<%BANNER%>

Followership Behaviors among Florida Community College Faculty

Permanent Link: http://ufdc.ufl.edu/UFE0024186/00001

Material Information

Title: Followership Behaviors among Florida Community College Faculty
Physical Description: 1 online resource (129 p.)
Language: english
Creator: Smith, John
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: college, community, faculty, follower, followership, leader, leadership
Educational Administration and Policy -- Dissertations, Academic -- UF
Genre: Higher Education Administration thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: As postsecondary institutions are confronted by the challenges of escalating accountability, shrinking budgets, and administrative downsizing, higher education leaders are expecting more from their faculty members. In this environment, an improved understanding of faculty followership behaviors is increasingly important. We examined the relationship between followership behaviors and individual variables, organizational culture, and institutional variables among Florida community college faculty. Followership behaviors consisted of a total followership score and five dimensional subscores; responsibility, service, challenge, transformation, and moral action. Individual variables included faculty rank, receipt of tenure, age, sex, race, education level, discipline, and duration of employment. Institutional variables included size of the institution, size of the population served, location, and degrees offered. An on-line questionnaire was completed by 661 faculty members from 27 of Florida's 28 community colleges. Analyses revealed significant effects for age, education level, and discipline for the responsibility dimension; sex, rank, and discipline for the service dimension; age and discipline for the challenge dimension; tenure, sex, and discipline for the transformation dimension; tenure and age for the moral action dimension; and age, sex, rank, and discipline for the total followership score. Further analyses indicated significant interactions in the responsibility dimension for tenure by duration of employment and academic discipline by duration of employment; and in the service dimension for age by tenure. Significant results for organizational culture were found for each followership dimension except responsibility. Also, statistically significant results were found among institutional variables for the moral action dimension for degree offered, with faculty from colleges that offered bachelor's degrees scoring higher for moral action than faculty from institutions that did not offer bachelor's degrees.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by John Smith.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Honeyman, David S.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024186:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024186/00001

Material Information

Title: Followership Behaviors among Florida Community College Faculty
Physical Description: 1 online resource (129 p.)
Language: english
Creator: Smith, John
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: college, community, faculty, follower, followership, leader, leadership
Educational Administration and Policy -- Dissertations, Academic -- UF
Genre: Higher Education Administration thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: As postsecondary institutions are confronted by the challenges of escalating accountability, shrinking budgets, and administrative downsizing, higher education leaders are expecting more from their faculty members. In this environment, an improved understanding of faculty followership behaviors is increasingly important. We examined the relationship between followership behaviors and individual variables, organizational culture, and institutional variables among Florida community college faculty. Followership behaviors consisted of a total followership score and five dimensional subscores; responsibility, service, challenge, transformation, and moral action. Individual variables included faculty rank, receipt of tenure, age, sex, race, education level, discipline, and duration of employment. Institutional variables included size of the institution, size of the population served, location, and degrees offered. An on-line questionnaire was completed by 661 faculty members from 27 of Florida's 28 community colleges. Analyses revealed significant effects for age, education level, and discipline for the responsibility dimension; sex, rank, and discipline for the service dimension; age and discipline for the challenge dimension; tenure, sex, and discipline for the transformation dimension; tenure and age for the moral action dimension; and age, sex, rank, and discipline for the total followership score. Further analyses indicated significant interactions in the responsibility dimension for tenure by duration of employment and academic discipline by duration of employment; and in the service dimension for age by tenure. Significant results for organizational culture were found for each followership dimension except responsibility. Also, statistically significant results were found among institutional variables for the moral action dimension for degree offered, with faculty from colleges that offered bachelor's degrees scoring higher for moral action than faculty from institutions that did not offer bachelor's degrees.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by John Smith.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Honeyman, David S.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024186:00001


This item has the following downloads:


Full Text

PAGE 1

1 FOLLOWERSHIP BEHAVIORS AMONG FLORIDA COMMUNITY COLLEGE FACULTY By JOHN SCOTT SMITH A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 John Scott Smith

PAGE 3

3 To my wife and daughter

PAGE 4

4 ACKNOWLEDGMENTS I express my sincere appreciation to a number of people who made this degree possible. F irst, I thank BG (Ret.) Barney Forsythe for selecting me to attend graduate school under the arm ys Advanced Civil Schooling fellowship program. Without this support, it is unlikely that I would have attempted this journey. Second, I express my gratitude t o Dr. James Doud for suggesting the topic of followership; Dr. David Quinn and Dr. Linda Behar -Horenstein for their assistance in developing my dissertation proposal; Dr. Gene Dixon and Dr. John Smart for allowing me the use of their measures; and the Flor ida community college faculty members who participated in this study. Third, I wish to acknowledge my dissertation committee; Dr. Dale Campbell, Dr. Walter Leite, and Dr. Craig Wood. Special credit must be given to Dr. David Honeyman who, as my committee c hair, helped me find my way through my graduate program and the dissertation process. Finally, I express my sincere thanks to my family who have stood beside me every step of the way. I have truly benefitted from their encouragement, patience, and unfailin g support. This journey would not have meant nearly as much without them by my side.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................................... 4 LIST OF TABLES ................................................................................................................................ 7 LIST OF FIGURES .............................................................................................................................. 9 LIST OF FIGURES .............................................................................................................................. 9 CHAPTER 1 INTRODUCTION ....................................................................................................................... 12 Statement of the Problem ............................................................................................................ 13 Purpose of this Study .................................................................................................................. 16 Hypotheses ................................................................................................................................... 18 Definition of Terms ..................................................................................................................... 18 Significance of the Study ............................................................................................................ 19 Limitations ................................................................................................................................... 20 2 REVIEW OF RELATED LITERATURE ................................................................................. 22 Background of Followership ...................................................................................................... 22 Characteristics of Exemplary Followers .................................................................................... 23 Types of Followers ...................................................................................................................... 28 Courageous Followership ........................................................................................................... 33 Organizational Culture ................................................................................................................ 37 Measures of Followership........................................................................................................... 39 Summary ...................................................................................................................................... 42 3 DESIGN OF THE STUDY ........................................................................................................ 45 Population .................................................................................................................................... 45 Sampling Frame .......................................................................................................................... 46 Sample Size ................................................................................................................................. 46 Instrumentation ............................................................................................................................ 48 Procedure for Data Collection .................................................................................................... 51 Variables ...................................................................................................................................... 52 Statistical Analyses ..................................................................................................................... 54 4 RESULTS AND DATA ANALYSES ....................................................................................... 57 Response Rate ............................................................................................................................. 57 Respondent Profile ...................................................................................................................... 57 Instrument Performance .............................................................................................................. 57

PAGE 6

6 Hypotheses Testing ..................................................................................................................... 59 Summary ...................................................................................................................................... 71 5 DISCUSSION .............................................................................................................................. 93 Findings ....................................................................................................................................... 93 Limitations and Future Research Considerations .................................................................... 101 Conclusions ............................................................................................................................... 104 APPENDIX A SAMPLE SIZE CALCULATION ........................................................................................... 108 B THE FOLLOWERSHIP PROFILE ABBREVIATED ........................................................ 109 C PERMISSION TO USE THE FOLLOWERSHIP PROFILE ................................................ 111 D IPS CULTURAL SCENARIOS ............................................................................................... 112 E PERMISSION TO USE CULTURAL SCE NARIOS ............................................................ 114 F INTRODUCTORY EMAIL AND CONSENT FORM .......................................................... 115 G INSTITUTIONAL REVIEW BOARD LETTER ................................................................... 116 H DEMOGRAPHICS QUESTIONNAIRE ................................................................................. 117 LIST OF REFERENCES ................................................................................................................. 119 BIOGRAPHICAL SKETCH ........................................................................................................... 129

PAGE 7

7 LIST OF TABLES Table page 4 1 Response rates by participating institution ........................................................................... 72 4 2 Percentage of sample N represented by d emographic variables of interest ....................... 73 4 3 Comparison of total population versus responding sample population .............................. 73 4 4 Rotated component matrix ..................................................................................................... 74 4 5 Pearson r correlation coefficients and significance levels for independent variables and TFP scores ....................................................................................................................... 75 4 6 Summary total followership factor score regression table for demographic variables ...... 76 4 7 Mean total followership factor scores and standard deviations grouped by sex ................ 76 4 8 Mean total followership factor scores and standard deviations grouped by d iscipline ..... 76 4 9 Mean total followership facto r scores and standard deviations grouped by faculty rank .......................................................................................................................................... 76 4 10 Summary responsibility factor score regression table for demographic variables ............ 78 4 11 Mean responsibility factor scores and standard deviations grouped by education level ... 78 4 12 Mean responsibility factor scores and standard deviations grouped by discip line ............ 78 4 13 Summary service factor score regression table for demographic variables ....................... 80 4 14 Mean service factor scores and stan dard deviations grouped by faculty rank ................... 80 4 15 Mean service factor scores and standard deviations grouped by sex .................................. 80 4 16 Mean s ervice factor scores and standard deviations grouped by discipline ....................... 80 4 17 Summary challenge factor score regression table for demographic variables ................... 81 4 18 Mean challenge factor scores and standard deviations grouped by discipline ................... 81 4 19 Summary transformation factor score regression table for demographic variables ........... 83 4 20 Mean transformation factor scores and standard deviations grouped by tenure ................ 83 4 21 Mean transformation factor scores and standard deviations grouped by sex ..................... 83 4 22 Mean transformation factor scores and standard deviations grouped by discipline .......... 83

PAGE 8

8 4 23 Summary moral action factor score regression table for demographic v ariables .............. 84 4 24 Mean moral action factor scores and standard deviations grouped by tenure .................... 84 4 25 Mean followership factor scores and standard deviations grouped by culture type .......... 85 4 26 Summary followership ANOVA table for culture type ....................................................... 86 4 27 Shaffer Holm procedure: followership factor scores and culture type ............................... 86 4 28 Mean service factor scores and standard deviations grou ped by culture type ................... 86 4 29 Summary service ANOVA table for culture type ................................................................ 86 4 30 Shaffer Holm procedure: service factor scores a nd culture type ........................................ 86 4 31 Mean challenge factor scores and standard deviations grouped by culture type ............... 87 4 32 Summary challenge A NOVA table for culture type ............................................................ 87 4 33 Shaffer Holm procedure: challenge factor scores and culture type .................................... 87 4 34 Mean transformation factor scores and standard deviations grouped by culture type ....... 87 4 35 Summary transformation ANOVA table for culture type ................................................... 87 4 36 Shaffer Holm procedure: transformation subscores and culture type ................................ 88 4 37 Mean moral action factor scores and standard deviations grouped by culture type .......... 88 4 38 Summary moral action ANOVA table for culture type ....................................................... 88 4 39 Shaffer Holm procedure: moral action factor scores and culture type ............................... 88 4 40 Pearson r correlation coefficients and significance levels for institutional variables and TFP factor scores ............................................................................................................. 89 4 41 Summary moral action factor scor e regression table for institutional variables ................ 89 4 42 Mean moral action factor scores and standard deviations grouped by degree offered ...... 89 4 43 Summary responsibility factor scores ANOVA table with interactions ............................. 90 4 44 Summary service factor scores ANOVA table with interactions ........................................ 91 5 1 TFP item changes to increase reliability ............................................................................. 107

PAGE 9

9 LIST OF FIGURES Figure page 2 1 Kelleys followership styles .................................................................................................. 44 2 2 Culture t ypes ........................................................................................................................... 44 4 1 Predicted total followership factor scores on age ................................................................. 77 4 2 Predicted responsibility factor scores on age ....................................................................... 79 4 3 Predicted challenge factor scores on age .............................................................................. 82 4 4 Predict ed moral action factor scores on age ......................................................................... 85 4 5 Predicted responsibility factor scores by tenure status and years working in higher education ................................................................................................................................. 90 4 6 Predicted responsibility factor scores by academic discipline and years working in higher education ..................................................................................................................... 91 4 7 Predicted service factor scores by tenure and age ................................................................ 92

PAGE 10

10 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy FOLLOWERSHIP BEHAVIORS AMONG FLORIDA COMMUNITY COLL EGE FACULTY By John Scott Smith May 2009 Chair: David S. Honeyman Major: Higher Education Administration As postsecondary institutions are confronted by the challenges of escalating accountability, shrinking budgets, and administrative downsizing, highe r education leaders are expecting more from their faculty members. In this environment, an improved understanding of faculty followership behaviors is increasingly important. We examined the relationship between followership behaviors and individual variables, organizational culture, and institutional variables among Florida community college faculty. Followership behaviors consisted of a total followership score and five dimensional subscores; responsibility, service, challenge, transformation, and moral a ction. Individual variables included faculty rank, receipt of tenure, age, sex, race, education level, discipline, and duration of employment. Institutional variables included size of the institution, size of the population served, location, and degrees of fered. An on -line questionnaire was completed by 661 faculty members from 27 of Floridas 28 community colleges. Analyses revealed significant effects for age, education level, and discipline for the responsibility dimension; sex, rank and discipline for the service dimension; age and discipline for the challenge dimension; tenure sex and discipline for the transformation dimension; tenure and age for the moral action dimension; and age sex, rank, and discipline for the total followership score. Further analyses indicated significant interaction s in the

PAGE 11

11 responsibility dimension for tenure by duration of employment and academic discipline by duration of employment; and in the service dimension for age by tenure Significant results for organizational cult ure were found for each followership dimension except responsibility. Also, statistically significant results were found among institutional variables for the moral action dimension for degree offered, with faculty from colleges that offered bachelors degrees scoring higher for moral action than faculty from institutions that did not offer bachelors degrees.

PAGE 12

12 CHAPTER 1 INTRODUCTION Followership w as traditionally neglected as a topic of interest (Lundin & Lancaster, 1990). H owever, recognition of its imp act on organizational success has recently grown (Gast, 2003). Over the past two decades, corporate hierarchies flattened workforces and eliminated mid level managers to increase efficiency (Rajan & Wulf, 2003). These transformations placed greater respo nsibility on the shoulders of followers. Reduced resourc ing and company downsizing resulted in followers pick ing up functions that were habitually performed by leaders (Hughes, Ginnett, & Curphy, 2001). Moreover, the traditional boundary between leaders an d followers disappear ed Followers and leaders found that they shared the responsibility for m aking institu tions better (Smith, 1996). Without responsible individuals who c ould be productive as both leaders and followers, organizations would not have succe ed ed These workplace changes mean t that extra attention needed to be given to followership. Researchers and practitioners began to ponder issues such as the qualities of exemplary followers and how to develop them (Pot ter, Rosenbach, & Pittman, 2001). Ev en higher education needed to more closely consider the role of followership. For many years, stakeholders calls for increased accountability had fallen on deaf ears. However, when the market g e t s its teeth into poor quality, new competitors begi n to tak e advantage of unguarded opportunities and even stodgy old-line universities begi n to increase their attention to performance. The role of followership in the effectiveness of all organizations is bec o m ing more and more prominent (Potter, et al., 2001, p. 164). However, little information on followership wa s available (Baker, 200 7 ). Few articles or books were written on the topic (Chaleff, 1995 ). While leadership was a topic in many textbooks, little recognition was given to followership (Baker 2007). Few researchers conducted empirical studies on followership (Densten & Gray, 2001) and only a small number of corporations or colleges offered courses on the topic (Kelley, 2008). Followership wa s an understudied topic in

PAGE 13

13 academic literature and an underappr eciated topic among practitioners (Bjugstad, Thach, Thompson, & Morris, 2006, p. 304). Accordingly, research on followership wa s timely, relevant, and appropriate. Statement of the Problem Despite Americas position as a world leader in higher education, the public [came] to believe quite strongly that our [postsecondary] institutions [we re ] not making the education of students a top priority (Bok, 1992, p. 15). Concerns about this and criticism about quality and effectiveness in education bec a me important matters on college campuses (Aburdene, 1993; Oldham, 2006). Colleges and universities fac ed increased calls for accountability (Lederman, 2006) Faculty were asked to relate their work and goals to the values and performance objective s of the org anization as a whole (Potter, et al., 2001 ). Exacerbating these tensions were issues related to funding, demographics, and political structures (Birnbaum, 1989). Similar issues had confronted American companies Businesses and industr ies responded with cor porate flattening, a restructuring of organizations that cut many mid -level managers and placed increased expectations and responsibility on followers. Both t he federal Commission on the Future of Higher Education and the National Conference of State Leg islators asked academe to follow suit (Field, 2006; Lederman, 2006). Alfred and Carter (2006) suggest ed that similar changes we re already taking place in higher education as colleges transitioned from bureaucratic, hierarchical institutions to flat, decen tralized, more accountable organizations. A number of author s recorded how tightening budgets led to administrative downsizing in public colleges (Diamond, 1996; Lee, 2005; Levine, 2004; Lively, 1993). The Office of Community College Research and Leadershi p noted the loss of mid level managers a cross the nations community college s (Bragg, 2004). Some academic

PAGE 14

14 leaders predicted that the middle -management tier of college administration w ould disappear altogether (Breneman 2002 ). In many ways however, high er education faculty already performed some of the leaders traditional roles. Two distinct but equally valid systems for institutional control and assertion of power exist ed in colleges and universities. One wa s the conventional administrative hierarchy s upported by the concept of legal and legitimate authority. The other wa s based on professional authority and wa s the structure by which faculty ma d e decisions on aspects over which they ha d control (Birnbaum, 1 989). Governance documents such as the Joint S tatement on Government of Colleges and Universities (American Association of University Professors, 1966) suggest ed that the proper relationship wa s one of shared authority among boards, administrators, faculty, and students. However, a system of shared g overnance is a system of mutual dependence in which we are all leaders and constituents in a process of social exchange (Birnbaum, 1989, p. 33). This type of system required faculty who we re competent, c ommitted, and responsible scholars that underst oo d t heir roles as organizational stewards what Chaleff (1995) define d as courageous follower s Chaleffs model of c ourageous followership (1995) focuse d on the roles effective followers played He believe d that exemplary followers exhibit ed the courage to s tand up for what they be lieved in, particularly when it wa s contrary to the views of others around them. The courageous follower share d the common purpose of the organization with the leader, believe d in the institutions direction, and want ed both the lea der and the organization to succeed. This put leadership and followership in a new perspective. Leaders and followers take on certain formal roles in the organization. What differentiate s them a re their roles. What unite s them i s the organizations purpos e (Davis, 2003, p. 11).

PAGE 15

15 In this model, leadership wa s not so much about using power as it wa s sharing power with followers to achieve a common purpose. A person could occupy a leadership position but find that she must also fulfill the follower role in answering to someone else her boss, a board of directors, stockholders, or another leader. Even when we have subordinates, we still have bosses. For every committee we chair, we sit as a member on several others (Kelley, 1988, p. 143). When examined in this light, the implication wa s that follower behaviors should be found at all organizational levels. The corollary wa s that followers c ould be identified at any organizational level (Dixon, 2003). Chaleff (1995) identifie d five unique behaviors in whi ch followers exhibit ed courageous followership: taking responsibility, service, challenging, participating in transformation, and taking moral action. Earlier studies of courageous followership among technology workers (Dixon, 2003) and college administrat ors (Ray, 2006) indicate d that these behaviors increase d at higher organizational levels. The importance of effective leadership in many organizations, to include higher education, was widely studied ( Baker, 2007; Gilbert & Hyde, 1988). Conversely, little empirical research was conducted on followership (Densten & Gray, 2001). Those studies that were done examined followership in blue collar workers (Alcorn, 1992; Brown & Thornborrow, 1996; Dixon, 2003), school administrators (Geist, 2001; Gouldner, 1957; Miller, 1992; Ray, 2006; Roe, 1989), nurses (Koo & Choi, 2000), college students (Tanoff & Barlow, 2002), and service members (Colangelo, 2000). However, no studies examined followership behaviors in community college faculty. As higher education flattened, college leaders look ed to faculty to assume larger institutional roles. Furthermore, shrinking budgets and rising accountability measures resulted in administrative demands for faculty to increasingly align themselves with institutional

PAGE 16

16 performance go als (Potter, at al., 2001). A greater understanding of followership could enhance organizational performance and continued learning about its members behaviors (Baker, 200 7 ). Knowing more about the roles of followers and how faculty members ascribe d to th ese roles could help community colleges better succeed in the complex and demanding environment facing todays higher education institutions. Purpose of this Study A number of authors (Alfred & Carter, 1999; Potter, et al., 2001; Williams & Ceci, 2007) ca lled for research on followership behaviors among faculty members. The purpose of this study wa s to test Chaleffs theory of c ourageous followership by relating followership to individual factors, institutional factors, and organizational culture for commu nity college faculty members. In doing so, it contribute d to the field of knowledge concerning followers and followership. The population of interest for this dissertation was Florida community college faculty. Moreover, some models of followership purpor t ed that attributions of followership vary by organizational level (Chaleff, 1995; Wortman, 1982) with greater attributions corresponding to higher organizational levels. Research by Dixon and Westbrook (2003) and Ray (2006) found support for this proposit ion. However, Steyers (2001) research did not support this conclusion. No study examined this claim with regard to community college faculty. Therefore, a purpose of this study was to investigate c ourageous followership by relating followership to faculty rank (i.e., instructor or assistant professor, associate professor, and full professors) within Florida community college scholars. Williams and Ceci (2007) found that only when faculty member s w ere promoted from associate professor s with tenure to full professor s was there a significant increase in the ir willingness to speak freely, to teach courses unpopular with one's colleagues, to publish controversial research, and to blow the whistle on ethical transgressions, behaviors that mirror

PAGE 17

17 what wa s expected from courageous follower s (Chaleff, 1995). However, concerns regarding the studys sample, the exclusion of non tenure track faculty, and methodology (Ceci, Williams, & Mueller Johnson, 2006) led to another purpose of this study; to investigate c ourageou s followership by relating followership behaviors to receipt of tenure for Florida community college faculty members. Several researchers related the importance of effective followers hip to organizational productivity (Bennis, 2008; Bjugstad, et al., 2006). Additionally, the study of organizational culture has been fueled by claims of prominent writers that culture is an essential construct in efforts to improve organizational performance (Smart & St. John, 1996, p. 219). However, while the relatio nships between culture type and organizational performance and between followership and organizational productivity have been examined, no study examined the relationship between organizational culture and followership. Therefore, a third purpose of this s tudy wa s to investigate c ourageous followership by relating followership to organizational culture for Florida community college faculty members. Finally, as research that explored followership utilizing the variables of age, sex race, education level, di scipline and number of years in the position yielded contradictory results due to methodological issues (Colangelo, 2000; Geist, 2001; Koo & Choi, 2000; Ray, 2006; Steyer, 2001), a final purpose of this study w as to examine followerships relationship to these variables with regards to community college faculty. Specifically, this study address ed the following questions using Chaleffs (1995) five identified variables of followership as an index: 1 Is faculty followership influenced by a respondent s: (a) rank ( b ) receipt of tenure, ( c ) age, (d ) sex ( e ) race, ( f ) education level, ( g ) discipline or ( h ) length of time working in higher education? 2 Is faculty followership influenced by organizational culture?

PAGE 18

18 3 Is faculty followership influenced by institutio n al : (a) size, (b) population served, (c) location, and ( d ) type (bachelor degree granting or not)? Hypotheses 1 There ar e no difference s in followership factor scores when compared by a respondent s: (a) rank (b) receipt of tenure, (c ) age, ( d ) sex ( e ) race, ( f) education level, ( g ) discipline or (h ) length of time working in higher education 2 There ar e no difference s in faculty followership factor scores when compared by organizational culture. 3 There ar e no difference s in faculty followership factor s cores when compared by institutional: ( a ) size, ( b ) population served, (c) location, and ( d ) type. Definition of Terms Challenging. Asking a leader who stray ed from the institutions purposes to be accountable (Davis, 2003). Continuing contract. A contra ct for full time employment with the implication that i t w ould be renewed each year as long as the position wa s needed and the employee perform ed satisfactorily Florida community colleges used this term in lieu of tenure. Courageous followership. Five un ique behaviors that Chaleff (1995) identified as: taking responsibility, service, challenging, participating in transformation, and taking moral action. Follower. A person who recognize d the leader as the key source of direction, regardless of how much formal authority the leader actually ha d over the person (Yukl, 2002). The terms followers, subordinates, workers, constituents, and employees wer e considered synonymous. Leader. An individual who deliberately influence d others to facilitate organizational activities and relationships (Yukl, 2002). Organizational culture. A pattern of shared basic assumptions that wa s learned by a group, worked well enough to be considered valid, and wa s taught to new group members (Schein, 2004).

PAGE 19

19 Participating in transfor mation. Standing by the leader and organization during change a s well as identifying and changing ones own behaviors that enable d dysfunctional actions. Service. Assuming new and additional responsibilities to unburden the leader and benefit the organiza tion (Dixon, 2008). Taking moral action. Leaving an organization for reasons of personal growth, exhaustion, when the organization would benefit from their departure, or as a response to immoral or illegal actions. When faced with unethical activities, fo llowers also ha d the options of disobeying or whistle -blowing. Taking responsibility. Show i n g ownership and initiative for oneself and the organization. Significance of the Study Though a growing number of scholars recognized the need for additional studies on followership, very little empirical research was conducted. This study provide d some recompense to this deficiency and advance d knowledge in this field. Second, important changes have occurred in higher education. Just as flattening increased expe ctations for followers in the business world, more was also anticipated of community college faculty members as their organizations under went restructuring (Alfred & Carter, 2006; Bragg, 2004; Bre n eman 2002 ). This was evidenced by calls for faculty to bet ter align their outcomes to the institutions values and goals (Potter, et al., 2001). A better knowledge of followership within community college faculty members helped demonstrate the capability of faculty to meet these calls Third, community college fa culty represented broad assortments of individuals of differing backgrounds. It wa s important to understand how this diversity affected followership. Fourth, studies have examined the relationships between organizational culture and performance and perform ance and followership. However, none have examined the relationship of organizational culture and followership. Finally, while the scope of this study did not specifically incorporate

PAGE 20

20 administrators, an improved understanding of community college faculty a s followers can help improve the interaction between these two groups. T his research increase d the understanding of faculty followership behaviors in Florida community colleges. Findings of this study advance d the body of knowledge on followership by test ing constructs of Chaleffs theory of c ourageous followership (1995) within the structure of community colleges. Additionally, this study helped define how faculty members view ed themselves as followers. This may help community college administrators devel op better work ing relationships with their faculty members and advance programmatic and institutional change agendas. R eports at both the national (Oldham, 2006) and state (Pound, 2006) levels have called on colleges to increase institutional accountability measures. With ever increasing demands of accountability in education and a greater emphasis on the creation and fostering of school culture, institutions need to generate greater attention on relational interventions in order to achieve more significan t and lasting change (Steyer, 2001, p. 6). Limitations 1 The population from which the sample was drawn wa s limited to Florida community college faculty members. R esults from this study should not be generalized beyond this population. 2 Participants needed to voluntarily agree to participate in this study. There may have be en unknown aspects of this study that attract ed certain types of participants while deterring other types of people from participating. 3 This study relied on self report data, which can b e distorted in a variety of ways, such as peoples desire to give socially approved information about themselves or the misunderstanding of questionnaire items. Though the instruments were pre tested before their use, distortions like these may have produc ed inaccurate results. 4 The sampling frame for this study was developed using information from Floridas community college websites. These websites do not list all faculty members for all institutions. In particular, part time faculty appeared to be underr epresented on the sites. 5 This study used an ex post facto design and looked only at the degree of association between multiple variables The problems of reverse causality and spurious correlations

PAGE 21

21 may have exist ed Therefore, the findings of this study di d not provide evidence of causal relationships, only information about the degree and shape of the relationship between the variables of interest.

PAGE 22

22 CHAPTER 2 REVIEW OF RELATED LI TERATURE The purpose of this chapter wa s to review the literature on followe rship that pertain ed to this study. Specifically, the chapter i s divided into sections relating to; (a) the background of followership, (b) characteristics of exemplary followers, (c) types of followers, (d) courageous followership, (e) organizational culture, and (f) measures of followership. The overview of relevant literature provided a rationale for the theoretical model, measurement instrument, and variables used in this study. Background of Followership In 1933, management scholar Mary Parker Follett called for more attention to followership. She suggested that leaders were responsible for teaching workers how to be followers, for engaging them in self -management, and for helping them develop an emotional commitment to the organization. In 1949, the S aturday Evening Post issued its own appeal for speakers and writers to address followership (Attridge, 1949). In 2003, followership made the Harvard Business Reviews top five list of breakthrough ideas (Gast, 2003). However, little information on followe rship was found in the literature. Though numerous articles were written on leadership, few journal articles were produced on followers (Lundin & Lancaster, 1990). Discounting those of a spiritual or political nature, only two books were written on the sub ject (Landino, 2006). Gilbert and Hyde note d that few organizational theory or behavior texts mention followership and there we re even fewer empirical studies in the literature about followership. Most of the articles where the term followership has been introduced or explored have been more normative and intuitively derived (1988, p. 963). Additionally, though many research articles were produced on leadership, few were created that

PAGE 23

23 address ed followership (Bjugstad, et al., 2006). Most research concerning followership w as limited to followers perceptions of, or as objects of, leadership (Densten & Gray, 2001). Well known studies such as Milgrams (1963) obedience to authority and Zimbardos (1971) Stanford Prison examined the tendency of some individuals to surrender to authority figures in experimental settings However, missing were investigators attempts to understand followership in normal settings (Vecchio, 1987). Only recently have researchers considered followership in industry, medical facilities the military and education Characteristics of Exemplary Followers As followership wa s still a relatively new concept, most authors sought to build a foundation for it by identifying characteristics common among exemplary followers (Gilbert & Hyde, 1988). This approach generally dominate d the reviewed followership literature. The greatest benefit to this approach w as that characteristics related to successful followership have been identified (Lussier & Achua, 2004). However, the possession of certain traits does not guarantee effectiveness, nor does their absence proscribe it (Bensimon, Neumann, & Birnbaum, 1989, p. 8). This trait approach to followership recognize d that there we re no universal traits which caused a person to be an exemplary follower nor we re all traits applicable in all situations. Examining the works of writers on exemplary followership, seven attributes appear ed most often in the literature. These we re: candor, competence, commitment, cooperation, responsibility, initiative, and f lexibility. Job competency, organizational commitment, and the ability to work well with others were characteristics that firms always sought Conversely initiative and candor were relatively new additions (Campbell, 2000). Each of these seven qualities w as explored in further detail.

PAGE 24

24 Candor The U.S. Army define d candor as honesty and fidelity to the truth (1986, p. 3) and list ed it as one of the institutions four individual core values. Bennis (2008) believe d that a followers most important character istic wa s a willingness to tell the truth. Others include d it as a critical follower trait (Alcorn, 1992; Banutu Gomez, 2004; Solovy, 2005). Gasaway (2006) believe d that candor wa s one of the qualities that differentiate d followers from employees. Dat a quoted by Brown (1995) suggested that 70% of subordinates would n o t protest even when they believed the leader was making a mistake. Research on higher education indicated that only when a faculty member was promoted from associate professor with tenure to full professor was there a significant increase in the willingness to speak freely, to teach courses unpopular with ones colleagues, to publish controversial research, and to blow the whistle on ethical transgressions (Williams & Ceci, 2007, p. 16). These figures suggest ed that candor may be hard to find in many organizations. However, researchers consistently stated that exemplary followers told their leaders the truth even when it gave voice to unpopular opinions (Goffee & Jones, 2006). They unders t oo d the importance of speaking out. Model followers understood that silence could cost their organizations dearly (Bennis, 2008). Competence Competence was define d as having necessary and sufficient talents (Mish, 1999). Gasaway (2006) suggested that exceptional followers differ ed from employees in their jobrelevant knowledge. Gilbert and Hyde (1988) included technical competence as one of their eight key dimensions they associated with followership. Blackshear (2002) named personal mastery among her cr itical characteristics of exemplary followership. Smith wrote that competence is based on a myriad of observations Success is probably the easiest to observe. Expertise in the groups field of operation is another dimension of competence. Expe rtise is the perception of skill s, craftsmanship, or artistry [ Expertise is often

PAGE 25

25 assumed by ] credentials or reputation without actually observing or experiencing them. Capability, a third aspect of competence, is the execution of expertise (1997, p. 3). Of the four essential qualities that Kelley (1988) proposed effective followers share, three related to competency. First, Kelley wrote that effective followers buil t their competence and focus ed their efforts for maximum impact They str o ve to reach higher levels of performance and expand themselves. Secondly, effective followers manage d themselves well. They were able to set goals and decide on roles appropriate to the larger context (Bjugstad, et al., 2006). Thirdly, effective followers demonstrate d independent and critical thinking skills. Commitment Kelley (1992) and Solovy (2005) both defined commitment as working beyond the expected to produce exemplary results. Banutu Gomez (2004) and Gasaway (2006) describe d commitment as organizational loyal ty, passion, persistence, and dedication. Etzioni (1986) saw commitment as a positive form of participation. He used the word involvement to refer to the orientation of an actor in terms of intensity and direction. He called those individuals with low or ganizational involvement alienative while those with a high intensity orientation were labeled as committed. Argyris (1957) believed individual s demonstrated commitment through their organizationally relevant act ion s. Gilbert and Hyde (1988) included commitment as one of the eight key dimensions they found associated with followership. For them, commitment consisted of loyalty, ambition, and a can do attitude. Smith (1997) noted that leaders functioned best with followers who we re loyally committed t o an organization. Loyal followers helped their leaders move the organization in the desired direction and they supported their leaders when confronted by obstacles or when redirection was necessary.

PAGE 26

26 Blackshear (2002) posited that followership was built o n: a belief in the organizations mission, vision, and purpose; a willingness to subjugate personal interests for the greater good; loyalty; and unity of focus. These characteristics helped followers develop and sustain the ir optimal organizational perform ance Lundin and Lancaster (1990) stated that followers must be enthusiastic about what they do to the point that roadblocks and repetition dont deter them from achieving their objectives. They need to feel a strong level of commitment, both to the organization and to their own work. And, finally, they must be highly responsible individuals who are willing to perform under stressful circumstances, motivated by a sense of a job well done (p. 18). Cooperation Smith (1997) included cooperation as a charact eristic that promote d leader -follower functioning. Cooperation allows the leader to focus on the group as a unit rather than devoting time to directing individuals and their independent behaviors. Cooperation also suggests a level of conformity that incre ases the efficiency and the power of the group (Smith, 1997, p. 3). Alcorn (1992) studied workers in Wyomings Lower Valley Power and Light to identify critical follower traits. Cooperation was one behavior he associated with high levels of performance p articipation, and job satisfaction. Cook (1998) believed that model followers were team players and cooperated with their leaders and coworkers in ways that benefit ted the organization. Gilbert and Hyde (1988) included working well with others as a key d imension of followership. Responsibility Several authors include d responsibility among the traits of model followers (Lundin & Lancaster, 1990; Banutu Gomez, 2004). Chaleff (1995) believed that followers must accept responsibility for themselves, their l eaders, their organization, and their customers. Exemplary followers kn e w that leaders stop leading when they grow tired of leading, so smart followers reinforce their best leaders by showing them that their leadership is not only needed but appreciated. By doing so, however, followers must accept responsibility for their leaders

PAGE 27

27 actions (Marino, 1999, p. 20). They also accept ed responsibility for the organization and those it serve d (Brown, 1995). Exemplary followers understood that their actions direct ly impacted unit performance (Vecchio, 1987). Dixon and Westbrook (2003) stated followers accepted responsibility for themselves and the organization. They assumed additional responsibilities to ease the leader s burden and serve the organization. Crocket t (1981) wrote that followers must assume responsibility for developing an effective understanding of the job, the boss, and themselves. They t ook responsibility for: knowing the mission, priorities, expectations, and the skills required; developing succes sful and effective leader -follower relationships; and acquiring self awareness, managing their feelings, and confronting their insecurities. Initiative and Flexibility A number of authors included initiative and flexibility in their critical traits of ex emplary followers (Lundin & Lancaster, 1990; Landino, 2006). Alcorn (1992) included flexibility, initiative, and problem -solving as critical follower traits. He believed that workers with ability and a willingness to act on their own could accomplish far m ore than their traditional job responsibilities. Blackshear (2002) identified initiative, problem -solving, and adaptability as critical exemplary followership characteristics. She believed that ideal followers displayed an internal locus of control and a w illingness to act. Goffee and Jones (2006) wrote that model followers developed an understanding of what was needed of them in different situations and adapted their behaviors appropriately. By doing so, these followers effectively complement ed the leader and promote d balance within the organization. Other Traits Authors have identified other traits of exemplary followers. However, these qualities tend to be named by only one writer. Cook (1998) believed th e best followers were self -motivated.

PAGE 28

28 Smith (1996) noted that model followers were dependable. Lundin and Lancaster (1990) stated that effective followers underst ood how their work contribute d to the big picture. Blackshear (2002) named persistence and optimism among her critical followership characteris tics. Kelley (1988) stated followers exhibited enthusiasm, intelligence, and self reliance. Gasaway (2006) believed that dynamic followers separated themselves from others in the qualities of ambition and energy. Gilbert and Hyde (1988) included proper com portment (dress, grammar, speech, and manners) among their qualities associated with ideal followers. Types of Followers Many of the characteristics described in the preceding paragraphs we re personality traits (candor, dependability, initiative, flexibility, etc.). Guilford (1959) defined a trait as any relatively enduring way in which individual s differ from one another. Research has suggested that personality traits we re heritable (Jang, McRae, Angleitner, Rieman, & Livesley, 1998), unaffected by external influences (Asendorpf & Wilpers, 1998), stable throughout a persons lifetime (McRae & Costa, 2003), and influence d how people react ed and respond ed to their environment (Olver & Mooradian, 2003). Traits serve d as the foundation upon which an individual s behavior is based (Hemsley, 2001). They also serve d as the basis for personality typology, the next step in the development of followership research. Personality types we re combinations of two or more traits. The fundamental difference between traits a nd types wa s that traits we re conceptualized as varying along a continuum whereas types we re usually thought of as discrete categories of trait combinations (Hughes, Ginnett, & Curphy, 2001). Though the common characteristic approach to followership tended to overshadow the followership literature (Gilbert & Hyde, 1988), some authors elected to explore the topic using personality typology. Among the follower type approaches, Kelleys (1992) model of e xemplary f ollowership dominate d the field (Densten & Gray 2001).

PAGE 29

29 Kelleys Model of Followership Kelley (1992) categorized followers by examining their behavior along the dimensions of thinking and acting. The thinking dimension ranged from uncritical, dependent thinking to critical, independent thinking. Critical thinking involves going beyond collecting information or observing activities passively. It implies an active mental debate with things or events we could otherwise process at face value (Latour & Rast, 2004, p. 104). Followers who were independent, critical thinkers considered the impact of their actions, were willing to be creative and innovative, and offered criticism. Dependent, uncritical thinkers only did what they were told and unquestionably accepted the leaders thinking. The second dimensio n, acting, determine d the followers sense of involvement. This dimension ranged from passive to active styles of participation. A follower engaged actively and comprehensively brings to mind an image of someone leaning forward into the situation at han d. This posture enables the person and those he or she affects to be in a position to anticipate requirements and plan accordingly (Latour & Rast, 2004, p. 104). An active follower took initiative in decision making while a passive followers contribution was limited to being told what to do. Based on this classification of follower behavior, Kelley (1992) developed a typology of five follower types; alienated follower, conformist follower, passive follower, effective follower, and pragmatic survivor (see figure 1). These typologies helped clarify a work er s contribution t o organizational success (Densten & Gray, 2001). The most p assive followers were sheep These were individuals who displayed uncritical thinking lack ed initiative and avoided responsib ility. They did only those tasks explicitly assigned to them and stop ped (Kelley, 1988). Conformist followers were yes people. These followers were more active but equally uncritical in their thinking. They depende d on a leader to inspir e them but were e xceedingly deferential in their

PAGE 30

30 activities (Kelley, 1988). Alienated followers were critical in their thinking but passive in executing their role. Their performance tended to be one of disgruntled acquiescence Survivors fell in the middle. They were adep t at withstanding change. Effective followers thought for themselves and energetically carr ied out their responsibilities. Because they we re self -starters, and problem solvers, they usually earned good ratings from supervisors (Kelley, 1988) Research Stud ies Few empirical studies wer e found in the literature that utilize d Kelleys theory of e xemplary followership to pragmatically investigate followership in specific populations. Only three studies were found in professional journals. Koo and Choi (2000) i nvestigated the relationship between followership, job satisfaction, and organizational commitment in Korean clinical nurses. They found significant differences among clinical nurses in their perceptions of followership according to demographic characteris tics such as sex age, education, position, and career field. They also found that followership was significantly related to job satisfaction and organizational commitment. Tanoff and Barlow (2002) examined the relationship of leadership and followership. They used Kelleys (1992) followership model along with the Five Factor Model (FFM) of trait personality (McRae & Costa, 2003). These authors utilized stepwise regression and found the FFMs three factors of conscientious, dynamic, and hardy combined to account for 31% of the explained variance in the active engagement area and 18% in the independent thinking area [while ] the collegial factor did not contribute in either model (Tanoff & Barlow, 2002, p. 161). Since achievement orientation is often considered a principle facet of the conscientious factor, the authors concluded that the relationship between leadership and followership can be found in the correlation between the achievement -oriented trait and the active engagement scale and the

PAGE 31

31 indepe ndent thinking scale Possibly the parallel between leadership and followersh ip is the convergence on doing ( behavior ) with a purpose ( goals ) (Tanoff & Barlow, 2002, p. 163). Brown and Thornborrow (1996) examined the relationship between follower st yle and organizational culture. These authors concluded that organizations tend to get the followers they deserve this means that not only is a particular leadership style favored by an organizations culture, but so is a specific type of followersh ip (Brown & Thornborrow, 1996, p. 10). T he authors also found that people are not born to be followers, but become followers as a result of other factors [This suggests that there is ] considerable support for the idea that followers can be taught h ow to be more effective followers (Brown & Thornborrow, 1996, p. 8). Though these three studies presented interesting findings demonstrating the importance of followership in organizations additional empirical studies of followers and followership we re largely absent from scholarly journals. However, three other studies using Kelleys model were found among doctoral dissertations. Colangelo (2000) examined the relation of immediate supervisors leadership styles to followership among junior enlisted airm en attending a basic leadership school. Participants were assessed on followership style and leadership effectiveness and adaptability. Results indicated that supervisors leadership styles were correlated to followership styles. Colangelo concluded that l eaders high in task and relationship behaviors promoted effective followership. Also, followers accredited these leaders with higher trust, commitment, and job satisfaction. T he author found a significant relationship between the active engagement subscore in Kelleys model and age and education level but not race and sex whereas the critical thinking subscore only showed a significant relationship with education level but not age, race, or sex However, no significant relationships were established betwe en the overall followership score and age, race, sex or education level.

PAGE 32

32 Geists (2001) study examined the followership abilities of NCAA Division II athletic directors. Middle managers' perceptions of the followership abilities of athletic directors pro vided one basis for comparison, while the variables of sex ethnic background, educational level, and experience were also utilized for comparison. Results indicated that both transformational and transactional leadership styles had a significant influence upon perceived followership abilities in athletic directors. However, unlike Koo and Choi (2000), the variables of sex educational level, and experience had no significant effect upon the followership levels of athletic directors. Steyer (2001) adapted K elleys (1992) Followership Questionnaire to education and renamed it the Teacher Sentiment Inventory (TSI ). She then used it to measure followership in 291 classroom teachers in Ohio public schools. She concluded that: the TSI was a reliable measure of fo llowership, elementary school teachers perceive independent thinking to be more important than middle school teachers, older teachers perceive independent thinking to be more important than younger teachers, elementary school teachers perceive active engagement to be more important than high school teachers, and female teachers perceive active engagement to be more important than male teachers. Steyer found no significant differences due to race or years of experience. S h e also determined that administrator s must recognize that teachers perceive themselves as being effective followers who should be allowed to think independently and become actively involved within the organization. Koo and Choi (2000), Colangelo (2000), Geist (2001) and Steyer (2001) all ex amined followerships relationship to multiple demographic variables sex age, education, experience, position, and career field with differing results. In addition, whereas Steyer concluded that the TSI a n adapted version of Kelleys (1992) Followers hip Questionnaire was a reliable measure

PAGE 33

33 of followership, both Colangelo (2000) and Geist (2001) raised concerns about the validity of Kelleys measure concluding that factor analysis (Colangelo, 2000) and item to item correlations (Geist, 2001) suggest ed that the scale wa s not solely measuring the variables of active engagement and critical thinking. Colangelos factor analysis identified four dimensions, which he entitled active engagement, critical thinking, passion, and team mindedness. Geist found tha t six of the twenty items in Kelleys questionnaire did not load highest on their own dimensions and reliability coefficients for Kelleys two dimensions were often below .70. Though several other authors offer ed followership models based on two -dimensiona l personality typology (Blackshear, 2002; Hersey, Blanchard, & Johnson, 2007; Potter, Rosenbach, & Pittman, 1996; Tagliere, 1972; Vecchio, 1987; Zaleznik, 1965), no empirical research was found that examine d these different models. Recently, however, researchers (Dixon, 2003; Ray, 2006) have quantitatively examine d the roles that effective followers play using Chaleffs (1995) model of c ourageous followers. Courageous Followership Researchers have discovered that many of the q ualities that make effective l eaders are paradoxically the same as those found in effective follower s What di fferentiates the two are the role s they play ( Davis, 2003; Kelley, 1988; Taylor & Rosenbach, 1996). However, most researchers examine d followers traits, behaviors, or personal ity types. Chaleff (1995) presented the only model of followership focused on the roles effective followers play. This wa s his theory of c ourageous follower ship. Chaleff noted that the tradition al leader -subordinate paradigm wa s based on power: Leaders h e ld the power to reward and punish through bonuses, choice assignments, promotions, firings, and negative evaluations. This led to followers who wanted to avoid upsetting their leaders. Followers who lacked courage abandon ed their unique perspectives and healthy

PAGE 34

34 dissension, which are at the heart of the creative process and innovation (Chaleff, 1995, p. 5). Chaleff believed that followers must exhibit the courage to stand up for what they believe in, particularly when it is contrary to what others around t hem believe. Exemplary f ollowers persevere d in what they kne w wa s right ma king it difficult for others to do wrong (Lee, 2006). Like leaders, followers serve d the organizations higher purpose and wer e not swayed by personal considerations (Goffee & Jones 2006). The courageous follower share d the common purpose with the leader, believe d in the direction of the organization, and want ed both the leader and organization to succeed. What differentiate d the two were their roles (Davis, 2003). Chaleff (1995) id entified five dimensions in which followers exhibit ed courageous followership. These we re the courage to: assume responsibility, serve, challenge, participate in transformation, and take moral action. A discussion of e ach of these five roles follows Cour age to Assume Responsibility In this role, courageous followers demonstrate d a sense of ownership for themselves and their organizations. Precisely because of the dangers of unconscious collaboration, they need to take some conscious responsibility for t he leaders they have helped create (Gast, 2003, p. 94). Followers d id not wait to be told what to do; they t ook initiative in their work and for their own self -development. Exemplary followers recognized that s e lf development wa s a growth process whose ke y was obtaining an accurate picture of themselves (U.S. Army, 1993). They took full advantage of all assessment opportunities ; self assessment, formal assessment by superiors, and assessments made by professional schools (U.S. Army, 1993). Courage to Serv e Courageous followers were prepared to assume additional responsibilities in service to the organization (Davis, 2003). They recognize d that they we re members of a team and we re willing to perform less favorable tasks to benefit that team. They took on n ew and additional tasks to

PAGE 35

35 unburden the leader (Dixon & Westbrook, 2003) They relay ed substantiated bad news to the leader and present ed workable options for handling tough situations. Courageous followers st oo d up for their leaders and the difficult deci sions they ma d e (Chaleff, 1995) Courage to Challenge Courageous followers recognize d that, once a leader had made her decision, they ha d the responsibility to implement the policies to the best of their abilities. However, the courageous follower wa s no t a subordinate. The follower stands up for the leader when support is warranted, but is not afraid to challenge the leader who strays from the institutions purposes (Davis, 2003, p. 12). The courageous follower underst oo d that there wa s a time and plac e to dissent just as there wa s a time and place to execute the leaders decision. We have the right to challenge policies in the policy -making process; we do not have the right to sabotage them in the implementation phase Those who sabotage their le aders efforts are no longer followers; they are opponents (Chaleff, 1995, p. 95). Courage to Participate in Transformation Courageous followers recognize d that they ha d a role in transformation. Exemplary followers stay ed with the leader and group duri ng the challenges of real change. They also recognize d that, in order to help a leader transform, they needed to examine their own roles and identify behaviors that enabled or colluded with dysfunctional actions. Courageous followers model ed appropriate be haviors, to include the openness to change, the risk taking inherent for change, persistence, and the empathy in understanding the difficulty of change (Chaleff, 1995) Courage to Take Moral Action Exemplary followers recognized that s elf or organizational growth may require them to leave a leader (Chaleff, 1995). They prepared to move on when it wa s the appropriate action to take. Followers saw that they may need to leave for reasons of personal exhaustion. At other

PAGE 36

36 times, they recognized that the organi zation would benefit from their departure Bringing in new followers may bring new ideas that help invigorate the leader or organization. Courageous followers were prepared to take action when organizational behaviors we re immoral or illegal. Followers h ad the options of disobeying, whistle -blowing, or leaving ( Kelley 1992) Research Studies As research on courageous followership was in its infancy only two empirical studies were found. Dixon and Westbrook (2003) utilized Chaleffs model of courageous f ollowership to examine engineering and technology workers in governmental agencies and industry. They sought to validate Chaleffs assertion that followership should be recognizable at all organizational levels. Results indicated that followership was evid ent across organizations and that statistically significant differences existed in self attributions of followership as a function of organizational level. The researchers concluded that: followership was measurable; followers were discernible within all l evels of technology-based organizations; and attributions of followership were related to organization level with increasing measures of followership at higher levels in the organization. This study implied that managers we re both leaders and followers and that follower roles c ould be described according to organization level. Ray (2006) also utilized Chaleffs model of c ourageous f ollowership to determine if followership behaviors varied by hierarchical level among North Carolina community college adminis trators. Utilizing the followership measure developed by Dixon (2003), Ray also found that mean responses were higher for all courageous follower behaviors as administrative responsibility increased. In addition, she found evidence that followership behavi ors increased with educational level, age, and work experience. Ray found that this supported the belief that leadership training was needed for all employees, not just currently identified leaders.

PAGE 37

37 Organizational Culture A number of researchers have stre ssed the relationship between effective followers and organization productivity (Buhler, 1993). Kelley wrote After controlling all the other variables the leaders effect on organizational success is only 10 to 20 percent Followership is the real people factor in the other 80 to 90 percent that makes for great success (1992, pp. 20 21). Kelleys assertions found support among other researchers (Bennis, 2008; Bjugstad, et al., 2006; Meindl, Ehrlich, & Dukerich, 1985; Pfeffer, 1977; Vecchio, 1987) who presented data substantiating the idea that leadership is less important to an institutions fate than traditional views would have us believe. Followership wa s shown to be an instrumental aspect of an organizations productivity. In the same vein, the study of organizational culture has been fueled by claims of prominent writers that culture is an essential construct in efforts to improve organizational performance (Smart & St. John, 1996, p. 219). Schein defined culture as a pattern of shar ed basic assumptions that was learned by a group as it solved its problems of external adaptation and internal integration, that has worked well enough to be considered valid and, therefore, to be taught to new members as the correct way to perceive, think and feel in relation to those problems (2004, p. 17). Cameron and Ettington (1988) proposed that cultures could be distinguished by examining their emphasis, orientation, and activity focus along a horizontal dimension and their flexibility, control, and spontaneity along the vertical dimension. From th is framework emerged four ideal culture types that we re consi stent with the organizational culture literature and with the ways that scholars view postsecondary institutions (Smart & St. John, 1996) (see figure 2). In Smarts and St. Johns (1996) model, the clan culture wa s characterized as having high flexibility, individuality, and spontaneity, as well as an internal emphasis, a short term time frame, and a focus on smoothing activities. Bureaucrat ic cultures also ha d an internal

PAGE 38

38 emphasis, short -term orientation, and a focus on smoothing activities. However, they differ ed in the emphasis on control, stability, and predictability. The adhocracy culture emphasized flexibility, individuality, and sp ontaneity but wa s also characterized by its external positioning, a long -term orientation, and achievement oriented activities. Like the adhocracy, the market culture wa s distinguish e d by its external positioning, a long -term orientation, and achievement oriented activities but it also value d control, stability, and predictability like the bureaucracy. Other researchers have found that community colleges tend ed to cluster around four types of cultures. Levin (1997) identified the dominant cultures as: tra ditional, in which the college focused on the development of its students ; service in which the school protect ed and promot ed the well -being of its students; hierarchical in which the college administration demonstrate d strong leadership in order to upho ld its ideal s of excellence; and business in which the community college adopt ed business practices and values for its operations. Bergquist (1992) also identified four cultures within higher education; collegial, managerial, developmental, and negotiatin g. The collegial culture wa s characterized by an emphasis on the generation, dissemination, and interpretation of knowledge in support of the institutions mission. The managerial culture wa s describ ed by clearly defined organizational structures, goals, a nd accountability procedures for achieving those goals. Bergquist (1992) distinguished the developmental culture as placing high value on the professional and personal growth of institutional members, to include students. The negotiating culture wa s typifi ed by the distribution of organizational resources according to equitable and democratic practices. Within this culture, bargaining and confrontation among constituents wa s expected. Bergquists system of classification was similar to Birnbaums (1988), w hich labeled higher education cultures as collegial, bureaucratic, anarchical, and political. While Birnbaums

PAGE 39

39 collegial, bureaucratic, and political cultures parallel ed Bergquists collegial, managerial, and negotiating cultures, Birnbaums anarchical cul ture wa s characterized by its attempts to gain consensus in an environment that may be irrational, unorganized, or lack a clear structure. In their study of colleges and universities, Smart and St. John (1996) found significant relationships between strong culture types and organizational effectiveness. They found that a three tier hierarchy exist ed among the four culture types in terms of their performance. The clan and adhocracy cultures exhibited the highest performance, m arket cultures fell in the middle, and bureaucratic cultures demonstrated the lowest performance (Smart & St. John, 1996). With the relationships between culture type and organizational performance and between organizational productivity and followership established, additional research needed to be undertaken to examine the relationship between organizational culture and followership. Measur es of Followership In their 1984 article, Nolan and Harty call ed for the development of followership measures similar to those available for the stu dy of leadership. Though many years passed, few measures of followership became available (Colangelo, 2000). Only five scales were described in the literature: The Followership Questionnaire (Kelley, 1992), the Teacher Sentiment Inventory (Steyer, 2001), t he Follower Maturity Index (Moore, 1976), the Performance and Relationship Questionnaire (Rosenbach, Pittman, & Potter, 1997), and The Followership Profile (Dixon, 2003). Of these five scales, only The Followership Profile provide d documented and acceptabl e levels of reliability and validity. Followership Questionnaire Kelleys (1992) Followership Questionnaire wa s a 20 -item Likert response scale and the most frequently used scale for measuring followership (Colangelo, 2000). However, no data concerning t he reliability and validity of the instrument wa s available from the developer. In

PAGE 40

40 addition, researchers who utilized this scale raised issues concerning its validity and reliability (Colangelo, 2000; Koo & Choi, 2000; Geist, 2001). Kelleys followership i nstrument was developed to measure two dimensions of behavior independent, critical thinking and degree of active engagement in task that contribute d to effective followership. Factor analysis by Colangelo (2000) suggest ed that it actually measure d fou r followership dimensions, which he named active engagement; critical, independent thinking; passion; and team mindedness. In Geists (2001) study, Cronbachs alpha ranged from .51 to .78. However, item to item correlations by Geist (2001) raised concerns because a number of items did not load highest on their own dimensions. In addition, Koos and Chois (2000) article raised issues with utilizing Kelleys questionnaire to produce a single followership score for use in statistical analysis. Teacher Sent iment Inventory Steyers (2001) Teacher Sentiment Inventory (TSI) wa s a version of Kelleys (1992) Followership Questionnaire adapted for use in elementary and secondary education. Steyer modified Kelleys original questions so that they more directly addressed the work of school administrators (Mertler, Steyer, & Petersen, 1997). Like The Followership Questionnaire, the TSI wa s a 20 -item, five -point Likert -response scale. Though the authors state d that internal reliability was as sessed using the data gat hered [ from ] pilot studies which resulted in further revisions of survey question wording and placement (Mertler, et al., 1997, p. 10), other than the calculation of Cronbach coefficient alpha (.84), no evidence of validity or reliability wa s availa ble for the TSI. Like Colangelo (2000), Steyer (2001) also raised concerns that Kelleys (1992) original questions actually measure d more than two followership dimensions. However, she did not submit the TSI to factor analysis to verify this issue.

PAGE 41

41 Follow er Maturity Index The Follower Maturity Index (FMI) wa s based on the situational leadership work of Hersey, Blanchard, and Johnson (2007). The instrument wa s used in observations of verbal and nonverbal behaviors in task groups. The scale was intended to provide a method for quantifying follower behavior in terms of the dimensions of maturity ; achievement motivation, willingness and ability to take responsibility, task relevant education or experience, activity level, dependency, behavior variety, interest perspective, position, and awareness (Moore, 1976). Moore tested the FMI using 788 people who were participants in 32 leadership seminars conducted by the researcher Analysis of data was conducted by discussions following group tasks and by post hoc re view of videotapes [ However ], no statistical tests were run because the purpose of the study was to provide a field test of the instruments utility (Moore, 1976, p. 207). Therefore, no information wa s available regarding the scales reliability or validity in measuring follower behaviors Performance and Relationship Questionnaire The Performance and Relationship Questionnaire (PRQ) was a 40 item questionnaire intended to provide an assessment of the users follower style and aptitude across eight dimensions (Rosenbach, et al., 1997). Believing that successful organizations were based on productive relationship between followers and leaders, the PRQ was built on two dimensions, the relationship initiative and the performance initiative. However, no information on the scales validity and reliability was available from the publisher. Additionally, use of the measure for research could be quite costly as Rosenbach and Associates administer ed the scale and provide d interpretive reports at a cost of $15 per assessment and a minimum charge of $90 for six assessments.

PAGE 42

42 The Followership Profile The final and most recently available measure of followership wa s The Followership Profile (TFP). For his doctoral dissertation, Dixon (2003) developed the TFP quest ionnaire to identify and measure the five behavioral categories of Chaleffs (1995) theory of c ourageous followership. Originally developed as a 56-item measure, Dixon further refined the measure and a 2 0 item abbreviated version wa s available. The TFP ha d the most evidence among followership measures for validity and reliability. A more detailed discussion of the TFPs validity and reliability is found in the instrumentation section of chapter 3. Summary This literature review confirmed the relative pauci ty of followership -specific research. Additional studies we re needed to not only develop a clearer understanding of the concept of followership, but also to test current theory and encourage further discussion of the related follower characteristics. Though the role of followership in education was considered in the dissertations of Geist (2001), Steyer (2001), and Ray (2006), none examined followership from the perspective of community college faculty. Research on courageous followership behaviors among fa culty members wa s needed to address the concerns raised by Alfred and Carter (1999); Potter, Rosenbach, and Pittman (2001); and Williams and Ceci (2007). In addition, the empirical studies described in the previous paragraph fall into what Densten and Gra y (2001) called previously established ruts t he research was generally limited to Kelleys (1992) original conceptualizations of followership or the focus was not on followers but on the followers relationships with their leaders. The exceptions were Dixons and Westbrooks (2003) and Rays (2006) exploration of Chaleffs concept of courageous followership. However, Dixon and Westbrook focused their research on management in the

PAGE 43

43 engineering and technology industry while Ray focused on administrators. P rior to this study, c ourageous followership w as not examined within the context of community college faculty. The idea that followership increase d with organizational level was first suggested by Chaleff (1995) and later found support in the work of Dixon and Westbrook (2003) and Ray (2006). However, Steyers (2001) research suggested differently. This disparity and the lack of information relating this idea to community college faculty suggest ed an area in which empirical stud y was needed. In addition, al l research that was based on Kelleys model of Exemplary Followership (Brown & Thornborrow, 1996; Colangelo, 2000; Geist, 2001; Koo & Choi, 2000; Steyer, 2001; Tanoff & Barlow, 2002) used Kelleys Followership Questionnaire (or an adaptation thereof) to collect data. As noted above, several researchers ( Colangelo, 2000; Geist, 2001; Koo & Choi, 2000) expressed concerns with the scales validity and reliability Followership needed to be examined using a measure shown to be both valid and reliable. Also, whi le empirical studies established the relationships between culture type and organizational performance and between organizational productivity and followership, none ha d examined the relationship between organizational culture and followership. Additional research wa s needed to address this shortcoming. Finally, Koo and Choi (2000), Colangelo (2000), Geist (2001), and Steyer (2001) all examined followerships relationship to multiple demographic variables sex age, education, experience, position, and ca reer field with differing results. It wa s apparent that additional evidence -based research on followership wa s needed; that an examination of followership as it pertains to community college faculty would contribute to the field; and that Chaleffs model of courageous followership and Dixons (2003) TFP offer ed the most valid approach to undertake this study.

PAGE 44

44 Figure 2 1 Kelleys f ollowership s tyles (Kelley, 1992) Figure 2 2 Culture t ypes (Smart & St. John, 1996) Form: Adhocracy Leader: Entrepreneur Bond: Innovation Emphasis: Growth Form: Clan Leader: Mentor Bond: Loyalty Emphasis: Cohesion Form: Market Leader: Producer Bond: Goal Emphasis: Achievement Form: Hierarchy Leader: Coordinator Bond: Rules Emphasis: Stability Flexibility, Individuality, Spontaneity Stability, Control, Predictability Exemplary Follower Alienated Follower Conformist Follower Passi ve Follower Pragmatist Follower Independent, Critical Thinking Dependent, Uncritical Passive ---------------Acting -----------------Active Internal emphasis -----------------External positioning, Short -term orientation, Longterm orientation, Smoothing activities Achievement -oriented activities

PAGE 45

45 CHAPTER 3 DESIGN OF THE STUDY The purpose of this chapter is to address the population, sampling frame, sample size, instrumentation, procedure for data collection, variables, and statistical analyses used in this study. This study w as done to investigate followership within community college faculty. To address this purpose, t he following research questions were posed Using Chaleffs (1995) five identified variables of followership as an index: 1 Is faculty followership influenced by a respondent s: (a) rank (b) receipt of tenure, ( c ) age, (d ) sex ( e ) race, ( f ) education level, ( g ) discipline or ( h ) length of time working in higher education? 2 Is faculty followership influenced by organizational culture? 3 Is faculty followership influenced by institution al : (a) size, (b) population served, (c) location, and ( d ) type (bachelor degree granting or not)? Population The participants for this study w ere community college faculty members in the S tate of Florida. Floridas 28 community c olleges were established to serve the states citizens by offering the first two years of a baccalaureate degree, vocational education, and adult continuing education. As of fall, 2006, these colleges served 793,517 students, of which funded full -time equi valent (FTE) student enrollment was 287,715. T he system employed 22,749 faculty members, with 5,359 working full -time and 17,390 part -time (Florida Department of Education, 2007). Floridas community college system follow ed the state -wide categorization of faculty positions. Categories include d professor, associate professor, assistant professor, and instructor (University of South Florida, 2004). Faculty members of all categories wer e on either tenure accruing (also called continuing contract) or nontenur e accruing track s Floridas 28 community colleges we re divided into three tiers or peer groups by sizes of enrollment Large -size institutions we re those with greater than an 11, 5 00 student FTE

PAGE 46

46 enrollment. Medium -size institutions we re those with an FTE enrollment between 6 0 00 and 11, 4 99 students, while small -size community colleges we re those with less than a 6 0 00 FTE student enrollment (Perrault, Madaus, Armbrister, Dixon, & Smith, 1999). The majority of schools in the large size tier are located in t he states large urban population centers while the m edium -sized schools are located in or near population centers or major state universities (Perrault, et al., 1999). In addition, as of 2008, ten Florida community colleges including Broward, Chipola, Da ytona Beach, Edison, Florida Community College at Jacksonville, Indian River, Miami Dade, Northwest Florida Palm Beach, and St. Petersburg had been approved by the State Board of Education to offer bachelor's degrees in a limited number of programs (Flori da Department of Education, 200 8 ). Sampling Frame A sampling frame is a list of potential participants in the population (Agresti & Finley, 1999) and is the source from which the sample is drawn. For this study, full -time and adjunct instructors, assistan t professors, associate professors, and full professors from Floridas community colleges were the intended population. On line faculty directories from each of Floridas community colleges were used to compile the sampling frame. However, Miami Dade Colle ge s website did not provide faculty e -mail addresses and it was the policy of this college to not provide outside parties with faculty e -mail or personal addresses or to distribute outside research surveys to its faculty (D. Kaiser, personal communication May 20, 2008). Using the available on line directories, the researcher produced a list of 5,852 faculty e -mail addresses representing 27 of Floridas 28 community colleges. Sample Size The sample of data collected is the information base from which all statistical decisions are made. It is therefore critical that the sample be collected properly and be of sufficient size to

PAGE 47

47 justify any decisions concerning hypotheses ( Brewer, 1996, p. 5 wa s set at than .20. Furthermore, Cohen (1969) considered a medium effect of .5 is visible to the naked eye of a careful observer. A small effect of .2 is noticeably smaller than medium but not so small as to be trivial. A large effect of .8 is the same distance above the medium as small is below it (Carson, n.d., p. 4). In determining sample size, the following values we re u se d po Based on these values, the sample size for this study w as determined to be at least as large as 480 participants (see Appendix A for the SAS power program output for between -subjects main effects that de termined sample size ). Previous research reported that response rates for web -based surveys tended to fall between 20 and 25 percent. After reviewing 31 web-based studies, Sheehan (2001) developed a model for return rates that predicted researchers utiliz ing web -based materials should expect a 21 to 23 percent return rate. Other researchers recorded similar findings on studies that relied on web based surveys. Phillips (2002) noted that her study obtained a 24.4% response rates from Florida community colle ge faculty. Roark (1988) also studied Florida community colleges and obtained a return rate of 20.9 % from part time instructors and their supervisors. Gerity (1999) surveyed 1,443 full and part time community college workforce development and training pro fessionals and reported a 20 % return rate. The researcher assumed that this study would achieve a response rate of approximately 15% The author assumed a slightly lower than normal response rate because the data was collected during the summer semester an d many full time and adjunct faculty members did not teach during that term and may not have looked at their e -mails (S. Sass, personal communication, May 15, 2008). Based on this response rate and a desired

PAGE 48

48 sample size of 480 participants, the researcher randomly selected 3,200 e -mail addresses from the developed sampling frame to ask to participate. Instrumentation The Followership Profile The Followership Profile (TFP) was developed by Dixon (2003) to measure courageous follower behaviors according to C haleffs (1995) theory. Chaleff (1995) identifie d five distinct categories of follower behaviors: responsibility, service, challenging, transforming, and moral action Dixon (2003) developed the TFP to assess each of these follower dimensions with a catego ry score and to also provide an overall follower score derived from the scores of the five behaviors. Dixons 56 -item measure was originally designed to assess levels of followership among engineers and technology workers in technology sectors of government and industry. Dixon also developed a 20 item abbreviated version of the questionnaire. On the 56 item measure, 20 questions reference d courage to assume responsibility, 10 questions referenced the courage to serve, seven questions reference d the courage to challenge, seven questions represent ed the courage to participate in transformation, and twelve questions represent ed the courage to take moral action. The relative number of items per behavior is indicative of the descriptive text Chaleff applies to each behavior (Dixon, 2003, p. 54). Factor analysis resulted in the deletion of all items with multiple loadings and all items lacking alignment with the dominant theoretical domain. This left an abbreviated version made up of 20 items remaining from the original 56 questions (see Appendix B). The researcher included i tems 21 23 for this study so that all dimensions consisted of at least four items (L. Behar Horenstein, personal communication, Ma rch 1 8 2008). These 23 items were divided among: six items for courage to assume responsibility, five questions referencing the courage to

PAGE 49

49 serve, four items for the courage to challenge, four questions representing the courage to participate in transformation, and four items under the courage to take moral action The TFP u s e d self ratings with a forced -choice Likert response category scale. Declarative statements we re followed by response options indicating varying degrees of agreement with the statements. Descriptor terms used with the five item Likert response options we re no extent, slight extent, moderate extent, great extent, and very great extent, and related to a numerical scale from 1 to 5, respectively. Lower extreme scores implied a minimum presence of courageous follower behaviors while the upper extrem e deno t ed complete follower behaviors. Dixon (2003) established evidence of validity for the TFP using four processes: test content, response processes, internal structure, and relations to other variables. Focus groups and multiple experts (to include Ch aleff) established validity evidence based on test content during the development process through item development, review, and modification. Validity evidence based on response processes was established using cognitive interviews and self administered que stionnaires to test subjects. Dixon (2003) conducted confirmatory factor analysis to provide evidence supporting the internal structure of the scale measuring five dimensions. Additionally, measurement of internal consistency yielded a Cronbachs alpha coe fficient of .956. Test -retest produced reliability correlations between .604 and .726 for the six assessed dimensions. Further evidence of validity through relations to other variables was established through convergent evidence. The PearsonProduct Moment for the TFP and The Self -Assessment Textual Instrument (an alternate form instrument based on detailed textual descriptions of each of the five courageous follower behaviors) was .739. Spearmans Rho was calculated as .697. Reliability was established in a number of ways. Dixons measurement of internal consistency yielded a Cronbachs alpha coefficient of .956. The calculated split -half Spearman -

PAGE 50

50 Brown reliability coefficient was .936 while the Guttman split -half was .934. Time stability was adequate with test retest reliability indicating correlations between .604 and .726 for the six assessed dimensions. For the 2 0 item abbreviated version, the Cronbachs alpha coefficient was .868. The calculated split -half Spearman -Brown reliability coefficient was .85 5 while the Guttman split -half was .853. Stability over time was not assessed. Permission was obtained from Dixon to use the TFP in this study (see Appendix C). Institutional Performance Survey Organizational culture was assessed through the use the Insti tutional Performance Survey (IPS) cultural scenarios (see Appendix D) that Smart and St. John (1996) state d indicate the extent to which their institutions evidenced attributes associated with four ideal culture types along four dimensions: institutional character, institutional leader, institutional cohesion, and institutional emphases (p. 226). 16 items we re divided equally among the four dimensions, giving each dimension one item that characterize d each culture type. Each presented scenario serve d as a word picture that help[ed] respondents convey not just the extent to which they are satisfied or dissatisfied with their organization (its climate) but the core values and orientations that characterize it (its culture) (Smart & St. John, 1996, p. 226). The IPS culture scenarios ask ed respondents to distribute 100 points among four institutional descriptors depending on how similar the description is to your school For example if Institution A seems very similar to mine, B seems somewhat si milar, and C and D do not seem similar at all, I might give 70 points to A and the remaining 30 points to B (Smart & St. John, 1996, p. 228). Scores we re obtained by averaging ratings for each culture type across the four dimensions. The highest culture s cale score wa s used to determine the dominant culture type for the institution.

PAGE 51

51 Smart and St. John report ed that there is consistent empirical evidence supporting the validity of the four dominant organizational culture types that evolve from responses to the IPS cultural scenarios (1996, p. 227). These authors report ed the following reliabilities with the items used to measure each culture type: clan culture coefficient alpha = .82; adhocracy culture coefficient alpha = .83; bureaucratic culture coeffici ent alpha = .67; and market culture coefficient alpha = .78. They conclude d that it is possible to develop a valid survey instrument to study organizational culture because their results using the IPS met the criteria of internal consistency, predictable relationships with other organizational phenomena, and discrimination among groups (Smart & St. John, 1996, p. 229). Permission was obtained from Smart to use the IPS cultural scenarios in this study (see Appendix E). Procedure for Data Collection Full ti me and adjunct instructors, assistant professors, associate professors, and full professors from Floridas community colleges were the intended population of this research. 3,200 faculty members from 27 of Floridas 28 community colleges were contacted via e -mail and asked to complete an on line survey of faculty followership behaviors. A web -based survey was chosen because web -responding tends to be less time -consuming for participants (Phillips, 2002). P otential participants receive d an e -mail to the acco unt listed in their faculty contact information. Upon opening the message, an introductory letter explained the study and asked for their participat ion (see Appendix F) The letter also explain ed the importance of their participation and that by completing and submitting the survey, participants were consenting to voluntarily participate in this study. Finally, the e -mail note d the endorsement of this research by the University of Florida Human Subjects Committee (see Appendix G) Faculty who considered p ar ticipat ing w er e directed to a linked website containing electronic versions of the abbreviated TFP, IPS, and a demographics questionnaire Permission from the authors to use the TFP and the

PAGE 52

52 IPS for the purposes of this research was obtained prior to study (Appendices C and E) Survey questions wer e presented sequentially R espondents did not need to complete any question that they were uncomfortable answering before mov ing on to the next question. Upon completion of the TFP and IPS participants move d immed iately to demographic information questions (see Appendix H ) that gather ed information about the respondents faculty rank tenure status, age, sex race, ethnicity, education level, discipline length of time working in higher education, institutional siz e, population served, location, and type. Variables Dependent Variables Factor scores were used as the dependent variables in this study. Factor scores are derived from the sum of the proportional weighting of items relevant in making the factor score. Fa ctor scores are standardized values; a oneunit difference in a factor score represents one standard deviation difference in the study population. Using factor scores allowed the researcher to avoid the assumption that all scale items equally affected the dependent variable scores. The dependent variables in this study we re as follows: 1 Total followership factor score a weighted summation of Chaleffs five distinct categories of follower behaviors as measured by the TFP: the courage to assume responsibili ty, the courage to serve, the courage to challenge, the courage to participate in transformation, and the courage to take moral action; 2 Responsibility factor score a weighted summation of TFP items 2, 6, 8, 16, 17, and 19; 3 Service factor score a weight ed summation of items 7, and 11 14; 4 Challenge factor score a weighted summation of items 15, 18, 21 and 22; 5 Transformation factor score a weighted summation of items 1, 3, 4, and 10; and 6 Moral action factor score a weighted summation of items 5, 9, 20, and 23.

PAGE 53

53 Independent Variables 1 Faculty rank. There wer e four levels of this variable: instructor assistant professor, associate professor, and full professor. 2 Tenure. A dichotomous variable that indicat ed whether the respondent wa s a tenured faculty m ember (on continuing contract) or not. 3 Age. A continuous variable that reflect ed the age of the respondent at the time of survey completion. 4 Sex A dichotomous variable that indicat ed whether the respondent wa s male or female. 5 Race. A dichotomous variab le that indicated whether the respondent wa s white or of a minority group, as defined u sing traditional U.S. Census Bureau (2008) classifications of : American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or Other Pacific Islan der, Hispanic/Latino and Other. 6 Education level. An ordinal variable that reflect ed the highest degree obtained by the respondent. F our options wer e available: associate (2 -year) degree, bachelor s (4 -year) degree, masters degree, or doctoral degree. 7 Di scipline A categorical variable based on Biglans (1973) classification of academic disciplines and community college workforce degrees that indicat ed whether respondents obtained their highest degree in (a) humanities business, education, or social scie nces (b) math, science, technology, or engineering or (c) health, culinary arts, public safety, or performing arts 8 Length of time working in higher education. A continuous variable that indicated the number of years the respondent ha d worked in educati on. 9 Culture. There wer e four levels of this variable: clan, adhocracy, hierarchy, and market, as measured by the IPS cultural scenarios. 10. Institutional size. A continuous variable that reflect ed total FTEs 11. Population served. A continuous variable that r eflect ed the population of the area intended to be served by the institution and based on 2006 census data (U.S. Census Bureau, 2006). 12. Institutional location. Three locations wer e available: large urban area, in or near a population center or state univers it y or not near a major population center. 13. Institutional type. A dichotomous variable of the institution either awarding the baccalaureate degree or not.

PAGE 54

54 Statistical Analys e s For all statistical techniques used in this study, an alpha level of .05 was us ed as the criterion significance level. Except for factor analysis and determination of factor scores a ll statistical analyses wer e conducted using SAS System for Windows version 8. Exploratory f actor analysis was conducted using SPSS Graduate Pack for Wi ndows version 15.0 on the collected data to ensure that the TFP wa s appropriate for use with a higher education faculty population since Dixon (2003) originally developed the TFP questionnaire for use with engineers and technology workers. Cronbachs alpha coefficient was also determined to address the TFPs reliability (internal consistency) characteristics with this different population. These analyses indicate d the appropriateness of the TFP for use in this study. Also, SPSS was also used to determine th e factor scores of all dependent variables used in this study. Question 1 The following regression model was used to test the first research questions hypothesis that total followership factor score (F) varie d by : (1) rank, (2) tenure status, (3 ) age, ( 4 ) sex ( 5 ) race, (6) education level, ( 7 ) discipline and ( 8 ) length of time working in higher education: [rank] + tenure + age + sex + race + [education] + [discipline] + time working In equation format: 1Z1 + 2Z2 + 3Z3] + 4Z4 + 5X5 + 6Z6 + 7Z7 + 8Z8 + 9Z9 + 10Z10] + 11Z11 + 12Z12] 13X13 Dummy coding of class variables for the equation was as follows: Rank: Z1 = 1 for assistant professors, 0 for all others; Z2 = 1 for associate professors, 0 for all others; Z3 = 1 for professors, 0 for all others Tenure: Z4 = 1 for tenure track; 0 for nontenure track Sex: Z6 = 1 for females; 0 for males Race: Z7 = 1 for non-minority status; 0 for minority status Education level: Z8 = 1 for associates degree; 0 for all others; Z9 = 1 for bachelors degree; 0 for all others; Z10 = 1 for masters degree; 0 for all others

PAGE 55

55 The hypothe ses that were tested for the first research question (stated in terms of the models parameters) were as follows : HO1: 1 = 2 = 3 = 0; HO2: 4 = 0; HO3: 5 = 0; HO4: 6 = 0 ; HO5: 7 = 0 ; HO6: 8 = 9 = 10 = 0 ; HO7: 11 = 12 = 0 ; HO8: 13 = 0 HA1: 1 2 3 HA2: 4 HA3: 5 HA4: 6 HA5: 7 HA6: 8 9 10 ; HA7: 11 12 ; HA8: 13 Similar regression models were used to test the hypotheses that the five followership component f actor scores (responsibility, service, challenging, transformation, and moral action) varie d by : (1) rank, (2) tenure status, (3 ) age, ( 4 ) sex ( 5 ) race, (6) education level, ( 7 ) discipline and ( 8 ) length of time working in higher education Question 2 O ne -way ANOVA was used to test the second research questions hypothesis that the total followership factor score varie d with organizational culture. One -way ANOVA was also used to test the hypotheses that the five followership component factor scores var ie d with organizational culture. The hypothesis that was tested for research question two wa s as follows: HO9: 1 2 3 4 = 0 HA9: 1 and/or 2 and/or 3 and/or 4 0 Question 3 The following regression model was used to test the third resear ch questions hypothes i s that the total followership factor score (F) varie d by institutional: (1) size, (2) population served, (3) location, and ( 4 ) type. size + population + [location] + type In equation format: 1X1 2X2 3Z3 + 4Z4] + 5Z5 +

PAGE 56

56 Dummy coding of the class variable (rank) for the equation was as follows: Z3 = 1 for large urban area, 0 for all others; Z4 = 1 for in or near a population center or state universit y, 0 for all others The hypotheses that were tested for the third research question (stated in terms of the models parameters) were as follows : HO10: 1 = 0 ; HO11: 2 = 0 ; HO12: 3 = 4 = 0; HO13: 5 = 0 HA10: 1 HA11: 2 HA12: 3 4 HA13: 5 Similar regression models were used t o test the hypotheses that the five followership component factor scores (responsibility, service, challenging, transformation, and moral action) varie d by institutional: (1) size, (2) population served, (3) location, and ( 4 ) type.

PAGE 57

57 CHAPTER 4 RESULTS AN D DATA ANALYSES The purpose of this chapter is to present the results of the analyses of the collected data. The chapter is divided into four major sections; (a) response rate, (b) respondent profile, (c) instrument performance, and (d) hypotheses testing. Response Rate Of the 3,200 surveys electronically mailed to the selected faculty members, 661 were returned for a response rate of 20.7% This rate fell within the predicted response rate of 20 to 25 percent suggested by previous studies (Gerity, 1999; Phillips, 2002; Roark, 1988; Sheehan, 2001). Table 1 provides the response rates of the participating institutions. Of the 661 returned surveys, 64 surveys were incomplete and were deleted. This left 597 surveys that were used in the data analyses. Respon dent Profile Table 2 is a presentation of the breakdown of the sample population corresponding to the demographic variables used in the study. Table 3 is a comparison of the composition of the sample population and the actual population of Florida communit y college faculty members. Instrument Performance Reliability The 23 items of the TFP were submitted to reliability analyses. A Cronbach coefficient alpha of .84 was obtained for the total score, .7 4 for the s ervice subscore .67 for the t ransformation su bscore, .60 for the c hallenge subscore .60 for the r esponsibility subscore and .5 4 for the m oral action subscore These analyses revealed one item whose deletion from the analysis would increase the alpha. The deletion of this item (question 19) from the subsequent

PAGE 58

58 reliability analysis would have increased the coefficient alpha to .84, a negligible increase. This item was also the only one to have less than a .30 correlation with the total. Exploratory Factor Analysis Exploratory f actor analysis was perf ormed to determine if the TFP performed on this sample in a manner similar to that in earlier applications. Dixon (2003) had previously identified the scale as having five factors. The researcher used the principal components method of factor analysis with varimax rotation using Kaiser n ormalization and ignored Eigenvalues less than 1.0. Rotation converged after six iterations. The first factor, whose loading were very similar to the courage to serve dimension described by Dixon (2003), accounted for 23.8 % of the total variance in TFP scores. The dimension courage to take moral action accounted for an additional 9.3 % of the total variance while the dimension courage to participate in transformation accounted for an additional 6.7 % of the total variance. The fourth factor represented Dixons theoretical domain of courage to challenge and accounted for 6.0 % of the total variance. The fifth factor included items from the courage to assume responsibility dimension and accounted for 4.8 % of the total variance. Th ese five factors accounted for 50.6 % of the total variance (as opposed to 62.79% in Dixons original study). As noted by Dixon (2003), while each factor did not purely consist of items from only one theoretical domain, behaviors from one theoretical domai n were the predominantly represented behaviors within each factor structure. These results were not unusual as factor analysis is considered an assessment of respondent patterns and not theory-based (Nunnally & Bernstein, 1994; Schrieschein, Clogliser, Sca ndura, Lankau, & Powers, 1999; Schrieschein, Powers, Scandura, Gardiner, & Lankau, 1993). The rotated component matrix is given in Table 4 Based

PAGE 59

59 on these results, the TFP was deemed appropriate to use for this population and items were judged to measure the six dimensions as defined by Dixon (2003). To determine factor scores, exploratory factor analysis was performed six times. Items hypothesized to contribute to an individual subscore (responsibility, service, challenging, transformation, or moral actio n ) were selected and a single factor solution was chosen Factor scores for the five subscores were then entered and a single solution was selected to determine factor scores for the total followership score. These factor scores were used for all analyses. Hypotheses Testing Question 1 Analysis of variance was conducted to determine the effect of institutional membership on the total followership factor score as well as its five component factor scores. The following variables were included in the analyses: institution, rank, tenure status, age, sex, race, education level, discipline, and length of time working in higher education Institution was not significant for any analyses. The effect of institution for total score was F (25, 428) = .7 8 p = .7 7 par tial 2 = .0 4 ; for the responsibility subscore, F (25, 428) = .9 6 p = 522 = .0 5 ; for the service subscore, F (25, 428) = 89, p = 622 = .0 5 ; for the challenge subscore, F (25, 428) = .9 8 p = .5 0 2 = .0 5 for the transformation s ubscore, F (25, 428) = 1. 24, p = 20, partial 2 = .0 7 ; and for the moral action subscore, F (25, 428) = .79, p = .7 6 2 = .0 4 Because the effect of institution was not significant in any analyses, multi level analysis was not used for the first r esearch questions analyses. The following regression model was used to test the first research questions hypothesis that the total followership factor score (F) varie d by : (1) rank, (2) tenure status, (3 ) age, ( 4 ) sex (5 ) race, (6) education level, ( 7 ) discipline and ( 8 ) length of time working in higher education: [rank] + tenure + age + sex + race + [education] + [discipline] + time working

PAGE 60

60 In equation format: 1Z1 + 2Z2 + 3Z3] + 4Z4 + 5X5 + 6Z6 + 7Z7 + 8Z8 + 9Z9 + 10Z10] + 11Z11 + 12Z12] 13X13 Dummy coding of class variables for the equation was as follows: Rank: Z1 = 1 for assistant professors, 0 for all others; Z2 = 1 for associate professors, 0 for all others; Z3 = 1 for professors, 0 for all others Tenure: Z4 = 1 for tenure track; 0 for nontenure track Sex: Z6 = 1 for females; 0 for males Race: Z7 = 1 for non-minority status; 0 for minority status Education level: Z8 = 1 for associates degree; 0 for all others; Z 9 = 1 for bachelors degree; 0 for all others; Z10 = 1 for masters degree; 0 for all others The hypothe ses that were tested for the first research question (stated in terms of the models parameters): HO1: 1 = 2 = 3 = 0; HO2: 4 = 0; HO3: 5 = 0; HO4: 6 = 0 ; HO5: 7 = 0 ; HO6: 8 = 9 = 10 = 0 ; HO7: 11 = 12 = 0 ; HO8: 13 = 0 HA1: 1 2 an 3 HA2: 4 HA3: 5 HA4: 6 HA5: 7 HA6: 8 9 10 ; HA7: 11 12 ; HA8: 13 Similar regression models were used to test the hypotheses that the five followership component factor scores (re sponsibility, service, challenging, transformation, and moral action) varie d by : (1) rank, (2) tenure status, (3 ) age, ( 4 ) sex ( 5 ) race, (6) education level, ( 7 ) discipline and ( 8 ) length of time working in higher education The Pearson r correlation co efficients and their associated levels of significance are presented in Table 5 for each of the dependent variables (followership factor scores) as they were related to the variables of interest. The issue of multicolinearity was examined. The largest corr elation between independent variables was .51 between age and years of work in higher education. This was less than + .7, which was recommended by Rumsey (2007) as a threshold criteria. All other correlations were less than .40. Additionally, no factor had a tolerance less than

PAGE 61

61 .1 or a variance inflation factor (VIF) greater than 10. Therefore, no factors were excluded due to multicolinearity. The model for the total followership factor score explained 5.9 6 % of the variance. Table 6 is the regression table for the analysis using total followership factor scores as the dependent variable. The data indicated that there was a significant positive relationship between total followership factor scores and age, t (498) = 3.15, p = .0 0 2 2 = .0 2 with olde r individuals scoring higher. Furthermore, there were statistically significant differences in total followership factor scores based on sex, t (498) = 2.12, p = .03, with females scoring higher than males. Cohens effect size for sex was .22, with a 95% co nfidence interval of [.05, 40]. In addition, t he data indicated that there were statistically significant differences in followership factor scores based on faculty rank between assistant professors and associate professors, t (498) = 5.60, p = .02, and be tween assistant professors and full professors, t (498) = 4.58, p = .03, with the followership factor scores of assistant professors being larger in each case. Cohens effect size for rank between assistant professors and associate professors was .2 8 with a 95% confidence interval of [.0 0 .57]. Cohens effect size for rank between assistant professors and full professors was .2 0 with a 95% confidence interval of [.00 .46]. Finally, the data indicated that there were statistically significant differences in total followership factor scores based on the discipline in which the faculty members earned their highest degrees, between Humanities, Business, E ducation, or Social sciences (HBES) and S cience, T echnology, Engineering, or Math (STEM), t (498) = 7.92, p = .005, with faculty having an HBES background scoring higher. Cohens effect size for discipline was .31, with a 95% confidence interval of [.12 .51]. Based on these results, HO1, HO3, HO4, and HO7 were rejected; there were significant total followershi p factor score differences based on rank, age, sex, and discipline. The means and standard

PAGE 62

62 deviations of followership factor scores grouped by sex, discipline, and faculty rank are presented in Tables 7 8 and 9 respectively. The regression line for predicted total followership factor scores on age is presented in Figure 3. No other comparisons utilizing total followership factor scores were significant. Plots of studentized residuals against the independent variables were examined. No violations of linea rity, conditional normality, or equal conditional variance were detected. The model for the responsibility factor score explained 5.66% of the variance. Table 10 is the regression table for the analysis using responsibility factor scores as the dependent v ariables. Statistically significant results were found for age, t (498) = 2.75, p = .0 06, 2 = .01, with older participants scoring higher. Statistically significant results were also found for education level between faculty whose highest degree was a master s and those with a doctorate, t (498) = 2.53, p = .01, with doctorate holders s coring higher. Cohens effect size for education level between faculty with a masters and those with a doctorate was .26 with a 95% confidence interval of [.0 7 .45]. Finally, statistically significant results were found for discipline between HBES and ST EM, t (498) = 10.85, p = .001, with HBES participants scoring higher. Cohens effect size for discipline between HBES and STEM was .32, with a 95% confidence interval of [.13, .52]. Based on these results, HO3R, HO6R, and HO7R were rejected; there were sign ificant responsibility factor score differences based on age, education level, and discipline. Tables 1 1 and 1 2 present the means and standard deviations of responsibility factor scores grouped by education level and discipline, respectively. The regressio n line for predicted responsibility factor scores on age is presented in Figure 4. No other comparisons utilizing responsibility factor scores were significant. Plots of studentized residuals against the independent variables were

PAGE 63

63 examined. No violations o f linearity, conditional normality, or equal conditional variance were detected. The model for the service factor score explained 6.06% of the variance. Table 1 3 is the regression table for the analysis using service factor scores as the dependent variable s. A s tatistically significant difference w as found for sex, t (498) = 2.82, p = .005, with females outscoring males Cohens effect size for sex was .25 with a 95% confidence interval between [.08, .43]. Statistically significant results were found for ran k between assistant professor and associate professor, t (498) = 9.31, p = .002, and assistant professor and full professor, t (498) = 5.54, p = .02 with assistant professors scoring higher in each case. Cohens effect size for rank between assistant profes sor and associate professor was .46 with a 95% confidence interval of [.16, .75] and Cohens effect size between assistant professor and full professor was .27 with a 95% confidence interval of [.02, .53] Finally, statistically significant differences wer e found in discipline between HBES and STEM, t (498) = 4.04, p = .04 with HBES outscoring STEM Cohens effect size for discipline between HBES and STEM was .22 with a 95% confidence interval of [.02, .42]. Based on these results, HO1S, HO4S, a nd HO 7 S were rejected; there were significant service factor score differences based on faculty rank sex and discipline Plots of studentized residuals against the independent variables were examined. No violations of linearity, conditional normality, or equal condi tional variance were detected. In addition, t he data indicated that there were statistically significant differences in service factor scores for rank when controlling the per comparison error rate between assistant professor and instructor, t (498) = 2.06 p = .04, However, these differences were no longer significant when controlling the familywise error rate ( t = 2.51). Tables 14 1 5 and 1 6 present the

PAGE 64

64 means and standard deviations of service factor scores grouped by rank, sex, and discipline, respectively. No other comparisons utilizing service factor scores were significant. The model for the ch allenge factor score explained 4.21 % of the variance. Table 1 7 is the regression table for the analysis using challenge factor scores as the dependent variables. Statistically significant results were found for age, t (498) = 2.99, p = .00 3 2 = .0 2, with older participants scoring higher. T he data also indicated that there were statistically significant differences in challenge factor scores for discipline between HBES and STEM, t (498) = 4.87, p = .03, with HBES scoring higher. Cohens effect size for discipline between HBES and STEM was .22 with a 95% confidence interval of [.03 .42]. Based on these results, HO3C and HO7C w ere rejected; there were significant challenge factor score differences based on age and discipline Table 18 presents the me ans and standard deviations of challenge factor scores, grouped by discipline. The regression line for predicted challenge factor scores on age is presented in Figure 5. No other comparisons utilizing challenge factor scores were significant. Plots of stud entized residuals against the independent variables were examined. No violations of linearity, conditional normality, or equal conditional variance were detected. The model for the transformation factor score explained 4.90% of the variance. Table 1 9 is th e regression table for the analysis using transformation factor scores as the dependent variables. Statistically significant results were found for tenure, t (498) = 2. 10, p = .0 4 with tenure -track faculty scoring higher Cohens effect size for tenure was .15 with a 95% confidence interval of [.00, .32]. Statistically significant results were found for sex, t (498) = 3.00, p = .00 3 with females outscoring males Cohens effect size for sex was .34 with a 95% confidence interval of [.16, .51]. Statistically significant differences were found for discipline between HBES and STEM, t (498) = 5.68, p = .02, with HBES scoring higher. Cohens effect size for

PAGE 65

65 discipline between HBES and STEM was .2 5 with a 95% confidence interval of [.05 .4 5 ]. Based on these result s, HO2T, HO4T, and HO7C were rejected; there were significant transformation factor score differences based on tenure sex and discipline Tables 20, 21, and 22 present the means and standard deviations of transformation factor scores, grouped by tenure, sex, and discipline, respectively. No other comparisons utilizing transformation factor scores were significant. Plots of studentized residuals against the independent variables were examined. No violations of linearity, conditional normality, or equal con ditional variance were detected. The model for the moral action factor score explained 6.3 5 % of the variance. Table 2 3 is the regression table for the analysis using moral action factor scores as the dependent variables. Statistically significant results w ere found for tenure, t (498) = 2.04, p = .04, with nontenured faculty scoring higher. Cohens effect size for tenure was .32 with a 95% confidence interval of [.1 6 .49]. Statistically significant results were also found for age, t (498) = 2.83, p = .00 5 part ial 2 = .0 2 with older participants scoring higher. Based on these results, HO2M and HO3M were rejected; there were significant moral action factor score differences based on tenure and age. Table 2 4 presents the means and standard deviations of moral ac tion factor scores grouped by tenure. The regression line for predicted challenge factor scores on age is presented in Figure 6. No other comparisons utilizing moral action factor scores were significant. Plots of studentized residuals against the independ ent variables were examined. No violations of linearity, conditional normality, or equal conditional variance were detected. In summary, significant differences due to faculty rank were found for the service and total followership factor scores ; due to te nure was for the transformation and moral action factor scores; due to age were for the responsibility, challenge, moral action, and total followership factor scores; due to sex were for the service, transformation, and total followership factor

PAGE 66

66 scores; du e to education level was only for the responsibility factor score; and due to discipline was for the responsibility service, challenge, transformation, and total followership factor scores. The variables of race and years working in higher education did not yield any significant differences. Question 2 One -way ANOVA was used to test the second research questions hypothesis that total followership factor score s varie d with organizational culture. The means and standard deviations of followership factor sc ores, grouped by culture types, are presented in Table 2 5 The data indicated that there were statistically significant differences in total followership factor scores based on culture type, F (3, 586) = 6.86, p 2 = .03. Table 2 6 is the ANOVA table for this analysis. Pairwise comparisons of culture types were planned and the familywise error rate was selected for control. The Shaffer Holm procedure was used to conduct pairwise comparisons of mar ginal means. The t statistics and critical values for the Shaffer Holm procedure are shown in Table 2 7 Significant differences were found for Clan vs. Hierarchy, t (586) = 4.44, p < .0001, and for Adhocracy vs. Hierarchy t (586) = 2.50, p = .01, but not fo r the other comparisons. Cohens effect size for Clan vs. Hierarchy was .42 with a 95% confidence interval of [.2 2 .6 1 ]. Cohens effect size for Adhocracy vs. Hierarchy was .28 with a 95% confidence interval of [.05, .51 ]. Based on these results, HO9 was rejected. One -way ANOVA was also used to test the hypotheses that the responsibility, service, challenging, transformation and moral action factor scores var ied with organizational culture. The data indicated that there were statistically significant diff erences based on culture type for the following factor scores ; service, challenging, transformation and moral action (but not for responsibility).

PAGE 67

6 7 The means and standard deviations of service factor scores, grouped by culture type, are presented in Table 2 8 The data indicated that there were statistically significant differences in service factor scores based on culture type, F (3, 586) = 6.28, p 2 = .03. Table 2 9 is the ANOVA table for this analysis. Pairwise comparisons of culture types were planned and the familywise error rate was selected for control. The Shaffer Holm procedure was used to conduct pairwise comparisons of marginal means. The t statistics and critical values for the Shaffer Holm procedure are shown in Table 30. Signific ant differences were found for Clan vs. Hierarchy, t (586) = 4.28, p < .0001, but not for the other comparisons. Cohens effect size for Clan vs. Hierarchy was .4 3 with a 95% confidenc e interval of [.23 .6 2 ]. Based on these results, HO9S was rejected. The means and standard deviations of challenging factor scores, grouped by culture type, are presented in Table 3 1 The data indicated that there were statistically significant differences in challenging factor scores based on culture type, F (3, 586) = 3.27 p = .02 2 = .02. Table 3 2 is the ANOVA table for this analysis. Pairwise comparisons of culture types were planned and the familywise error rate was selected for control. The Shaffer Holm procedure was used to conduct pairwise comparisons of marginal me ans. The t statistics and critical values for the Shaffer Holm procedure are shown in Table 3 3 Significant differences were found for Clan vs. Hierarchy, t (586) = 2.74, p = .0 1 but not for the other comparisons. Cohens effect size for Clan vs. Hierarchy was .27 with a 95% confidence interval of [.08 .46]. Based on these results, HO9C was rejected. The means and standard deviations of transformation factor scores, grouped by culture type, are presented in Table 3 4 The data indicated that there were stat istically significant differences in transformation factor scores based on culture type, F (3, 586) = 8.57, p < .0001,

PAGE 68

68 2 = .04. Table 3 5 is the ANOVA table for this analysis. Pairwise comparisons of culture types were planned and the familywise err or rate was selected for control. The Shaffer Holm procedure was used to conduct pairwise comparisons of marginal means. The t statistics and critical values for the Shaffer Holm procedure are shown in Table 3 6 Significant differences were found for Clan vs. Hierarchy, t (586) = 4.58, p < .0001, for Clan vs. Market, t (586) = 3.30, p = .001, and for Adhocracy vs. Hierarchy, t (586) = 2.68, p = .0 1 but not for the other three comparisons. Cohens effect size for Clan vs. Hierarchy was .45 with a 95% confidence interval of [.25, .64]. Cohens effect size for Clan vs. Market was .53 with a 95% confidence interval of [.2 4 .81]. Cohens effect size for Adhocracy vs. Hierarchy was .30 with a 95% confidence interval of [.0 6 .53]. Based on these results, HO9T was rejected. The means and standard deviations of moral action factor scores, grouped by culture type, are presented in Table 3 7 The data indicated that there were statistically significant differences in moral action factor scores based on culture type, F (3 586) = 2.75, p 2 = .01. Table 3 8 is the ANOVA table for this analysis. Pairwise comparisons of culture types were planned and the familywise error rate was selected for control. The Shaffer Holm procedure was used to conduct pairwise comparisons of marginal means. The t statistics and critical values for the Shaffer Holm procedure are shown in Table 3 9 Significant differences were found for Clan vs. Hierarchy, t (586) = 2.63, p = .0 1 but not for the other comparisons. Cohens effect size for Clan vs. Hierarchy was .2 6 with a 95% confidence interval of [.07 .45]. Based on these results, HO9M was rejected. Question 3 The following regression model was used to test the third research question s hypothes i s that total followership factor scor e s (F) varie d by institutional: (1) size, (2) population served, (3) location, and ( 4 ) type (bachelor degree -granting or not)

PAGE 69

69 size + population + [location] + type In equation format: = 1X1 2X2 3Z3 + 4Z4] + 5Z5 + Dummy codin g of the class variable for the equation was as follows: Rank: Z3 = 1 for large urban area, 0 for all others; Z4 = 1 for in or near a population center or state universit y, 0 for all others The hypotheses that were tested for the first research question (s tated in terms of the models parameters): HO10: 1 = 0 ; HO11: 2 = 0 ; HO12: 3 = 4 = 0; HO13: 5 = 0 HA10: 1 HA11: 2 HA12: 3 4 HA13: 5 Similar regression models were used to test the hypotheses that the five followership factor scores (responsibility, service, challeng ing, transformation, and moral action) varie d by institutional: (1) size, (2) population served, (3) location, and ( 4 ) type. The Pearson r correlation coefficients and their associated levels of significance are presented in Table 40 for each of the dependent variables (followership factor scores) as they were related to the variables of interest. The only significant difference among factor scores was found in the moral action dimension for the independent variable of degree offered. This model explained 1 .44% of the variance. Table 41 is the regression table for the analysis using moral action factor scores as the dependent variable. Statistically significant results were found for degree offered, t (528) = 2.55, p = .01, with faculty from institutions that offered bachelors degrees scoring higher for moral action than faculty from institutions that did not offer bachelors degrees. Cohens effect size for degree offered was .23 with a 95% confidence interval of [.05, .42]. Table 4 2 presents the means and s tandard deviations of moral action subscores, grouped by degree offered. No other comparisons of institutional variables were

PAGE 70

70 significant for followership factor scores. Based on these results, only HO 13M was rejected under the third research question. Additional Analys e s Analyses of all potential two -way interactions were performed. ANOVA indicated several significant interactions. In the responsibility dimension, ANOVA indicated a significant tenure by years working interaction, F (1, 495) = 7.05, p = .01, 2 = .01, and a significant discipline by years working interaction, F (2, 495) = 3.23, p 2 = .01. Table 43 is the ANOVA tables for this analysis. Figures 7 and 8 present the predicted responsibility factor scores by tenure and disc ipline across years working in higher education institutions respectively The tenure by years working interaction suggest ed that tenured faculty responsibility factor scores decline d over years working at a faster rate than non tenured faculty factor sco res. The academic discipline by years working interaction suggest ed that, while faculty in the math, science, technology, or engineering disciplines show ed improvement in responsibility scores as their years working increase d faculty in humanities busine ss, education, or social sciences and in health, culinary arts, public safety, or performing arts show ed declining responsibility scores over time. In the service dimension, ANOVA indicated a significant age by tenure interaction, F (1, 497) = 5.33, p = .02 2 = .01. Table 44 is the ANOVA table for this analysis. Figure 9 presents the predicted service factor scores by tenure across age. The analysis indicated that among young er faculty members, tenured faculty ha d lower service factor scores than no n tenured faculty. However, tenured faculty ma d e service factor score gains as they age d while non tenured faculty remain ed almost constant.

PAGE 71

71 Summary This chapter began with a description of the sample population followed by an examination of the performanc e of The Followership Profile to explore whether it had performed in a fashion consistent with Dixons (2003) intent. Having determined that the TFP performed in a manner suitable for this study, analyses of the data proceeded and the results of hypotheses testing were reported. In Chapter 5, a discussion of these results, the implications drawn, and suggestions for future research are presented.

PAGE 72

72 Table 4 1 Response rates by p articipating i nstitution Institution Response Rate Brev ard Community College 17.0% Broward College 18.9% Central Florida Community College 23.1% Chipola College 10.0% Daytona Beach College 19.5% Edison College 18.9% Florida Community College at Jacksonville 12.5% Florida Key s Community College 9.5% Gulf Coast Community College 15.4% Hillsborough Community College 41.2% Indian River State College 10.0% Lake City Community College 17.5% Lake Sumter Community College 20.6% Manatee Community College 39.2% North Florida Community College 15.6% Northwest Florida State College 11.1% Palm Beach Community College 21.8% Pasco Hernando Community College 18.1% Pensacola Junior College 26.6% Polk Community College 11.0% Santa Fe Com munity College 17.3% Seminole Community College 14.3% South Florida Community College 13.6% St. Johns River Community College 6.3% St. Petersburg College 20.0% Tallahassee Community College 17.3% Valencia Community College 18.8% Overall response rate 20.7% 63 respondents did not identify their institution. Miami Dade College did not participate.

PAGE 73

73 Table 4 2 Percentage of s ample N r epresented by d emographic v ariables of i nterest Rank % of Total Discipline % o f Total Professor 40.2% Humanities, business, Associate Professor 19.3% education, or social sciences 57.6% Assistant Professor 13.1% Math, science, technology, Instructor 27.4% or engineering 24.9% H ealth, culinary arts, public safety, or performing arts 17.5% Tenured % of Total Highest Degree Earned % of Total Yes 61.7% Doctorate 27.4% No 38.3% Masters 62.8% Bachelors 6.9% Associates 2.9% Min ority Status % of Total Sex % of Total Caucasian 85.7% Female 66.6% Minority 14.3% Male 33.4% Age % of Total Years in Education % of Total 20 29 years 4.0% 1 5 years 20.8% 30 39 years 14.5% 6 10 years 22.5% 40 49 years 24.9% 11 20 years 31.8% 50 59 years 38.7% 21 30 years 17.6% > 60 years 17.9% > 31 years 7.3% Table 4 3 Comparison of t otal p opulation versus responding s ample p opu lation Sex Minority Status Population Female % Male % Caucasian % Minority % Total* 55.5% 44.5% 84.9% 15.1% Respondent 66.6% 33.4% 85.7% 14.3% Highest Degree Earned Population Doctorate Masters Bac helors Associates Total* 23.5% 66.5% 6.1% 2.0% Respondent 27.4% 62.8% 6.9% 2.9% Tenure Status Population Tenured Adjunct Total* 23.6% 76.4% Respondent 61.7% 38.3% Source: Florida Department of Education, 20 07

PAGE 74

74 Table 4 4 Rotated c omponent m atrix Component 1 2 3 4 5 Item 1 0 .64 .1 3 .18 .01 .13 Item 12 .62 .29 .0 1 .3 1 .05 Item 13 .62 .3 4 .14 20 .3 1 Item 4 .55 .42 .0 1 .2 6 .14 Item 11 .5 5 .29 .0 2 .33 .0 1 Item 18 .53 .3 4 .1 6 .20 .15 Item 16 .52 .2 6 .2 1 .26 .3 3 Item 14 .52 .0 2 .1 4 30 .2 4 Item 7 .5 1 .37 .21 .12 .08 Item 8 50 .21 .38 .2 1 .1 3 Item 15 .49 .17 .13 .3 1 .0 4 Item 3 .48 .44 .01 .39 .2 3 Item 9 .45 .0 2 .2 9 .02 .1 7 Item 5 .44 .06 .0 8 .1 1 .2 3 Item 20 .4 4 .4 2 .4 2 .15 .0 2 Item 22 .40 .04 .38 .11 .27 Item 21 .43 .58 20 .1 1 .1 1 Item 23 .4 1 .49 .5 2 .13 .0 5 Item 17 .40 30 .44 .01 .03 Item 2 .46 .01 .16 .5 1 .0 6 Item 1 .4 4 .2 4 .2 8 .47 .13 Item 6 .38 .15 .0 6 .0 4 .51 Item 19 .2 2 .3 9 .4 4 .0 1 .51

PAGE 75

75 Table 4 5 Pearson r c orrelation c oefficients and s ignificance l evels for i ndependent v ariables and TFP s cores total_ r_subscore s_subscore c_subscore t_subscore m_subscore score Asst_ p rof -.0 4 09 05 05 0 8 06 40 0 3 2 8 20 06 12 587 587 587 587 587 587 Assoc_ p rof -. 0 2 -. 09 -. 03 0 1 -. 0 7 05 7 0 0 3 50 8 3 11 20 587 587 587 587 587 587 Full_ p rof 05 -. 02 -. 0 1 -. 0 3 -. 1 1 03 27 5 6 82 49 0 1 4 3 587 587 587 587 587 587 Tenure 0 1 -. 0 8 -. 03 07 -. 1 6 0 5 8 2 07 54 08 0002* 25 587 587 587 587 587 587 Age 12 07 14 0 4 06 12 004* 10 001* 37 17 005 530 530 530 530 530 530 Sex 02 12 01 1 6 0 8 10 58 004* 79 .0002* 07 01* 572 572 572 572 572 572 Race 0 2 01 -. 0 1 -. 01 -. 0 6 01 7 1 79 88 80 15 76 592 592 592 592 592 592 Assoc d egree -. 03 04 03 02 03 001 53 34 46 70 46 97 578 578 578 578 578 578 Bac_ d egree -. 03 06 0 4 0 2 0 3 03 4 7 13 38 67 52 45 578 578 578 578 578 578 Mas_ d egree -. 09 -. 02 -. 02 004 04 02 04* 61 68 92 30 59 578 578 578 578 578 578 Hum_discipline 15 06 0 7 07 -. 03 09 0003* 1 8 1 2 0 9 5 3 .03 576 576 576 576 576 576 STEM_ d isc ip 11 -. 1 0 -. 1 0 -.11 -. 06 -. 13 0 1 0 2 0 2 .0 1 17 .00 1* 576 576 576 576 576 576 Years_ w ork ing 0 7 02 0 5 -. 0 1 -. 0 8 0 2 07 6 2 1 9 8 4 06 6 4 5 7 3 5 7 3 5 7 3 5 7 3 5 7 3 5 7 3 significant at p < .05

PAGE 76

76 Table 4 6 Summary t otal followership f actor s core r egression t able for d emographic v ariables Variable B SE B Asst_prof 32 21 .08 Assoc_prof 20 20 .06 Full_prof 12 19 .04 Tenure 07 15 .02 Age .0 2 01 .16 Sex 2 9 13 1 0 Race .1 6 19 .02 Assoc_degree .14 .39 .02 Bac_degree 1 0 26 .0 2 Mas_degree 09 15 .0 3 Hum_discipline 09 17 .03 STEM_discipline 34 20 .10 Years_working .0 1 .01 .0 3 Table 4 7 Mean t otal followership f actor s cores and s tandard d eviations g rouped by s ex Sex N Mean SD Male 191 21 1.45 Female 381 .1 0 1.37 Table 4 8 Mean t otal followership f actor s cores and s tandard d eviations g rouped by d iscipline Discipline N Mean SD Humanities, business, education, 332 .1 0 1.37 or social sciences Math, science, technology, 143 33 1.39 or engineering Health, culinary arts, public safety, 101 .1 0 1.38 or performing arts Table 4 9 Mean t otal followership f actor s cores and s tandard d eviations g rouped by f aculty rank Rank N Mean SD Instructor 161 07 1.43 Assis tant Professor 77 22 1.22 Associate Professor 113 16 1.45 Professor 236 06 1.42

PAGE 77

77 + ----------------------------------------------------------------------------------------+ PRED 1.5 P r e d i 1 c 1.0 t 1 e 1 d 1 11 1 1 1 2 1 1 V 1 1 111 11 12 1 1 a 0.5 32 21 2 12 1 21 l 1 2 1 11 22 1 2 3 11 1 u 1 11 1 1 1 141 21 213 4 1 1 1 e 1 1 11 1 1 122 31 461 5 222 21 11 1 1 2 21 122 1 11 4 413 26 22 2 232 1 1 o 2 2 2 23 2 151 22 4 6 3 113 f 0.0 1 1 1 1 2 11 21 341 35 16 12 1 3 1 32 1 31 2 23 12 21 31 34 22 21 23 2 411 42 22 2 1 t 1 1 11 1 2 321 3 31 12 2 13 412 21 12 1 1 1 o 11 11 1 3 2 4 122 1 22 1 12 2 112 t 1 11 1 1 1 1 1 2 13 1 211 21 11 1 a 1 1 2 11 21 1 2 1 1 1 12 1 11 3 l 0.5 1 1 1 1 1 2 221 1 1 1 11 12 2 21 1 2 1 1 s 1 1 12 1 1 2 1 c 1 1 1 1 1 o 1 1 r e 1.0 2 11 1 1.5 + ----------------------------------------------------------------------------------------+ 20 25 30 35 40 45 50 55 60 65 70 75 Age Figure 4 1 P redicted t otal followership f actor s cores on a ge

PAGE 78

78 Table 4 10. Summary r esponsibility f actor s core r egression t able for d emographic v ariables Variable B SE B Asst_prof .0 1 15 .002 Assoc_prof .09 .1 4 .0 4 Full_prof .0 5 .1 3 .02 Tenure .04 .1 1 .02 Age .0 1 .0 1 .1 4 Sex .05 10 .0 3 Race .0 4 .1 4 .01 Assoc _degree 18 27 .0 3 Bac_degree 28 1 9 .0 7 Mas_degree 27 11 13 Hum_discipline 2 4 12 1 2 STEM_discipline 1 2 14 05 Years_working .0 1 .0 1 .06 Table 4 1 1 Mean r esponsibility f actor s cores and s tandard d eviations g rouped by e ducation l evel Level N Mean SD Associates 17 .15 1.12 Bachelors 40 .11 0.94 Masters 363 .07 0.96 Doctorate 158 .19 1.07 Table 4 1 2 Mean r esponsibility f actor s cores and s tandard d eviations g rouped by d iscipline Discipline N Mean SD Humani ties, business, education, 332 .13 0.96 or social sciences Math, science, technology, 143 .1 9 1.06 or engineering Health, culinary arts, public safety, 101 .14 0.93 or performing arts

PAGE 79

79 + ----------------------------------------------------------------------------------------PRED 0.6 1 P 1 r 1 e 1 1 1 d 11 1 1 111 i 0.4 1 2 3 2 1 1 c 2 2 11 3 1 1 1 t 1 1 12 12 111 3 1 1 1 11 e 11 1 3 1 411 1 12 11 d 1 12 11 21 221 1 1 2 1 1 1 0.2 1 1 1 1 1 11 2 22 1 212 2 2 1 V 12 1 11 1 22 14 12 11 12 1 a 1 1 1 1 11 13 31 121 21 21 21 1 l 1 11 113 1 12 216 12 11 u 1 12 11 1 2 11 2 1 2 311 1 1 e 0.0 11 1 2 32 32 11 12 12 31 1 1 1 1 12 22 1 11 11 13 2 311 12 2 13 1 o 2 11 1 4 12 11 1 1 231 31 2 11 1 311 f 1 2 1 1 24 2 1 22 2 1 1 111 1 21 11 3 1 1 4 1 4 11 12 12 1 322 3 2 1 1 r 0.2 21 111 1 21 12 3 1 12 1 11 21 11 1 1 1 1 1 4 11 13 2 1 11 11 1 1 s 1 1 12 11 1 2 2 2 u 1 11 1 1 11 1 1 11 1 1 1 b 1 1 1 1 11 1 s 0.4 2 2 2 3 1 1 c 1 1 1 1 1 1 1 1 1 o 1 1 13 r 1 1 1 1 1 1 1 e 1 1 1 0.6 1 0.8 + ---------------------------------------------------------------------------------------20 25 30 35 40 45 50 55 60 65 70 75 Age Figure 4 2 Predicted r esponsibility f actor s cores on a ge

PAGE 80

80 Table 4 1 3 Summary s ervice f actor s core r egression t able for d emograph ic v ariables Variable B SE B Asst_prof 31 15 1 1 Assoc_prof 17 14 .0 7 Full_prof .0 4 1 4 .0 2 Tenure 16 .1 1 .0 8 Age .01 .01 .0 7 Sex 27 1 0 1 3 Race .01 .1 4 .00 2 Assoc_degree 42 28 07 Bac_degree 2 5 19 .0 7 Mas_degree 01 11 0 1 Hum_discipline 07 1 3 03 STEM_discipline 15 14 .0 6 Years_working .01 .0 1 .06 Table 4 1 4 Mean s ervice f actor s cores and s tandard d eviations g rouped by faculty rank Rank N Mean SD Instructor 161 .0 5 1. 04 Assistant Professor 77 2 3 .74 Associate Professor 113 -. 19 1. 03 Professor 236 -.0 3 1. 01 Table 4 1 5 Mean s ervice f actor s cores and s tandard d eviati ons g rouped by s ex Sex N Mean SD Male 191 -. 17 1. 02 Female 381 08 .98 Table 4 1 6 Mean s ervice f actor s cores and s tandard d eviations g rouped by d iscipline Discipline N Mean SD Humanities, business, education, 332 .0 4 1. 03 or social sciences Math, science, technology, 143 -. 18 94 or engineering Health, culinary arts, public safety, 101 08 95 or performing arts

PAGE 81

81 Table 4 1 7 Summary c hallenge f actor s core regression t able for d emographic v ariab les Variable B SE B Asst_prof 22 15 07 Assoc_prof 0 5 1 4 .0 2 Full_prof .0 2 1 3 .0 1 Tenure 06 .1 1 .0 3 Age .0 1 .0 1 .1 6 Sex .0 3 .1 0 .01 Race .0 3 1 4 .0 1 Assoc_degree 09 28 02 Bac_degree 15 19 .0 4 Mas_degree .003 11 -. 0 01 Hum_discipline 0 1 12 0 1 STEM_discipline -. 2 3 1 4 1 0 Years_working .0003 .0 1 .00 3 Table 4 1 8 Mean c hallenge f actor s cores and s tandard d eviations g rouped by d iscipline Discipline N Mean SD Humanities, business, education, 332 .0 5 .95 or social sciences Math, science, technology, 143 -. 17 1. 05 or engineering Health, culinary arts, public safety, 101 .0 5 1. 01 or performing arts

PAGE 82

82 + ---------------------------------------------------------------------------------------PRED 0.8 P r e d 1 1 i 0.6 c 1 t e d 11 1 0.4 1 1 1 1 1 1 1 V 2 1 1 a 1 1 11 1 1 2 1 1 l 11 1 21 1 1 312 1 1 u 1 1 11 22 1 1 131 1 41 e 0.2 1 1 11 3 341 32 331 1 1 1 1 1 2 1 131 5 11 o 1 1 11 1 1 1 44 43 538 1 1 f 1 1 1 11 1 122 11 857 4* 1 1 1 11 31 1 121 45 638 1 1 1 c 0.0 1 1 1 33 161 25 21 1 1 1 1 1 31 1 2 27 39 1 1 3 2 s 1 1 1 1 42 163 53 22 1 1 2 12 1 u 1 1 23 11 9 63 2 1 21 221 b 111 1 34 3 11 1 1 2 21 s 0.2 32 22 32 211 1 1 1 1 31 c 11 2 2 1 1 1 1 111 1 21 o 22 1 3 11 11 1 11 3 r 11 11 1 1 2 221 1 1 e 1 1 21 0.4 1 1 1 1 2 11 2 1 1 21 1 11 1 1 0.6 1 + ----------------------------------------------------------------------------------------20 25 30 35 40 45 50 55 60 65 70 75 Age Figure 4 3 Predicted c hallenge f actor s cores on a ge

PAGE 83

83 Table 4 1 9 Summary t ransformation f actor s core r egression t able for d emographic v ariables Variable B SE B Asst_prof 16 15 0 6 Assoc_prof 05 14 .0 2 Full_prof 1 0 13 .0 5 Tenure 22 1 1 .1 1 Age .01 .01 .07 Sex 29 1 0 .1 4 Race .0 3 14 -.01 Assoc_degree 02 28 -. 01 Bac_degree 07 1 9 .0 2 Mas_degree 06 11 0 3 Hum_discipline .0 3 12 .0 1 STEM_discipline 23 14 10 Years_working .0 1 .01 .0 5 Table 4 20. Mean t ransformation f actor s cores and s tandard d eviations g rouped by t enure Tenure Status N Mean SD Tenure t rack 362 05 .95 Non -tenure track 225 -. 1 0 1.06 Table 4 2 1 Mean t ransformation f actor s cores and s tandard d eviations g rouped by s ex Sex N Mean SD Male 191 22 1.03 Female 381 11 .96 Table 4 2 2 Mean t ransformation f acto r s cores and s tandard d eviations g rouped by d iscipline Discipline N Mean SD Humanities, business, education, 332 05 1.02 or social sciences Math, science, technology, 143 -. 2 0 .95 or engineering Health, culinary arts, public sa fety, 101 07 .93 or performing arts

PAGE 84

84 Table 4 2 3 Summary m oral a ction f actor s core regression t able for d emographic v ariables Variable B SE B A sst_prof 15 15 05 Assoc_prof 15 14 .0 6 Full_prof 11 13 .05 Tenure 22 11 .11 Age .0 1 .01 .1 5 Sex 17 10 .0 8 Race 21 14 .0 7 Assoc_degree 21 27 0 4 Bac_degree 06 19 0 1 Mas _degree 09 11 05 Hum_discipline 13 12 .06 STEM_discipline 15 14 06 Years_working .01 .01 .06 Table 4 2 4 Mean m oral a ction f actor s cores and s tandard d eviations g rouped by t enure Tenure Status N Mean SD Tenure track 362 13 1.03 Non -tenure track 225 19 .92

PAGE 85

85 + ---------------------------------------------------------------------------------------PRED 0.8 P r 1 1 e 1 d i 0.6 1 1 c 11 1 2 t 1 2 1 1 1 1 e 1 21 1 1 d 1 1 1 11 11 1 0.4 1 12 1 1 1 2 V 1 1 1 1 1 21 21 1 1 1 1 1 a 1 1 2 2 1 21 1 22 1 l 1 1 11 2 12 11 1 11 1 1 u 1 1 22 11 11 12 1 13 1 1 1 e 0.2 1 1 1 1 3 22 2 11 111 11 1 1 1 1 1 11 22 2 2 1 11 1 o 1 11 1 1 1 1 11 1 1 12 1 2 112 1 f 1 1 1 1 12 11 12 1 2 2 1 12 2 111 1 11 21 1 12 2 1 1 13 1 342 12 1 m 0.0 11 11 1 12 2 12 1 1 212 12 1 1 2 1 1 1 1 1 11 2 1 211 25 123 2 12 s 1 1 1 1 121 2 11 11 315 21 2 2 2 u 1 1 1 11 22 1 1 3 1 11 13 12 11 1 b 1 11 12 11 1 3 211 22 11 2 21 11 21 21 1 s 0.2 1 1 11 1 1 1 3 1 21 2 212 1 21 2 1 c 1 2 31 1 42 11 2 2 2 1 1 1 1 o 1 11 1 11 2 1 1 1 1 1 1 r 1 1 1 321 1 6 1 2 1 1 1 e 1 1 1 1 12 1 111 1 1 0.4 1 1 3 1 21 1 11 2 12 1 1 2 1 1 11 1 1 0.6 + ----------------------------------------------------------------------------------------20 25 30 35 40 45 50 55 60 65 70 75 Age Figure 4 4 Predicted m oral a ction f actor s cores on a ge Table 4 2 5 Mean followership f actor s cores and s tandard d eviations g rouped by c ulture t ype Culture N Mean SD Clan 189 30 1.33 Adhocracy 102 .1 1 1.36 Hierarchy 234 30 1.51 Market 65 .0 4 1.25

PAGE 86

86 Table 4 2 6 Summary followership ANOVA t able for c ulture t ype Source df MS F p Culture 3 13.40 6.86 .0002 S/RT 586 1.95 significant at p < .05 Table 4 2 7 Shaffer Holm p rocedure: followership f actor s cores and c ulture t ype Rank t C + t fw /2, C, N J Comparison 1 4.44 3 2.40 Clan vs. Hierarchy 2 2.50 3 2.40 Adhocracy vs. Hierarchy 3 1. 70 3 2.40 Clan vs. Market 4 1. 35 3 2.40 Hierarchy vs. Market 5 1. 12 2 2.25 Clan vs. Adhocracy 6 0. 6 7 1 1.96 Adhocracy vs. Market significant at p < .05 Table 4 2 8 Mean s ervice f actor s cores and s tandard d eviations g rouped by c ulture t ype Culture N Mean SD Clan 189 22 .89 Adhocracy 102 .0 6 .94 Hierarchy 234 .2 0 1. 06 Market 65 .0 4 1. 05 Table 4 2 9 Summary s ervice ANOVA t able for c ulture t ype Source df MS F p Culture 3 6.14 6.28 0.0003 S/RT 586 .98 significant at p < .05 Table 4 30. Shaffer Holm p rocedure: s ervice f actor s cores and c ulture t ype Rank t C + t fw /2, C, N J Comparison 1 4.28 3 2.40 Clan vs. Hierarchy 2 2.20 3 2.40 Adhocracy vs. Hierarchy 3 1.81 3 2.40 Clan vs. Market 4 1.28 3 2.40 Clan vs. Adhocracy 5 1.1 3 2 2.25 Hierarchy vs. Market 6 0. 65 1 1.96 Adhocracy vs. Market significant at p < .05

PAGE 87

87 Table 4 3 1 Mean c hallenge f actor s cores and s tandard d eviations g rouped by c ulture t ype Culture N Mean SD Clan 189 12 .97 Adhocr acy 102 .0 5 .99 Hierarchy 234 .1 5 1. 02 Market 65 17 .95 Table 4 3 2 Summary c hallenge ANOVA t able for c ulture t ype Source df MS F p Culture 3 3.22 3.2 7 0.02 S/RT 586 .98 significant at p < .05 Table 4 3 3 Shaffer Holm p rocedure: c hallenge f actor s cores and c ulture t ype Rank t C + t fw /2, C, N J Comparison 1 2.74 3 2.40 Clan vs. Hierarchy 2 2.24 3 2.40 Hierarchy vs. Market 3 1.37 3 2.40 Clan vs. Adhocracy 4 1.35 3 2.40 Adhocracy vs. Market 5 0.84 2 2.25 Adhocracy vs. Hierarchy 6 0. 32 1 1.96 Clan vs. Market significant at p < .05 Table 4 3 4 Mean t ransformation f actor s cores and s tandard d eviations g rouped by c ulture t ype Culture N Mean SD Clan 189 25 .86 Adhocracy 102 12 .97 Hierarchy 234 19 1.07 Market 65 22 .99 Table 4 3 5 Summary t ransformation ANOVA t able for c ulture t ype Source df MS F p Culture 3 8.24 8.5 8 < 0.0001 S/RT 586 .96 significant at p < .05

PAGE 88

88 Table 4 3 6 Shaffer Holm p rocedure: t ransformation s ubscores and c ulture t ype Rank t C + t fw /2, C, N J Comparison 1 4.58 3 2.40 Clan vs. Hierarchy 2 3.30 3 2.40 Clan vs. Market 3 2.68 3 2.40 Adhocracy vs. Hierarchy 4 2.17 3 2.40 Adhocracy vs. Market 5 1.06 2 2.25 Clan vs. Adhocracy 6 0. 19 1 1.96 Hierarchy vs. Market significant at p < .05 Table 4 3 7 Mean m oral a ction f acto r s cores and s tandard d eviations g rouped by c ulture t ype Culture N Mean SD Clan 189 14 .98 Adhocracy 102 0 9 .94 Hierarchy 234 12 1.03 Market 65 09 1.02 Table 4 3 8 Summary m oral a ction ANOVA t able for c ulture t ype Source df MS F p Culture 3 2.73 2.75 0.04 S/RT 586 .99 significant at p < .05 Table 4 3 9 Shaffer Holm p rocedure: m oral a ction f actor s cores and c ulture t ype Rank t C + t fw /2, C, N J Comparison 1 2.63 3 2.40 Clan vs. Hierarchy 2 1.73 3 2.40 Adhocracy vs. Hierarchy 3 1.62 3 2.40 Clan vs. Market 4 1.14 3 2.40 Adhocracy vs. Market 5 0.42 2 2.25 Clan vs. Adhocracy 6 0. 17 1 1.96 Hierarchy vs. Market signifi cant at p < .05

PAGE 89

89 Table 4 40. Pearson -r c orrelation c oefficients and s ignificance l evels for i nstitutional v ariables and TFP f actor s cores total_ r_subscore s_subscore c_subscore t_subscore m_subscore score Size -.0 1 -.0 6 .0 1 -.0 5 .0 01 -.0 3 .8 1 .18 87 .2 5 98 4 9 534 534 534 534 534 534 Population -.0 1 -.04 .0 2 -.0 4 .0 08 -.0 2 .8 2 35 62 .3 5 85 71 534 534 534 534 534 534 Urban _area -.0 3 -.0 1 .0 1 -. 06 04 -. 01 4 4 .8 2 7 5 19 30 80 534 534 534 534 534 534 Pop_center .0 3 01 0 3 003 -. 0 4 01 48 8 2 53 95 39 8 2 534 534 534 534 534 534 Degree_Offer -.0 5 02 01 -. 05 1 1 01 30 60 87 2 3 01* 81 534 534 534 534 534 534 significant at p < .05 Table 4 4 1 Summary m oral a ction f actor s core re gression t able for i nstitutional v ariables Variable B SE B Size 1.66E 5 1.47E 5 1 0 Population 8.17E 8 1.90E 7 .03 Urban_area 09 15 .0 5 Pop_center 10 16 .05 Degree_ o ffered 28 11 13 Table 4 4 2 Mean m oral a ction f actor s cores and s tandard d eviations g rouped by d egree o ffered Bachelor Degree Offered N Mean SD No 371 08 1.00 Yes 163 15 .99

PAGE 90

90 Table 4 4 3 Summary r esponsibility f actor s cores ANOVA t able with i nt eractions Source df MS F p Rank 3 .39 .42 74 Tenure 1 5.57 5.91 0 2 Age 1 7.26 7.70 .0 1* Sex 1 .57 .60 44 Race 1 13 .1 3 71 High degree 3 1.66 1. 76 15 Disc ipline 2 5.12 5.43 01* Years work 1 .28 .30 59 Tenure Years_Work 1 6.64 7.05 .0 1* Discipline Years_Work 2 3.05 3. 23 .0 4 Error 4 95 .94 significant at p < .05 -1.4 -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 Years Working in Higher Education Responsibility Factor Scores Tenured Non-tenured Figure 4 5 Predicted r espons ibility f actor s cores by t enure s tatus and y ears w orking in h igher e ducation

PAGE 91

91 -1.4 -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 Years Working in Higher Education Responsibility Factor Scores HEBS STEM HCPP Figure 4 6 Predicted r esponsibility f actor s cores by a cademic d iscipline and y ears w orking in h igher e ducation Table 4 44. Summary s ervice f actor s cores ANOVA t able with i nt eractions Source df MS F p Rank 3 3.20 3.27 .02* Tenure 1 6.65 6.77 .01* Age 1 3.11 3.18 .08 Sex 1 7.24 7.38 .01* Race 1 .00 .00 .95 High degree 3 1.1 5 1.16 .32 Disci pline 2 1.81 1.84 .16 Years work 1 .26 26 61 Age Tenure 1 5.23 5.33 .0 2 Error 497 .9 8 significant at p < .05

PAGE 92

92 -0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 Age Service Scores Tenured Non-tenured Figure 4 7 Predicted s ervice f actor s cores by t enure and a ge

PAGE 93

93 CHAPTER 5 DISCUSSION Th e purpose of this study wa s to test Chaleffs (1995) theory of c ourageous followership among Florida community college faculty members by relating followership : to the variables of faculty rank, receipt of tenure, age, sex race, education level, disciplin e and number of years working in higher education; to the way by which faculty members perceived the organizational cult ure of their institution; and to the community colleges size, population, location, and highest degree -offered This chapter presents a summary of the studys findings, limitations and future research considerations, and conclusions that can be drawn from these findings. Findings Survey Response 3,200 surveys were electronically mailed to a randomly selected sample of Florida community c ollege faculty members. 661 surveys were returned for a response rate of 20.7% This rate fell within the predicted response rate of 20 to 25 percent suggested by previous studies (Gerity, 1999; Phillips, 2002; Roark, 1988; Sheehan, 2001). Of the 661 retur ned surveys, 64 surveys were incomplete and were deleted. This left 597 surveys that were used in the data analyses. Respondents were similar to the Florida community college faculty population with respect to the factors of sex, race, and highest degree earned while tenured faculty were over -sampled (see Table 2). This difference may have been the result of collecting data during the summer semester when very few adjunct faculty teach (B. Sloan, personal communication, May 13, 2008) as well as relying on the institutions on -line faculty directories to build the sampling frame. Adjunct faculty as more likely than full time faculty to be missing from these listings (W. Leite, personal communication, July 10, 2008)

PAGE 94

94 Question 1 Faculty rank. Previous researc h studies indicated that individuals of higher professional status (of which academic rank served as an indicator) would more likely exhibit exemplary conduct (Abbott, 1983; Braxton & Bayer, 1999) and take responsibility to challenge known incidents of wrongdoing (Braxton & Bayer, 1996; Knight & Auster, 1999). Additionally, existing models of followership purport ed that attributions of followership var ied by organizational level with greater attributions corresponding to higher organizational levels (Chaleff, 1995; Wortman, 1982). Research by Dixon and Westbrook (2003) and Ray (2006) found support for this proposition. However, n o study had examined this claim with regard to community college faculty. The findings of this research contradicted those of the aforementioned studies In general, faculty of low rank (i.e., assistant professors) scored higher across all dimensions of followership than more senior faculty (i.e., associate and full professors). S tatistically significant differences were achieved for the service dimension between assistant and associate professors, t (498) = 9.31, p = .002, d = 4 6 and between and assistant professor and full professor, t (498) = 5.54, p = .02, d = .27, and for total followership factor scores between assistant profess ors and associate professors, t (498) = 5.60, p = .02, d = .28, and between assistant professors and full professors, t (498) = 4.58, p = .03, d = .20 The only dimension where senior faculty outscored junior academics was for the responsibility factor score s This difference was not significant. These findings suggested that junior faculty were more willing to assume additional tasks in service to the institution. Tenure. Prior research on tenure indicated that tenured faculty were more likely to sanction i nappropriate behaviors by other faculty (Braxton, Eimers, & Bayer, 1996) and were more likely to voice stronger disapproval of inviolable behaviors than untenured academics

PAGE 95

95 (Braxton & Bayer, 1999). Researchers also found that untenured faculty were more fe arful of being labeled a whistleblower, since the stigmatization could damage their professional standing and harm their chances for career advancement (Braxton & Bayer, 1999; OToole, 1978). However, more recent research by Williams and Ceci (2007) found that only when faculty member s w ere promoted from associate professor s with tenure to full professor s was there a significant increase in the ir willingness to speak freely, to teach courses unpopular with one's colleagues, to publish controversial research and to blow the whi stle on ethical transgressions. The current study found that untenured Florida community college faculty members tended to score higher than tenured faculty in all followership dimensions, except responsibility and transformation. Howe ver, s ignificant differences due to tenure status were found only for the dimensions of transformation, t (498) = 2.10, p = .04, d = .15 (tenured faculty scored higher than non tenured), and moral action, t (498) = 2.04, p = .04 d = .32 (nontenured facult y scored higher than tenured). This suggested that tenured faculty were willing to create environments supportive of change while untenured faculty members were more likely to speak out against inappropriate behaviors and play the role of whistleblower. Th ese findings more closely aligned to the findings of Williams and Ceci (2007) than those of the earlier studies (Braxton & Bayer, 1999; Braxton, Eimers, & Bayer, 1996). Age. Studies that considered age as an independent variable found that it exerted a co nsiderable influence on followership behaviors, with older participants scoring significantly higher than younger subjects (Colangelo, 2000; Dixon, 2003; Koo & Choi, 2000; Steyer, 2001). The findings of this study supported those of earlier research. Acros s all dimensions, older faculty scored higher than younger faculty with significant differences due to age found for the dimensions of respons ibility, t (498) = 2.75, p 2 = .01 ; challenge, t (498) = 2.99,

PAGE 96

96 p 2 = .02 ; moral a ction, t (498) = 2.83, p 2 = .02 ; and total followership, t (498) = 3.15, p 2 = .02 Older faculty members were more likely to take initiative without formal authority challenge inappropriate behaviors and remain devoted t o their personal ethics Sex. Researchers previously found that women faculty members express ed a greater commitment to teaching than their male counterparts (Bayer & Astin, 1975; Boice, 1992; Boyer, 1990; Finkelstein, 1984; Tierney & Rhoads, 1993). Among college faculty, researchers have noted that women were more likely than men to experience significant increases in their service obligations at their universities following tenure (Terosky, Phifer, & Neumann, 2008). Furthermore, researchers have found tha t women tend to voice stronger disapproval for condescending negativism and personal disregard, as well as a greater disdain for colleagues who demean co -workers (Braxton & Bayer, 1999). These findings suggested that women should exhibit higher followershi p scores than men. However, previous research conducted on followership behaviors generally did not discern any statistically significant differences due to gender (Colangelo, 2000; Dixon, 2003; Geist, 2001). The one exception was Steyer (2001), who exami ned followership behaviors among primary and secondary school teachers. She found that women consistently displayed statistically greater followership behaviors than men. Our findings supported those of Steyer (2001). Female faculty members scored higher than males across all followership dimensions, with statistically significant differences for service, t (498) = 2.82, p = .005, d = .26 ; transformation, t (498) = 3.00, p = .003, d = .3 4 ; and total followership, t (498) = 2.12, p = .03, d = .22. These findings suggest female community college

PAGE 97

97 faculty were more likely to demonstrate a stronger willingness to express support for their supervisors and co -workers, as found in the earlier research of Braxton and Bragg (1999). Race. Though minority group member fa culty tended to score higher than non-minority group member faculty in all followership dimensions except responsibility and service, none of the differences were significant. These results were similar to those of Geist (2001). Education level. In resear ch on followership behaviors, Colangelo (2000), Pierce (2002), and Koo and Choi (2000) found significant differences due to respondents educational level. Ackerman (1985) linked increased education to corresponding gains in followership. Alternatively, findings from Geists (2001) and Dixons (2003) studies contested these earlier results. This study found no discernible pattern in followership scores due to education level. Faculty with an associates degree scored highest among the groups in the service and moral action dimensions but lowest for the responsibility, challenge, and transformation dimensions. Those with a bachelors degree scored highest in the challenge and transformation and total followership dimensions. The masters degree group scored lowest for the total followership score dimension. Faculty with a doctorate scored highest in the responsibility dimension but lowest for the service and moral action dimensions. Significant differences due to educational level were found only for the responsibility dimension between faculty with a doctorate and those with a masters degree t (498) = 2.53, p = .01, d = .26. Small samples of associates degree holders ( n = 17) and those with bachelors degrees ( n = 40) may have influenced these results. Acad emic discipline. In earlier research, discipline accounted for a larger proportion of the variance explained in faculty disapproval of inappropriate behaviors than individual faculty characteristics such as administrative experience, gender, and research a ctivity (Braxton &

PAGE 98

98 Bayer, 1999). In followership research, both Steyer (2001) and Koo and Choi (2000) attributed significant differences in followership behaviors to academic discipline. Our study found that faculty members with a background in math, scie nce, engineering, or technology scored the lowest across all followership dimensions. Faculty members with a background in humanities, business, education, or social sciences scored the highest in the dimensions of responsibility, challenge, and total foll owership while faculty with a background in health, culinary arts, public safety, or performing arts scored highest in service, transformation, and moral action. Significant differences due to discipline appeared between faculty in humanities, business, education, or social sciences and faculty in math, science, engineering, or technology for the dimensions of responsibility, t (498) = 10.85, p = .001, d = .32 ; service, t (498) = 4.04, p = .04, d = .22; challenging, t (498) = 4.87, p = .03, d = 2 2; transforma tion, t (498) = 5.68, p = .02, d = 25; and total followership, t (498) = 7.92, p = .005, d = .31 In each case, humanities, business, education, or social sciences faculty scored significantly higher than math, science, engineering, or technology faculty. T hese findings suggested humanities, business, education, or social sciences faculty were more likely to be self -managed, assume initiative, and keep commitments. Years working in higher education. Earlier research on followership was split on the effect o f time spent with an organization. Dixon (2003) and Koo and Choi (2000) reported a significant effect due to time employed while Colangelo (2000), Geist (2001), and Steyer (2001) did not find employment duration to have a significant effect on followership behaviors. The results of the current study did not find any significant relationships in any of the followership factor scores associated with the number of years working in higher education institutions.

PAGE 99

99 Question 2 Though several studies examined the r elationship between leadership styles and followership behaviors (Brown & Thornborrow, 1996; Colangelo, 2000; Geist, 2001) no study examined the relationship between organizational culture and courageous followership. Each of the earlier studies found sig nificant relationships between preferred leadership styles and followership. In a similar manner, t he current study found significant relationships between organizational culture and followership behaviors Faculty members who rated their institutional culture as a Clan scored the highest across all followership dimensions, except responsibility, while faculty who scored their culture as a Hierarchy scored the lowest across all dimensions. Statistically significant differences were found in comparing Clan v s. Hierarchy cultures in service, t (586) = 4.28, p < .0001, d = .4 3 ; challenging, t (586) = 2.74, p = .01, d = .27 ; transformation, t (586) = 4.58, p < .0001, d = 45; moral action, t (586) = 2.63, p = .01, d = .2 6 ; and total followership, t (586) = 4.44, p < .0001, d = .42. Additionally, statistically significant differences were found for Adhocracy vs. Hierarchy cultures in transformation t (586) = 2.68, p = .01, d = .30 ; and total followership, t (586) = 2.50, p = .01, d = .28. Finally, statistically signifi cant differences were found for Clan vs. Market cultures in transformation, t (586) = 3.30, p = .001, d = .5 3 These findings suggested that courageous followership behaviors are promoted in institutions exhibiting Clan culture characteristics while suppres sed in Hierarchical cultures. Question 3 The only significant difference among factor scores due to institutional factors was found in the moral action dimension for the variable of degree offered. Faculty from institutions that offered bachelors degrees scored significantly higher for moral action than faculty from institutions that did not offer bachelors degrees. In earlier studies, Braxton and Bayer (1999)

PAGE 100

100 found that academics in two -year colleges tended to exhibit significantly lower levels of disapp roval for inappropriate behaviors than faculty members in other institutional settings. Steyer (2001) found that school level, which she defined as either elementary, middle, or high school, had a significant effect on followership behaviors among primary and secondary school teachers. In the present study, significant differences for moral action factor scores were found for the variable of degree offered, t (528) = 2.55, p = .01, d = .23. Faculty from institutions that offered bachelors degrees scored sig nificantly higher for moral action than faculty from institutions that did not offer bachelors degrees. These findings suggested that faculty from bachelors degree -granting institutions would be less willing to compromise their personal ethics for continued employment Additional Analyses Potential two -way interactions were tested. Significant interactions in the responsibility dimension for tenure by years working, F (1, 495) = 7.05, p 2 = .01, and for discipline by years working, F (2, 495) = 3.23, p 2 = .01, were found. Analysis indicated that responsibility factor scores of tenured faculty declined over time faster than nontenured faculty factor scores Analysis also indicated that responsibility factor scores improved as employment duration grew among faculty with backgrounds in the math, science, technology, and engineering disciplines while faculty with backgrounds in humanities business, education, or social sciences and in health, culinary arts, public safety, or performing arts showed declining responsibility factor scores as their years in higher education increased. In the service dimension, a significant age by tenure interaction, F (1, 497) = 5.33, p = .02, 2 = .01 was found A nalysis indicated that among young faculty members, tenured faculty ha d lower service factor scores than non tenured faculty. As age increased tenured faculty service factor score increased while nontenured fac ulty remain ed almost constant.

PAGE 101

101 Limitations and Future Research Considerations The major limitation of this study was the inability to establish causation due to its ex post facto design. It looked only at the degree of association between multiple variable s T herefore, inferences of causality were inappropriate. Although significant relationships existed between followership behaviors and a number of variables, the direction of influence was not apparent. For example, do individuals who obtain a degree in h umanities, business, education, or social sciences develop stronger followership behaviors than their peers who obtain degrees in other fields or are individuals with stronger followership behavior s drawn to the fields of humanities, business, education, o r social sciences? This question and questions like it were unanswerable in this study as the problems of reverse causality and spurious correlations may have exist ed Therefore, our findings did not provide evidence of causal relationships, only informati on about the degree and shape of the relationship between the variables of interest. A more controlled experimental design with manipulation of variables may help understand the directional influences between followership behaviors and the individual and i nstitutional factors examined in this study. Additionally, the use of structural equation modeling could increase the understanding of direction of influence between followership and individual and institutional variables. S everal issues related to partici pants created limitations to this study. Since all participants were Florida community college faculty members, the generalizability of this study is limited. The possibility of cultural implications unique to this population might have influenced the fi ndings. Although this suggests continued research on faculty members from other educational institutions is needed it should also be noted that the individual differences found in this study may not be unique to this setting. As sta ted earlier in this cha pter, many of this studys findings supported those of other followership studies. In other words, discounting the relationship

PAGE 102

102 between age and followership behaviors in other institutions simply because they are not Florida community colleges seems unwise Participation in this study was limited to Florida community college faculty. Numerous differences between community college and university faculty have been identified in the literature. The proportion of men is lower than in universities, higher than in secondary schools. Most of the faculty members hold academic masters degrees they are less likely to hold advanced graduate degrees than university professors are. Their primary responsibility is to teach. They rarely conduct research or scholarl y inquiry. They are more concerned with subject matter than are their counterparts in secondary schools, less so than university professors. On a full time basis they conduct four or five classes per term Sixty percent are part time employees at thei r colleges. Many, both full and part timers, sustain other jobs in addition to their teaching (Cohen & Brawer, 1989). Future research could continue to explore these differences and their effects on followership behaviors. In particular, some Florida co mmunity colleges are starting to offer four -year bachelors of applied science degrees. In t his study faculty from community colleges that offered bachelors degrees scored significantly higher for moral action than faculty from institutions that did not offer bachelors degrees. Other researchers found that faculty in two -year colleges tended to exhibit significantly lower levels of disapproval for inappropriate behaviors than faculty members in other institutions (Braxton & Bayer, 1999). Further research to explore additional potential differences within institutions between faculty working in four year programs and those working in two -year and certificate programs could provide additional clarity to the emerging picture of community college faculty. S ixty -four surveys were deleted from analyses due to incompleteness. Most of the incomplete surveys went unfinished beginning with TFP questions on courage to take moral action or during the IPS cultural scenarios. It would appear that these individuals cea sed answering questions because they either did not understand what was being asked (a likely

PAGE 103

103 explanation for the IPS cultural scenarios) or felt threatened by what the questions asked about their followership behaviors (a likely explanation for the moral action questions). Stopping at these points meant that no biographical information on the respondents was collected. Had the data from these surveys been available, the findings of this study may have been altered. Although experts recommend beginning ques tionnaires with the most interesting items and placing duller, demographic data (age, gender, and the like) at the end of a self administered questionnaire (Babbie, 2007, p. 256), it may have resulted in addition al data for analyses in this instance if the demographic data had been collected during the first portion of the questionnaire. Also, the coefficient alphas obtained for responsibility, challenge, transformation, and moral action were below the traditional target of .7 noted by Fraenkel and Wall en (1996) and Hair, Anderson, Tathem, and Black (1998). Increasing the number of items in the dimensions falling short of the target would increase their coefficient alphas. Table 44 provides the number of items needed to reach the target .7 coefficient al pha for each TFP dimension. An additional limitation concerning the participants of this study was the small sample of racial minority groups ( n = 68) and holders of associates ( n = 17) and bachelors degree ( n = 40). Other followership researchers have e xperienced similar issues regarding small samples of minority group members (Dixon, 2003; Geist, 2001). Small sample sizes limit the ability of researchers to detect anything but relatively large differences. Though the data appeared to indicate that betwe en -group differences may have existed for some followership dimensions, the small samples likely contributed to these differences not reaching significance. Also, the tenure status of the sample was not representative of Floridas community college faculty population, possibly contributing to the existence of a sampling bias (Weiten, 2001). Collecting data during the summer semester when many adjunct faculty were not working as well as relying on

PAGE 104

104 institutional websites to build the sampling frame may have c ontributed to this shortfall. Other researchers studying college faculty may realize more representative samples by collecting data during the fall or spring semesters. Nonetheless, this study was able to collect sufficient data that described several sign ificant relationships and differences between followership behaviors and the Florida community college faculty population. A final limitation of the current study concerns the biases of mono method (Cook & Campbell, 1979) and social desirability (Weiten, 2001). This study utilized self report measures for the TFP followership dimensions and IPS cultural scenarios for gathering sample data. No other means to assess followership or institutional culture were utilized. Weiten (2001) reported that self reports may be plagued by a tendency for respondents to give socially approved answers to questions about oneself. Future studies might incorporate observer ratings or supervisor evaluations in assessing followership behaviors and address the relation of self and observers ratings (Funder & West, 1993) to minimize the possibility of mono method and self -report biases. A measure such as the Marlow Crowne Social Desirability Scale might also be utilized to assess for social desirability bias. Conclusions This resea rch provided additional support to several existing studies on followership and its relation to demographic variables such as age, sex, educational level, duration of employment, and educational discipline. In these areas, this study suggested that Florida community college faculty members were similar to workers in health care, the military, technology, administration, and primary and secondary education. However, this study also contradicted earlier studies in the areas of rank, tenure, and employment dur ation. These findings have implications for community college leaders as they work to increase the performance and accountability of their organizations. Understanding faculty followership

PAGE 105

105 behaviors can help administrators manage the challenges of reduced resourcing, changing faculty demographics, rising accountability, and worker downsizing that educational institutions face in the evolving environment of higher education. Community college leaders whose institutions are undergoing change should understan d that tenured and female faculty members are more likely to provide greater levels of support then non -tenured or male faculty. In addition, leaders should recognize that female faculty and assistant professors are more likely to voluntarily assume additi onal responsibilities in service to the institution while older faculty, those with masters degrees, or whose background is in h umanities, b usiness, e ducation, or social sciences are more likely to demonstrate a sense of ownership for themselves and their organizations as opposed to younger scholars, those with doctorates, or those who studied math, science, engineering, or technology. Also, community college administrators should acknowledge that their institutions cultures may influence the followershi p behaviors of their faculty. Colleges whose cultures were characterized as having high flexibility, individuality, and spontaneity earned higher followership scores across all dimensions than institutions whose cultures were known for their stability, con trol, and predictability. Schein (2004) asserted that one of leaderships crucial roles is to shape the culture of an organization. Following this contention, community college administrators should attempt to build institutional cultures that value flexib ility, individuality, and spontaneity by recognizing, acknowledging, and rewarding such behaviors by organizational members, and by recruiting, selecting, and promoting internal administrators who appreciate these same characteristics. Finally, community college leaders must recognize that their institutional members followership behaviors may evolve as their institutions become bachelors degree -granting

PAGE 106

106 institutions. Earlier studies found that four year college faculty tended to exhibit stronger levels of disapproval for inappropriate behaviors than their two -year college counterparts (Braxton & Bayer, 1999). The present study made similar findings. Administrators must remember that changes to one area are often followed by unintended changes to other ar eas. When these changes are desirable, the organization benefits. However, unwanted outcomes may also be the result of these alterations. For example, increasing the number of untenured faculty relative to tenured scholars may result in the college having members more willing to speak out against inappropriate behaviors and play the role of whistleblower while concurrently decreasing the number of constituents who are more strongly supportive of institutional change. The role of followership among community college faculty is increasing as workforces downsize and requirements escalate. Administrators continue to look to their faculty members to do more in support of institutional requirements. These leaders must develop a better understanding of followership and how their institutional faculty respond. This study made an important contribution to understanding followership in the community college environment by helping to define how faculty members view ed themselves as followers. Th ese contributions may help community college leaders develop more positive relationships with their faculty members that in turn, can assist in advanc ing programmatic and institutional change agendas.

PAGE 107

107 Table 5 1 TFP i tem c hanges to i ncrease r eliability Coefficient Desired Incr ease in Initial no. Required no. a lpha reliability n needed for of items of items for desired K desired K Service .7 4 .7 .83 5 5 Transform ation .67 .7 1.16 4 5 Challeng ing .60 .7 1.54 4 7 Responsib ility .60 .7 1.55 6 10 Moral action .5 4 .7 2.01 4 9 Total items 23 36

PAGE 108

108 APPENDIX A SAMPLE SIZE CALCULAT ION (A lgina 2008) *This program can be used for power calculations for betweensubjects main effects in completely betweensubjects and in splitplot designs. Approximations are used to avoid having to specify all variances and covariances for the repeated measures. Compound symmetry is assumed. As a result of the appro ximations, the n will not be exactly correct for split plot designs; data one; es=.35 ; *Effect size for largest and smallest mean; alpha=.05 ; *Type I error rate; j=3 ; *Number of levels for the betweensubjects factor for which es is defined. j is limited to 6 or less; tpower= .80 ; *Target power; minn= 002 ; *Minimum sample size for calculations; maxn= 2000 ; *Maximum sample size for calculations; prodw=1 ; *Product of number of levels of all withinsubjects factors. Equals p (the number of levels for the withinsubjects factor) when there is just one within subjects factor. Equals 1 when there are no withinsubjects factors; prodb=2 ; *product of number of levels of the other betweensubjects factors. Equals 1 for a design with one betweensubjects factor design and equals the number of level in the second factor if there are two between subjects factors; rho= .0 ; *correlation for withinsubjects variables. Equals zero if there are no within subjects factors; Note that mus are in standard deviation units Obs comment 1 analysis ok 2 analysis ok Obs config 1 maximum range, one extreme mean, and two extreme means 2 minimum range a nd equally spaced Obs prodw prodb rho es j alpha tpower power 1 1 2 0 0.35 3 0.05 0.8 0.8 2 1 2 0 0.35 3 0.05 0.8 0.8 Obs mu1 mu2 mu3 cell_n lambda cval 1 0. 18 0.18 0.18 60 9.8 3.02 2 0.18 0.00 0.18 80 9.8 3.01 Total participants = 3 rank levels 2 tenure levels 80 participants per cell = 480 Total Participants

PAGE 109

109 APPENDIX B T HE FOLLOWERSHIP PROF ILE ABBREVIATED (Dixon, 2003) No Slight Moderate Great Very great E xtent Extent Extent Extent Extent 1. I create a supportive environment at my workplace in which change can occur. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 2. I am passionate about my work ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 3. I acknowledge improvements that my supervisor has made. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 4. I provide support to my supervisor for experimentation and learning. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 5. I prepare to transfer responsibilities should severance become ne cessary. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 6. I am self -managed in meeting deadlines and keeping commitments. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 7. I reflect my supervisors val ues to the organization without injecting my own personal agenda. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 8. I assume responsibility in dilemmas where rules impede service. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 9. I would resign to protect my supervisor from the repercussions of my actions. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 10. I establish coping mechanisms that reinforce transformational progress. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 11. I summarize communications for my ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) supervisor. 12. I minimize unnecessary pressure on my ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) supervisor. 13. I defend my supervisor from unwarranted ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) attacks. 14. I encourage complainers to communicate concerns not emotions. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 )

PAGE 110

110 15. I confront groupthink. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 16. I assess my own performance. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 17. I take initiative without formal authority. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 18. I challenge inappropriate behaviors. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 19. I am willing t o bend the rules to get the right things done. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 20. I will not compromise my personal ethics for continued employment. ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) 21. I do what I believe is right even when there ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) may be negative repercussions. 22. I give honest feedback to my supervisor on ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) his/her behaviors. 23. I would resign rather than violate important ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) personal values.

PAGE 111

111 APPENDIX C PERMISSION TO USE THE FOLLOWERSHIP PROFI LE On Fri Nov 10 16:27: 29 EST 2006, "Dixon, Gene" wrote: You have permission to use TFP (The Followership Profile), provided you forward a copy of your collected raw data to me. The data will be compiled with other data received from users like yourself in an on -going effort to update supportive metrics. Additional studies beyond your doctoral study will require additional permission. TFP along with the original validation and reliability estimates are available from: UMI Corporation, 300 North Zeeb Road, Ann Arbor MI 48106 8005210600 x7020 7349974113 (F) disspub@umi.com Registration number TX5 848482 Gene Dixon, MBA, PhD Director, ECU, Inc. College of Technology and Computer Science Department of Engineering East Carolina Un iversity Greenville NC 27858 2527371031 (O) 2527371041 (F)

PAGE 112

112 APPENDIX D IPS CULTURAL SCENARIOS (Smart & St. John, 1996) These questions relate to the type of organization your institution is most like. Each of these items contains four descriptions of institutions of higher education. Please distribute 100 points among the four descriptions depending on how similar the description is to your school. None of the descriptions is any better than the others; they are just different. For each description u se all 100 points. For example, in question 1, if Institution A seems very similar to mine, B seems somewhat similar, and C and D do not seem similar at all, I might give 70 points to A 30 points to B 0 points to C, and 0 points to D 1. Institutional characteristics (Please distribute 100 points). __________ Institution A is a very personal place. It is like an extended family. People Points for A seem to share a lot of themselves. __________ Institution B is a very dynamic and entrepreneurial pla ce. People are willing to Points for B stick their necks out and take risks. __________ Institution C is a very formalized and structured place. Bureaucratic procedures Points for C generally govern what people do. __________ Institution D is productio n oriented. A major concern is with getting the job Points for D done. People arent personally involved. 2. Institutional leader (Please distribute 100 points). __________ The head of Institution A is generally considered to be a mentor, sage, or a P oints for A father or mother figure. __________ The head of Institution B is generally considered to be an entrepreneur, an Points for B innovator, or a risk taker. __________ The head of Institution C is generally considered to be a coordinator, an Po ints for C organizer, or an administrator. __________ The head of Institution D is generally considered to be a producer, a technician, Points for D or a hard driver.

PAGE 113

113 3. Institutional glue (Please distribute 100 points). __________ The glue that holds Institution A together is loyalty and tradition. Points for A Commitment to this school runs high. __________ The glue that holds Institution B together is a commitment to innovation and Points for B development. There is an emphasis on being fi rst. __________ The glue that holds Institution C together is formal rules and policies. Points for C Maintaining a smooth running institution is important here. __________ The glue that holds Institution D together is the emphasis on tasks and goal Points for D accomplishment. A production orientation is commonly shared. 4. Institutional image (Please distribute 100 points). __________ Institution A emphasizes human resources. High cohesion and morale in the Points for A school are important. __________ Institution B emphasizes growth and acquiring new resources. Readiness to Points for B meet new challenges is important. __________ Institution C emphasizes permanence and stability. Efficient, smooth Points for C operations are important. __________ Institution D emphasizes competitive actions and achievement. Measurement Points for D goals are important.

PAGE 114

114 APPENDIX E PERMISSION TO USE CULTURAL SCENARIOS On Tue Feb 05 12:51:45 EST 2008, John C Smart wrote: Greetin gs, The items I used (from the work of Kim Cameron at the University of Michigan) are not copyrighted and thus you should feel free to use them in your dissertation research. Good luck on your dissertation research. John Smart

PAGE 115

115 APPENDIX F INTRODUCTOR Y EMAIL AND CONSENT FORM Dear faculty member, Please participate in an online survey that Scott Smith, a University of Florida doctoral candidate, is using to collect data for his dissertation. The survey examines followership behaviors among community c ollege faculty members. Completing this survey should take no more than 10 minutes of your time. By clicking the link below and submitting the survey, you are consenting to voluntarily participate in the study: http://www.surveymonkey.com/s.aspx?sm= Your responses will be anonymous and there will be no identifying information associated with them. We are using a host provider that does not maintain the Internet addresses of respondents, so we will h ave no way of knowing who has responded and who has not. There are no risks or compensation involved with participation. If you decide to take part in the study, you do not have to answer any questions that you do not wish to answer; just skip to the next question. If you have any questions about this study, please contact Scott Smith at jssmit13@ufl.edu or Dr. David Honeyman at daveh@coe.ufl.edu Thank you in advance for participating in this survey. Your time and help are greatly appreciated. Sincerely Scott Smith and David Honeyman, Ph.D. Department of Educational Administration and Policy University of Florida (352) 3922391 ext. 272 If you have any questions about your rights as a participant in a research project, you can contact the UFIRB Office Box 112250, University of Florida, Gainesville, FL 326112250; tel. (352) 3920433.

PAGE 116

116 APPENDIX G INSTITUTIONAL REVIEW BOARD LETTER

PAGE 117

117 APPENDIX H DEMOGRAPHICS QUESTIONNAIRE Please provide the following demographic information. 1. What is your title? a Inst ructor b Assistant Professor c Associate Professor d Professor e Other (please list ____________________ ) 2. Are you on continuing contract (tenured) ? a Yes b No 3. What is your age? __________ 4. What is your sex ? a Male b Female 5. What is your race? a American India n or Alaska Native b Asian c Black or African American d Native Hawaiian or Other Pacific Islander e White f Other 6. Are you Hispanic or Latino? a Yes b No 7. What is your highest attained education level? a High school graduate or GED b Associate (2 year) degree c Bachel ors (4 -year) degree d Masters degree e Doctoral degree 8. From what field did you obtain your highest degree? a Humanities business, education, or social sciences b Math, science, technology, or engineering c Health, culinary arts, public safety, or performing arts d Other (please list ____________________ )

PAGE 118

118 9. How many years have you worked in higher education? __________ 10. What is the name of your institution? ____________________

PAGE 119

119 LIST OF REFERENCES Abbott, A. (1983). Professional ethics. American Journal of Sociology, 88(5), 855885. Aburdene, P. (1993). An American imperative: Higher expectations for higher education. Report of the Wingspread Group on Higher Education. Racine, WI: Johnson Foundation. Ackerman, L. (1985). Leadership vs. managership. Leader ship and Organization Development Journal, 6(2), 17 19. Agresti, A. & Finley, B. (1999). Statistical methods for the social sciences (3rd ed.). Upper Saddle Rive, NJ: Prentice Hall. Alcorn, D. S. (1992). Dynamic followership: Empowerment at work. Management Quarterly, 33(1), 9 13. Alfred, R. L. & Carter, P. (1999). New colleges for a new century: Organizational change and development in community colleges. In J. Smart (Ed.), Higher e ducation: Handbook of t heory and r esearch (pp. 240287), New York: Agathon Press. Alfred, R. & Carter, P. (2006). Contradictory colleges: Thriving in an era of continuous change (Issue paper #5). Washington, DC: American Association of Community Colleges. Algina, J. (2008). SAS p ower p rogram: Between -s ubjects m ain e ffects. Unive rsity of Florida, Gainesville, FL. Retrieved February 7, 2008 from http://plaza.ufl.edu/algina/pwr.between.factors.sas American Association of University Professors (1966). Joint statem ent on government of colleges and universities. AAUP Policy Documents and Reports. Washington, DC: AAUP. Argyris, C. (1957). Personality and organization. New York: Harper & Row. Asendorpf, J. B., & Wilpers, S, (1998). Personality effects on social relati onships. Journal of Personality and Social Psychology, 74, 15311544. Attridge, R. (1949, March 5). Children need training in followership; but most educators prefer leadership. The Saturday Evening Post, 221, 12. Babbie, E. (2007). The practice of social research (11th ed.). Belmont, CA: Wadsworth/Thomson Learning. Baker, S. D. (200 7 ). Followership: The theoretical foundation of a contemporary construct. Journal of Leadership and Organizational Studies, 14(1), 50 60. Banutu Gomez, M. B. (2004). Great leaders teach exemplary followership and serve as servant leaders. The Journal of American Academy of Business, 4(1/2), 143151.

PAGE 120

120 Bayer, A. E., & Astin, H. S. (1975). Sex differentials in the academic reward system. Science, 188(4190), 796802. Bennis, W. (2008). Introduction. In R. E. Riggio, I. C. Chaleff, & J. L. Blumen (Eds.), The art of followership (pp. xxiii -xxvii), San Francisco, CA: JosseyBass. Bensimon, E., Neumann, A., & Birnbaum, R. (1989). Making sense of administrative leadership: The L word i n higher education (ASHE ERIC Higher Education Report No. 1). Washington, DC: George Washington University. Bergquist, W. H. (1992). The four cultures of the academy: Insights and strategies for improving leadership in collegiate organizations. San Francis co, CA: Jossey Bass. Biglan, A. (1973). The characteristics of subject matter in different academic areas. Journal of Applied Psychology, 57(3), 195203. Birnbaum, R. (1988). How colleges work: The cybernetics of academic organization and leadership. San Francisco, CA: Jossey Bass. Birnbaum, R. (1989). Leadership and followership: The cybernetics of university governance. In J. H. Schuster & L. H. Miller (Eds.), Governing t omorrows c ampus: Perspective and a gendas (pp. 27 41), New York: Macmillan. Bjugstad K., Thach, E. C., Thompson, K. J., & Morris, A. (2006). A fresh look at followership: A model for matching followership and leadership styles. Institute of Behavioral and Applied Management 7 (3), 304319. Blackshear, P. B. (2002). The followership conti nuum: A model for increasing organizational productivity. Unpublished manuscript. Retrieved December 10, 2006 from http://www.innovation.cc/discussion -papers/blackshear emp.pdf Blackshear, P. B. (2003). The followership continuum: A model for fine -tuning the workforce. Public Manager, 32(2), 2530. Boice, R. (1992). The new faculty member: Supporting and fostering professional development. San Francisco CA : Jossey -Bass. Bok, D. (1992). Reclaiming the public trust. Change, 24(4), 13 20. Boyer, E. L. (1990). Scholarship reconsidered: Priorities for the professoriate. Princeton, NJ: Carnegie Foundation for the Advancement of Teaching. Bragg, D. D. (2004). Leading forward: An interview with Nan Ottenritter. On Research and Leadership Update, 16(1), 1 4. Braxton, J. M., & Bayer, A. E. (1996). Personal experiences of research misconduct and the response of individual academic scientists. Science, Technology, and Human Values, 21(2), 19 8 213.

PAGE 121

121 Braxton, J. M., & Bayer, A. E. (1999). Faculty misconduct in collegiate teaching. Baltimore, MD: Johns Hopkins Press. Braxton, J. M., Eimers, M. T., & Bayer, A. E. (1996). The implications of teaching norms for the improvement of undergraduate educa tion. Journal of Higher Education, 67(6), 603625. Breneman, D. W. (2002, June 14). For colleges, this is not just another recession. The Chronicle of Higher Education, 48(14), B7 Brewer, J. K. (1996). Introductory statistics for researchers (6th ed.). Edina, MN: Burgess. Brown, A. D., & Thornborrow, W. T. (1996). Do organizations get the followers they deserve? Leadership and Organization Development Journal, 17(1), 5 11. Brown, T. (1995). Great leaders need great followers. Industry Weekly, 244 (25), 283 0. Buhler, P. (1993). The flip side of leadership cultivating followers. Supervision, 54(3), 1719. Cameron, K. S., & Ettington, D. R. (1988). The conceptual foundation of organizational culture. In J. C. Smart (Ed.), Higher e ducation: Handbook of t heory and r esearch (Vol. 4, pp. 356396), New York: Agathon Press. Campbell, D. J. (2000). The proactive employee: Managing workplace initiative. The Academy of Management Executive, 14(3), 52 66. Carson, C. (n.d.). The effective use of effect size indices in institutional research. Keene State College, Keene, NH. Retrieved January 7, 2008 from http://www.keene.edu/ir/effect_size.pdf Ceci, S. J., Williams, W. M., & Mueller Johnson, K. (2006). Is tenure justified: An experimental study of faculty beliefs about tenure, promotion, and academic freedom. Behavioral and Brain Sciences, 29(6), 553 594. Chaleff, I. (1995). The courageous follower: Standing up to and for our leaders. San Francisco: Berrett -Koeh ler. Cohen, A. M., & Brawer, F. B. (1989). The American community college (2nd ed.). San Francisco CA: Jossey Bass. Cohen, J. (1969). Statistical power analysis for the behavioral sciences. New York: Academic Press. Colangelo, A. J. (2000). Followership and leadership styles. Unpublished doctoral dissertation. University of Oklahoma. Cook, J. L. (1998). Change and the lost art of followership. Fire Engineering, 151(9), 150153. Cook, T. D., & Campbell, D. T. (1979). Quasi -experimentation: Design and analy sis issues for field settings. Boston: Houghton Mifflin.

PAGE 122

122 Crockett, W. J. (1981). Dynamic subordinancy. Training and Development Journal, 35(5), 155164. Davis, J. R. (2003). Learning to lead: A handbook for postsecondary administrators. Westport, CT: Praeg er. Densten, I. L., & Gray, J. H., (2001). The links between followership and the experiential learning model: Followership coming of age. The Journal of Leadership Studies, 8(1), 6976. Diamond, R. M. (1996, January 5). What it takes to lead a department The Chronicle of Higher Education, B1 B2. Dixon, E. N. (2003). An exploration of the relationship of organizational level and measures of follower behaviors. Unpublished doctoral dissertation. The University of Alabama in Huntsville. Dixon, G. (2008). Ge tting together. In R. E. Riggio, I. C. Chaleff, & J. L. Blumen (Eds.), The art of followership (pp. 155176), San Francisco, CA: JosseyBass. Dixon, G., & Westbrook, J. (2003). Followers revealed. Engineering Management Journal, 15(1), 1925. Etzioni, A. ( 1986). Leaders control and members compliance. In M. T. Matteson & J. M. Ivancevich (Eds.) M anagement c lassics (3rd ed., pp. 227233). Plano, TX: Business Publications. Field, K. (2006, June 2). A Texas millionaire plots the future of higher education. The Chronicle of Higher Education, 52(39), A16. Finkelstein, M. J. (1984). The American academic profession. Columbus, OH: Ohio State University Press. Florida Department of Education. (2007). The fact book: Report for the Florida community college system Tallahassee, FL: Department of Education. Florida Department of Education. (200 8 ). Community college bachelors degree programs. Florida Department of Education Tallahassee, FL Retrieved April 23, 2008 from http://www.fldoe.org/cc/students/bach_degree.asp Follett, M. P. (1949). The essentials of leadership. In L. Urwick (Ed.) Freedom & c oordination: Lectures in b usiness o rganisation (pp. 47 60). London: Management Publications Trust. F raenkel, J. R., & Wallen, N. E. (1996). How to design and evaluate research in education (3rd ed.). New York: McGraw Hill. Funder, D. C., & West, S. G. (1993). Consensus, self -other agreement, and accuracy in personality judgment: An introduction. Journal of Personality, 61(4), 457476.

PAGE 123

123 Gasaway, R. B. (2006). The leader -follower relationship. Fire Engineering, 159(7), 1213. Gast, J. (2003). The 2003 HBR List: Breakthrough ideas for tomorrows business agenda. Harvard Business Review, 81 (4), 92 98. Geist, A. L. (2001). Leadership and followership in NCAA Division II athletic directors. Unpublished doctoral dissertation. The Ohio State University. Gerity, P. (1999). A study to identify community college workforce training and development professionals percei ved competencies and their perceived professional development needs. Unpublished doctoral dissertation. The Pennsylvania State University. Gilbert, G. R., & Hyde, A. C. (1988). Followership and the federal worker. Public Administration Review, 48 (6), 962968. Goffee, R. & Jones, G. (2006). Why should anyone be led by you? Boston: Harvard Business School Press. Gouldner, A. W. (1957). Cosmopolitans and locals: Toward an analysis of latent social roles. Administrative Science Quarterly, 2 (3), 281306. Guilfo rd, J. P. (1959). Personality. New York: McGraw Hill. Hair, J. F., Anderson, R. E., Tathem, R. L., & Black, W. C. (1998). Multivariate data analysis (5th ed.). Upper Saddle River, NJ: Prentice Hall. Hemsley, A. (2001). Willpower vs. skill power: Defeating self -doubts and restoring self confidence are essential to achieve greater success and happiness. Research, 24 (11), 26. Hersey, P., Blanchard, K. H., & Johnson, D. E. (2007). Management of organizational behavior: Leading human resources (8th ed.). Upper S addle River, NJ: Prentice Hall. Hughes, R. L., Ginnett, R. C., & Curphy, G. J. (2001). Leadership: Enhancing the lessons of experience (3rd ed.). Boston: McGraw Hill. Jang, K. L., McRae, R. R., Angleitner, A., Rieman, R., & Livesley, W. J. (1998). Heritabi lity of facet level traits in a cross -cultural twin sample: Support for a hierarchical model of personality. Journal of Personality Traits and Social Psychology, 74, 15561565. Kelley, R. E. (1988). In praise of followers. Harvard Business Review, 66 (6), 1 42148. Kelley, R. E. (1992). The power of followership: How to create leaders people want to follow, and followers who lead themselves. New York: Doubleday. Kelley, R. E. (2008). Rethinking leadership. In R. E. Riggio, I. C. Chaleff, & J. L. Blumen (Eds.) The art of followership (pp. 5 15), San Francisco, CA: Jossey-Bass. Knight, J., & Auster, C. J. (1999). Faculty conduct: An empirical study of ethical activism. Journal of Higher Education, 70(2), 188210.

PAGE 124

124 Koo, O. H., & Choi, O. S. (2000). The relations hip between the nurses followership, job satisfaction, and organizational commitment. Journal of Korean Academy of Nursing, 30(5), 12541264. Landino, R. J. (2006). Followership: A literature review of a rising power beyond leadership. Unpublished manuscr ipt, Lock Haven University, Lock Haven, PA. Retrieved November 20, 2006 from http://www.natcom.org/nca/files/ccLibraryFiles/FILENAME/ 000000000885/INDV B.pdf Latour, S. M., & Rast, V. J. (2004, Winter). Dynamic followership: The prerequisite for effective leadership. Air & Space Power Journal, 102110. Lederman, D. (2006, November 27). Fixing higher ed, legislator -style. Inside Higher Ed Washington, DC Retr ieved November 25, 2007 from http://insidehighered.com/news/2006/11/28/ncsl Lee, G. (2006). Courage: The backbone of leadership. San Francisco CA : Jossey Bass. Lee, J. H. (2005, May 20). Tro uble at City Colleges of Chicago. The Chronicle of Higher Education, 51(37), A47. Levin, J. S. (1997). The cultures of the community college. Paper presented at the Annual Meeting of the Association for the Study of Higher Education, Albuquerque, NM. Levin e, A. (2004, October 29). The biggest challenge for community colleges. The Chronicle of Higher Education, 51(10), B10. Lively, K. ( 1993, February 3). State colleges face tough decisions on downsizing. The Chronicle of Higher Education, 39(22), A23. Lun din, S. C., & Lancaster, L. C. (1990). Beyond leadership The importance of followership. The Futurist, 24(3), 18 24. Lussier, R. N., & Achua, C. F. (2004). Leadership: Theory, application, skill development (2nd ed.). Eagan, MN: Thomson West. Marino, S. ( 1998). Followers create the best leaders. Industry Week, 247(20), 20. McRae, R. R., & Costa, P. T. (2003). Personality in adulthood: A five -factor theory perspective. New York: Guilford. Meindl, J. R., Ehrlich, S. B., & Dukerich, J. M. (1985). The romance of leadership. Administrative Science Quarterly, 30 (1), 78 102. Mertler, C. A., Steyer, S., & Petersen, G. J. (1997). Teachers perceptions of the leadership/followership dialectic. Paper presented at the Annual Meeting of the Midwestern Educational Resear ch Association, Chicago, IL.

PAGE 125

125 Milgram, S. (1963). Behavioral study of obedience. Journal of Abnormal and Social Psychology, 67(4), 371378. Miller, R. E. (1992). Followership: A prescription for human resource development in public schools. Unpublished doct oral dissertation. Boston University. Mish, F. C. (Ed.). (1999). MerriamWebsters c ollegiate d ictionary (10th ed.). Springfield, MA: Merriam Webster. Moore, L. I. (1976). The FMI: Dimensions of follower maturity. Group & Organization Studies, 1 (2), 203222. Nolan, J. S., & Harty, H. F. ( 1984). Followership Education, 104(3), 311312. Nunnally, J. C. & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). New York: McGraw Hill. Oldham, C. (2006). A test of leadership: Charting the future of U.S. higher education. Jessup, MD : U.S. Department of Education. Olver, J. M. & Mooradian, T. A. (2003). Personality traits and personal values: A conceptual and empirical integration. Personality and Individual Differences, 35(1), 109125. OToole, J. (1978). Tenure: A conscientious obje ction. Change, 10(6), 24 31. Perrault, A. H., Madaus, J. R., Armbrister, A., Dixon, J., & Smith, R. (1999). Florida community college library collection assessment. Tallahassee, FL: Florida State Board of Community Colleges. Pfeffer, J. (1977). The ambigui ty of leadership. Academy of Management Review, 77(1), 104112. Phillips, K. D. (2002). Faculty development goals and activities as perceived by full -time and adjunct mathematics and communications instructors in Florida community colleges. Unpublished do ctoral dissertation. University of Florida Pierce, B. J. (2002). Do educational levels make a difference in leadership styles? Unpublished masters thesis. Embry Riddle Aeronautical University Potter, E. H., Rosenbach, W. E., & Pittman, T. S. (1996). Lea ding the new professional. In R. L. Taylor & W. E. Rosenbach (Eds.), Military l eadership: In p ursuit of e xcellence (3rd ed., pp. 145152). Boulder, CO: Westview Press. Potter, E. H., Rosenbach, W. E., & Pittman, T. S. (2001). Followers for the times: Engag ing employees in a winning partnership. In W. E. Rosenbach & R. L. Taylor (Eds.), Contemporary i ssues in l eadership (5th ed., pp. 163 181). Boulder, CO: Westview.

PAGE 126

126 Pound, W. T. (2006). Transforming h igher e ducation: National i mperative s tate r esponsibility. Washington, DC: National Conference of State Legislators. Rajan, R. & Wulf, J. (2003). The flattening of the firm: Evidence from panel data on the changing nature of corporate hierarchies. Cambridge, MA: National Bureau of Economic Research. Ray, L. K. (2006). Follow the leader: An investigation of the relationship between hierarchical levels and measures of follower behaviors of selected North Carolina community college employees. Unpublished doctoral dissertation. East Carolina University. Roark, D. B. (1988). In service faculty development needs of part -time faculty as perceived by part -time faculty and their supervisors at three selected community colleges in Florida. Unpublished doctoral dissertation. Florida State University Roe, M. A. (1989). Iden tifying readiness for leadership: The behavioral competencies associated with outstanding community college presidents and executive administrators. Unpublished doctoral dissertation. University of Texas. Rosenbach, W. E., Pittman, T. S., & Potter, E. H. ( 1997). The performance and relationship questionnaire. Gettysburg, PA: Leading and Following. Rumsey, D. (2007). Intermediate statistics for dummies. Indianapolis, IN: Wiley. Schein, E. H. (2004). Organizational culture and leadership (3rd ed.). San Franci sco CA : Jossey Bass. Schrieschein, C. A., Clogliser, C. C., Scandura, T. A., Lankau, M. J., & Powers, K. J. (1999). An empirical comparison of approaches for quantitatively assessing the content adequacy of paper and -pencil measurement instruments. Organi zational Research Method, 2(2), 140156. Schrieschein, C. A., Powers, K. J., Scandura, T. A., Gardiner, C. C., & Lankau, M. J. (1993). Improving construct measurement in management research: Comments and a quantitative approach for assessing the theoretica l content adequacy of paper and -pencil surveytype instruments. Journal of Management, 19(2), 385417. Sheehan, K. (2001). Email survey r e sponse rates: A review. Journal of Computer -mediated Communication, 6(2), 0 0 Retrieved April 27, 2008 from http://www.blackwell synergy.com/doi/full/10.1111/j.10836101.2001.tb00117.x Smart, J. C. & St. John, E. P. (1996). Organizational culture and effectiveness in higher edu cation: A test of the culture type and strong culture hypotheses. Educational Evaluation and Policy Analysis 18 (3), 219241. Smith, D. K. (1996). The following part of leading. In F. Hesselbein, R. Beckhard, & M. Goldsmith (Eds.) Leader of the futur e: New visions, strategies, and practices for the next era. San Francisco CA : Jossey Bass.

PAGE 127

127 Smith, R. M. (1997, November 19). Defining leadership through followership: Concepts for approaching leader development. Paper presented at the Annual Meeting of the National Communications Association, Chicago, IL Retrieved November 20, 2006 from http://eric.ed.gov/ERICWebPortal/Home.portal?_nfpb=true&_pageLabel= RecordDetails&ERICExtSearch_SearchValue_0=ED416547&ERICExtSearch_SearchTy pe_0=eric_accno&objectId=0900000b80133421. Solovy, A. (2005). Followership: Leading is a skill; so is following. Hospitals & Health Networks, 78 (5), 32. Steyer, S. A. (2001). An examination of public school teachers' perceptions of followership through the refinement of the Teacher Sentiment Inventory. Unpublished doctoral dissertation. Bow ling Green State University. Tagliere, D. A. (1972). Organalysis. Training and Development Journal, 26(4), 3036. Tanoff, G. F., & Barlow, C. B. (2002). Leadership and followership: Same animal, different spots? Consulting Psychology Journal: Practice and Research, 54 (3), 157 167. Taylor, R. L., & Rosenbach, W. E. (1996). Military l eadership: In p ursuit of e xcellence (3rd ed.). Boulder, CO: Westview Press. Terosky, A. L., Phifer, T., & Neumann, A. (2008). Shattering P lexiglas: Continuing c hallenges for w ome n p rofessors in r esearch u niversities In J. Glazer -Raymo (Ed.) Women in a cademe: The u nfinished a genda. Baltimore MD : The Johns Hopkins Press Tierney, W. G., & Rhoads, R. A. (1993). Enhancing promotion, tenure, and beyond: Faculty socialization as a cultural process. ASHE ERIC Higher Education Report No. 6. Washington, DC: George Washington University. University of South Florida. (2004). Infomart glossary. Infomart, Tampa, FL. Retrieved January 21, 2008 from http://usfweb2.usf.edu/infomart/infomartapps/Glossary.aspx. U.S. Army. (1986). Department of the Army pamphlet 600-68: The bedrock of our profession white paper. Washington, D.C.: Headquarters, Department of the Army. U.S. A rmy. (1993). Soldier training publications (STP) 21-IIIMQS: Military qualification standards III, leader development manual for majors and lieutenant colonels. Washington, D.C.: Headquarters, Department of the Army. U.S. Census Bureau. (2006). State and c ounty quick facts. U.S. Census Bureau, Washington, DC. Retrieved June 28, 2008 from http://quickfacts.census.gov/qfd/states/12/12001.html U.S. Census Bureau. (2008). Racial and ethnic classifications used in census 2000 and beyond. U.S. Census Bureau, Washington, DC. Retrieved April 23, 2008 from http://www.census.gov/population/www/socdemo/race/racefactc b.html

PAGE 128

128 Vecchio, R. P. (1987). Effective followership: Leadership turned upside down. Journal of Business Strategies, 4 (1), 3947. Weiten, W. (2001). Psychology: Themes and variations (5th ed.). Belmont, CA: Wadsworth/Thomson Learning. Williams, W. M., & Ceci, S. J. (2007, March 9). Does tenure really work? The Chronicle of Higher Education, 53(27), B16. Wortman, M. S. (1982). Strategic management and changing leader -follower roles. The Journal of Applied Behavioral Science, 18(3), 371383. Yukl, G. (2002). Leadership in organizations (5th ed.). Upper Saddle River, NJ: Prentice Hall. Zaleznik, A. (1965). The dynamics of subordinancy. Harvard Business Review, 43 (3), 119131. Zimbardo, P. G., Haney, C., Banks, W. C., & Jaffe, D. (1973, April 8). The mind is a formidable jailer: A Pirandellian prison. New York Times Magazine, 6(36), 3860.

PAGE 129

129 BIOGRAPHICAL SKETCH Scott Smith was born and raised in Pittsburgh, Pennsylvania. After graduating from West Point in 1991, he was commissioned as a field artillery offic er in the United States Army. He served in Korea as a company fire support officer, battery fire direction officer, platoon leader, battalion ammunition officer, division artillery fire control officer, battery commander, and group personnel management off icer; at Fort Sill, Oklahoma as a brigade training officer; and at West Point, New York as a psychology instructor and assistant professor. He holds a BS degree in National Security and Public Affairs from the United States Military Academy, an MBA from O klahoma City University, and MS and EdS degrees in Counseling and Human Systems from Florida State University. Upon completion of the University of Floridas PhD program in Higher Education Administration, Scott returns to West Point to serve as the Deputy Director of the Center for Enhanced Performance.