paid peanuts” (p. 29).

Toutkoushian and Bellas (2003) analyzed the data from the 1993 National Survey

of Post-secondary Faculty (NSOPF) and found that contrary to Terry-Sharp, part-time

faculty are satisfied with their salaries and jobs overall. While part-time faculty may be

paid lower for teaching classes, when all sources of income are combined, part-time

faculty salaries compare quite favorably with their full-time counterparts. Mello (2007)

disagrees with those who claim adjuncts are dissatisfied. According to Mello, adjuncts

are not teaching for the money. “Many part-time faculty have careers outside of academia

and teach part-time primarily for the intrinsic rewards of teaching” (Toutkoushian &

Bellas, 2003, p. 191). According to Noble (2000), “adjunct faculty are often more willing

to accept symbolic instead of hard currency rewards for their contributions” (p. 90).

Flexibility in Scheduling

In addition, the face of the college student has changed (Ochoa, 2011). According

to the National Center for Education Statistics (2002), 73% of all undergraduate students

can be classified as non-traditional. This means that universities must look for new and

creative delivery methods to accommodate these working adults. Teaching evening and

weekend classes typically does not appeal to full-time, tenured faculty (Leatherman,

1997).

Some believe that administrators welcomed this increase in the use of part-timers

not only because it saved money but also because it provided for flexibility in scheduling

(Barnetson, 2001; Bradley, 2004; Fulton, 2000; German, 1996; Ivey et al., 2005; Marcus,

1997). This flexibility is important as it allows administrators to cope with the fluctuating

enrollment trends (Leatherman, 2000b). Adjuncts are much more likely to accept evening

and weekend teaching assignments than their full-time counterparts (Leatherman, 1997).

Supply and Demand

The age-old, free-market concept of supply and demand is another leading cause

in the rise of part-time faculty appointments. There is simply an abundance of qualified

individuals; it makes sense that administrators would take “sensible advantage” of this

increase in supply (German, 1996, p. 236). Others echo this sentiment. Bowen and

Schuster (1986) noted that in the 70s and 80s, part-time faculty appointments increased

because of the large numbers of individuals with advanced degrees who did not seek or

could not find full-time employment. Not enough jobs to go around meant that job

seekers were “forced to piece together a living from a patchwork of part-time positions”

(Bradley, 2004, p. 29; Fulton, 2000). This phenomenon led to the coining of such phrases

as Roads Scholars (Tillyer, 2005), Add-Ons (Beem, 2002), Freeway Flyers (Will, 1997),

and Gypsy Faculty (Bowen & Schuster, 1986). The transitory nature of part-time faculty

appointments has contributed to the debate regarding the over-reliance on part-time

faculty members. The following section will discuss this debate.

The Debate Related to the Use of Part-time Faculty

Bowen and Schuster (1986) firmly assert that the excellence of higher education

is a function of the type of faculty each university is able to recruit and retain. Smith

(1980) could not agree more; he claims that the use of part-time faculty actually

undermines academic excellence. So, it should come as no surprise that the debate

regarding the use and over-reliance on part-time faculty continues. The review of the

literature suggests that concerns over the use of adjuncts center on the quality of

instruction they provide, the lack of academic freedom afforded to part-timers, and job

security coupled with other financial considerations. This section concludes with a brief

look at the benefits that using part-time faculty members can bring to the post-secondary

classroom.

Quality of Instruction

Much of the literature regarding the use of part-time faculty members centers on

the quality of instruction (AAUP, 2003, 2006). The literature assumes that full-time

faculty members provide better instruction and hold students to a higher academic

standard. Ochoa (2011) asserts that the overabundant use of continent or part-time faculty

could potentially threaten the quality of undergraduate education in America. Ochoa cites

the Study Group on the Conditions of Excellence in Higher Education (1984) who

concluded that “strong faculty identification with the institution and intense faculty

involvement with students requires a primary commitment” (Study Group, 1984, p. 36)

and are necessary to facilitate excellence in undergraduate educational programs. This

primary commitment is jeopardized by the very nature of part-time faculty appointments

(Ochoa, 2011).

Another fact that could contribute to the debate regarding the quality of

instruction is the fact that universities recruit and select full-time faculty with more care

than they do for part-timers (Shakeshaft, 2002). One author noted that the recruiting of

part-timers is haphazard, last minute, and accomplished without consulting full-time

faculty (Balch, 1999). It seems evident that a direct correlation exists between the

excellence in higher education and the quality of the instructional staff (Bowen &

Schuster, 1986).

Some authors believe that questions of the quality of instruction provided by part-

time faculty are raised by unions (like the AAUP) and accrediting agencies who are run

by and for the benefit of tenured faculty (Selingo, 2008). In fact, Smith (1980) found that

full-time professors do not welcome the use of part-timers and see the use of adjuncts as a

very serious problem. This is a somewhat ironic observation as Barnetson (2001)

suggests that “full-time, permanent faculty has (intentionally or inadvertently) facilitated

the growing use of part-time and limited-term faculty” (p. 97) by refusing to accept the

less desirous teaching assignments scheduled during the evening and on weekends. The

only way for full-timers to protect their stake in the institution is to declare non-tenured

faculty as incompetent (Selingo, 2008).

Full-time faculty members believe their institutions hire too many part-timers and

that the instruction provided by these adjuncts is marginal (Bowen & Schuster, 1986).

Balch (1999) states that 80% of full-time faculty expressed concern about the large

number of part-time faculty who teach courses in their colleges and universities. The

controversies are related to the quality of instruction, availability of part-timers to meet

with students outside of class, and the lack of continuity in academic planning (AAUP,

2006; Balch, 1999; Ochoa, 2011).

Klein et al. (1996) confirm this conclusion in their study of social work programs.

They found that while part-timers were enthusiastic and exposed students to the reality of

the social work profession, they lacked integration with the full-time faculty, were

isolated from curriculum and research discussions, and lacked quality in their teaching

when compared to full-timers (Klein et al., 1996). Klein and his associates do admit that

research regarding the effectiveness of part-timers is “scant” (Klein et al., 1996, p. 2).

The Role of Academic Freedom

The conclusion that full-timers provide better quality instruction is largely due to

the academic freedom afforded full-time faculty (Ochoa, 2011). Benjamin (2002)

concluded that the institutional practice regarding the excessive use of part-time faculty

members erodes academic freedom. Bradley (2004) contends that tenure is the best

protector of academic freedom and, by extension, instructional excellence. The AAUP

(2003, 2006) agrees that tenure is the only secure protection for academic freedom. Chait,

a professor of higher education at Harvard University, studied the faculty handbooks of

250 colleges and found that only 10 had policies related to academic freedom that

explicitly mentioned adjuncts (Schneider, 1999). Without this protection, it becomes all

too easy to drop part-timers at the stroke of a pen (Schneider, 1999; Sonner, 2000).

Adjuncts feel less confident and, as a result, less empowered to voice opinions

(AAUP, 2006; Bradley, 2004). As a result, part-time instructors tend to “fly under the

radar” and go to great lengths not to “rock the boat” (Tillyer, 2005, para. 26). Adjuncts

are reluctant to complain or to bring concerns to the administration because of the lack of

job security (Ochoa, 2011; Thompson, 2003). Thus, the part-time faculty member is

required to be circumspect in order to keep the right people satisfied.

This would be in direct contrast to full-time, tenured faculty members who,

because of job security, have the freedom to address issues and concerns without fear of

reprisals. Bradley (2004) observed that contingent faculty have every incentive to avoid

taking risks and/or tackling controversial topics. Part-time faculty are vulnerable to

student complaints and, as a result, may not teach as rigorously or give low grades to

students who earned them (Bradley, 2004). Part-time faculty are “beholden to individual

administrators for their jobs,” so these contingent faculty members “avoid any actions

that might offend either administrators or students” (AAUP, 2006, pp. 9-10). In essence,

“open mouths lead to closed doors” and this results in a lack of confidence on the part of

adjunct faculty members to confront issues of academic quality as they relate to student

performance (Schneider, 1999, para. 5).

Job Security and Other Financial Considerations

As outlined in the previous section, the lack of job security is a major factor that

contributes to the belief that the use of part-time faculty members negatively affects

quality instruction and student learning (Ochoa, 2011). Cost-savings is one major factor

indicated in the literature as to why the use of part-time instructors has peaked and

remains high (Magner, 1999; Noble, 2000). The savings in terms of salary is evident.

However, other savings occur, as institutions are reluctant to invest in the professional

development of adjunct faculty members.

A study of 421 anthropology departments, conducted by Terry-Sharp (2001),

found that only 23.6% of part-time faculty members receive support to attend workshops.

Terry-Sharp (2001) contends that universities provide insufficient support to provide

necessary faculty development opportunities to part-time instructors. The failure to invest

in faculty development directly influences the institution’s ability to improve instruction.

Institutions refuse to make long-term commitment to the academic careers of part-timers

(Lerber, 2006). Tillyer (2005) observed that administrators see no benefit to allocating

faculty development funds for use by part-timers. One common theme recommended

throughout the literature to improve instructional quality is professional development

(AAUP, 2006; Balch, 1999; Crannell, 1998; Feldman & Turnley, 2001; Howell &

Tweedell, 2007; Irvin, 2003; Klein et al., 1996; Selingo, 2008; Shakeshaft, 2002; Terry-

Sharp, 2001).

Contributions of Part-time Faculty

Certainly, the literature reports that the rising use of adjunct faculty reflects, at

least in part, a need to cut costs and increase flexibility in scheduling. Although there is

much rhetoric surrounding the lack of quality, there are many voices raised to negate

these conclusions. Many authors note the real-world experience and expertise in the field

that adjuncts bring to the classroom (Shakeshaft, 2002; Sonner, 2000). This infusion of

faculty, who have not been protected by the ivory tower of academia allows for a fresh

perspective and a chance to cover new topics in the curriculum (Mello, 2007). Balch

(1999) believes that part-time faculty members are highly committed, conscientious, and

provide strong links to the community. In addition, they offer “a wide range of variety in

experience” (Balch, 1999, p. 33). Irvin (2003) studied the use of part-timers in graduate,

non-profit programs and concluded, “Part-time faculty ensure a more diverse teaching

force with stronger representation from the people in direct practitioner roles” (p. 181).

Regardless of which side one takes in this debate, the conclusion that Adams

(1995b) drew from her study of part-time nursing faculty makes the most sense. “Since

quality education is the ultimate goal of the institution, the faculty member who can

provide the student with the best guidance is the faculty who should be hired, evaluated

and retained” (Adams, 1995b, p. 296). It is important now, as the debate continues, that

administrators and professors decide how to define the term “best.” It is evident that part-

timers are overworked, underpaid, underappreciated, and have little input into decisions

(Adams, 1995b). Fulton (2000) agrees that at the heart of the debate is the belief that part-

time faculty members are overused and abused. Still, we cannot forget that “adjuncts are

a source of expertise that greatly enriches the educational experience of students” (Arden,

1989, p. A48).

It is for these reasons that this dissertation seeks to explore the nature of this

dependence on part-time faculty and the critical issues identified in the literature of

instructor confidence, institutional loyalty, and the quality of instruction that adjuncts

provide as measured by student satisfaction scores.

Issues Related to Part-time Faculty

Instructor Confidence

Bowen and Schuster (1986) define faculty as the professional person with

substantial learning employed by higher educational institutions to teach, to conduct

research, and to serve. After reviewing the literature regarding the use of part-time

faculty members, it was evident that some scholars see a “perceptual divide” and that

part-time faculty members are viewed as outcast members of the scholarly community

(Benjamin, 2002; Rice, 2004). Foster and Foster (1998) refer to part-time faculty as

“second-class citizens” (p. 30). It seems only logical that the uncertain nature of part-time

faculty teaching assignments and the lack of professional development and support would

result in faculty who are uncertain and lack confidence to fulfill their roles as easily as

full-time faculty do.

Considering Bowen and Schuster’s (1986) lofty definition of the professoriate, it

is no wonder that some have asked whether corporate America would continue to prosper

and provide quality services and products if 65% of its workforce were part-time (Balch,

1999). Marcus (1997) echoes these thoughts: “It is hard to think of any industry that uses

so many part-time professionals as a regular part of their staff” (p. 9). Giamartino, the

dean for the business school at the University of Detroit Mercy, asked, “How can you

build an entire curriculum and academic program around part-time faculty” (Leatherman,

1998, p. A14)?

According to Arden (1989), higher education’s use of part-time faculty members

is the major scandal in American higher education. Arden (1989) continues further and

refers to part-time or adjunct faculty members as “the invisible people of higher

education” (p. A48). One author went so far as to state that the increased use of part-time

faculty members constitutes “a sneak attack on academic values and on the stability of

the faculty as a whole” (Bradley, 2004, p. 28). These phrases hardly conjure up images of

a highly trained, confident, and effective labor force. But what do we mean by instructor

confidence?

Instructor Confidence Defined

A popular online dictionary defines an instructor as one who teaches (The

Essential American Heritage Dictionary, 2011). The editors particularly note that the

individual teaches college and is a rank below that of assistant professor. Confidence is

defined as the state of being certain (The Essential American Heritage Dictionary, 2011).

So, instructor confidence is the degree to which a faculty member is certain about the

skills and knowledge related to the act of instructing or teaching.

While there appears to be no consensus on what constitutes effective instruction at

the post-secondary level (Kelly et al., 2007; Smith & Welicker-Pollack, 2008), some

common themes do emerge from the literature. Reviewing books and articles related to

effective instruction, particularly with regard to non-traditional students, it appears that

understanding the course content and the ability to apply it to practical situations is a key

component (Beder & Darkenwald, 1982; Brookfield, 1986; James & Maher, 2004;

Pearson, 2005). In addition, understanding and accommodating the unique needs of

students through lesson planning and learning methodologies is also important (Bash,

2005; Cross, 1981; Fleming & Garner, 2009; Knowles & Associates, 1984; Knowles

et al., 2005; Merriam, 2001). Finally, it would seem that effective instructors, who

operate within an organizational setting, should have an understanding of the

organization in which they teach, which includes its mission, purpose, organizational

policies, as well as the ability to enforce them appropriately (ASHE-ERIC, 1995; Boyle

& Boice, 1998; Kram & Isabella, 1985; Mullen, 2000; Simmons, 1998).

So, for the purposes of this study, instructor confidence can be understood to be

the degree of certitude that a part-time faculty has with regard to his or her ability to

understand course content, to apply that content pragmatically, to understand student

needs, to provide developmentally appropriate lessons and methodology, and to

understand, interpret, and apply key organizational polices.

Role of Inclusion in Instructor

Confidence

Santovec (2004) observes that building a collaborative and inclusionary teaching

culture is a way to prevent feelings of isolation and second-class citizenship. Feldman

and Turnley (2001) studied 105 non-tenured-track faculty members concerning job

satisfaction and professional commitment. One participant noted, “I have not had any

contact with any other members of the department, nor have I ever been invited to take

part in faculty meetings” (p. 14). Five years earlier, German (1996) studied the part-time

faculty trend and noted that little time is placed on the development of part-time faculty

and no institutional commitment is made to them. He concludes by saying, they “have

become invisible in our midst” (German, 1996, p. 237). The reality is that the

professional development of part-time faculty members has been ignored (AAUP, 2003).

Rhoades (1996) examined the contracts of 183 institutions and found that while

part-timers deliver instruction, there are few offers for professional growth and

development. He specifically noted that this practice was in direct contrast to the

literature, which stated that to improve instruction, part-timers need inclusion. Rhoades

(1996) found that out of the 183 contracts, only 10 provided some sort of professional

development and none included part-timers in the academic decision-making process. To

improve quality instruction we need to enable part-time faculty members to contribute to

the quality of the educational program and integrate them into the lives of departments in

which they teach (Rhoades, 1996). The AAUP (2006) notes that part-time faculty have

limited interaction with colleagues, they teach in isolation, do not understand how the

courses they teach fit into the overall instructional design, and are excluded from the

“broader departmental or institutional governance” (p. 9). This results in a “substantial

limitation” as part-timers fulfill their roles as teachers (p. 9). The literature could not be

clearer; yet, it is evident that our practice continues to lag behind our knowledge.

Many authorities believe that instructional quality is associated with adjuncts

belonging to the larger community, providing them with support and professional

development opportunities, and including them in curriculum development and other key

decision-making discussions (Ochoa, 2011; Thompson, 2003). Arden (1989) noted that

there are “important educational advantages for making adjunct members a part of the

team” (p. A48). The University uses part-time faculty almost exclusively in their degree-

completion programs. They contend that properly leading and managing the adjunct

faculty pool can “maximize student learning, ensure consistency in outcomes, and build a

culture of excellence” (Howell & Tweedell, 2007, para. 1). As a result, the University has

found that this training and support results in part-time faculty members who are

confident and “enthusiastically support” the mission of the University (Howell &

Tweedell, 2007, para. 2).

The conclusion is obvious: Providing training and professional development

opportunities to part-time faculty has the potential to increase confidence and quality of

instruction (Ochoa, 2011). Adams (1995b) in her study on part-time nursing faculty noted

that an adjunct’s overall job satisfaction increased when administrators expressed

appreciation regarding their contributions and when the adjunct was included in

orientation and other professional development activities. There is the belief that “quality

instructors will want to teach for institutions that demonstrate a commitment to quality

and to its employees’ professional and personal development and overall satisfaction”

(Santovec, 2004, p. 5). In addition to building confidence and competence, these

inclusionary overtures have the potential to result in an adjunct’s sense of belonging to

the institution.

Institutional Loyalty

If it is true, as some of the literature claims, that adjuncts bring richness to the

educational experience of students, how do we include these “invisible people” (Arden,

1989, p. A48) into the larger scholarly community? The matter becomes even more

complex when we consider that many adjuncts teach at more than one institution (AAUP,

2006; Reichard, 1998; Tillyer, 2005) and that 66% of them report being employed full-

time elsewhere (Fulton, 2000). Since a large number of part-timers are employed

elsewhere and the literature has identified that inclusion into the larger community is

essential to improving instruction, increasing an adjunct’s sense of loyalty to the

university in which he or she teaches is of paramount importance.

Transitory Nature of Part-time

Appointments

As early as 1986, the literature began recognizing the transitory nature of part-

time appointments by coining such phrases as Gypsy Faculty (Bowen & Schuster, 1986),

Freeway Flyers (Will, 1997), and Roads Scholars (Tillyer, 2005). These terms hardly

conjure up a vision of professionals who are loyal and committed to an organization. The

sense is that, like migrant farm workers, part-time faculty move with the seasons and

follow the crops to the next paid teaching assignment.

This pattern contributes only to the sense of isolation that many researchers note

as a concern. This isolation means that those who are teaching the most classes are least

familiar with campus culture (Leatherman, 1997). Balch (1999) noted that most post-

secondary institutions are remiss in integrating part-time faculty into campus life; we

“need to work to ensure their integration and participation in each institution’s faculty

community” (pp. 33-34). Jones (2000) agrees that we do not have a systematic plan for

supporting and including part-time faculty.

Institutional Loyalty Defined

Ashforth et al. (2008) observe that “commitment represents a positive attitude

toward the organization” (p. 333). Meyer and Allen (1988) found in their work on

organizational commitment that employees who are favorably disposed to an institution

will remain and work to contribute to its success. They claim that increasing commitment

is the first step to reducing turnover. In fact, Mello (2007) claims that if a part-time

faculty member develops a “deep connection to one school, they’re unlikely to be

tempted away by better salaries” (p. 44).

While it is true that the organization plays an important role in employee

commitment, the type of work an employee does is also essential. The work that

employees do must be satisfying and meet basic needs (Meyer & Allen, 1988).

According to Edmonson, Fisher, Brown, Irby, and Lununburg (2002), these basic needs

include control, meaning in situations, and positive support. When organizations and

leadership meet these basic human needs, commitment and loyalty are increased (Angle

& Perry, 1986).

The nature of the attachment that forms between individuals and their

organizations is the essence of institutional loyalty (Ketchland & Strawser, 2001). The

fact is that individuals do form opinions and beliefs regarding the degree to which an

organization values their contributions and their well-being (LaMastro, 2000). The

stronger the individual’s positive regard in those areas, the stronger his or her

identification and involvement with a particular organization. The result is more loyalty

to the institution and more energy put forth to remain connected (Hafer & Martin, 2006;

Ketchland & Strawser, 2001). Angle and Perry (1986) refer to the dynamics of this

relationship as the “norm of reciprocity” (p. 128).

Simply put, when people are on the receiving end of kind regard and good works,

they are strongly prone to reciprocate in kind (Angle & Perry, 1986). Angle and Perry

describe this as a “psychological contract” (p. 128); when members of the organization

offer their commitment in exchange for the organization’s support and acknowledgment

of contributions, this contract is fulfilled. When organizational leaders ensure that the

employees’ needs are met, their loyalty and commitment increase. Angle and Perry

studied bus operators from 24 mass transit systems in the United States (n = 1099). They

discovered that the most significant predictor of organizational commitment was the

institution’s treatment of the employee. Essentially, they observed that “organizational

commitment is largely a matter of reciprocation between individual and organization. . . .

Our data suggest that the principal antecedents of commitment may be well within

management’s capacity to influence” (Angle & Perry, 1986, p. 144).

Organizational commitment is the phraseology adopted by Mowday, Porter, and

Steers (1982) to refer to this idea of being loyal to an institution. Mowday et al. define

this concept as “(a) strong belief in and acceptance of organizational goals and values,

(b) a willingness to exert considerable effort on behalf of the organization, and (c) a

strong desire to maintain membership in the organization” (p. 27). Ashforth et al. (2008)

agree that through “close intimate personal cooperation” (p. 326) the interests of the

employee and company become identical; the outcome of this mutuality is a willingness

on the part of the individual to increase his or her efforts on behalf of the organization.

This idea of reciprocation with regard to institutional loyalty is an important one. The

AAUP (2006) articulates the paradox that exists when institutions ask teachers to commit

to “them, their mission, and their students” (p. 16) yet withhold institutional commitment

to the faculty employee in terms of inclusion, support, and professional development

opportunities.

Types of Institutional Loyalty

Meyer, Allen, and Smith (1993) are quick to note that not all forms of

organizational commitment are equal. In their research, they have conceptualized the

following three components to this construct: affective, continuance, and normative.

Affective commitment is an emotional attachment and occurs when the employee

identifies with the organization and exhibits a willingness to help achieve the institution’s

goals.

Continuance commitment develops gradually, over time, as the employee

accumulates investments (such as promotion, salary increases, seniority). The greater and

larger the investments, the greater the threat of loss and the less willing the employee is

to risk losing them (Meyer et al., 1993). Continuance commitment typically grows out of

economic factors (Meyer & Allen, 1984). When the employee develops feelings of moral

obligation and responsibility, they remain with the organization because they have a

normative commitment (Meyer et al., 1993). LaMastro (2000) summarized the

definitions of these three concepts by noting that affective commitment means the

employee stays because he or she wants to. If the individual stays because he or she

needs to, that is continuance commitment. Finally, employees who exhibit normative

commitment stay because they ought to.

According to Meyer et al. (1993) all three forms of commitment are negatively

correlated with turnover. Affective and, to a lesser degree, normative commitment should

be positively related to job performance. This means that an employee’s commitment to

an organization should result in improved quality in his or her job performance and a

willingness to remain connected. Ketchland and Strawser (2001) support these

conclusions; they note a significant relationship exists between affective commitment and

organizational commitment. Employees with high levels of affective commitment are less

likely to consider other job offers or explore alternative job opportunities. This is one key

reason why increasing the loyalty of part-time adjuncts to their universities is important.

LaMastro (2000) supports this conclusion in her study of full-time, K-12 teachers

in the state of New Jersey. She found a correlation between perceived level of

organizational support and affective commitment. She also found a positive correlation

between perceived organizational support and daily mood, suggesting that feelings of

support and feelings of being valued can lead to greater excitement and enthusiasm for

teaching. She concludes her study by saying, “Administrators who take steps to support

employees will reap the benefit of strongly committed teachers willing to ‘go the extra

mile’ for their students and their schools” (LaMastro, 2000, p.9).

While LaMastro (2000) studied elementary and secondary school teachers,

Edmonson et al. (2002) report similar results as they describe the collaborative culture of

Sam Houston State University’s Department of Educational Leadership. They observed

that higher morale, an enhanced commitment to teaching, and a desire to continue in the

profession are all outcomes of an increased collaborative culture. They describe a

collaborative culture as one with mutual trust and respect. The environment must present

a safe venue for honest and open communication. In addition, leaders need to recognize

the contributions of their members. These authors stress that you cannot leave such a

culture to “chance” (Edmonson et al., 2002, p. 11). Tyrrel and Stine (1997) support these

conclusions:

Leadership practices that emphasize cooperative relationships and a shared vision can

create schools that aim for excellence. . . . The emphasis on cooperative relationships

can help bring about a work climate in which self-esteem, commitment, and task

accomplishment are so significant that they raise people to a higher level. (p. 34)

The benefits of strategically and systematical seeking ways to include part-time

faculty members in the larger, educational community are numerous. In addition to those

noted above, Ashforth et al. (2008) note that the increasing organizational commitment

can result in the following outcomes: “cooperation, effort, participation and

organizationally beneficial decision-making, low turnover, higher retention, job

satisfaction, work adjustment, increased social support and helping behaviors, customer

orientation, positive evaluation of organization, and defense of the organization” (p. 337).

When we consider the large number of positive outcomes associated with high

levels of institutional loyalty, it is no wonder that the insight gleaned from the literature

regarding part-time faculty and organizational commitment underline the importance of

seeking ways to increase adjuncts’ inclusion in the larger scholarly community. It is

likely that issues of quality instruction and isolation may be ameliorated simply by

increasing the part-timers’ perceived level of organizational support. Administrators who

treat part-timers well as “critical assets will find their adjunct professors are loyal,

productive, and insightful members of the faculty team” (Mello, 2007, p. 44). It seems

obvious that organizations will thrive only when “the individuals within the organization

flourish” (Krajewski, as cited by Edmonson et al., 2002). It seems reasonable, then, to

assume that part-time faculty members who are flourishing in their roles as educators,

who exhibit high levels of self-esteem, and who are loyal to their institutions would have

students who express a high degree of satisfaction with their learning experiences.

Student Satisfaction

As mentioned previously, it seems only logical that when members of an

organization feel supported and are flourishing, the results will be highly satisfied

customers (Likert, 1967). In the post-secondary setting, this means students, who are the

consumers of the educational product we provide, are highly content with the quality of

their educational experience. This is particularly true because student evaluation of

instructor teaching effectiveness is the tool most frequently used when the decision to

assign additional classes to part-time faculty members is made (Kelly et al., 2007;

Sproule, 2002). For this reason, measures of student satisfaction are an important factor

to consider when discussing issues related to part-time faculty members.

Student Satisfaction Defined

What do we mean by satisfaction? Oliver (1999) defines satisfaction as the

perception that the fulfillment of a service or product has resulted in feelings of pleasure.

Appleton-Knapp and Krentler (2006) build on this concept when they explain that a good

predictor of student satisfaction is the extent to which an instructor meets the student’s

expectation. Typically, one gauges satisfaction with a product or service after the fact

(Appleton-Knapp & Krentler, 2006). In the university or college setting, this means that

when students are successful, they are learning and they are satisfied with their

educational experiences (Kariotis, 2000). Student perceptions of their educational

experiences are most frequently collected using student evaluation of teaching (SET)

forms.

Students complete SETs using standardized forms at the conclusion of a course.

University representatives provide the forms or surveys to students in a well-defined and

controlled process (Otto, Sanford, & Ross, 2008). This controlled process usually has

students completing the form in class and a paid staff member processing the results

(Arden, 1989; Sproule, 2002). Because students may have radically different perceptions

of similar experiences and they may not recall the past correctly, some question whether

such a tool is a viable means for evaluating instruction (Appleton-Knapp & Krentler,

2006).

Sproule (2002) believes that SETs are widely used because they are inexpensive

to administer. Questions arise as to the validity of these measures, particularly in light of

the fact that administrators and leaders base important personnel decisions, such as

promotion, tenure, and additional teaching assignments, on the findings (Kelly et al.,

2007; Sproule, 2002). Otto, Sanford, and Ross (2008) agree that the SET has become an

ever-present measure of teacher effectiveness and will continue to be the major

consideration for promotion, tenure, and merit pay. Sproule (2002) believes SETs, while

riddled with errors and hopelessly flawed, “will continue to be a cornerstone of teaching

assignments” (Sproule, 2002, p. 292).

Validity of Using Student

Satisfaction Surveys

Part of the reason that SETs are so controversial is that a lack of clarity exists

regarding what the standards are that comprise excellence in post-secondary teaching

(Kelly et al., 2007; Smith & Welicker-Pollack, 2008). Others cite the lack of validity of

SETs as a measure of instructional effectiveness because such tools are essentially

popularity contests and that crowd-pleasing instructors get high evaluation marks

regardless of whether or not students learned (Clayson & Sheffet, 2006). Felder (1995)

disagrees and states that the idea that SET constitutes a popularity contest is a myth. In

addition, Boice (1992) observes that measures of “personality and popularity correlate

low, usually at insignificant levels, with SETs” (p. 2).

Some researchers have conducted studies in an effort to answer these questions

and others. The findings provide no clear direction. Clayson and Sheffet (2006) surveyed

the SETs of 727 students at their university. They collected data on measures of teacher

effectiveness at week 0, 10, and 16, which was the end of the term. They compared

measures of personality (agreeableness, conscientiousness, emotional stability,

extroversion, and imagination) with SETs. These researchers found that all the

personality dimensions were significant predictors of a positive final SET except

extroversion. Clayson and Sheffet (2006) concluded, “Students universally associate

personality with instructional effectiveness” (p. 156). In addition, they found that

combining the expected grade, a halo effect, and personality/likability resulted in 73% of

variance in the final SET. The halo effect occurs when individuals mistakenly conclude

that an unrelated factor contributed to or resulted in the desired outcome (Clayson &

Sheffet, 2006; Feeley, 2002; Otto et al., 2008). For example, physically attractive

individuals tend to be perceived by others as being happier and more successful. In

reality, the happiness and success are more likely related to factors other than one’s

physical appearance. In the above mentioned study, the halo affect likely occurred when

student-raters failed to discriminate between relevant and irrelevant teaching behaviors

(Feeley, 2002; Otto et al., 2008). Clayson and Sheffet (2006) concluded that SET follows

a seriously flawed paradigm, and those who utilize the results of SETs as indicators of

teacher quality should closely monitor their use.

Clayson and Sheffet (2006) concluded that while a teacher’s likability is an

important piece of information, using only this piece of information reduces the

“evaluation process” down to a “beauty pageant” and students are the “ultimate losers”

(p. 159). They state, “It may be that some teachers never receive consistently high

evaluations in certain environments, irrespective of anything they do or possibly could

do” (p. 158). Machina (1987) came to similar conclusions almost 20 years earlier. “Some

teachers will not receive the high marks they deserve despite brilliantly conceived

lectures” (p. 22).

Students and SETs

Centra (1980) believes that student evaluation of teacher effectiveness and

instructional quality is inexplicitly linked to the quality of interactions between the

student and teacher as well as the interpersonal rapport students experience. These have

been historically prominent variables in many studies. Regardless of how we view the

SET system, students regard their relationship with faculty as one of the most critical

aspects of the teaching-learning experience; student perception affects both learning and

motivation (Aguinis, Nelser, Quigley, Lee, & Tedeschi, 1996).

These findings seem to support the idea that SETs are highly correlated to

personality traits, that students form an opinion early on about an instructor, and that

there is little evidence to suggest that changes occur over the duration of the course

(Clayson & Sheffet, 2006). Decision-makers would be wise to remember that teachers

may score high on evaluations “not because they teach well, but simply because they get

along with students” (Foote, Harmon, & Mayo, 2003, p. 17).

Instructors and SETs

It is evident that no clear path emerges from the literature as to how instructors

feel about SETs. Do they perceive them as a valuable source of information they can use

to improve teaching? A Smith and Welicker-Pollack (2008) study found that professors

were generally favorable to student feedback and felt that students were capable of

assessing quality instruction, were honest in their responses, and that students have the

right to provide feedback on the learning experiences in the classroom.

The professors in Smith and Welicker-Pollack’s (2008) study indicated that they

wanted student feedback on their teaching, that they took the feedback seriously, and that

they desired to improve. Despite this finding, few teachers sought out counsel on how to

improve their teaching based on the feedback from SET (Smith & Welicker-Pollack,

2008). In addition, most university administrators fail to provide a useful platform for

real reflection and feedback—an element that is essential if the individual teacher and

institution as a whole is to achieve instructional improvement. They conclude by noting

that in order

to minimize the danger that formalized feedback remains solely a managerial

instrument for the purposes of accountability, teacher education institutions need to

develop a comprehensive follow-up support system, which enable educators to

engage in supportive dialogues with colleagues—a factor which has been found to be

of major importance in professional development processes for practitioners. (Smith

& Welicker-Pollack, 2008, p. 212)

A decade earlier, Baer (1999) came to a similar conclusion; he noted that a faculty

member’s acceptance of the feedback provided by student evaluations depends on the

administration not using the results of these surveys for reward or punishment. Baer

suggests that student feedback be a part of the overall instructor evaluation process. This

process should include the following elements: course syllabi, course materials, active

learning techniques, clarity of course requirements and expectations, the instructor’s

contributions and service, as well as other duties.

SETs as Measure of Instructor

Effectiveness

Based on the literature, a legitimate question arises as to the appropriateness of

this, or any study, utilizing student satisfaction to measure instructor effectiveness. There

is sufficient support on each side of the argument. Despite the controversy, it is evident

that the use of SETs is on the rise (Cochran-Smith, 2004). Grenzke (1998) estimates that

students in some form or fashion evaluate 89% of all part-time faculty members. The

highly competitive marketplace of higher education has resulted in teachers and

administrators at all levels finding themselves in an era of accountability (Cochran-Smith,

2004).

If we seek feedback from our student-consumers, administrators and teachers

need to take such feedback seriously. Failing to do so could have serious repercussions

(Baer, 1999). SETs are built on the premise that a student who learns is satisfied. Studies

suggest that students give low satisfaction scores to instructors they do not learn from and

high scores to those they do (Kelly et al., 2007). Because of this, researchers,

administrators, and teachers consider SETs to be the single most valid source of data on

teaching effectiveness.

Summary of Issues Related to Part-time Faculty

The historical trend of relying on part-time faculty members, beginning with the

passage of the G.I. Bill in 1944, has resulted in an ongoing debate among various factions

within the post-secondary academy. The temporary nature of part-time faculty teaching

assignments leaves them vulnerable. Part-time faculty members who desire continued

teaching appointments find themselves aligned with institutions and administrations that

largely ignore their contribution and fail to provide support. They are often left to sink or

swim with little or no orientation or opportunities for professional development. They are

isolated from the full-time faculty and have little or no input into curriculum design and

academic policies. Continued teaching contracts are often predicated upon positive

student evaluations.

Part-time faculty members, many who teach at multiple institutions, find

themselves in a precarious situation. Or more accurately, they find themselves caught

between university leaders, who need them to boost the bottom line and staff less popular

class times declined by full-time counterparts, and students, who must be satisfied with

the teaching experience before further teaching contracts would be offered.

This literature is clear. Conservatively, at least 50% of college classes are taught

by part-time faculty. Yet, they are largely ignored, given little or no opportunities to

increase in competence, no feedback is provided on their teaching performance, and

when faced with questions about academic quality, must wrestle with the competing

interest of pleasing students or future teaching contracts. How can any industry not

professionally invest and support 50% of its workforce and yet expect to deliver a quality

product to its customers? Businesses have learned that mentoring is an effective means

for supporting new employees, for bolstering self-confidence, and for building positive

regard and commitment to the organization that results in decreased turnover.

Mentoring

In order to set the framework, this portion of the literature review relates to

mentoring and will focus on defining the term mentoring. I will discuss the concepts of

natural and spontaneous mentoring relationships, as opposed to the more structured form

known as systematic mentoring. The principles involved in developing systematic

mentoring programs will be explored along with some empirical studies, which examine

the effectiveness of such programs. The advantages of peer-to-peer mentoring, over other

kinds of relationships, will be included as well. Finally, the benefits of participating in a

mentoring program will be discussed from the perspectives of the mentor, the mentee,

and the institution.

Mentoring: A Definition

Young and Wright (2001) define mentoring as the creation of a practical

relationship that focuses on providing the individual with opportunities for personal and

professional growth through the advice, coaching, and nurturing by a more experienced

individual. They identify mentors as individuals who are “skilled, knowledgeable,

visionary, dynamic and committed” to the process of investing in the development and

improvement of others (Young & Wright, 2001, p. 202). At its heart, mentoring is a

relationship (ASHE-ERIC, 1995; Biehl, 1996; Kaye, 2000; Phi Delta Kappa, 2002; Price,

Graham, & Hobbs, 1997; Young & Wright, 2001).

Mentoring is an honorable and necessary means by which the “wisdom of the

past” connects to the creative future (Loeffler, 2004, p. 23). Routman (2000) in her book

entitled Conversations states that “mentoring encompasses coaching, supporting, hand

holding, actively listening, gently suggesting, sharing materials, dialoguing with empathy

and understanding, and much more” (p. 3). The activities typically engaged in by a

mentor (coaching, nurturing, guiding, and modeling) are purposeful, with the result being

the advancement of the mentor or protégé (Davis, 2001; Hansman, 2001; Washburn &

LaLopa, 2003; Young & Wright, 2001). Young and Wright (2001) define mentoring as

a nurturing process in which a more skilled or more experienced person, serving as a

role model, teaches, sponsors, encourages, counsels, and befriends a less skilled or

less experienced person for the purpose of promoting the latter’s professional and/or

personal development. (p. 202)

There exists some agreement among scholars that the traditional mentoring

relationship involves some form of hierarchy in that those with more experience help

those with less (Cunningham, 1999; Grant-Vallone & Ensher, 2000; Price et al., 1997;

Semeniuk & Worall, 2000; Street, 2004). In the classical sense, the mentor is older,

wiser, and more mature (Levinson, Darrow, Klein, Levinson, & McKee, 1978). In fact,

Davis (2001) insists that mentors are always more experienced and knowledgeable than

the mentee. Ellison and Scribner (2001) describe a mentor as

someone who has already been where you are going. When you walk down the path,

you do not have to fall into the potholes and wait for someone to pull you out of the

mire. You can see the way ahead through the eyes of those who have traveled that

way, and you can avoid their mistakes. (p. 27)

Mullen (2000) expands on this idea by noting that traditional mentors guide,

facilitate, and help to transfer knowledge. Pierce (1998) extends this definition and

describes mentors as essentially acting in a parental role; mentors are role models with

demonstrated competence in teaching. Kram, in a 1985 landmark study (Kram &

Isabella, 1985), noted that previous research on mentoring tended to focus on this

hierarchical relationship and noted that “mentors tended to be older and several

organizational levels higher and that the mentoring relationship was characterized as a

one-way helping relationship” (p. 115).

Typically, the mentoring relationship is a dyad. One common theme found

throughout the literature is that of relationship, and it would appear that a recipe for

success carved in stone does not exist (Young & Wright, 2001). As a result, the matching

of the mentor and the mentee is of critical importance. Many mentees self-select their

mentors in what is known as a natural or spontaneous mentoring relationship. The

matching of mentor and mentee is a much more complex decision when the mentoring

program is systematic or developed, sanctioned, and implemented by the institution. The

following section discusses these two types of mentoring programs.

Types of Mentoring Programs

There are two camps when it comes to mentoring programs—informal or natural

and formal or systematic (Hansman, 2001). Informal mentoring occurs spontaneously,

whereas formal mentoring programs are sanctioned and planned by the organization. The

nature of how mentors and mentees elect to begin a mentoring relationship determines

the nature of the relationship and whether it would be classified as a formal or informal

one. This particular section will explore the similarities and differences between these

types of mentoring programs. In addition, I share the stages of mentoring program

development and what areas an organization should include as they create and implement

a systematic mentoring program.

Since mentoring has its basis in relationships, the stronger the relationship, the

more effective the mentoring experience. Some concur that the end result of the

mentoring relationship is key, regardless of how it is formed. Mentoring is an influential

relationship that results in an individual’s growth and change. Wang (2001) notes the

promise of change is inherent in the mentoring process and lies in the nature of the

mentor-mentee relationship. This growth can occur in both structured and unstructured

relationships and in planned or unplanned ones (ASHE-ERIC, 1995).

Spontaneous Mentoring

There are those who firmly assert that a natural mentor is best (Phi Delta Kappa,

2002). Biehl (1996) believes that mentoring relationships require a “positive, natural

chemistry” (p. 18) and cannot be guaranteed; he continues by saying that attempts to

artificially match mentor to mentee can produce a deep frustration. One of IBM’s six

principles of mentoring states that the employee must be the driving force behind the

mentoring relationship (Price et al., 1997). Semeniuk and Worall (2000) agree and

believe that planned and systematic mentoring programs have many drawbacks. They

believe that

relationships which arise spontaneously and gradually allow teachers to enter more

deeply into considering what it means to be teachers, to teach better and to engage

more fruitfully with students. Because the associations are more personally created,

their likelihood of success maybe greater. (Semeniuk & Worall, 2000, p. 425)

Croake (1996) worked to develop a systematic mentoring program in her

community college and learned that new faculty already considered themselves to be

experienced teachers; they questioned the need for a formal mentoring program. These

new faculty members suggested that the mentoring program be less formal, provided to

all faculty (not just the new ones), and be more individualized. Wang, Strong, and Odell

(2004), in their comparison study of the mentoring relationships between Chinese and

U.S. teachers, found that any kind of contrived collegiality, where partnerships are

imposed by administrators, poses challenges to the creation and development of

continued collaboration among faculty.

While many who engage in the debate do so in a diplomatic manner, Davis (2001)

speaks plainly. In order for an authentic mentoring relationship to develop, it must be “a

voluntary and ultimately personal relationship between two individuals. No one can

mandate and monitor such a relationship” (p. 2). For Davis, the idea of personal choice is

so essential to the mentoring relationship that if this element is mandated and devoid of

personal choice, it is “not mentorship” (p. 3). Davis criticizes administrators and

organizations that seek to artificially create these relationships; he concludes by saying

that this contrived effort will “yield empty results” (p. 3).

Vincent and Seymour (1995) concur, “Mentoring should never be forced but only

encouraged in organizations” (p. 9). It would seem that the participants of Goodwin,

Stevens, and Belamy’s (1998) study agree. Their exploratory and descriptive study of

full-time faculty members in 13 public and private universities found that faculty

members valued voluntary and informal mentoring relationships over those that were

formal and mandated (Goodwin et al., 1998).

Systematic Mentoring

While some believe that the spontaneous nature of the mentoring relationship

works best; one author noted that, for better or worse, organizational leaders routinely

institutionalize mentoring programs (Kaye, 2000). Many note that in the business world

formalized mentoring programs exist and have worked quite adequately (Gaskill, 1993).

While Gaskill acknowledges that disagreements exist on “the value of mandatory

programs,” the reality is that “formalized mentoring programs are successful” (p. 148).

Because these programs are well designed and not left to chance, they offer the greatest

opportunity for personal and organizational benefit. Gaskill defines a formal mentoring

program as one that “involves setting explicit goals and practices for linking less

experienced and more experienced managers, encouraging mentoring by arranging

relationships that serve developmental purposes for both individuals” (p. 147).

Boyle and Boice (1998) are two authors who believe that the traditional view of

spontaneous mentoring is, at best, a myth. They found that spontaneous mentoring tends

to be irregular and short-lived. In addition, oftentimes those who desire mentoring go un-

mentored. In fact, they state that those who benefit the most from “spontaneous

mentoring tend to be white males” (p. 159). Women and minorities are least likely to find

a mentor.

Boyle and Boice (1998) are not quick to praise these systematic mentoring

models; they note that oftentimes these programs are seen as remedial and a hindrance

rather than an opportunity to enhance professional growth. In addition, participants fear

that administrators will use the mentoring relationship for evaluation purposes; this is

particularly true if the mentor and mentee are from the same department. Their pilot

study sought to evaluate the effectiveness of a university’s systematic mentoring

program.

Boyle and Boice’s (1998) study focused on pairing mentors and recording their

interactions. Mentoring pairs agreed to meet for one academic year, to meet weekly, and

to attend monthly meetings. The researchers collected weekly phone logs and journals,

submitted by the participants, in order to rate the frequency, content, and compatibly of

the mentoring interactions. The study also utilized a score of 65 or higher on a mentoring

index to indicate an acceptable level of mentoring.

The results showed that individuals with mentors across departmental lines

reported a higher mentoring index when compared to mentors from the same department.

Mentees rated administrative mentors more highly than they did regular faculty members

(Boyle & Boice, 1998). Not surprisingly, mentoring pairs that met weekly, attended the

group meetings, and enthusiastically worked to build compatibility and reciprocity were

the most successful. Mentoring pairs agreed that the impetus for change came from the

monthly group meetings.

Boyle and Boice’s (1998) pilot study also compared mentoring pairs who

participated in the systematic mentoring program with a group of faculty members who

self-reported having experienced a spontaneous mentoring relationship. The authors

found that systematic mentoring works better than spontaneous, natural mentoring. Boyce

and Boice concluded, “Even the best-treated of natural mentees fared less well than most

of the counterparts in the formal program” (p. 169).

Gaia, Corts, Tatum, and Allen (2001) reported the results of their study on a

graduate teaching assistant mentoring program. This systematic mentoring program

recognized that few programs exist that prepare graduate students for the professoriate.

This particular program combined regular meetings of small teams (one outstanding,

experienced faculty member working with 5-7 teaching assistants) with large group

meetings. The purpose of the program was to provide an open forum for discussing issues

related to research and effective teaching. Program organizers hoped that because of

increased professional support, the GTAs (graduate teaching assistant) would experience

greater self-confidence about classroom teaching (Gaia et al., 2001). These authors

conclude the study by noting that creating systematic mentoring programs that provide

opportunities for junior- and senior-level faculty members to interact across disciplines

have the potential to greatly contribute to the quality of undergraduate education (Gaia

et al., 2001).

Cunningham (1999) notes that the intentional and effective mentoring among

colleagues can enhance professional growth and development. The literature concerning

systematic mentoring is encouraging and seems to indicate that it can be as effective as a

natural and spontaneous mentor pairing. The above studies regarding both natural and

systematic mentoring have focused on the relationship between a more experienced

individual with one who is less experienced or confident. The following section will

explore the idea of peer mentoring.

Peer Mentoring

Typically, research regarding mentoring investigates a one-way helping

relationship (Kram & Isabella, 1985). Peer relationships may offer “unique

developmental opportunities” that should not be overlooked (p. 120). Previous studies

have hinted at the potential of peer relationships. Due to current organizational structure

trends, individuals are much more likely to have more peers than supervisors or bosses.

In addition, those who have looked at peer mentoring conclude that the absence of a

hierarchical relationship means that communication, mutual support, and collaboration

will result.

Because of the potential benefit of peer mentoring, Kram and Isabella (1985)

conducted an exploratory study to investigate peer mentoring. The study focused on

identifying the purpose of a peer mentor, to explore the varying kinds of peer mentors,

and to understand the function of peer relationships at different career stages. This

grounded-theory study sought patterns and themes from the participants’ responses. The

researchers interviewed 15 individuals in three different age groups. During the first

interviews, the voluntary participants were asked to name “two relationships with

colleagues that you feel supported your personal and professional growth” (p. 114).

Participants then gave permission for the researchers to contact these individuals for

another guided interview.

The results showed that peer relationships are an important alternative to the

traditional mentoring dyad. These peer relationships provided a range of support that

spanned different career stages. The psychosocial support, information sharing, and

technical support resulted in a sense of competence and confidence (Kram & Isabella,

1985). In addition, these peer-mentoring relationships tended to be more intimate and

have a higher degree of self-disclosure and trust.

Peer relationships offer a degree of mutuality and enable both individuals to

experience being a giver as well as the receiver of these functions. In contrast to a

mentoring relationship, where one individual specializes in the role of guide or

sponsor, in a peer relationship both assume both kinds of roles. (p. 118)

While traditional and peer mentoring have several common attributes, peer

relationships have a longevity that exceeds most traditional mentoring relationships.

While a conventional mentor may be most useful at the beginning stage of a career, Kram

and Isabella (1985) conclude that peer relationships are important at all stages.

Mullen (2000) agrees with this assessment and notes that the traditional model of

mentorship falls short of satisfying the deeper and multiple demands required of a

professional partnership. Peer mentoring, or co-mentoring as Mullen calls it, is

empowering, proactively supportive, and synergistic. The lack of a hierarchical

relationship encourages professional learning among equals.

Despite the potential for peer mentoring, little is known about this kind of

mentoring experience and few studies have examined how formally assigned peer

mentors support their mentees and what affect this has on important outcomes (Grant-

Vallone & Ensher, 2000). As a result, Grant-Vallone and Ensher examined 29 peer

mentors. Both mentors and mentees volunteered and were required to attend an

orientation session, a large-group function at the beginning of each semester, and meet at

least two times during the semester. The researcher-developed instrument collected data

relative to support functions, perceived stress, and the degree of satisfaction with the

program. The researchers hypothesized that increased high-contact relationships would

report more support and less stress. Support was identified as psychosocial

(interpersonal) and instrumental (task-related support).

The findings revealed that high-contact peers (HCP) reported significantly higher

levels of psychosocial support and instrumental support than did the low-contact peers

(LCP). The HCP pairs also reported a greater degree of overall satisfaction with the

program than did the LCP (Grant-Vallone & Ensher, 2000). Grant-Vallone and Ensher

(2000) also learned that peer mentors provided more psychosocial support than

instrumental support.

Because of their study, Grant-Vallone and Ensher (2000) recommend that peer

mentors receive intensive training, that a more extensive pairing strategy be designed or

allow peers to choose their mentors, and that an orientation be provided so that mentees

can understand the expectations. In addition, the authors recommend rewarding peer

mentors for their efforts. Because contact was high at the beginning of the program and

tapered off at the end, Grant-Vallone and Ensher recommend organizing formal events,

which ensure that the mentoring pairs have regular contact.

Whether one engages a traditional or peer mentor, most agree that organizations

that desire to establish this collaborative relationship are best served by designing and

implementing a systematic mentoring program. The literature is replete with examples of

institutional mentoring programs and best practices with regard to how these programs

should be developed. The following section will explore this literature.

Developing Systematic Mentoring Programs

In 1993, Gaskill studied 90 executive development directors across the United

States who employed 10 or more executives, in order to propose a conceptual framework

for the development of systematic mentoring programs. Only 51 of the 90 executives

responded to his initial inquiry and of those, only nine had formal mentoring programs.

Not surprising, Gaskill (1993) learned that systematic mentoring begins with

development and moves toward the implementation phase. Finally, the organization

needs to evaluate the program’s effectiveness in order to determine if each mentor and

mentee pair achieved the stated outcomes of the program.

All the executive development directors noted the importance of obtaining the

support and buy-in from those at the highest leadership levels in the organization.

Without this support, important resources may not be available. Pierce (1998) supports

the importance of administrative and institutional support and notes that without this buy-

in “a viable program is not possible” (p. 36). In addition, he notes that this support needs

to demonstrate “concretely” by committing human, physical, and psychological

resources, which includes, but is not limited to, meeting space and time (p. 37).

After gaining institutional support, other critical elements of the development

phase include the selection of mentors. While the mentees’ participation can be either

voluntary or mandated, this should not be the case with the mentors. “Mentor candidates

. . . need to be carefully reviewed for their qualifications, willingness, and desire to

participate” (Gaskill, 1993, p. 155). Gaskill advises those who are developing mentor

programs to select mentors wisely.

Mentor selection process should begin with the identification of a pool of potential

candidates obtained either on a voluntary basis or by targeting key individuals in the

organization. Once potential mentors are identified, candidates should then be

evaluated using pre-determined selection criteria. (p. 155)

Gaskill (1993) also recommends that organizers provide training to both the

mentor and mentee prior to entering into the mentoring relationship. The purpose of the

training is to orientate participants to the program’s purposes and objectives. Training

also serves to minimize any ambiguity in roles, duties, and responsibilities. Developing

essential mentoring skills, such as active listening and problem-solving techniques, is

also an important aspect of this pre-service training.

Perhaps the most important element to consider in this planning phase is how one

will actually pair each mentor and mentee. Gaskill (1993) refers to this process as

“linkage” (p. 155). He suggests that linkage be based on commonalities like shared

interest and geographic proximity. It is essential that once the mentoring link has been

made that the program’s facilitator monitors the match to “determine if the individuals

are compatible and functioning adequately” (p. 156).

Once the pre-implementation steps have been accomplished and mentees have

been linked to a compatible mentor, the mentoring program moves into the

implementation phase (Gaskill, 1993). The participants of this study recommended that

the details related to the specifics of each mentoring pair’s contact be “left to the

discretion of the parties involved” (p. 156). This contact should include elements of

career function and enhancement and such psychosocial support as building confidence

and self-worth.

Finally, the systematic mentoring program needs to be evaluated for its

effectiveness and to determine whether the objectives were achieved. This evaluation

should include feedback from both the mentor and the mentee; thus, using pre-developed

questionnaires or non-structured interviews are two methods he recommends. Gaskill

(1993) recommends that data be collected on the participants’ impressions of the overall

effectiveness of the program and the mentoring relationship, personal benefits achieved,

and suggestions for improvement. Gaskill concludes his study by observing that

“formalized mentoring programs build confidence and competence, provide emotional

support, and, in general, help to develop productive, fast-tracking executives with

improved levels of career commitment” (p. 159).

Boyle and Boice (1998) noted that effective systematic mentoring programs need

planning, structure, and intentional assessment. Boyle and Boice also recommend that

you recruit mentors and mentees (early), carefully pair the mentoring assignments (cross-

department pairings worked best), and communicate clearly the mentoring program’s

goals to participants. They also note that the monthly meetings were critical to the

program’s success. The researchers regularly followed up and provided incentives for

pairs to meet. In conclusion, Boyle and Boice note that “effective mentoring, as far as we

can tell, begins with institution-wide programs that coach departments in ways to

systematically immerse newcomers in support programs and provide them with a sense of

connectiveness” (p. 177).

Recognizing that mentoring programs are an effective way to integrate employees

into the workplace, Price et al. (1997) offer a comprehensive model for developing a

systematic mentoring program. The model includes program design and encompasses the

establishment of goals and objectives, selection criteria for mentors, as well as

developing and training the mentors to be successful in their roles. These authors also

suggest that both the mentor and mentee receive training that includes an understanding

of their different roles. For the mentor, this means how to use meetings to build

relationship and to create trust with the mentee, how to write objectives as part of a

growth plan, and how to observe and evaluate the mentee. The mentee also needs to be

orientated to his or her roles and responsibilities. These can include how to accept advice,

the importance of commitment, and following through on suggestions for improving.

Price et al. (1997) also note that assessing the mentoring program is of critical

importance. This should be formative as well as summative. The assessment plan should

include a means whereby the mentee can evaluate the mentor. Participants should

evaluate all components of the program, as this feedback is essential for the

implementation of a continuous improvement plan. These steps are essential. The benefit

is well worth the effort. “An employee who experiences an intensive mentoring program

under the tutelage of a competent mentor will become an asset to the organization and

will ideally become a role model for future employees” (para. 24).

In conclusion, research seems to suggest that systematic mentoring is as effective

as and, in some cases, more effective than mentoring relationships, which evolve from a

more natural and spontaneous pairing. The literature also seems to agree that effective

mentoring programs need to be intentional and the result of careful planning that includes

a pre-implementation phase where goals and objectives are established. Strong

administrative support is critical (Pierce, 1998). The qualifications of the mentor must be

carefully considered as well as how the mentor and mentee pair will be linked (ASHE-

ERIC, 1995; Grant-Vallone & Ensher, 2000; Price et al., 1997).

Contrary to conventional wisdom, the literature supports the idea that

interdisciplinary pairings are more effective than assigning mentors and mentees from the

same department. Organizers should also provide some form of training and orientation

for both mentors and mentees (Phi Delta Kappa, 2002; Price et al., 1997; Wang, 2001). In

addition, during the actual mentoring phase it is important to provide regular monitoring

to ensure that the linkage is effective and that the mentoring dyad is meeting on a regular

basis. Mentor pairs left alone were ineffective (Boice, 1992).

Finally, the creators and authors of the program must plan for and provide a

means whereby ongoing and summative assessment occurs (Gray, 1989). Feedback, from

all levels and all participants, will ensure that the program is continually revised and

improved so that the stated goals and objectives continue to be met. When all these

components are effectively put into place, a successful program will result, which

empowers the institution’s human capital and supports the institution’s vision and

mission (ASHE-ERIC, 1995). In addition to these two important outcomes, the literature

suggests there are many more benefits to mentoring.

Benefits of Mentoring

It should come as no surprise that engaging in a mentoring relationship, whether

natural and spontaneous or systematic and formalized, will benefit the mentee. However,

the literature is also replete with data that suggest that mentors and institutions benefit as

well. The development and maintenance of these personal and professional relationships

provide a forum for discussing what we mean by quality teaching. The dialogue alone

should raise our awareness and spur us on to excellence.

For mentees, the benefits are numerous and as individualized as each person. Gaia

et al. (2001) noted in their study that mentees benefited from the GTA mentoring

program as they built relationships with professors and other graduate assistants across

disciplines. This increased sense of community resulted in a positive and supportive

culture, which encouraged risk-taking that might not otherwise have occurred. The one-

on-one meetings with the mentor, as well as the group meetings, fostered professionalism

and a venue to think and talk about teaching as a profession (Gaia et al., 2001). As are

result of this community discussion on excellence in teaching, the mentors “reengaged in

issues of everyday teaching” (p. 64).

On a practical level, Simmons (1998) reported that university professors who

engage in a mentoring relationship are better prepared for the tasks of teaching. These

include, but are not limited to, the following: creating course syllabi, meeting promotion

and tenure requirements, preparing for annual performance reviews, setting class

expectations, holding students to high academic standards, securing class materials, and

understanding the institutional lore. New faculty members do not have the time to pore

over faculty manuals and handbooks. Mentors who support, advise, and affirm their

mentees will find that the protégé has gained the confidence necessary to be comfortable

and successful.

In addition to the task-related support outlined above, the personal or

psychosocial support a mentee receives can be invaluable. Wickman and Sjodin (1997)

suggest that mentoring opens doors for the protégé, reduces frustration by saving time,

increases success and productivity, and improves career satisfaction. The result of this

personal support and encouragement is a teacher with higher levels of institutional

commitment and loyalty.

Goodwin et al. (1998) echo this conclusion. They note that the major outcomes of

mentoring are gained confidence. A close relationship with a mentor will help the protégé

successfully navigate the institution’s tenure process and assist the protégé in engaging in

scholarly activities, such as research and service. Finally, an increased commitment to the

program is also common.

While the benefits to the mentee or protégé can be easily understood, the reality is

that the “mentors themselves become more professional, less anxious and better engaged

and positive about their work—there is a greater sense of responsibility and security”

(Semeniuk & Worall, 2000). Wickman and Sjodin (1997) further explain this outcome as

they note that those who serve as mentors benefit as they pass their legacy on to others.

Serving as a role model and guide to another forces the mentor to set an example, which

can often result in more creativity and effectiveness concerning the mentor’s own

teaching style. The reality is that as you impart your values to others, you “get by giving”

(p. 89).

While both mentee and mentor benefit and grow from an effective mentoring

relationship, this positive outcome is not lost on the institution (Cunningham, 1999). The

institution realizes an enhanced teaching effectiveness on the part of the professoriate.

This results in a vibrant and productive faculty who are keenly satisfied in their positions

and who are loyal and remain committed to the institution’s purpose and mission. There

can be no doubt that mentoring strengthens the organization and supports its vision

(ASHE-ERIC, 1995).

Conclusion

This chapter explored the rising dependence on part-time faculty in America’s

post-secondary institutions. While the Bureau of Educational Statistics indicates that the

trend has remained a steady 44% of the professoriate over the past decade, some

researchers believe this number is misleading due to the varied categories and ranks that

colleges and universities use to define these faculty members.

Despite the steady use of part-time faculty members over the past 50 years, the

debate continues to rage across college campus classrooms and faculty meetings. Some

believe this over-reliance on part-time faculty members threatens the very fabric of

academic quality and excellence. Others believe that the part-time faculty bring a much-

needed link between theoretical ivory towers and the application of these theories in a

pragmatic manner within the fields of professional practice. No clear conclusion emerges

and the debate continues.

While the discussion continues, the literature appears to agree on several key

factors. Part-time faculty members are isolated from the larger academic community and

are not provided the same level of support and professional development opportunities as

their full-time counterparts. This isolation and feelings of belonging to a second-class

citizenry have resulted in a lack of confidence and loyalty to the institution. Tenured

faculty and professional organizations like the AAUP believe that part-time faculty

provide a less than stellar educational experience and that students ultimately suffer from

this lack of quality.

The literature review clearly supports the evidence that part-time faculty members

are here to stay and that university administrators and leaders need to begin to address

how best to meet the professional development and inclusion needs of these vital

members of the faculty. Mentoring has proved effective in the business arena and at all

levels of the educational spectrum. There is no evidence to suggest that such an

intervention would not be effective or that participants who receive mentoring would do

worse than those who do not.

Of the many studies examined, only two specifically addressed the use of peer

mentors. Consequently, it seems logical that a study designed to explore the impact of

peer mentoring on a part-time faculty member’s confidence, organizational loyalty, and

student satisfaction scores is needed to help address the existing gap in the literature.

Chapter 3 presents the research design and methodology, including the data collection

and analysis plans.

CHAPTER THREE

METHODOLOGY

The purpose of this study was to examine what relationship self-reported

perceptions on the quality of a peer-to-peer mentoring experience would have on a part-

time faculty member’s instructor confidence, institutional loyalty, and student satisfaction

at a very large mid-west university. This chapter informs the reader about the research

protocol used to both collect and analyze the data. Specifically, I have organized this

chapter into the following sections: (a) description of participants, (b) data collection

procedures, (c) design of the study, (d) statement of hypotheses, (e) the variable list,

(f) the instrumentation, (g) data analysis plan, and (h) summary of methodology.

Description of Participants

Part-time faculty members who teach in the School of Business in the College of

Adult and Professional Studies (CAPS) programs from the University served as the

sample for this ex post facto study. It is important to note that I did not use random

sampling techniques in this study. Because the independent variable cannot be

manipulated or controlled, the variation was achieved by selecting participants in which

the variable, in this case, a peer-mentoring experience, is “present, absent, strong, or

weak” (Ary et al., 2010, p. 334). For this reason, the following list details the criteria for

determining who was eligible to participate in the study:

1. Eligible participants must be part-time faculty members and teach in the School

of Business in the University’s CAPS program.

2. Eligible participants must have taught at least three courses so that the mean

from end-of-course surveys can be calculated.

3. Eligible participants may not hold a full-time teaching or administrative

position with the University.

To identify eligible faculty, I coordinated with the Associate Vice-president for

Academic Administration and Operations to identify the sample who meets the above

criteria. Because the University changed the end-of-course survey form and the manner

in which they collect and analyze this information, participants must have taught at least

three courses between June 1, 2009, and December 31, 2010.

In order to maintain the confidentiality and anonymity of the participants and their

individual end-of-course survey results, the University handled the dissemination of the

survey. The statistics department matched survey responses with the corresponding end-

of-course survey means. Each participant was given a unique identifier before

disseminating survey and end-of-course results to the researcher. Since the identifier

contained no personal information, no one was able to match survey responses and end-

of-course data to a particular respondent.

Babbie (2007) notes that there is an important distinction between confidentiality

and anonymity, and researchers must be clear with participants as to how individual

identity will be handled. Since I am not a member of the University’s academic

community, this additional step was deemed prudent to further protect the confidentiality

of the study’s participants.

Data Collection Procedures

The CAPS Assistant Director for Research and Program Development was

responsible for putting the survey into the University’s Internet survey site, launching the

survey, and compiling the results. The Associate Vice-president for Academic

Administration and Operations and the Associate Dean for the School of Business sent

email communications to the School of Business Faculty informing them of the research,

stressing to them the importance and value of the research, and encouraging them to

complete the survey in a timely manner. All parties believed this introduction would

increase both the response rate and timeliness of completion. The survey remained open

for 2 weeks; after 1 week, a reminder was sent to encourage those who had not yet

responded to access the survey’s URL and to complete the survey. Once the data from the

survey were complete, the responses were compiled and the unique identifier was created

and attached to the end-of-course survey means. The results were then emailed to me.

The data were analyzed using the statistical software package PASW 18.0 (formerly

SPSS).

Design of the Study

Kerlinger and Lee (2000) note that the cardinal rule of planning any research

study is that the research questions should dictate the research design. Since this study

sought to explore the quality of a previous peer-mentoring experience and its relationship

to several dependent variables, it was determined that an ex post facto study was the most

appropriate research design to use in order to answer the research questions and to test

the hypotheses. Ex post facto research, by its very design, investigates “the world as it

naturally occurs” and explores phenomena that have already occurred (Johnson &

Christensen, 2008, p. 257).

According to Newman et al. (2006), there are three types of ex post facto

research. The first design uses no hypothesis and generally considered the weakest of the

three. Other ex post facto designs included the testing of hypotheses and have a bit more

scientific value. The third type of ex post facto research design includes the test of

hypotheses along with test for alternative hypotheses are “considerably more powerful in

terms of internal validity” (Newman, et al., 2006, p. 101).

Ary et al. (2010) note that an ex post facto research design is useful when one

wants to investigate the relationship between the dependent and independent variables

when randomization or manipulation of the independent variable is not possible. While

both an experimental and ex post facto design test relationships between variables and

test hypotheses, the ex post facto design does not provide adequate safeguards;

consequently, less evidence exists to infer a causal relationship (Ary et al., 2010). Despite

these limitations, ex post facto research contributes valuable information to the field of

education and other social sciences.

Internal Validity

At the beginning of any research, it is important to note the threats to both the

internal and external validity. Internal validity, a term coined by Campbell and Stanley

(1963), concludes that the relationship between two variables is causal. Campbell and

Stanley indicate that threats to the internal validity of any study occur when researchers

draw inaccurate conclusions from the research results; this happens whenever anything

other than the independent variable affects the dependent variables. So, for a study to

have internal validity it must control for all other possible explanations for the

relationship (Johnson & Christensen, 2008).

Since ex post facto research does not manipulate the independent variable,

researchers must be particularly cautious when making inferences from the study’s

findings (Ary et al., 2010; Johnson & Christensen, 2008; Kerlinger & Lee, 2000).

Regardless of these restrictions, researchers can take steps in the implementation and data

analysis phases to minimize the effects of these limitations. For the purposes of this

study, several factors may influence the conclusions drawn from the study’s findings.

Perhaps the most significant threat to the internal validity of an ex post facto study

is the post hoc fallacy, which occurs when the investigator incorrectly concludes that

causation exists simply because evidence of a relationship is found (Ary et al., 2010). In

order to conclude that a causation relationship exists, researchers must be able to meet the

following three standards for a non-spurious relationship: (a) a statistical relationship

between X and Y has been established, (b) X preceded Y in time, and (c) other factors

did not determine Y (Ary et al., 2010, p. 333).

It is with this last condition that ex post facto studies have the most difficulty

(Johnson & Christensen, 2008). Because the independent variable is not controlled, a

causation conclusion would be a mistake. The relationship could exist due to the presence

of an undetermined third variable. One way to combat this is to control for extraneous

variables, which could provide for an alternate conclusion, into the research design.

For this reason, age, gender, ethnicity, educational level, primary teaching area,

and teaching experience were included as independent variables in statements of alternate

hypotheses. Kerlinger and Lee (2000) note that the credibility of ex post facto research

can be improved when designing the study that a number of hypotheses be considered.

This is because the greater number of alternative hypotheses investigators can eliminate,

the greater likelihood the study can approximate internal validity (Newman, Newman,

Brown, & McNeely, 2006). Newman et al. support this conclusion when they state, “Ex

post facto research with hypotheses and tests for alternative hypotheses is considerably

more powerful in terms of internal validity” (p. 101).

In addition, the primary means of data analysis will be a general linear regression

model, which can be used to statistically control for some extraneous variables. This

technique, according to Johnson and Christensen (2008), can be used to remove some of

the influence of the extraneous variables. Kerlinger and Lee (2000) note that controlling

for extraneous variables in a research design makes it possible to extract additional

research information on the possible relationship the extraneous variables may have with

the dependent variables.

Another threat to internal validity is maturation, which refers to changes that can

occur in individuals over time. Babbie (2007) states that it is possible that a participant is

not completely truthful in answering the survey questions or may respond by choosing

what he or she perceives to be the preferred or ideal answer as opposed to choosing a

response that accurately reflects the current situation. Since this study asked instructors to

recall a previous past mentoring experience, their memory may be faulty. The changes in

instructor confidence and loyalty may increase over time, regardless of the mentoring

experience. It is for this reason that the general linear model controls for length of

teaching experience.

External Validity

While internal validity refers to the ability of researchers to infer causation,

external validity relates to the ability for one to generalize based on the research findings

(Campbell & Stanley, 1963). Several factors significantly affect the ability to generalize

the findings of this research to the broader population. The sample population limits the

ability of the research findings to be generalized. The sample consists of part-time faculty

from one private, mid-western Christian university. It would be inappropriate to draw

conclusions about part-time faculty members in the public and private sectors across the

United States. At best, the study’s results can be used to make generalizations about the

relationship between peer-to-peer mentoring on part-time instructors in the University’s

College of Adult and Professional Studies. While care must be taken to limit

recommendations for practice to the University, certainly recommendations for future

research can include implications for the broader part-time faculty community.

Statement of Hypotheses

Hypothesis statements are frequently used when the investigator wishes to

compare groups with regard to the relationship among the different variables (Creswell,

2003). The reader will note that the hypothesis statements are directional; it is assumed

that those who perceive a higher degree of quality within the peer-mentoring experience

will score higher on the dependent variable measures than those who perceive a lesser

quality experience. The literature on mentoring seems to strongly support the supposition

that recipients of mentoring have increased confidence and greater loyalty (Boyle &

Boice, 1998). At least the literature is clear that those who receive mentoring will not do

worse than those who have not received mentoring. The primary and secondary

directional hypotheses or prediction statements that form the basis for this study are

detailed in the sections that follow.

Hypothesis Statements Related to Instructor Confidence

Hypothesis Statements Related to Mentoring

and Instructor Confidence

Hypothesis 1a: Subjects who receive mentoring will score higher on indicators of

instructor confidence than those who receive no mentoring.

Hypothesis 1b: Subjects who receive mentoring will score higher on indicators of

instructor confidence, independent of age, than those who receive no mentoring.

Hypothesis 1c: Subjects who receive mentoring will score higher on indicators of

instructor confidence, independent of gender, than those who receive no mentoring.

Hypothesis 1d: Subjects who receive mentoring will score higher on indicators of

instructor confidence, independent of ethnicity, than those who receive no mentoring.

Hypothesis 1e: Subjects who receive mentoring will score higher on indicators of

instructor confidence, independent of length of service, than those who receive no

mentoring.

Hypothesis 1f: Subjects who receive mentoring will score higher on indicators of

instructor confidence, independent of modules taught, than those who receive no

mentoring.

Hypothesis 1g: Subjects who receive mentoring will score higher on indicators of

instructor confidence, independent of educational level, than those who receive no

mentoring.

Hypothesis 1h: Subjects who receive mentoring will score higher on indicators of

instructor confidence, independent of primary teaching level, than those who receive no

mentoring.

Hypothesis 1i: Subjects who receive mentoring will score higher on indicators of

instructor confidence, independent of whether or not the faculty member teaches at other

universities, than those who receive no mentoring.

Hypothesis Statements Related to Mentoring,

Fidelity, and Instructor Confidence

Hypothesis 1j: Subjects who perceive a higher quality mentoring experience will

score higher on indicators of instructor confidence than those who report a lesser quality

mentoring experience.

Hypothesis 1k: Subjects who perceive a higher quality mentoring experience will

score higher on indicators of instructor confidence, independent of age, than those who

report a lesser quality mentoring experience.

Hypothesis 1l: Subjects who perceive a higher quality mentoring experience will

score higher on indicators of instructor confidence, independent of gender, than those

who report a lesser quality mentoring experience.

Hypothesis 1m: Subjects who perceive a higher quality mentoring experience will

score higher on indicators of instructor confidence, independent of ethnicity, than those

who report a lesser quality mentoring experience.

Hypothesis 1n: Subjects who perceive a higher quality mentoring experience will

score higher on indicators of instructor confidence, independent of length of service, than

those who report a lesser quality mentoring experience.

Hypothesis 1o: Subjects who perceive a higher quality mentoring experience will

score higher on indicators of instructor confidence, independent of number of modules

taught, than those who report a lesser quality mentoring experience.

Hypothesis 1p: Subjects who perceive a higher quality mentoring experience will

score higher on indicators of instructor confidence, independent of educational level, than

those who report a lesser quality mentoring experience.

Hypothesis 1q: Subjects who perceive a higher quality mentoring experience will

score higher on indicators of instructor confidence, independent of primary teaching

level, than those who report a lesser quality mentoring experience.

Hypothesis 1r: Subjects who perceive a higher quality mentoring experience will

score higher on indicators of instructor confidence, independent of whether or not the

faculty member teaches at other universities, than those who report a lesser quality

mentoring experience.

Hypothesis Statements Related to Mentoring,

Type of Mentor, and Instructor Confidence

Hypothesis 1s: Subjects who receive peer mentoring will score higher on

indicators of instructor confidence than those with an administrator or full-time faculty

mentor.

Hypothesis 1t: Subjects who receive peer mentoring will score higher on

indicators of instructor confidence, independent of age, than those with an administrator

or full-time faculty mentor.

Hypothesis 1u: Subjects who receive peer mentoring will score higher on

indicators of instructor confidence, independent of gender, than those with an

administrator or full-time faculty mentor.

Hypothesis 1v: Subjects who receive peer mentoring will score higher on

indicators of instructor confidence, independent of ethnicity, than those with an

administrator or full-time faculty mentor.

Hypothesis 1w: Subjects who receive peer mentoring will score higher on

indicators of instructor confidence, independent of length of service, than those with an

administrator or full-time faculty mentor.

Hypothesis 1x: Subjects who receive peer mentoring will score higher on

indicators of instructor confidence, independent of number of modules taught, than those

with an administrator or full-time faculty mentor.

Hypothesis 1y: Subjects who receive peer mentoring will score higher on

indicators of instructor confidence, independent of educational level, than those with an

administrator or full-time faculty mentor.

Hypothesis 1z: Subjects who receive peer mentoring will score higher on

indicators of instructor confidence, independent of primary teaching level, than those

with an administrator or full-time faculty mentor.

Hypothesis 1aa: Subjects who receive peer mentoring will score higher on

indicators of instructor confidence, independent of whether or not the faculty member

teaches at other universities, than those with an administrator or full-time faculty mentor.

Hypothesis Statements Related to Institutional Loyalty

Hypotheses Related to Mentoring and

Institutional Loyalty

Hypothesis 2a: Subjects who receive mentoring will score higher on indicators of

institutional loyalty than those who receive no mentoring.

Hypothesis 2b: Subjects who receive mentoring will score higher on indicators of

institutional loyalty, independent of age, than those who receive no mentoring.

Hypothesis 2c: Subjects who receive mentoring will score higher on indicators of

institutional loyalty, independent of gender, than those who receive no mentoring.

Hypothesis 2d: Subjects who receive mentoring will score higher on indicators of

institutional loyalty, independent of ethnicity, than those who receive no mentoring.

Hypothesis 2e: Subjects who receive mentoring will score higher on indicators of

institutional loyalty, independent of length of service, than those who receive no

mentoring.

Hypothesis 2f: Subjects who receive mentoring will score higher on indicators of

institutional loyalty, independent of modules taught, than those who receive no

mentoring.

Hypothesis 2g: Subjects who receive mentoring will score higher on indicators of

institutional loyalty, independent of educational level, than those who receive no

mentoring.

Hypothesis 2h: Subjects who receive mentoring will score higher on indicators of

institutional loyalty, independent of primary teaching level, than those who receive no

mentoring.

Hypothesis 2i: Subjects who receive mentoring will score higher on indicators of

institutional loyalty, independent of whether or not the faculty member teaches at other

universities, than those who receive no mentoring.

Hypotheses Related to Fidelity and

Institutional Loyalty

Hypothesis 2j: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of institutional loyalty than those who report a

lesser quality mentoring experience.

Hypothesis 2k: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of institutional loyalty, independent of age,

than those who report a lesser quality mentoring experience.

Hypothesis 2l: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of institutional loyalty, independent of gender,

than those who report a lesser quality mentoring experience.

Hypothesis 2m: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of institutional loyalty, independent of

ethnicity, than those who report a lesser quality mentoring experience.

Hypothesis 2n: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of institutional loyalty, independent of length

of service, than those who report a lesser quality mentoring experience.

Hypothesis 2o: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of institutional loyalty, independent of number

of modules taught, than those who report a lesser quality mentoring experience.

Hypothesis 2p: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of institutional loyalty, independent of

educational level, than those who report a lesser quality mentoring experience.

Hypothesis 2q: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of institutional loyalty, independent of primary

teaching level, than those who report a lesser quality mentoring experience.

Hypothesis 2r: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of institutional loyalty, independent of whether

or not the faculty member teaches at other universities, than those who report a lesser

quality mentoring experience.

Hypothesis Statements Related to Type of

Mentor and Institutional Loyalty

Hypothesis 2s: Subjects who receive a peer-mentoring experience will score

higher on indicators of institutional loyalty than those who report a lesser quality

mentoring experience.

Hypothesis 2t: Subjects who receive a peer-mentoring experience will score

higher on indicators of institutional loyalty, independent of age, than those who report a

lesser quality mentoring experience.

Hypothesis 2u: Subjects who receive a peer-mentoring experience will score

higher on indicators of institutional loyalty, independent of gender, than those who report

a lesser quality mentoring experience.

Hypothesis 2v: Subjects who receive a peer-mentoring experience will score

higher on indicators of institutional loyalty, independent of ethnicity, than those who

report a lesser quality mentoring experience.

Hypothesis 2w: Subjects who receive a peer-mentoring experience will score

higher on indicators of institutional loyalty, independent of length of service, than those

who report a lesser quality mentoring experience.

Hypothesis 2x: Subjects who receive a peer-mentoring experience will score

higher on indicators of institutional loyalty, independent of number of modules taught,

than those who report a lesser quality mentoring experience.

Hypothesis 2y: Subjects who receive a peer-mentoring experience will score

higher on indicators of institutional loyalty, independent of educational level, than those

who report a lesser quality mentoring experience.

Hypothesis 2z: Subjects who receive a peer-mentoring experience will score

higher on indicators of institutional loyalty, independent of primary teaching level, than

those who report a lesser quality mentoring experience.

Hypothesis 2aa: Subjects who receive a peer-mentoring experience will score

higher on indicators of institutional loyalty, independent of whether or not the faculty

member teaches at other universities, than those who report a lesser quality mentoring

experience.

Hypothesis Statements Related to Student Satisfaction

Hypotheses Related to Mentoring and

Student Satisfaction

Hypothesis 3a: Subjects who receive mentoring will score higher on indicators of

student satisfaction than those who receive no mentoring.

Hypothesis 3b: Subjects who receive mentoring will score higher on indicators of

student satisfaction, independent of age, than those who receive no mentoring.

Hypothesis 3c: Subjects who receive mentoring will score higher on indicators of

student satisfaction, independent of gender, than those who receive no mentoring.

Hypothesis 3d: Subjects who receive mentoring will score higher on indicators of

student satisfaction, independent of ethnicity, than those who receive no mentoring.

Hypothesis 3e: Subjects who receive mentoring will score higher on indicators of

student satisfaction, independent of length of service, than those who receive no

mentoring.

Hypothesis 3f: Subjects who receive mentoring will score higher on indicators of

student satisfaction, independent of modules taught, than those who receive no

mentoring.

Hypothesis 3g: Subjects who receive mentoring will score higher on indicators of

student satisfaction, independent of educational level, than those who receive no

mentoring.

Hypothesis 3h: Subjects who receive mentoring will score higher on indicators of

student satisfaction, independent of primary teaching level, than those who receive no

mentoring.

Hypothesis 3i: Subjects who receive mentoring will score higher on indicators of

student satisfaction, independent of whether or not the faculty member teaches at other

universities, than those who receive no mentoring.

Hypotheses Related to Fidelity and

Student Satisfaction

Hypothesis 3j: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of student satisfaction than those who report a

lesser quality mentoring experience.

Hypothesis 3k: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of student satisfaction, independent of age,

than those who report a lesser quality mentoring experience.

Hypothesis 3l: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of student satisfaction, independent of gender,

than those who report a lesser quality mentoring experience.

Hypothesis 3m: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of student satisfaction, independent of

ethnicity, than those who report a lesser quality mentoring experience.

Hypothesis 3n: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of student satisfaction, independent of length

of service, than those who report a lesser quality mentoring experience.

Hypothesis 3o: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of student satisfaction, independent of number

of modules taught, than those who report a lesser quality mentoring experience.

Hypothesis 3p: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of student satisfaction, independent of

educational level, than those who report a lesser quality mentoring experience.

Hypothesis 3q: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of student satisfaction, independent of primary

teaching level, than those who report a lesser quality mentoring experience.

Hypothesis 3r: Subjects who perceive a higher quality of peer-mentoring

experience will score higher on indicators of student satisfaction, independent of whether

or not the faculty member teaches at other universities, than those who report a lesser

quality mentoring experience.

Hypothesis Statements Related to Type of

Mentor and Student Satisfaction

Hypothesis 3s: Subjects who receive a peer-mentoring experience will score

higher on indicators of student satisfaction than those with an administrator or full-time

faculty mentor.

Hypothesis 3t: Subjects who receive a peer-mentoring experience will score

higher on indicators of student satisfaction, independent of age, than those with an

administrator or full-time faculty mentor.

Hypothesis 3u: Subjects who receive a peer-mentoring experience will score

higher on indicators of student satisfaction, independent of gender, than those with an

administrator or full-time faculty mentor.

Hypothesis 3v: Subjects who receive a peer-mentoring experience will score

higher on indicators of student satisfaction, independent of ethnicity, than those with an

administrator or full-time faculty mentor.

Hypothesis 3w: Subjects who receive a peer-mentoring experience will score

higher on indicators of student satisfaction, independent of length of service, than those

with an administrator or full-time faculty mentor.

Hypothesis 3x: Subjects who receive a peer-mentoring experience will score

higher on indicators of student satisfaction, independent of number of modules taught,

than those with an administrator or full-time faculty mentor.

Hypothesis 3y: Subjects who receive a peer-mentoring experience will score

higher on indicators of student satisfaction, independent of educational level, than those

with an administrator or full-time faculty mentor.

Hypothesis 3z: Subjects who receive a peer-mentoring experience will score

higher on indicators of student satisfaction, independent of primary teaching level, than

those who with an administrator or full-time faculty mentor.

Hypothesis 3aa: Subjects who receive a peer-mentoring experience will score

higher on indicators of student satisfaction, independent of whether or not the faculty

member teaches at other universities, than those with an administrator or full-time faculty

mentor.

The Variable List

Below is a list of the operational definitions for the independent (IV) and

dependent variables (DV), which was used in this study. Table 42 (Appendix A) provides

the reader with information regarding the conceptual, instrumental, and operational

definition of each variable.

1. Peer-to-Peer Mentoring (IV): Refers to the perceived quality of the peer-

mentoring experience. Peer-to-peer refers to the relationship between the mentor and

mentee and denotes that one part-time faculty member mentors another part-time faculty

member. This variable was measured using the 25 questions in Part V of the survey

instrument, “Tell us about your mentoring experience.” Questions 1-9 are answered using

a “yes” or “no” response (yes = 1; no = 0). Questions 10-24 are forced-choice using a

Likert scale of Strongly Agree (SA = 3), Agree (A = 2), Disagree (D = 1), and Strongly

Disagree (SD = 0).

2. Instructor Confidence (DV): Refers to a self-reported measure in which a part-

time instructor indicates his or her level of confidence with the course content, adult

learning methodology, and institutional policies and procedures. This variable was

measured using the 39 questions in Part IV of the survey instrument, “Tell us what you

know.” All questions use a forced-choice Likert scale of Strongly Agree (SA = 3), Agree

(A = 2), Disagree (D = 1), and Strongly Disagree (SD = 0).

3. Institutional Loyalty (DV): Refers to a self-reported measure in which a part-

time instructor indicates his or her preference for accepting a teaching assignment at the

institution in which they were mentored over teaching opportunities at other universities.

(Frequently one finds the term organizational commitment used almost synonymously

with institutional loyalty. Consequently, these terms are used interchangeably.) This

variable was measured using the 25 questions in Part III of the survey instrument, “Tell

us how you feel about the University.” These items were based on the Allen and Meyer

(1990) previously validated organizational commitment scale. Questions for this section

were used with permission. All questions use a forced-choice Likert scale of Strongly

Agree (SA = 3), Agree (A = 2), Disagree (D = 1), and Strongly Disagree (SD = 0).

4. Student Satisfaction (DV): Refers to the self-reported perception by students of

the instructor’s effectiveness in terms of overall course quality, teaching skill, and

availability. Nine items listed under the INSTRUCTOR section of the University’s end-

of-course survey forms were used to measure this variable. (The complete list of

statements can be found in the Instrumentation section of this chapter.) Please note that

the mean for each statement for the three most current end-of-course survey forms is

calculated. The aggregate of 27 mean scores was used to measure this variable. The

assumption has been made that all items are equally important.

5. Age (IV): Refers to the chronological age of each participant and was measured

in years.

6. Gender (IV): Refers to the sex of the participant and was measured as follows:

1 = male, 0 = all others; 1 = female, 0 = all others.

7. Ethnicity (IV): Refers to the ethnic group in which the participant belongs and

was measured as follows: 1 = African-American, 0 = all others; 1 = Asian, 0 = all others;

1 = Caucasian, 0 = all others; 1 = Hispanic, 0 = all others; 1 = Middle Eastern, 0 = all

others;1 = Mixed, 0 = all others; 1 = all others.

8. Educational Level (IV): Refers to the highest educational level achieved by

each participant and was measured as follows: 1 = Bachelor’s Degree; 2 = Master’s

Degree; 3 = Post-Masters; 4 = Specialist degree and, 5 = Terminal Degree (Ph.D.,

D.Min., J.D., Ed.D., or other).

9. Primary Area of Teaching Responsibility (IV): Refers to level where the faculty

member does the majority of his or her teaching. Primary Area of Teaching was

measured as follows: 1 = undergraduate, 2 = graduate, 3 = online.

10. Teaching Experience (IV): Refers to the number of module/courses that each

participant has taught and was measured by reviewing the answer to question 4, which is

found in Part I of the survey, “How many courses/modules have you taught?” This

variable is operationally defined in increments of five (0-5 modules = 0; 6-10 modules

= 1; and more than 10 modules = 2).

11. Teach at Other Universities (IV): Refers to whether or not the participant

teaches at universities other than The University, which is the setting for this research.

This is found in Part I of the survey, “Do you teach (either face to face or online) for

other universities?” Respondents indicate Yes (= 0) or No (= 1). If they answered yes, the

participant is prompted to enter the number of other universities.

The Instrumentation

An online survey instrument was used to gather information that aided in

answering both the research questions and whether to accept or reject the hypotheses. I

developed the survey after an extensive review of the literature. One portion of the survey

utilized the organizational commitment research conducted by Meyers and Allen (1984,

1988). This instrument was deployed to participants using the University’s online survey

site.

Online Survey Instrument

The online survey was developed to address the relationship between the

independent variable, peer-to-peer mentoring, and the twin dependent variables of

institutional loyalty and instructor confidence. The survey was organized into five parts.

Following is a description of each section.

Part I—Informed Consent

This section was the informed consent. Participants were introduced to the

purpose of the study and invited to complete the survey. The benefits and risks were

included; the participant was also informed that the institution would be accessing his or

her most recent end-of-course survey forms. Respondents were told they would remain

anonymous. Respondents had to agree to the conditions detailed in the informed consent.

Those who disagreed were immediately taken to a fade-out page that thanked them for

participating.

Part II—Tell Us about You

A series of drop-down boxes was utilized to collect information with regard to the

participants’ age, gender, ethnicity, educational level, primary instructional area, and

teaching experience.

Part III—Tell Us How You Feel About

the University

This section consisted of 25 questions, developed by Meyer and Allen (1988,

1990), which gathered information related to each instructor’s commitment to the

University. Research suggests that this commitment scale is positively correlated to

institutional loyalty and improved job performance. The institutional loyalty subscale has

been studied and its validity estimate was based on previous research conducted by

Meyer and Allen (1997) and others (Ketchland & Strawser, 2001; LaMastro, 2000;

Meyer et al., 1993). The 22 questions, which were used with permission, are listed below.

Note that due to the University’s desire to remain anonymous, the actual name of the

university, which was included in the survey, has been changed to the University to

protect the identity of the research setting.

1. I do not feel like a part of a family at the University.

2. I am not concerned about what might happen if I left the University without

having another part-time teaching position lined up.

3. I feel emotionally attached to the University.

4. Working at the University has a great deal of personal meaning for me.

5. I feel a strong sense of belonging to the University.

6. I cannot financially afford to leave the University right now.

7. The University does not deserve my loyalty.

8. I would feel guilty if I left the University now.

9. I am proud to tell others I work for the University.

10. I attend faculty meetings and other professional development opportunities on

a regular basis.

11. I would be happy to continue teaching at the University.

12. Right now, staying at the University is a matter of necessity as much as desire.

13. It would be very hard for me to leave the University right now; even if I

wanted to.

14. I really feel that the problems faced by the University are also my problems.

15. I do not feel any obligation to remain at the University.

16. I serve on a curriculum development team.

17. The University deserves my loyalty.

18. It would be wrong to leave the University right now because of my obligation

to the people in it.

19. One reason I continue to work for the University is that leaving would require

considerable sacrifice—another organization may not match the overall benefits I have

there.

20. Too much in my life would be disrupted if I decided I wanted to leave the

University right now.

21. I enjoy discussing the University with people outside of it.

22. I would be willing to serve as a mentor to another part-time faculty member.

23. One of the few serious consequences of leaving the University would be the

scarcity of available alternatives.

24. Even if it were to my advantage, I do not feel it would be right to leave the

University right now.

25. I owe a great deal to the University.

Part IV—Tell Us What You Know

A series of 39 questions was designed to gather data about each respondent’s

confidence level concerning his or her knowledge of course content, adult learning

methodology, and the University’s policies and procedures. Administrators and others

who work closely with full-time and part-time faculty in a supervisory role were

consulted as these questions were constructed. Questions were also taken from

measurement tools used to evaluate instructor performance, such as the student

evaluation and instructor observation forms.

I assumed that an increase in the instructor's knowledge and understanding of the

needs and characteristics of adult learners and in the understanding and implementation

of each school's policies would result in a corresponding increase in confidence. The

literature supports this conclusion. Cross (1981), Merriam ( 2001), Galbraith (2004),

Bash (2005), and Fleming and Garner (2009) include chapters on understanding adult

learners and adult learning theory in their books, which were written for teachers of adult

learners with the expressed intent of increasing instructor effectiveness. The questions

were as follows:

Twenty-five of the 39 items on this subscale relate to effective instructional

behaviors as identified in the literature. In some statements, the opposite wording is

given. For example, the statement on lecture as a primary instructional strategy when

teaching adults is contrary to effective strategies noted in the literature. These statements,

along with corresponding citations, are noted below:

1. When teaching adults, it is important to share practical work/life experiences

(Beder & Darkenwald, 1982; Brookfield, 1986; Pearson, 2005).

2. Adult learners are more disciplined and need less direction from the instructor

(Knowles & Associates, 1984; Knowles et al., 2005).

3. I can verbally explain my philosophy of education and teaching to anyone who

asked me (Heimlich & Norlund, 1994).

100

4. Adult learners are uniquely different from traditional age college students

(Knowles & Associates, 1984; Knowles et al., 2005).

5. I use the time that students are in small groups to review my notes and lesson

plans (Cross, 1981; Galbraith, 2004).

6. When teaching adults, it is important to spend a lot of time lecturing (Cross,

1981; Galbraith, 2004).

7. Overall, adult learners are motivated to learn (Bash, 2005; Cross, 1981;

Fleming & Garner, 2009).

8. I am comfortable facilitating a large group discussion (Brookfield, 1986).

9. I have thought about my philosophy of education and teaching and could write

it down if asked to (Heimlich & Norlund, 1994).

10. I clearly explain course requirements (Brookfield, 1986; James & Maher,

2004; Pearson, 2005).

11. The criteria I use to evaluate student work are clearly defined for the student

(Brookfield, 1986; James & Maher, 2004; Pearson, 2005).

12. The expected learning outcomes for my course(s) are clearly defined

(Brookfield, 1986; James & Maher, 2004; Pearson, 2005).

13. I return assignments and/or exams in a timely manner (James & Maher, 2004;

Pearson, 2005).

14. I provide helpful feedback on assignments (James & Maher, 2004; Pearson,

2005).

15. I provide students with contact information so they may reach me during the

week when questions and problems arise (James & Maher, 2004; Pearson, 2005).

101

16. I am careful to relate course content and material to practical situations (Beder

& Darkenwald, 1982; Brookfield, 1986; Pearson, 2005).

17. I am respectful and open to students’ differing perspectives (Brookfield, 1986;

Donaldson, 1988; Pearson, 2005).

18. I make clear and understandable presentations (Donaldson, 1988; Pearson,

2005).

19. I am knowledgeable about the course content that I teach (Donaldson, 1988;

Knowles et al., 2005).

20. I use a variety of learning approaches (Beder & Darkenwald, 1982; Cross,

1981; James & Maher, 2004; Pearson, 2005).

21. I am helpful and responsive to students (James & Maher, 2004; Pearson,

2005).

22. I am prepared for each class period (Donaldson, 1988; James & Maher, 2004;

Pearson, 2005).

23. I use experiential and applied learning strategies to enhance student learning

(Knowles & Associates, 1984).

24. I integrate relevant theory and knowledge into real-world situations (Knowles

& Associates, 1984).

25. I return graded papers and final grades to students within 2 weeks (James &

Maher, 2004; Pearson, 2005).

The remaining 14 items were designed to measure how well each participant

understood the University’s institutional policies and procedures. The Assistant Dean for

Teaching Learning, who was responsible for orientating new faculty members and

102

implementing the peer-mentoring program, was consulted. She agreed that the following

items represented important expectations regarding the administrative duties of faculty

members. No citations are provided for these questions since they reflect the internal

priorities of the University.

1. I know what to do when a student does not turn the final paper in on time.

2. I am fair and accurate when evaluating a student’s written work and

consistently assign the appropriate grade.

3. I know APA style and consistently insist my students use it when writing

papers.

4. I am familiar with the University’s attendance policy and know how to handle

student absences.

5. I am aware of the faculty handbook.

6. I can recognize student plagiarism and know the correct procedure to follow for

addressing the situation.

7. I can create appropriate makeup assignments for any student who is absent in

my class.

8. I use a grading rubric for major written assignments.

9. I am confident in my ability to handle a student who challenges me in a

classroom setting.

10. I know what to do in an emergency and how to report them.

11. I am confident that if another instructor graded my students’ work, he or she

would assign a similar grade.

12. I am a role model for the Christian life.

103

13. I provide my classes with a written information sheet.

14. I am very familiar with the University’s policies and procedures.

Part V—Tell Us About Your

Mentoring Experience

This subscale was designed to gather information on each participant’s perception

on the quality of a previous peer-mentoring relationship. A fade-out function was used to

exclude participants who indicated they had not been mentored. Based on a review of the

mentoring literature, the following 24 questions were developed (the corresponding

citations are also provided). Questions 1-9 require a yes or no response, except for

question 2, which asks if the mentor was an administrator or full- or part-time faculty

member; and question 3, which asks the length of the mentoring relationship. Questions

10-24 used the following Likert scale: SA = Strongly Agree, A = Agree, D = Disagree

and SD = Strongly Disagree.

1. I have received mentoring (Grant-Vallone & Ensher, 2000; Kram & Isabella,

1985; Mullen, 2000). Yes or No [If respondent selects “no”; he or she was automatically

faded from the survey].

2. My mentor was

a) A full-time faculty member

b) A part-time faculty member

c) An administrator

3. Please indicate how many weeks your mentoring relationship lasted.

4. I observed my mentor teaching, either face-to-face or online (Boyle & Boice,

1998; Le Maistre, Boudreau, & Paré, 2006; Wang, 2001). Yes or No

104

5. My mentor observed me teaching (Le Maistre et al., 2006; Young & Wright,

2001). Yes or No

6. My mentor shared written notes about my teaching with me (Davis, 2001;

Hansman, 2001; Le Maistre et al., 2006; Washburn & LaLopa, 2003). Yes or No

7. My mentor's feedback helped me improve my teaching (Wang, 2001; Young &

Wright, 2001). Yes or No

8. My mentor explained my university's polices to me (Washburn & LaLopa,

2003). Yes or No

9. My mentor helped me understand what was expected of me (Washburn &

LaLopa, 2003) Yes or No

Questions 10-24 used the following Likert scale: SA = Strongly Agree;

A = Agree; D = Disagree; and SD = Strongly Disagree.

1. I was motivated to grow and improve through my mentoring relationship

(Semeniuk & Worall, 2000; Wickman & Sjodin, 1997; J. Wilson & Elman, 1990).

2. I was satisfied with the number of times that my mentor and I met or

communicated (either face-to-face, phone or via email, or other technologies) (Cohen,

1995; Hawkey, 1997; Phi Delta Kappa, 2002).

3. I was satisfied that my mentor was knowledgeable and skilled. He or she was

someone who could help me grow (Goodwin et al., 1998; Young & Wright, 2001).

4. I believe my mentor cared about me as a person and as a teacher (Cohen, 1995;

Hawkey, 1997; Routman, 2000).

5. My mentor was motivated and wanted me to succeed (Pierce, 1998).

105

6. I was satisfied that my mentoring partner and I spent time getting to know each

other (Cohen, 1995; Hawkey, 1997; Routman, 2000).

7. Our mentoring partnership set professional and personal development goals for

the mentoring experience (Young & Wright, 2001).

8. The mentoring goals were written down (Levinson et al., 1978).

9. The mentoring goals were achieved (Levinson et al., 1978).

10. My mentor helped me understand the University’s polices.

11. I was satisfied with how my mentor responded to my questions and concerns

(Goodwin et al., 1998).

12. My mentoring experience has helped me become more confident in my

teaching (Simmons, 1998; Wickman & Sjodin, 1997).

13. I was more confident about my teaching ability after participating in the

mentoring experience (Hawkey, 1997; Simmons, 1998).

14. My mentoring experience helped me feel more loyalty to the University

(Semeniuk & Worall, 2000; Simmons, 1998; Young & Wright, 2001).

15. I would recommend my mentor to another new instructor.

Part VI—Conclusion

The survey concluded with a brief statement thanking the instructor for

participating.

This section described in detail the development and content of the online survey

tool, which was used to collect data related to peer mentoring (IV) and the twin

dependent variables of institutional loyalty and instructor confidence. Historical data

106

from the University end-of-course survey forms were used to measure the third

dependent variable, Student Satisfaction.

End-of-Course Surveys

To measure the student satisfaction variable, I utilized the University’s School of

Business end-of-course survey forms. This form asks students to provide feedback related

to course learning outcomes and instructor effectiveness. For the purposes of this study,

only the nine questions directly related to instructor performance were used. The

questions are as follows:

1. The instructor was very knowledgeable about the subject.

2. The instructor’s Christian faith was clear.

3. The instructor was well prepared for this course.

4. The instructor motivated me to learn about the subject.

5. The instructor was available to help when I needed it.

6. The instructor provided timely feedback on my work.

7. The instructor graded my work fairly.

8. The instructor treated me like an adult professional.

9. Overall, I think the instructor did a good job.

The University provided me with each student’s anonymous ranking on each of

these items. The mean was calculated based on the three most current end-of-course

surveys received.

107

Estimates of Validity and Reliability

Estimates of Validity

Researchers agree that it is not possible to calculate the precise reliability or

validity of any instrument (Newman et al., 2006; Trochim, 2006). However, statistical

processes and procedures can be utilized to estimate the validity of a research tool. When

considering the validity and reliability of an instrument, validity must be considered the

most important characteristic (Newman et al., 2006). This is because validity attempts to

determine the degree to which any instrument accurately measures the construct it was

designed to measure.

The online survey, which I developed, used to gather data related to the research

questions and alternative hypotheses, was divided into the following subscales: Instructor

Confidence, Institutional Loyalty, and Fidelity of the Mentoring Experience. The

Institutional Loyalty subscale was based on previous research and an instrument tool that

has already had its validity estimated; this is a form of construct validity (Newman et al.,

2006). Space is not dedicated in this study to report the results of these tests; the reader

can refer to such studies as Meyer et al. (1993), Meyer and Allen (1997), LaMastro

(2000), and Ketchland and Strawser (2001) for a full discussion.

The other two subscales, Instructor Confidence and Fidelity of Mentoring, used a

form of face validity known as expert judges. After the scale was developed, selected

experts were asked for their feedback. The qualifications of these judges can be found in

Appendix C. In addition to expert judges, the Instructor Confidence and Fidelity of

Mentoring subscales were developed based on content or logical validity. The survey

items for the Instructor Confidence and Fidelity of Mentoring subscales were constructed

108

based on previous research and review of the scholarly literature; consequently, they have

the assumption of a sound estimate of content validity. Each scale was described in detail

earlier in this chapter.

Field Testing

After the initial development of the survey, I consulted several expert judges.

Newman et al. (2006) suggest that expert judges is a type of face validity that is “a little

more sophisticated” than simple face validity (p. 48). I invited the three Regional

Coordinators of Faculty Services from Spring Arbor and the directors of the various

programs to give feedback on the survey’s readability and content validity. The

coordinators and directors supervise and monitor the performance of the adjunct faculty.

Because of this supervisory relationship, the coordinators and directors are uniquely

qualified to provide insight into the reliability of the instrument. In order to ensure the

confidence scale accurately reflected the University practices and policies, the Associate

Vice-president and Associate Dean for Teaching and Learning were also consulted.

In an email, these judges were provided information on the purpose of the survey

as well as the operational definitions of the dependent variables, specifically instructor

confidence and fidelity of mentoring experience. Copies of the subscales were attached

and these expert judges were asked to evaluate how closely, in their opinion, each survey

item approximated the construct being tested. Reviewers were asked to pay particular

attention to the subscale, which was designed to measure Instructor Confidence and was

labeled Part IV—Tell Us What You Know and the Fidelity of Mentoring subscale

(Part V).

109

All five judges (100%) agreed that the Instructor Confidence subscale appeared to

be an appropriate measurement of the instructor confidence variable. One of the five

judges suggested rewording all statements contained in the Instructor Confidence

subscale so respondents could answer using the Strongly Agree to Strongly Disagree

scale. In addition, two questions were added to the Instructor Confidence subscale based

on the recommendation of three of the five judges (60%). They were as follows: “I am

comfortable integrating my faith into the classroom experience” and “I make full use of

the allotted instructional time in each of the courses I teach (i.e. I teach for four hours if

the class is designed to run from 6-10 pm).”

These questions were an important addition; the former question goes to the heart

of the Christian university’s commitment to Jesus Christ. The latter question relates to a

core behavioral expectation of all faculty members who teach in an accelerated adult-

degree completion program like the University’s. In addition, some minor typographical

and formatting errors were corrected.

All (100%) of the expert judges felt that the Fidelity of Mentoring subscale was

an appropriate measure for assessing the perceptions of a previous peer-mentoring

experience. This portion of the survey took less than 5 minutes to complete. All the

judges noted the unclear wording of question 3, "How long did your mentoring

relationship last?" As a result, the question was re-phased to ask how many weeks the

mentoring relationship lasted. Question 7, which asked if the mentor had reviewed the

mentee’s syllabus, was also deleted, based on feedback from the University. Part-time

faculty members are not required to submit a syllabus. Also, both questions 8 and 19

asked if respondents understood the University’s policies, so question 19 was deleted.

110

The directions for completing the survey were also revised based on feedback from one

of the judges.

Estimates of Reliability

After the revisions were completed, approximately 150 part-time faculty members

from Spring Arbor University were invited to field test the Instructor Confidence and

Fidelity of Mentoring subscales using an Internet survey site. This field test was used to

estimate the initial reliability of the survey instrument. The field test also allowed me to

determine how long, on average, it took participants to complete the survey. In order to

encourage maximum participation, the goal was for the survey to take no more than 20

minutes to complete.

Newman et al. (2006) note that reliability estimates can be increased by using the

following procedures: reviewing the instrument for clarity, by standardizing the

administration, increasing the number of items, and eliminating items that are unclear.

All of these strategies were utilized in the development and pre-testing phase. The survey

was field tested for clarity. Since all participants received the same written survey

questions and the same written instructions, administration of the online survey was

standardized and each subscale had a large number of items (22, 39, and 24,

respectively).

The Cronbach’s Alpha was calculated to evaluate whether survey items had

psychometric properties (Trochim, 2006). The Cronbach’s Alpha (α

) for the Loyalty

Scale was .777 (25 items) and the Confidence Scale was α

c =

.937 with 39 items.

Finally, the Fidelity of Mentoring subscale was field-tested using the East Region

Faculty and the Cronbach’s Alpha was calculated (α

c =

.813 with 24 items). Using

111

established procedures to increase reliability resulted in a Cronbach’s Alpha scores of

over .75 on all subscales, and I estimated that each of the subscales was sufficiently

reliable.

Fidelity Measures

All participants responded to the four main sections of the survey and were asked

if they had received mentoring from an administrator, or full-time or other (peer) part-

time faculty member (non-full-time faculty person). Boyle and Boice (1998) suggest that

faculty members believe mentoring received from administrators was more effective than

mentoring received from a peer. For this reason, all participants who participated in a

mentoring experience were asked to complete a fifth section entitled, “Tell Us About

Your Mentoring Experience.” Results from this section were used to arrive at a fidelity

estimate with regard to a common understanding of what is meant by a “quality

mentoring experience.”

Essentially, fidelity measures are used to ascertain how well implementation of an

intervention matched its prescribed protocol (Blase, 2005). Fidelity measures speak not

only to how well one conformed to the “prescribed elements” of a particular intervention

but also the “absence of non-prescribed elements” (Freeman, 2001, para. 2). Blase (2005)

indicates that fidelity measures typically are used to increase confidence in the outcomes

achieved by the proposed intervention. Typically, fidelity measures do not occur in real-

time but tend to be retrospective in nature (Trabin & Minden, 2006).

The fifth section of the survey was comprised of 24 questions; only 21 items were

used to estimate the fidelity. Question 1 asked whether or not the respondent had received

mentoring and question 2 asked if the mentor was full-time faculty, part-time faculty, or

112

an administrator; question 3 asked how many weeks the mentoring relationship lasted.

Questions 4-9 required a yes or no response; questions 10-24 used a Likert scale from

0 (Strongly Disagree) to 3 (Strongly Agree).

Following the completion of the survey, the estimated fidelity was calculated.

Respondents answered the first nine questions. For questions 4-9, yes responses were

coded as 1; no responses were coded 0. For the remaining 15 forced-choice statements,

the responses indicating Strongly Agree were coded as 3 and represents responses at the

top of the scale. Responses for Strongly Disagree were coded as 0. The remaining two

forced-choice responses were coded as 2 (Agree) and 1 (Disagree). The total raw score

was arrived at by summing the individual’s responses for each question. The highest

possible individual raw score for questions 4-9 was 6, and the highest possible individual

raw score for questions 10-24 was 45 (15 * 3). Consequently, the highest possible

individual raw score on Part V of the survey was 51. In order to estimate the fidelity of

each participant’s mentoring experience, the individual raw score was summed. Raw

scores that ranged from 90–100% (46 to 51) were considered as having a high level of

fidelity; scores that ranged from 75-89% (38.25-45.9) were considered as having a

satisfactory level of fidelity. Scores that fell below 38.249 were categorized as having an

unsatisfactory fidelity measure. This perceived quality mentoring score was used to test

the research hypotheses.

This estimate of the fidelity of the peer-mentoring relationship will assist in

interpreting the outcomes of the study. Since participants are recalling past mentoring

experience, estimating the fidelity helps determine how effectively the peer-mentoring

experiences were implemented. If the estimate of fidelity is low, this could imply there is

113

a need to provide mentors and mentees with more effective training and follow-up. If the

estimate of fidelity is high, I can be somewhat assured that the actual peer-mentoring

relationship mirrors the effectiveness noted in the literature. Whether the fidelity estimate

is high or low, the outcomes inform and help interpret the data.

Data Analysis Plan

Both descriptive and inferential statistics were utilized to analyze the data

collected to answer the research questions. Demographic data were collected in order to

best describe the sample population. The hypotheses related to the participants’ age,

gender, ethnicity, educational level, and one of the following primary instructional areas:

graduate or undergraduate and on-site or online. Participants self-reported this

information on the survey.

The length of each participant’s teaching experience was gathered using the

survey and was reported in terms of modules/courses taught and the years the participant

had been affiliated with the University. Many part-time instructors teach one or two

courses each year; therefore, reporting only years of service rather than modules taught

could be misleading and could make it appear the part-time instructor had more teaching

experience than he or she actually had.

In order to clearly and completely report the description of the sample and for the

split cases (mentored and non-mentored), frequency distribution tables were generated

that recorded descriptive statistics for each of the six demographic categories listed

above. Both of the percentages for each of the nominal variables (gender, ethnicity,

educational level, and primary instructional area) were calculated. For the age and

114

teaching experience variables, a table was generated that provided the following

descriptive statistics: mean, minimum, maximum, and the standard deviation.

In order to determine whether to accept or reject the research hypotheses, and to

determine the significance of the independent variable (perceived quality of the peer-

mentoring experience) in predicting the dependent variables (institutional loyalty,

instructor confidence, and student satisfaction) multiple linear regression was used to

analyze the survey responses. This form of statistical analysis is appropriate when there is

a single criterion variable (Y) and multiple predictor variables (X) (Hinkle, Wiersma, &

Jurs, 2003). In the case of this study, the hypotheses stated that the quality of peer

mentoring would predict higher scores for instructor confidence, institutional loyalty, and

student satisfaction. In addition, other independent variables, such as age, gender,

educational level, and experience, are hypothesized not to be significantly related to the

predictor variable.

PASW 18 for Windows® (SPSS, Inc., 2009) was used to analyze the data and to

generate the full and restricted multiple regression models. An alpha level of .05 was

chosen as the criteria for determining whether to accept or reject the directional

hypotheses. The full and restricted models illustrated below were generated in order to

address the research hypotheses. Newman et al. (2006) note that when utilizing linear

regression models, any number of questions can be asked. In addition, regression models

can be generated that will specifically address the research questions. Consequently, the

researcher is not limited in the number of research questions that he or she can ask. “This

is the most compelling reasons for using the general linear model” (McNeil, Newman, &

Kelly, 1996, p. 2).

115

Since the possible variation of models that can be generated is infinite, the full

and restricted models that follow are not meant to be exhaustive. The full and restricted

multiple regression models for the first dependent variable, instructor confidence, are

presented. The same model was used for analyzing the remaining two dependent

variables, institutional loyalty and student satisfaction. Essentially, the full model, shown

below, asks whether or not all the variables together can significantly predict instructor

confidence.

Model One (M1): Full Model

con

= a

U + a

Xmen + a

Xgender + a

Xage + a

+ a

ASN

CAU

+ a

HIS

+ a

oth

+ a

post

+ a

spec

+ a

PhD

+ a

under

+ a

grad

+ a

online

+ a

TeEX

+ E

The first restricted model is written below and was tested against the full model.

All of the variables are removed and the question is asked, “Do all of the variables

identified in the full model account for a significant amount of variance in predicting

confidence, beyond what one would expect by chance alone?” Remember that Y = the

dependent variable confidence, a

U = the constant, and E

= the errors. The restricted

model is written as follows:

Model Two (M2): Restricted Model 2

con

= a

U + E

The final restricted model (M3) removes the independent variable, peer-to-peer

mentoring, and asks whether or not peer mentoring, when it is controlled for everything

else, is a significant predictor of instructor confidence, institutional loyalty, and student

satisfaction. You will note that this model is similar to the full model illustrated above;

116

only the a

men

has been removed. Model three was tested against Model one and asks

whether or not mentoring is a significant predictor of confidence when all the variables in

the restricted model are controlled.

Model (M3): Restricted Model 3

con

= a

U + a

Xgender + a

Xage + a

+ a

ASN

CAU

+ a

HIS

+ a

oth

+ a

post

+ a

spec

+ a

PhD

+ a

under

+ a

grad

+ a

online

TeEX

+ E

An alpha level of .05 (α = .05) was chosen for this particular study. An alpha level

of .05 means that if a relationship is found between peer mentoring and the three

dependent variables, the likelihood of such a relationship occurring by chance is 5 times

out of 100 (Newman et al., 2006). As previously mentioned, all the hypotheses in this

study are directional, since I predict higher scores on the dependent variables from those

participants who indicate a high fidelity measure of the mentoring relationship. As

discussed in the literature review, there is significant support in the literature to predict

that those who receive mentoring will do better than those who do not; at least it seems

reasonable to assume that those who receive mentoring will not do significantly worse.

The Bonferroni test was used to maintain for Type I inflation errors.

Summary of Methodology

This chapter presented the reader with details regarding the study’s methodology

and research design. This ex post facto study primarily used a researcher-constructed

questionnaire to collect data with regard to the respondents’ perception of the quality of a

peer-mentoring relationship as well as instructor confidence and institutional loyalty.

Items from the University’s end-of-course survey forms were used to measure the

117

dependent variable of student satisfaction. Both full and restricted linear regression

models were used to test the 81 research hypotheses and to determine whether the

relationship between the perceived quality of a peer-mentoring experience is related to

increased instructor confidence, institutional loyalty, and student satisfaction. Chapter 4

provides the reader with a description of the research setting as well as a detailed

explanation of the University’s institutional mentoring program.

118

CHAPTER FOUR

SETTING OF THE RESEARCH

This dissertation sought to ascertain whether a relationship exists between

mentoring and a part-time instructor’s level of instructor confidence, institutional loyalty,

and student satisfaction scores, as measured by end-of-course survey forms. Previous

chapters have outlined the perceived over-reliance on part-time faculty and the issues this

dependence raises for universities and colleges. This chapter provides the reader with a

description of the University where the research was conducted. In addition, the

University’s mentoring model is described in detail. The program’s philosophy, purpose,

and outcomes are delineated. The qualifications of faculty mentors are described as well

as the expectations of the faculty mentee.

The University

The University, which is the subject of this research, is a large, private evangelical

Christian university located in the mid-west. The University began offering courses in the

fall of 1920. Several years prior to this, a group of local citizens worked to raise the funds

needed to endow the school. Between the years of 1920 and 1988, the University

operated as a liberal arts school and continually worked to develop a variety of both

Bachelor of Arts and Bachelor of Science degrees. In the mid-80s, the University started

119

offering classes in the evenings and on the weekends in order to attract working adults

who desired to complete their bachelor’s degrees.

Currently, the traditional undergraduate, on-campus programs have an enrollment

of over 3,200 students; the off-campus programs have been hugely successful, having

educated thousands of students. The campus is situated in an urban setting and was

ranked in the top quartile in US News and World Report’s 2012 Best Colleges in the Mid-

West (US News, 2012).

Founded on Christian principles, the school’s website identifies it as a

“Christ-centered academic community” (the University website). The school’s

commitment to excellence and creative programming has resulted in a number of honors

and awards including receipt of the Foundation of Excellence Award, granted in

recognition of the school’s quality programs directed at first-year college students. In

2008, the University was selected as one of the top three Best Christian Workplaces in

America; two faculty members have received Fulbright Scholarships (the University

website). The University is also an active member in the Council for Christian Colleges

and Universities (CCCU).

The huge success of the University programs designed to reach working adults

resulted in the University offering courses at a number of off-site campuses across

several states in the mid-west. The growth of these programs resulted in the University

turning to a large pool of part-time faculty members to staff the thousands of courses

offered each semester. The program was developed and implemented in 2005 and revised

in 2008 (C. Fleming, personal communication, July 9, 2010). The following section

discusses the mentoring program and its protocols.

120

The University Mentoring Program

The administration of the University recognized the need to provide orientation

and mentoring support for its teaching faculty. The growing number of part-time faculty

used to staff courses, as well as the growing number of external sites (both in-state and

out-of-state), contributed to a growing need to implement strategies to connect the

growing number of part-time faculty members to the University and to ensure that quality

instruction was being consistently delivered. This section discusses the University’s

mentoring program’s philosophy, as well as the outcomes of the mentoring experience.

The mentoring model or process is described, and the qualifications of mentors and

mentees are provided. The information provided in theses sections was learned through a

series of emails and conversations with the developer of the mentoring program and from

internal University documents provided to me. These documents can be found in

Appendix B.

The Philosophy of the Mentoring Program

Much of the philosophy of the program was based upon research completed by

Veenman in 1984. Veenman (1984) reviewed over 100 empirical studies and found that

most beginning teachers, who had taught less than 3 years, experienced the following

common problems: (a) classroom discipline/management, (b) motivating students,

(e) organization and planning of course work, and (f) knowledge of university policies

and procedures.

The philosophy of the University’s mentoring program is based on the belief that

the faculty want to be skilled and it is the University’s responsibility to provide them with

121

a comprehensive plan for development. The University strongly believes that the faculty

are “one faculty” and work hard to minimize any distinctions and felt differences

between full- and part-time faculty members. A continuous improvement model is the

basis for developing excellence in the “delivery of instruction for adult learners” (see

Appendix B, Induction Program Overview). In order to achieve this both full and affiliate

(part-time) faculty are utilized to provide support and mentoring to new faculty. The goal

is for the program to be ongoing and to utilize a variety of formats, which include on-site

and online training, classroom observations, and individual coaching sessions.

The Mentoring Program Outcomes

It is hoped that upon completion of the mentoring experience, the entire faculty

(full-time and part-time) will function as one faculty in order to provide quality

educational experiences for all the University’s adult learners, regardless of delivery

system (face-to-face, online, on campus, offsite). The purposes of the mentoring program

are found in the “Adult and Professional Studies Mentoring Program” handout in

Appendix B and are outlined as follows:

1. Provide all faculty members with the proper materials, training, and support to

enable them to succeed.

2. Provide all new faculty members with a mentor, who will discuss strengths and

weaknesses with them, as well as assist them with teaching strategies, preparation of

course syllabi, and development of lesson plans.

3. Provide new faculty members with mentors from the same discipline,

recognizing that a full-time faculty mentor or experienced adjunct from another area

could still provide mentoring.

122

4. Provide orientation sessions that must be attended by all new faculty members.

5. Provide new faculty members with an introduction to andragogy and pedagogy.

6. Provide new faculty with the basic concepts of evaluating student work.

7. Provide a thorough understanding of the University’s grading criteria.

Qualifications of the Mentor

Before detailing the specific procedures that mentors and mentees follow, it is

important to note how the University selects mentors. The literature provides a plethora

of qualities the effective mentors must possess. Prospective mentors are required to

complete and submit an application (see Appendix B). The University has identified the

following qualifications.

The faculty mentor must either be a full-time faculty member or be a selected

part-time faculty member. The mentor, regardless of employment status, must receive

consistently positive evaluations on his or her teaching observations and on the

University’s end-of-course survey form. Faculty members are observed by University

administration and are evaluated on a number of elements. In addition, students assess

instructor effectiveness using a standardized student satisfaction form. This qualification

addresses a key trait that the literature suggests mentors must have; that is, the ability to

lead by example (Phi Delta Kappa, 2002). The University expects that mentors are good

role models and have the respect of others at the University.

The prospective mentor must also meet the University guidelines for attendance at

annual faculty development events. This highlights another characteristic found in the

literature. Mentors must possess the ability to help others assess their development and to

help them create opportunities for growth (La Maistre et al., 2006). Mentors must show a

123

willingness to share their knowledge, show the ability to motivate and encourage others,

and possess the knowledge to be an influential mentor.

Mentors must also exhibit a willingness to fulfill the duties and responsibilities of

a mentor. It seems quite elementary to state the obvious. However, the literature is clear

in this regard. Successful mentors possess a desire to give back to others (Biehl, 1996;

Pierce, 1998). Gaskill (1993) also notes that when using a systematic mentoring program

it is essential that mentor qualifications be reviewed and that the mentors exhibit a

willingness and desire to participate as mentors.

Finally, all mentors must successfully complete the mentoring training. During

this training, mentors receive information on how to create a trusting relationship and

how to provide constructive feedback. In addition, the mentors receive training on what is

expected of them. Gaskill (1993) recommends that mentor training include an orientation

to the program’s purposes and objectives. The mentor training should also serve as time

to minimize any ambiguity regarding the roles, duties, and responsibilities of the mentor.

During training, the University provides the following information regarding the

expectations of the mentoring role and is outlined in the “Mentoring Program for New

Adjunct Faculty” in Appendix B.

1. Mentors are to establish contact with the new faculty member.

2. Mentors are to welcome the new faculty to their classroom.

3. Mentors are to review the new faculty member’s syllabus/faculty information

sheet.

4. Mentors are to utilize a variety of teaching strategies during the adjunct faculty

member’s visit.

124

5. Mentors are to designate, after discussion with the new faculty member, a

segment of the class session to be taught by new adjunct (devotion, prayer, mini-lecture,

small-group activity, class discussion).

6. Mentors are to observe the new adjunct faculty member during his or her

teaching segment.

7. Mentors are to complete the observation form.

8. Mentors are to spend time as needed to discuss the observation form with the

new adjunct faculty member, highlighting strengths and making suggestions.

9. Mentors are to submit the observation to the Faculty Development Assistant.

10. Mentors are to maintain contact with the new adjunct as needed to assist

during the new adjunct’s first teaching assignment.

The Mentoring Program Model

The procedure or model the University uses is outlined in the “Induction Program

Overview,” and this document can be found in Appendix B as well. The forms noted

below can also be found in this appendix. Because the mentoring program was developed

by University administration and it is expected that all new faculty complete the program

prior to their first teaching assignment, the model qualifies as a systematic mentoring

program. Systematic mentoring programs are found primarily in the business sector and

occur when organizational leaders institutionalize the mentoring program (Kaye, 2000).

These programs are characterized by written goals, specific practices are outlined, and

strategies for linking mentor and mentee are identified (Gaskill, 1993).

The University has also recognized the value of having part-time faculty members

serve in the role of mentor. Due to the size of the University’s programs and the large

125

number of part-time faculty used to staff courses, there are insufficient numbers of full-

time faculty or administrators to meet the mentoring demand (Kram & Isabella, 1985).

Research suggests that peer mentoring has the potential to provide a wide range of

support that results in increased competence and confidence (Kram & Isabella, 1985).

The mentoring experience of new faculty begins after the new faculty member has

attended the two orientation sessions and completed the “New Adjunct Faculty

Mentoring Assignment.” The form asks for the new faculty’s name and contact

information, the desired location and nights for the mentoring experience, and the courses

he or she expects to teach. The form reminds new faculty that they must complete the

mentoring program before an initial teaching assignment was given.

Upon completion of the form, the administration matches the mentee with a

suitable mentor. The mentor is selected from a list of qualified and available mentors that

is kept by the University’s administration. When the mentor accepts the mentoring

assignment, he or she is provided with the mentee’s contact information. A “Mentoring

Assignment Checklist” is then begun; this form records the new faculty member’s

information and the mentor assigned to him or her. This form is also used to track the

mentoring activity and to ensure that the classroom observations occur and the necessary

forms are completed. The secretary then mails the forms that the mentor must complete

as part of the mentoring experience to the mentor at this time.

At this point, the mentor and mentee make contact in order to make arrangements

to meet. At this initial meeting, the mentor and mentee meet and collaborate on goals and

a time for the mentee to visit and observe the mentor’s classroom. At this juncture, the

mentee also completes a self-assessment form to aid in the formulation of goals for the

126

mentoring relationship. The mentee is also provided information on how to develop his or

her course syllabi and lesson plan. In addition, the mentor will discuss with the mentee

which segment of the observed class time he or she would like to facilitate (devotion,

prayer, mini-lecture, small-group activity, class discussion, etc.).

The next activity involves the mentee’s actual visit to the mentor’s classroom.

During this time, the mentor is expected to utilize a variety of teaching strategies and this

allows the mentee to facilitate the previously agreed-upon activity. The mentor completes

the “Mentee Observation Form” during the time the mentee facilitates. Upon completing

the classroom visit, the mentee and mentor complete page 2 of this form together. Page 2

asks the mentee to record his or her observations and insights from observing the

mentor’s teaching. Specifically, the mentee is asked to identify behaviors and strategies

that were observed that he or she will incorporate his or her teaching and to list any

questions he or she has about the classroom experience.

On the same form, the mentor is asked to critique the strengths he or she observed

while the mentee was conducting his or her portion of the evening’s lesson. The mentor

also reviews the mentee’s proposed syllabus and lesson plan. A copy of this completed

form is sent to the Faculty Development Assistant who sees that the mentee is given a

copy, and that one copy is kept in the mentee’s permanent employment file. The

submission of this completed form begins the payroll process and the compensation is

then paid to the mentor. However, mentors are expected to remain available to the mentee

throughout the first teaching assignment to provide support and to answer any questions.

Sometimes, through mutual agreement, the mentor and mentee elect to continue the

relationship.

127

Conclusion

Chapter 4 provided the reader with a description of the setting of the research.

This University was chosen because it has a strong program designed for the adult

learner. A large number of part-time faculty members are employed to staff classrooms in

over 80 off-site locations. In addition, the University has developed an organizational

mentor program that utilizes part-time faculty members to mentor other part-time faculty

members. This chapter outlined the philosophy of the mentoring program and the

qualifications of the mentors, and described in detail the mentor program protocols.

Chapter 5 presents the results of the study.

128

CHAPTER FIVE

RESULTS

The purpose of this study was to examine what relationship, if any, exists between

the self-reported perceptions on the quality of a mentoring experience on a part-time

faculty member’s instructor confidence, institutional loyalty, and student satisfaction

scores at a large, private Christian University in the mid-west. The study also sought to

determine if a peer mentor significantly related the relationship to the three dependent

variables mentioned previously. This chapter presents the findings of the study.

First, the response rate is given as well as the number of cases used in the analysis

of the data. Following this, descriptive statistics are provided on a number of variables.

The sample is described in its entirety and then descriptive statistics are presented

according to whether or not the participant received mentoring or not. Correlational

matrixes were generated to determine if any significant relationships existed between the

independent and dependent variables. Linear regression models were then generated in

order to test the hypotheses.

Response Rate

Six hundred part-time faculty members who teach in the School of Business in the

College of Adult and Professional Studies (CAPS) programs from a large, private

midwest University met the eligibility requirements outlined in the research protocol

129

described in chapter 3. To be eligible, participants must have taught at least three courses

between June 1, 2009, and December 31, 2010.

The Associate Vice-president sent an invitation email to the 600 eligible

participants on May 13, 2011. The survey was deployed through the school’s Office of

Research and Program Development. A reminder email was sent prior to the survey being

closed on July 18, 2011 (A copy of the emails and the survey can be found in Appendices

C and D). Of the 600 participants, 150 completed the survey for a response rate of 25%.

Of the 150 who responded, 3 did not agree to the conditions outlined in the informed

consent, which left 147 usable cases. See Table 2.

Table 2

Descriptive Statistics for Response Rate

Response Rate

I have read and agree to

the conditions.

147

98.0

I have read and do not

agree with the

conditions.

2.0

Total

150

100.0

Descriptive Statistics

Of the 150 respondents, 147 usable cases were identified. This section provides

the reader with descriptive statistics relative to the sample. In addition, descriptive

statistics are further analyzed based on whether the respondent received mentoring or not.

130

Of the 147 respondents, 63 or 42.9% indicated they had received mentoring, and 57.1%

had not received mentoring. With regard to what type of mentor they had (full-time

faculty, part-time faculty, or administrator), 31 (49.2%) of faculty report they were

mentored by a peer (another part-time faculty member). More than 36% were mentored

by a full-time faculty person, and 8 or 12.7% received mentoring from an administrator.

Tables 3 and 4 illustrate these findings.

Table 3

Descriptive Statistics for Mentoring Status

Mentoring Status

Mentored

42.9

Non-mentored

57.1

Total

147

100.0

Table 4

Descriptive Statistics for Type of Mentoring Experiences

Type of Mentor

Part-time Faculty

49.2

Full-time Faculty

36.5

Administrator

12.7

Missing

1.6

Total

100.0

131

Of the total number (147) of respondents, 109 (74.1%) were male; 37 (25.2%)

were female. One person elected not to respond to that item (7%). As mentioned earlier,

the cases were split into two groups: those who had received mentoring and those who

had not. In the mentoring subgroup, 69.8% or 44 were male while 18 (28.6%) were

female (n = 63). One case (1.6%) did not respond to the gender question. From the non-

mentored subgroup, 65 or 77.4% were male; 22.6% (n = 19) were female (n = 84). See

Table 5, which summarizes these results.

The ethnic distribution of the sample and its subgroups can be found in Table 6,

Descriptive Statistics for Ethnicity. In terms of ethnic background, 11.6% (n = 17) were

African-American; 122 (83%) were Caucasian, and 1 (7%) each for the following

ethnicities: Asian/Pacific Islander, Hispanic, Middle Eastern, Alaskan Native, Greek, and

Mixed Race. There was one missing case (7%). Reviewing the frequency distribution for

ethnicity by group, the mentoring group (n = 63) noted 10 (15.9%) were African

American, 52 Caucasian (83.5%), and 1 Hispanic (1.6%). From the non-mentored

subgroup, the large majority were Caucasian (n = 70; 83.3%), African-Americans had the

second highest percentage with 8.3% (n = 7). This group also reported 1 (1.2%) for each

of the following: Asian/Pacific Islander, Middle Eastern, Greek, and Alaskan Native. In

addition 2 (2.4%) indicated a mixed ethnic heritage.

It was determined that using single cases of the other categories (Asian/Pacific,

Middle Eastern, Greek, Hispanic, Alaskan, etc.) for ethnicity in the testing of the

hypotheses could create outlier data that could skew the results. Consequently, I decided

to use only the African-American and Caucasian variables in the analysis. These results

for ethnicity are summarized in Table 6.

132

Table 5

Descriptive Statistics for Gender

Gender

Entire Sample

Male

109

74.1

Female

25.2

Missing

0.7

Total

147

100.0

Mentored Group

Male

69.8

Female

28.6

Missing

1.6

Total

100.0

Non-mentored Group

Male

77.4

Female

22.6

Missing

0.0

Total

100.0

133

Table 6

Descriptive Statistics for Ethnicity

Ethnicity

Entire Sample

African-American

11.6

Asian-Pacific Islander

0.7

Caucasian

122

83.0

Hispanic

0.7

Middle Eastern

0.7

Alaskan Native

0.7

Greek

0.7

Mixed Race

2.0

Total

147

100.0

Mentored Group

African-American

15.9

Caucasian

82.5

Hispanic

1.6

Total

100.0

Non-mentored Group

African-American (AA)

8.3

Asian/Pacific Islander

1.2

Caucasian

83.3

Hispanic

1.2

Middle Eastern

1.2

Alaskan Native

1.2

Greek

1.2

Mixed Race

2.4

Total

100.0

134

Respondents also were asked to indicate the highest level of education they had

completed (master’s, post-masters, Ph.D., or other terminal degree). Over half (55.8%) of

the sample had a master’s degree (n = 82); 16.3% had post-master’s education (n = 24),

and 41 (27.9%) had some kind of terminal degree (Ph.D., D.Min., Ed.D., J.D., or other).

Master’s degrees were held by 25 or 39.7% of those in the mentored subgroup, 14

(22.2%) had completed some level of post-master’s degree coursework, and 24 or 38.1%

had terminal degrees. In the non-mentored subgroup, 57 (67.9%) held master’s degrees,

10 (11.98%) had work at the post-master’s level, and 17 (20.2%) had terminal degrees.

These results are summarized in Table 7.

The mean age for the entire sample was 52.87 (SD = 9.935) with a range of 31 to

79 years. For the mentored subgroup, the mean age was 50.9 (SD = 10.922) with a range

between 31 and 79; for the non-mentored subgroup, the mean age was 54.36 (SD =

8.896) and the range was between 33 and 77. Table 8 illustrates these results.

Based on the results for the entire sample, the mean length of time that a part-time

faculty has taught for the University was 6.83 years (SD = 3.928) with a range between 2

to 25 years. The mentored subgroup had a mean length of service of 5.52 years (SD =

2.911) with a range of 2 to 16 years. The non-mentored subgroup had a range between 2

and 25 years of service (M = 7.81; SD = 4.312). Table 9 summarizes these results.

Due to the nature of part-time teaching assignments, years of service can provide

an inflated sense of seniority. As a result, respondents were also asked to indicate how

many courses/modules they had taught at the University. For the entire sample, 16 or

10.9% taught 5 or less modules; 18.4% (n = 27) have taught between 6 and 10 modules,

and over 70% (70.7%; n = 104) have taught more than 10 modules. In the mentored

135

subgroup, 4 (6.3%) taught between 0–5 modules, 17.5% (n = 11) taught between 6 and

10 courses, and 48 (76.2%) had taught more than 10 modules. From the non-mentored

subgroup, 12 (14.3%) taught five or fewer modules, 16 (19%) taught between 6 and 10

courses, and 56 (66.7%) taught over 10 classes. Table 10 provides a summary of this

result.

Table 7

Descriptive Statistics for Educational Level

Educational Level

Entire Sample

Master’s

55.8

Post-Master’s

16.3

Terminal Degree

27.9

Total

147

100.0

Mentored Group

Master’s

39.7

Post-Master’s

22.2

Terminal Degree

38.1

Total

100.0

Non-mentored Group

Master’s

69.9

Post-Master’s

11.9

Terminal Degree

20.2

Total

100.0

136

Table 8

Descriptive Statistics for Age

Item

Min

Max

Entire Sample

147

52.87

9.935

Mentored Group

50.9

10.900

Non-mentored Group

54.36

8.896

Table 9

Descriptive Statistics for Years of Service

Item

Min

Max

Entire Sample

147

6.83

3.928

Mentored Group

5.52

2.911

Non-mentored Group

7.81

4.312

When asked if they taught at other universities, 36.1% (n = 53) of the entire

sample indicated yes, while 94 (63.9%) said no. From the mentored group, 46%

responded yes to this question (n = 29) and 54% or 34 said no. Twenty-four respondents

in the non-mentored subgroup (28.6%) indicated they did teach for other universities,

while over 71.4% (n = 60) said no. Table 11 illustrates these results. In addition, those

who indicated they taught at other universities were asked to indicate how many. Those

results are found in Table 12.

Respondents also were asked to indicate whether they taught undergraduate

classes, graduate classes, or online courses. Because respondents were not forced to

137

indicate only one preference, the total number of responses (n = 199) is greater than the

number of participants in the sample (n = 147). Table 13 provides a summary of the

findings.

Table 10

Descriptive Statistics for Number of Modules Taught

# of Modules Taught

Entire Sample

0–5

10.9

6–10

18.4

More than 10

104

70.7

Total

147

100.0

Mentored Group

0–5

6.3

6–10

17.5

More than 10

76.2

Total

100.0

Non-mentored Group

0–5

14.3

6–10

19.0

More than 10

20.2

Total

100.0

138

Table 11

Descriptive Statistics for Taught at Other Universities

Taught at Other

Universities

Entire Sample

Yes

36.1

63.9

Total

147

100.0

Mentored Group

Yes

46.0

54.0

Total

100.0

Non-mentored Group

Yes

28.6

71.4

Total

100.0

Fidelity Measure

In order to test the hypotheses related to the fidelity measure, only those

participants who responded to all of the questions in the Instructor Confidence,

Institutional Loyalty, and Fidelity subscales were used in the testing of the hypotheses

related to the quality of the mentoring experience. Of the 63 who indicated they had

received mentoring, only 50 answered all the questions in the Fidelity subscale, and of

those 50, only 46 answered all the questions in all the subscales and were included in the

testing of the hypotheses related to the fidelity measure.

139

Table 12

Descriptive Statistics for How Many Other Universities

# of Other Universities

Entire Sample

44.4

27.8

22.2

5.6

Total

100.0

Mentored Group

35.0

20.0

10.0

Total

100.0

Non-mentored Group

56.3

18.8

25.0

0.0

Total

100.0

140

Table 13

Descriptive Statistics for Teaching Level

Teaching Level

Entire Sample

Undergraduate Classes

103

52.0

Graduate Classes

25.0

Online Classes

23.0

Total

199

100.0

Mentored Group

Undergraduate Classes

46.0

Graduate Classes

22.0

Online Classes

32.0

Total

100.0

Non-mentored Group

Undergraduate Classes

57.0

Graduate Classes

24.0

Online Classes

19.0

Total

102

100.0

According to the research protocol, the fidelity score was assigned based on the

total raw score obtained from the Fidelity subscale; the highest possible raw score was

72. A raw score which fell in the 90–100% range (65-72) was coded as 3, represented the

highest fidelity score possible, and was considered excellent. Raw scores that ranged

from 54-64 (75-89%) were coded as a 2 and were considered to have a satisfactory level

of quality. Any raw score that fell below 53 (or less than74%) was coded as a 1 and was

141

considered to have a poor level of fidelity. Of the 50 respondents who answered all the

questions in the Fidelity subscale, 26 (56.5%) scored a 3; 19 (41.3%) scored a 2, and

1 (2.20%) scored a 1. In generating the regression models, only the raw score on each

scale was used (see Table 14).

Table 14

Fidelity Score Frequencies

Raw Score

Fidelity Rank

0-53

2.20

54-64

41.30

65-72

56.50

Total

100.00

Correlation Matrix

The Pearson R correlation coefficient was initially run in order to determine if any

significant relationship existed between the independent variables (mentoring, fidelity of

the mentoring experience, and type of mentor) and the three dependent variables

(instructor confidence, institutional loyalty, and student satisfaction). The results show a

significant relationship exists between mentoring and instructor confidence and between

institutional loyalty and confidence. These relationships were significant at the .01 level.

Table 15 provides the correlational matrix for these variables.

A second correlational matrix was generated in order to test what relationships

may exist between the second cluster of hypotheses and variables. Namely, does a

142

significant relationship exist among these variables when one considers the quality or

fidelity of the mentoring experience? The results show a strong, positive correlation

(r = .517; p = .000) exists between the fidelity of the mentoring program and institutional

loyalty. Table 16 provides the correlational matrix for these variables.

Table 15

Correlation Matrix for Confidence, Loyalty, Student Satisfaction, and Received

Mentoring

Mentored Confidence Loyalty Student Satisfaction

Mentored

Confidence

.273

Loyalty

.073

.342

Student Satisfaction

-.058

.035

.048

Correlation is significant at the .01 level; mentored coded as received mentoring = 1;

all others = 0.

Table 16

Correlation Matrix for Confidence, Loyalty, Student Satisfaction, and Fidelity

Fidelity Confidence Loyalty Student Satisfaction

Fidelity

Confidence

-.069

Loyalty

.517

.342

Student Satisfaction

-.003

.035

.048

Correlation is significant at the .01 level; fidelity coded as unsatisfactory = 1;

satisfactory = 2; excellent = 3.

143

Finally, a third correlational matrix was generated to determine whether or not the

type of mentor a faculty member had (full-time faculty, part-time faculty, or an

administrator) was significantly related to the three dependent variables. The hypothesis

statements postulated that mentees who were mentored by a peer would do as well or

better than those mentored by either a full-time faculty member or administrator. The

results show no significant relationship exists between the type of mentor and the

instructor’s confidence, loyalty, or student satisfaction scores. Table 17 provides the

correlational matrix for this cluster of variables.

Multiple Regression Analysis

According to the data analysis plan outlined in chapter 3, linear regression models

were generated in order to test a number of hypotheses. The hypothesis statements were

organized in clusters by the dependent variables, predictor variable, and a number of

independent variables. This section highlights the results of the linear regression models

that were generated. It is important to note that the research hypotheses predict that

mentoring will result in increased confidence, loyalty, and student satisfaction.

Additional hypotheses were generated that stated the higher the fidelity score of the

mentoring experience, the higher the measures on the dependent variables and that those

with a peer mentor would do better than those without a peer mentor.

Instructor Confidence Variables

The first linear regression model tested the ability of mentoring (PV) to predict a

part-time faculty member’s instructor confidence (DV). Participants were grouped into

those who had received mentoring and those who had not. Additional hypothesis

144

statements were generated that asked whether or not the significant relationship between

mentoring and instructor confidence was independent of age, gender, ethnicity, length of

service, number of courses taught, educational level, primary teaching level

(undergraduate, graduate, and online), and whether faculty taught at other universities or

not. A second model was generated, similar to the first, but used the fidelity score of the

mentoring program, rather than the receipt of mentoring, as the predictor variable.

Finally, a third model was generated that used the type of mentor as the predictor

variable.

Table 17

Correlation Matrix for Confidence, Loyalty, Student Satisfaction, and Type of Mentor

Confidence

Loyalty

Student Satisfaction

Full-Time

Part-Time

Confidence

Loyalty

.342

Student

Satisfaction

.035

.048

Full-Time

.168

-.046

-.056

Part-Time

.061

.011

-.012

.223

Administrator

.165

.152

-.028

-.103

-.124

Note. Type of Mentor coded as follows: FT = 1; all others = 0; PT = 1; all others = 0;

Admin = 1; all others = 0.

Correlation is significant at the .01 level.

In order to report the results, the hypothesis statements for each cluster of

predictor variables (mentoring, fidelity, and type of mentor) are given followed by the

results of the regression model.

145