Citation
Faculty Perceptions of Student-Athlete Deviance

Material Information

Title:
Faculty Perceptions of Student-Athlete Deviance A Big Ten and SEC Comparison
Creator:
Kuhn, Ashley Price
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (294 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Criminology, Law, and Society
Sociology and Criminology & Law
Committee Chair:
WILSON,JODI LANE
Committee Co-Chair:
HOLLINGER,RICHARD C
Committee Members:
LEVETT,LORA M
SAGAS,MICHAEL

Subjects

Subjects / Keywords:
athletics -- deviance -- perceptions -- student-athletes
Sociology and Criminology & Law -- Dissertations, Academic -- UF
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Criminology, Law, and Society thesis, Ph.D.

Notes

Abstract:
Faculty are a primary source of learning in college, and they directly impact the success of students. Student-athletes are a special subpopulation of students on campus that interact with faculty. This study uses labeling theory as a theoretical backdrop to understand faculty perceptions of student-athletes. The study looks at the label as the dependent variable, to determine what attributes influence the deviant label of student-athletes by faculty. Faculty at four NCAA Division I institutions located in the South and Midwest regions of the country were surveyed to examine their views of student-athlete academic and criminal behavior. Faculty were randomly assigned to answer questions about either men's football, men's baseball, or women's basketball. I found that faculty have low perceptions of student-athlete deviance overall. However, there were some group differences. The most negative perceptions of deviance among men's football student-athletes and the least negative perceptions about women's basketball student-athletes. In addition, the more familiarity or closeness faculty have with student-athletes the less likely they are to have negative perceptions. Implications for theory and practice are discussed. ( en )
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.
Thesis:
Thesis (Ph.D.)--University of Florida, 2017.
Local:
Adviser: WILSON,JODI LANE.
Local:
Co-adviser: HOLLINGER,RICHARD C.
Statement of Responsibility:
by Ashley Price Kuhn.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
LD1780 2017 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

FACULTY PERCEPTIONS OF STUDENT ATHLETE DEVIANCE : A SEC AND BIG TEN COMPARISON By ASHLEY PRICE KUHN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FO R THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2017

PAGE 2

2017 Ashley Price Kuhn

PAGE 3

To Kennedy and Payton

PAGE 4

4 ACKNOWLEDGMENTS First, I would like to thank my chair, Jodi Lane, for all the time and e nergy she has spent on me over the last six years. She has stuck with me through the ups and downs of graduate school and life. She has truly been a mentor and so much of my growth as a person can been attributed to her. I am also grateful to the members o f my committee, Mike Sagas, Lora Levett, and Dick Hollinger. Their expertise and knowledge improved my dissertation project and overall learning experience. I would also l ike to thank my husband, Alton, for his support and patience We have talked about r eaching this point for so long and it is finally here. Thank you for putting up with me through it all. We made it! Finally, I would like to thank m y children, Kennedy and Payton. I wrote my first major draft of this document at 36 weeks pregnant with yo u both You have been my driving force to finish. I hope if there is one thing you learn from me; it is to value education.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ ........ 13 ABSTRACT ................................ ................................ ................................ ................... 14 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 15 2 LIT ERATURE REVIEW ................................ ................................ .......................... 19 Perceptions of Student Athletes Generally ................................ ............................. 19 Dumb Jock ................................ ................................ ................................ ....... 20 Violence t owards Women ................................ ................................ ................. 21 Substance Use ................................ ................................ ................................ 23 Faculty Perceptions of Student Athletes ................................ ................................ 25 Theoretical Background ................................ ................................ .......................... 28 Deviance and Deviant Labels ................................ ................................ ........... 28 Theoretical Highlights ................................ ................................ ....................... 30 Labeling Perspective Applied to the Present Study ................................ .......... 32 Unique Contributions of the Current Study ................................ ............................. 34 Research Questions ................................ ................................ ............................... 34 Research Questions Regarding Academic Deviance Label of Student Athletes ................................ ................................ ................................ ......... 34 Research Questions R egarding Normative Deviance Label of Student Athletes ................................ ................................ ................................ ......... 35 Research Hypotheses ................................ ................................ ............................. 35 3 METHODOLOGY ................................ ................................ ................................ ... 38 Research Design ................................ ................................ ................................ .... 38 Sample Procedure and Recruitment ................................ ................................ ....... 42 Target P opulation ................................ ................................ ............................. 42 Sampling Frame ................................ ................................ ............................... 42 Recruitment and Data Collection Procedures ................................ ................... 43 Survey Design ................................ ................................ ................................ ........ 45 Sample Characteristics ................................ ................................ ........................... 52 Sample Size ................................ ................................ ................................ ..... 52 Response Rate ................................ ................................ ................................ 53 Respondents versus Sampling Frame ................................ ............................. 55 Early versus Late Respondents ................................ ................................ ........ 57 Measures ................................ ................................ ................................ ................ 58

PAGE 6

6 Independent Variables ................................ ................................ ..................... 58 Dependent Variables ................................ ................................ ........................ 61 Analysis ................................ ................................ ................................ .................. 67 4 DESCRIPTIVE STATISTICS ................................ ................................ .................. 83 Independent Variables ................................ ................................ ............................ 83 Faculty Status Attributes ................................ ................................ ................... 83 University Status Attributes ................................ ................................ .............. 85 Perceptions of Student Athlete Attributes ................................ ......................... 86 University Athletic Status Attributes ................................ ................................ .. 86 Dependent Variables ................................ ................................ .............................. 87 Academic Deviance ................................ ................................ .......................... 87 Normative Deviance ................................ ................................ ......................... 87 Diagnosing Missing Data ................................ ................................ .................. 88 5 RESULTS PREDICTING ACADEMIC DEVIANCE ................................ ............... 148 Academic Deviance by University and Sporting Group ................................ ......... 148 Factorial ANOVA for General Cheating ................................ .......................... 148 Factorial ANOVA for Relying on Others ................................ ......................... 149 Bivariate Relationships for the Entire Sample ................................ ....................... 150 General Cheatin g ................................ ................................ ........................... 150 Relying on Others ................................ ................................ ........................... 151 OLS Regression Models for Entire Sample ................................ .......................... 153 Predicting General Cheating ................................ ................................ .......... 153 Predicting Relying on Others ................................ ................................ .......... 154 Academic Deviance by University ................................ ................................ ......... 155 Bivariate Relationships for Individual Universities ................................ .......... 155 OLS Regression by University and Clogg Coefficient Comparison Test ........ 159 Academic Deviance by Student Athlete Sporting Group ................................ ....... 164 Bivariate Relationships by Sporting Group ................................ ..................... 164 OLS Regression by Sporting Group and Clogg Coefficient Comparison Test 168 6 RESULTS PREDICTING NORMATIVE DEVIANCE ................................ ............. 198 Nor mative Deviance by University and Sporting Group ................................ ........ 198 Factorial ANOVA for Criminal Deviance ................................ ......................... 198 Factorial ANOVA for Drinking Relate d Deviance ................................ ........... 198 Bivariate Relationships for the Entire Sample ................................ ....................... 200 Criminal Deviance ................................ ................................ .......................... 200 Drinking Related Deviance ................................ ................................ ............. 201 OLS Regression Models for the Entire Sample ................................ .................... 203 Predicting Criminal Deviance ................................ ................................ ......... 203 Predicting Drinking Related Deviance ................................ ............................ 203 Normative Deviance by Individual University ................................ ........................ 204 Bivariate Relationships for Individual Universities ................................ .......... 204

PAGE 7

7 OLS Regression by University and Clogg Coefficient Comparison Test ........ 208 Normative Deviance by Student Athlete Sporting Group ................................ ...... 213 Bivariate Relationships by Sporting Group ................................ ..................... 213 OLS Reg ression by University and Clogg Coefficient Comparison Test ........ 216 7 DISCUSSION ................................ ................................ ................................ ....... 243 Revisiting the Hypotheses ................................ ................................ .................... 243 Intergroup Contact ................................ ................................ .......................... 243 Sporting Group ................................ ................................ ............................... 249 Dumb Jock ................................ ................................ ................................ ..... 251 Race of Faculty ................................ ................................ .............................. 253 Non Response Patterns ................................ ................................ ................. 253 Limitations ................................ ................................ ................................ ............. 255 Implications for Theory ................................ ................................ .......................... 258 Implications for Practice ................................ ................................ ........................ 260 Conclusion ................................ ................................ ................................ ............ 262 APPENDIX A EMAIL TEMPLATES ................................ ................................ ............................. 268 B IRB PROTOCOL ................................ ................................ ................................ ... 271 C INFORMED CONSENT ................................ ................................ ........................ 274 D ONLINE SURVEY INSTRUMENT ................................ ................................ ........ 276 LIST OF REFERENCES ................................ ................................ ............................. 282 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 294

PAGE 8

8 LIST OF TABLES Table page 3 1 Demographics of Top 25 Public Universities ................................ ...................... 69 3 2 Demographic of final four institutions ................................ ................................ .. 70 3 3 Timeline of recruitment emails ................................ ................................ ............ 72 3 4 Faculty reasons for declining to participate themes ................................ ............ 72 3 5 Outcome Rate Categories ................................ ................................ .................. 73 3 6 Rate Estimates ................................ ................................ ................................ ... 73 3 7 Sampling fr ame data versus sample data ................................ .......................... 74 3 8 Early versus late respondents ................................ ................................ ............ 75 3 9 Principal components analysis for fandom variable ................................ ............ 77 3 10 Principal component analysis for student athlete interaction variable ................ 77 3 11 Principal Components Analysis for all academic d eviance items in instrument .. 78 3 12 Principal Components Analysis for individual components of academic deviance ................................ ................................ ................................ ............. 79 3 13 Princip al Components Analysis for all normative deviance items in instrument deviance variable ................................ ................................ ................................ 80 3 14 Principal Components Analysis for individual components of normative deviance ................................ ................................ ................................ ............. 82 4 1 Independent variables descriptive statistics (N = 1,100) ................................ .... 93 4 2 Faculty athletic service themes (N = 353) ................................ ........................... 95 4 3 Individual status attributes of student athletes descriptive statistics ................... 96 4 4 Descriptives of academic deviance scales and items for the entire sample ....... 97 4 5 Descriptives of normative deviance scales and items for the entire sample ....... 98 4 6 Missing data analysis for item 1 of academ ic deviance scale (N = 492) ........... 100 4 7 Missing data analysis for item 2 of academic deviance scale (N = 492) ........... 102

PAGE 9

9 4 8 Missing data a nalysis for item 3 of academic deviance scale (N = 488) ........... 104 4 9 Missing data analysis for item 4 of academic deviance scale (N = 492) ........... 106 4 10 Missing data analysis for item 5 of academic deviance scale (N = 491) ........... 108 4 11 Missing data analysis for item 6 of academic deviance scale (N = 496) ........... 110 4 12 Missing data analysis for item 1 of relying on others scale (N = 516) ............... 112 4 13 Missing data analysis for item 2 of relying on others scale (N = 503 ) ............... 114 4 14 Missing data analysis for item 3 of relying on others scale (N = 496) ............... 116 4 15 Missing data analysis for Item 1 of cr iminal deviance scale (N = 507) ............. 11 8 4 16 Missing data analysis for Item 2 of criminal deviance scale (N = 503) ............. 120 4 17 Missing data analysis for Item 3 of criminal deviance scale (N = 497) ............. 122 4 18 Missing data analysis for item 4 of criminal deviance scale (N = 493) .............. 124 4 19 Missing data analysis for item 5 of criminal deviance scale (N = 470) .............. 126 4 20 Missing data analysis for item 6 of criminal deviance scale (N = 476) .............. 128 4 21 Missing data analysis for item 7 of criminal deviance scale (N = 471) .............. 130 4 22 Missing data analysis for item 8 of criminal deviance scale ( N = 472) .............. 132 4 23 Missing data analysis for item 9 of criminal deviance scale (N = 483) .............. 134 4 24 Missing data analysis for item 1 of drinking related deviance scale (N = 505) 136 4 25 Missing data analysis for item 2 of drinking related deviance scale (N = 516) 138 4 26 Missing data analysis for item 3 of drinking related deviance scale (N = 492) 140 4 27 Missing data analysis for item 4 of drinking related deviance scale (N = 468) 142 4 28 Missing data analysis for item 5 of drinking related deviance scale (N = 479) 144 4 29 Missing data analysis for item 6 of drinking related deviance scale (N = 507) 146 5 1 Factorial ANOVA of university and sporting group for general cheating ........... 174 5 2 Mean dif ferences and confidence intervals of perceptions of general cheating by university ................................ ................................ ................................ ...... 174

PAGE 10

10 5 3 Mean differences and confidence intervals of perceptions of general cheating by sport ................................ ................................ ................................ ............. 175 5 4 Factorial ANOVA of university and sporting group for relying on others ........... 175 5 5 Mean differences and confidence intervals of perceptions of relying on others by university ................................ ................................ ................................ ...... 175 5 6 Mean differences and confidence intervals of perceptions of relying on others by sport ................................ ................................ ................................ ............. 176 5 7 Correlations of academic deviance and faculty status attributes ...................... 177 5 8 Correlations of academic deviance and university status attributes ................. 178 5 9 Correlations of academic deviance variables and student athlete status attributes ................................ ................................ ................................ ........... 178 5 10 Correlations of academic deviance and university athletic status attrib utes ..... 179 5 11 OLS regression predicting general cheating academic deviance for the entire sample ................................ ................................ ................................ .............. 179 5 12 OLS regression predict ing relying on others academic deviance for the entire sample ................................ ................................ ................................ .............. 180 5 13 Correlations of academic deviance and faculty status attributes ...................... 181 5 14 Correlations of academic deviance and perceptions of student athlete status attributes by university ................................ ................................ ...................... 183 5 15 OLS regression predicting general cheating academic deviance by un iversity 184 5 16 Z values comparing beta coefficients predicting general cheating by university ................................ ................................ ................................ .......... 185 5 17 OLS regression predict ing relying on others academic deviance by university 186 5 18 Z values comparing beta coefficients predicting relying on others by university ................................ ................................ ................................ .......... 187 5 19 Correlations of academic deviance and faculty status attributes by sporting group ................................ ................................ ................................ ................ 188 5 20 Correlations of academic deviance and student athlete status attributes by sp orting group ................................ ................................ ................................ ... 190 5 21 Correlations of academic deviance and university status attributes by sporting group ................................ ................................ ................................ ... 191

PAGE 11

11 5 22 Correlations of academic deviance and university athletic status attributes by sporting group ................................ ................................ ................................ ... 191 5 23 OLS regression predicting general cheating academic deviance by sporting group ................................ ................................ ................................ ................ 192 5 24 Z values comparing beta coefficients by sporting group predicting general cheating ................................ ................................ ................................ ............ 193 5 25 OLS regression predicting relying on others acade mic deviance by sporting group ................................ ................................ ................................ ................ 194 5 26 Z values comparing beta coefficients by sporting group predicting relying on others ................................ ................................ ................................ ............... 195 6 1 Factorial ANOVA of university and sporting group for criminal deviance .......... 220 6 2 Mean differences and confidence intervals of perceptions of criminal deviance by university ................................ ................................ ...................... 220 6 3 Mean differences and confidence intervals of perceptions of criminal deviance by sport ................................ ................................ ............................. 221 6 4 Factorial ANOVA of university and sporting group for alcohol related deviance ................................ ................................ ................................ ........... 221 6 5 Mean differences and confidence intervals of perceptions of drinking related deviance by university ................................ ................................ ...................... 221 6 6 Mean differences and confidence intervals of perceptions of drinking related deviance by sport ................................ ................................ ............................. 222 6 7 Correlations of normative deviance and faculty status attributes ...................... 223 6 8 Correlations of normative deviance and university status attributes ................. 224 6 9 Correlations of normative deviance variables and student athlete status attributes ................................ ................................ ................................ ........... 224 6 10 Correlations of normative deviance and athletic status attributes ..................... 225 6 11 OLS r egression predicting criminal deviance for the entire sample .................. 225 6 12 OLS regression models predicting drinking related deviance for the entire sample ................................ ................................ ................................ .............. 226 6 13 Correlations of normative deviance and faculty status attributes by university 227

PAGE 12

12 6 14 Correlations of normative deviance and student athlete status attributes by university ................................ ................................ ................................ .......... 229 6 15 OLS regression predicting criminal deviance by university ............................... 230 6 16 Z values comparing beta coefficients predicting criminal deviance by university ................................ ................................ ................................ .......... 231 6 17 OLS regression predicting drinking related deviance by university .................. 232 6 18 Z values comparing beta coefficients predicting drinking related deviance by university ................................ ................................ ................................ .......... 233 6 19 Correlations of normative deviance and faculty status attributes by sporting group ................................ ................................ ................................ ................ 234 6 20 Correlations of normative deviance and student athlete status attributes by sporting group ................................ ................................ ................................ ... 236 6 21 Correlations of normative devian ce and university status attributes by sporting group ................................ ................................ ................................ ... 237 6 22 Correlations of normative deviance and university athletic status attributes by sporting group ................................ ................................ ................................ ... 237 6 23 OLS regression predicting criminal deviance by sporting group ....................... 238 6 24 Z values comparing beta coefficients predicting criminal deviance by sporting grou p ................................ ................................ ................................ ................ 239 6 25 OLS regression predicting drinking related deviance by sporting group ........... 240 6 26 Z values comparing beta coefficients p redicting drinking related deviance by sporting group ................................ ................................ ................................ ... 241 7 1 Summary of results for the entire sample ................................ ......................... 265 7 2 Summary of result s by university ................................ ................................ ...... 266 7 3 Summary of results by sport ................................ ................................ ............. 267

PAGE 13

13 LIST OF FIGURES Figure page 2 1 Perspectives in labeling ................................ ................................ ...................... 33 5 1 Main effects of university for general cheating academic deviance .................. 196 5 2 Main effects of spor ting group for general cheating academic deviance .......... 196 5 3 Main effect of university for relying on others academic deviance .................... 197 5 4 Main effect of sport for relying on others academic deviance ........................... 197 6 1 Main effect of sport for criminal deviance ................................ ......................... 242 6 2 Interact ion effect of university and sport for alcohol related deviance .............. 242

PAGE 14

14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for th e Degree of Doctor of Philosophy FACULTY PERCEPTIONS OF STUDENT ATHLETE DEVIANCE: A SEC AND BIG TEN COMPARISON By Ashley Price Kuhn May 2017 Chair: Jodi Lane Major: Criminology, Law and Society Faculty are a primary source of learning in college, and t hey directly impact the success of students. Student athletes are a special subpopulation of students on campus that interact with faculty. This study uses labeling theory as a theoretical backdrop to understand faculty perceptions of student athletes. The study look s at the label as the dependent variable, to determine what attributes influence the deviant label of student athletes by faculty. Faculty at four NCAA Division I institutions located in the South and Midwest regions of the country were surveyed to examine their views of student athlete academic and criminal behavior. F aculty were randomly assigned to answer questions about I found that faculty have low perceptions of student athlete d eviance overall. However, there were some group differences. T he most negative perceptions of deviance among athletes and the least negative perceptions athletes. In addition the more familiarity or closeness faculty have with student athletes the less likely they are to have negative perceptions Implications for theory and practice are discussed.

PAGE 15

15 CHAPTER 1 INTRODUCTION Stud ent athletes are a special sub population of students on campus because they represent the unive rsity as a whole in activities that bring the university a lot of attention They are often associated with the image or brand of the university due to athletic success and visibility to the public through the media (Sternberg, 2016) T hat is, student athletes are often well known on and off campus. Their team success can bring a lot of money and financial donations to the university from alumni and boosters (Letawsky et al., 2003). Additionally, universities with winning teams can have higher numbers and caliber of unde rgraduate applicants (Chung, 201 3; Toma & Cross, 1998). Student athletes can represent the university in both positive and negative ways. Student athletes have the opportunity to have a positive impact as a representative of the university community. Through the large amount of media attention they bring with athletic success, student athletes can also be highlighted for success in the classroom and involvement in the community. Additionally, student athletes bring divers ity to the campus community, which may increase the educational experiences for all students on campus (Hirko, 2009). However, student athletes also have the opportunity to make the university look very bad. They may do this through academic deviance or n ormative deviance. An example of academic deviance is cheating on an exam or getting credit for work that is not their own. An example of normative deviance is underage drinking or drug use. R egular student s could partake in these same activities, but the image/brand of the university is not at the same level of risk as it is for a student athlete partaking in these deviant activities and getting caught because of the limelight that generally surrounds

PAGE 16

16 this high profile groups (Sternberg, 2016). Specifical ly, t he frequency of these behaviors for student athletes may not be that different if compared to the student population as a whole. However, these behaviors, when performed by athletes as opposed to non athletes, are more newsworthy because of the ir sta tus as student athletes. Negative incidents may influence attitudes towards athletes, both among the public and among people on campus, including other students and faculty. perceptions of student athletes and intercollegiate athletic s is often controversial ( Eitzen, 2012). Some believe these students add to the atmosphere and pride of the university (Putler & Wolfe, 1999). Some see it as an opportunity for athletes to receive funding and an education they may not otherwise have the op portunity to receive (Toma, 1999). However, opponents of intercollegiate athletics may believe athletic programs value winning at any cost (Putler & Wolfe, 1999). For example, many people fear the integrity and intellectual environment of institutions are severely affected by the commercialization of National Collegiate Athletic Association ( NCAA ) Division I athletic programs (Watt & Moore, 2001). Student athletes may be recruited to play college sports and receive an undergraduate education regardless of t heir academic ability, background, or motivation. If so, t hese factors may increase the perception among the general public that student athletes do not belong and lead to beliefs that they are more likely to be deviant. However, the general public is not the only group that may have negative perceptions of student athletes. Faculty are one group of individuals who are part of the campus community who may also have negative attitudes and beliefs about student athletes Faculty attitudes matter because they are a critical part of the undergraduate

PAGE 17

17 student experience including that of student athletes Faculty are a primary source of learning in college and they directly impact the success of students. Research shows that f aculty attitudes and perceptions o engagement in the classroom, motivation for learning, and self concept ( Arbaugh, 2001; Umbach & Wawrzynski, 2005 ; Urdan & Schoenfelder, 2006 ). That is faculty can affect behavior and outcomes of students. Studies in the education literature also show that children rise to the expectation of their teachers (Rubie Davies, 2006 van den Bergh et al., 2010 ). S tudies show that s tudent athletes can be acutely aware that faculty might be biased against them due to their ascribed student athlete status resulting in what Steele (1997) refers to as a stereotype threat where student athletes can feel negative view s by others and conform to it Therefore, if student athletes know faculty have negative stereotypes about th em, in settings where their identity as an athlete is heightened, they may be more vulnerable to conforming to the stereotype. Faculty and instructors also have a stake in the image or brand of the university as employees who work hard to advance the ins titution through research, grants, and instruction. That is, faculty may care a lot about how the university is perceived, because it impacts their own careers. If they believe that student athletes are somehow different than other students and that they r eflect on the university, and if these opinions are negative, instructors may be more inclined to dismiss them as serious students and possibly treat them differently. athletes have been examined by academics f aculty perceptions of student athlete deviance has not. This is true, e ven though there have been some recent cases where faculty have spoken out about

PAGE 18

18 student in 2006 a group of faculty responded to allegations of sexual violence by Duke lacrosse members putting out an ad in the Lacrosse team members were later found innocent (Copeland, 2006; Johnson, 2008). This event shows that these faculty paid att ention and made judgments regarding the stude nt athletes alleged actions. Given the possible consequences of negative faculty opinions on student athletes it is important to investigate what factors contribute to perceptions of student athlete deviance. This study will use labeling theory as a theoretical backdrop to understand faculty perceptions of student athletes. The study will look at the label as the dependent variable, to determine what attributes influence the deviant label of student athletes by faculty. Faculty at four NCAA Division I institutions located in the South and Midwest regions of the country are surveyed to examine their views of student athlete academic and criminal behavior. Additionally, characteristics of the institution, athletic department, and individual faculty will be assessed to establish their impact on the label. The following general research questions will be examined: What, if any, individual status attributes of faculty increase their perceptions of student athlete acad emic and normative deviance? What university status attributes may increase faculty perceptions of student athlete academic and normative deviance? More specific research questions are addressed in the methods section.

PAGE 19

1 9 CHAPTER 2 LITERATURE REVIEW Student athletes are a distinct student population on college campuses, isolated from regular student groups because of practice and competition schedules (Watt & Moore, 2001). Mor e specifically, their classes are arranged around athletic priorities, like practice and workouts. Additionally, several athletes may group (e.g., sit) together in classes away from non athlete students making them stand out as a distinct group (Sparent, 1 988). Student athletes may also stand out functionally, psychologically, and physically from non athletes (Nishimoto, 1997). Each of these characteristics makes it very easy for student athletes to be noticeable on campus, in the classroom, and the communi ty. Perceptions of Student Athletes G enerally There are two main negative perceptions of student athletes that are discussed in the academic literature; these perceptions reflect both academic and normative deviance. The oldest stereotype of student athlet es is that of the reflects some level of academic deviance (Coakley, 1990). Mor e recent perceptions of student athletes involve normative deviance, with beliefs that they are violent towards women and are substance use rs ( Humphrey & Kahn 2000; Gage, 2008 ; LaBrie, Gossbard, & Hummer, 2009 ; McCray, 2014 ; Page & Roland, 2004) Although these stereotypes to date have only been studied and found among the general public not faculty in particular -they may also be perceptions that faculty hold

PAGE 20

20 Dumb Jock athletes are not as academically since the beginning of sports in 500 BC, where the Greeks believed athletes were u seless because they neglected their intellectual development and had dull minds (Coakley, 1990). Today, this stereotype persists with several media reports of universities lowering academic standards for student athletes, lower student athlete graduation r ates compared to the rest of the student body, and high profile academic fraud cases (Eitzen, 2012). For example, o ne prestigious university and sports program that has recently received a lot of media attention for academic fraud is the University of Nor th Carolina at Chapel Hill (UNC). The NCAA and Federal Bureau of Investigations (FBI) are American Studies Department for having phantom courses that distributed inflated grades to students (Tracy, 2014). These classes had a athletes enrolled, with reports that academic advisors encouraged their athletes to take the classes knowing minimal work would be required. Additionally, a former learning specialist for UN basketball players between 2004 and 2012 read below a third grade level (Wilson, 2014). In her report, she links the academic scandal of phantom courses to the under pre paredness of student athletes at UNC. Although this whistleblower report has raised Family Educational Rights and Privacy Act (FERPA) concerns regarding the identifiable co mmitting deviant behavior (i.e., taking phantom classes and cheating) (Beard, 2014).

PAGE 21

21 Violence t owards W omen Another perception of student athletes explored in the academic literature is sexual violence towards women. This topic has been looked at extensiv ely with the prevalence of sexual assaults among all male groups (i.e., student athletes and fraternity members) but findings have been mixed (McCray, 2014) One of the first studies to examine this relationship between sexual assault and student athletes at Division I institutions used official records ( Crosset, Benedict, & McDonald, 1995). In their sample of 20 institutions, Crosset et al. (1995) found that athletes constitute about 3.3% of the total male student population, but were involved in 19% of se xual assaults reported to university judicial affairs. However, of those sexual assault cases officially reported to the campus poli ce for a criminal investigation, there were no significant differences in the percentage of athlete and non athlete offender s. This study has some limitations because of the underreporting of s exual assault s generally and especially on college campuses. There may be other variables involved here that are not taken into account because of underreporting or other issues. It is pr oblematic to blame a particular group of students for an issue that has multiple layers. Other studies examining college athletes and sexual violence show that student athletes on particular teams or those involved in other high risk behaviors are more li kely to be involved in sexually aggressive behaviors. For example, Gage (2008) found student athletes on high profile sport teams, such as football, self reported higher levels of sexual aggression and sexual activity than non athletes and student athletes on lower profile teams, such as tennis and track. Also, Humphrey and Kahn (2000) found that student athletes who engaged in other high risk activities (e.g., party atmosphere) were significantly more likely to have attitudes supporting sexual aggression a nd

PAGE 22

22 hostility towards women. The findings of Gage (2008) and Humphrey and Kahn (2000) may explain the inconsistent results of previous studies. That is, characteristics of particular student athletes and teams may explain the relationship between athletics and sexual violence rather than simply student athlete status alone In other words, there are likely contextual factors that affect the relationship between student athlete status and deviance. While there are only a few studies of student athlete violenc e toward women in the academic literature, there have been a number of scandal s involving student athletes using violence against women, which may contribute to deviant labels by the public. One of the most publicized was the 2006 Duke University Lacrosse team incident where an African American female stripper falsely accused three white male Duke Lacrosse athletes of raping and kidnapping her. The case created a swarm of media attention for issues dealing with racism, sexism, and politicization of the jud icial system (Johnson, 2008). The prosecutor for the case, Micha el Nifong, admitted to withholding evidence from the defense, misleading the court, and intensifying media reports to the public about the case, for which he was later disbarred (Liptak, 2007) Even though there was not misconduct involved, the case connected student athletes to sexual violence and was broadcast across the nation regularly (Cohan, 2016) Although the student athletes in the Duke Lacrosse situation were eventually cleared after being falsely accused, there are some sexual assault incidents that make the news where athletes may get off easily possibly leading to perceptions that athletes are treated differently. For example, r ecently, a class action suit was filed against the Un iversity of Tennessee (UT), where several female sexual assault victims claim the

PAGE 23

23 university used a biased adjudication process and created a hostile sexual environment, which involved primarily student athlete defendants (Wadhwani, 2016a). The women have come together to expose a larger Title IX gender discrimination issue at UT. There has been a lot of media attention surrounding this case because of the student athletes named in the lawsuit, including six football players. The head football coach and ath letic director were also named in the suit as key witnesses for their knowledge about the cases and culture of sexual assault with the football team (Wadhwani, 2016b) Student athlete actions towards women is such a concern that t he NCAA has partnered with creating an environment where sexual violence is un acceptable and victims are supported (NCAA, 2014a). Additionally, the NCAA released a handbook for athletic Addres sing Sexual Assault and Interpersonal Violence to guide athletic departments in changing the culture regarding sexual assault among student athletes. T he handbook provides best practices for the role of athletic departments in cases involving st udent handbook for athletic departments demonstrate the perception that student athlete involvem ent with sexual assault is real, is recognized by the larger organization governing college athletics and needs to be addressed (NCAA, 2014b). Substance Use Another deviant behavior of student athletes that has been explored in the academic literature is increased substance use. This research has focused on the prevalence of substance use among student athletes. A maj ority of studies show that student athletes use illicit drugs about as often as or less often than non athletes but not necessarily more often (Wechsler et al., 1997; Page & Roland, 2004; Yusko et al.,

PAGE 24

24 2008). Also, the risk factors associated with substan ce use are the same for both athletes and non athlete students (Buckman et al., 2011). Studies examining perceptions of substance use and student athletes have only used samples of student athletes to assess the prevalence among their athlete peers (Page & Roland, 2004; LaBrie et al., 2009). Like other peer perceptions studies, athletes reported higher prevalence of use by their peers than by themselves (Page & Roland, 2004; LaBrie et al., 2009). It is possible that student es may be highlighted in local and national media outlets, when a similar student who was not involved in athletics would be ignored by the media. Student athletes are randomly tested for illegal substances by their athletic departments and the NCAA (NCAA, 2014c). Institutional athletic departments have policies regarding consequences, and these vary based on the substance and number of positive tests. However, most consequences result in game suspensions. Testing positive on an NCAA drug test for any subst ance immediately results in suspension (NCAA, 2014c). These events are newsworthy because student athletes are high profile in large athletic programs and also represent the university (Sternberg, 2016). In addition, it may be obvious if a high profile ath lete is not suiting up for a game, and so the absence must generally be explained. If deviance by athletes is shown in the media, it may create a perception in the public that student athletes engage in more substance use and abuse than non athlete student s, even though a majority of academic research does not support this conclusion To summarize, the literature shows that perceptions of student athletes as dumb jocks, more likely to commit violence towards women, and substance users reflect both

PAGE 25

25 academic and normative deviance. These perceptions are prevalent among the general public, but need to also be considered for faculty who have more potential impact on the daily lives of student athletes Facul ty Perceptions of Student A thletes Faculty members ar e part of the public, but their perceptions and attitudes about students are perhaps even more important than the ideas of the population at large. Professors have a direct impact on the lives of students, including the student athletes they encounter thr ough their classes, and their perceptions and attitudes toward student athletes (good or bad) may influence how they treat them. For example, a football player more break s on grades or attendance than he might if the student were not on the team. In contrast, a professor who believes that student athletes are not good students, are criminals, or get too many perks they do not deserve may be harsher toward athletes in thei r classes than toward students more generally. Yet, we do not know for sure, because research on faculty perceptions of student athletes specifically is very limited. A majority of the research that exists focuses on faculty satisfaction with control and administration of athletic programs (Co ckley & Roswell, 1994; Engstrom, Sedlack, & McEwen 1995). These studies indicate that faculty at Division I institutions, also administrat ion of athletic programs compared to faculty at Division II or III institutions. Some of these studies ask faculty questions about student athletes being academically prepared enough for college, but these questions are not the mai n focus of the study (Eng strom et al.

PAGE 26

26 no studies regarding faculty perceptions of student athlete normative deviance (i.e., crime, substance use or violence towards women). What we do know about faculty perception s from the academic research is that faculty hold more prejudicial attitudes towards both male revenue and non revenue athletes compared to other students and female athletes (Engstrom et al., 1995). Engstrom et al. (1995) sampled faculty at a large public university with a Division I NCAA athletic department using the adapted Situational Attitude Scale (SAS). This scale is used to measure prejudice and differential attitudes towards certain groups (i.e., different gender, racial/ethnic groups, and student groups). The scale in Engstrom et athletes (revenue and non revenue producing male sports) and non athlete students. The researchers found that faculty members had higher levels of disapproval for student athletes that receive full scholarships and are admitted with low SAT scores. The respondents also expressed more surprise and concern about cheating in situations where student chievement by a normal student. These findings may support that faculty label student athlete engage in classes. There is also research indicating that student et al., 2013; Stone, Harrison, & Mottley, 2012; Wininger & White, 2008), meaning t hese groups all viewed student athletes negatively in regards to their ability and effort towards academics. Also, Yopyk and Prentice (2005) fou nd when priming student

PAGE 27

27 athlete respondents with their athletic identity (i.e., called them athletes when they t ook the questionnaire), compared to their student identity (i.e., calling them students) or no identity (i.e., not indicating an identity) they had lower self regard and performed most abels may lead to self fulfilling prophecies with low academic achievement and academic integrity issues, which is referred to as stereotype threat in the literature (Feltz et al., 2013; Steele, 1997). In the context of this study, this means that if resul ts show that faculty characterize student athletes as participating in deviant behaviors and create a negative label or stereotype of them, the implications may be that these student athletes may in turn be at risk for confirming this characteristic or wor se. Although there is some scholarly knowledge about faculty perceptions of student athlete academic performance, which may relate to academic deviance, there is no research on faculty perceptions of student beha viors that are socially unacceptable and usually illegal (i.e., sexual assault, drug use, etc.). However, there are public examples of faculty expressing concern about student athletes allegedly engaging in these types of behaviors. For example, during the 2006 Duke University Lacrosse Team scandal mentioned above, a group of faculty, social an African and African American Studies (AAAS) forum about frustration with the case begin on

PAGE 28

28 perceptions that problems with athletes were an ongoing issue at that university. Additiona moment when some of the most vulnerable among us are being asked to quiet down lacrosse te am shows that at least some faculty members pay attention and make judgments regarding actions of student athletes. This particular response from faculty is concerning because the lacrosse players were later cleared of all charges (Wilson, 2007) That is, faculty perceptions of student athletes do not always correspond to the reality of athlete behavior. Yet, because there is no published research on faculty attitudes about criminal and other deviant behaviors (beyond academic issues) by student athletes, it is important to conduct more studies to better understand how faculty view students who also are involved in athletic programs. This study is designed to further our knowledge of faculty perceptions of student athletes and expand the field by examining faculty perceptions of both academic and normative deviance among university athletes. Specifically, this study aims to identify what factors at the institutional and individual level are associated with faculty members giving student athletes a deviant l abel. Theoretical B ackground Deviance and Deviant Labels The most common definition of deviance is a violation of the standards of behavior, or norms, in a social group (Cohen, 1959). Faculty and students, more specifically student athletes, are all part of a widespread social group of the university. The norms and expectations of appropriate and deviant behavior are defined by the

PAGE 29

29 university social group, but mostly by those with status and pow er (i.e., faculty) (Becker, 1963 ). Therefore, in a university setting, there are well established rules regarding academics to protect the integrity of the institution. Students who engage in certain types of behavior that compromise the integrity would violate the social norms and be seen as deviants. Behaviors, be liefs, and characteristics of persons can all be labeled deviant. However, committing deviant behavior is not the same as being labeled a deviant. Individuals are labeled deviant because of their behavior or how they look (Clinard & Meier, 2011). Certain c onditions may lead others to label individuals deviant through coding schemes. These coding schemes may include violations of appearance norms, like a physical handicap, having tattoos, or being obese (Hawkins & Tiedeman, 1975; Durkin & Houghton, 2000). Th e same coding schemes may also be used for student athletes, where they may be judged as deviant for what they do and/or the group to which they belong. The idea of student athletes as deviants, for simply being a member of their social group, resonates w ith According to Skolnick (1993), the police have quick ways to identify potential criminals efficiently to prevent dangerous situations. Police develop coding schemes using nguage, and attire. More often than not, young black males are police context, it can extend to people general ly. Like police do with people on the s treet, f aculty may develop coding schemes for students they believe are

PAGE 30

30 deviants, where student athletes may be more likely to be viewed with suspicion for violating norms compared to other students. Faculty may use representations of crime that are conve yed in our culture to criminal is disproportionately portrayed as a young, black male (Drummond, 1990; Barlow, 1998; Russell, 2002). This typification of crime is compound ed for athletes because the stereotype of a talented athlete is also a young, black male (Sailes, 1998). Additionally, student athletes who are easily identified as athletes may be perceived as more deviant compared to those who are not. For example, an at hlete in a revenue compared to non revenue generating sport with little media attention and fan support Theoretical Highlights This study is situated in part in Labeling Theory, which focuses on the informal and formal application of deviant labels by society on certain members (Cullen & Agnew, 2003). Unlike other criminological theories, labeling theory proposes that the focus be on those who create and react to the label of offenders, rather than the behavior of the offender him/herself (Tannenbaum, 1938; Becker, 1963; Lemert, 1972). rather a consequence o f the application by others of rules and sanctions to the 9). There are two main hypotheses that come out of the labeling perspective, which include the deviance amplification hypothesis and differential enforcement hypothesis (Paternoste r & Iovanni, 1989; Triplett, 1993). The deviance amplification hypothesis

PAGE 31

31 claims that labeling an individual deviant will increase his/her probability of criminal label is the independent variable having an impact on future criminal behavior, which is the dependent variable. This is the most popular hypothesis explored using the labeling framework because it has the most implications for the criminal justice system (Paternos ter & Iovanni, 1989). The differential enforcement/status characteristics hypothesis explains why certain people are labeled deviants and others are not. In this case, the label is the dependent variable. The differential enforcement/status characteristics hypothesis will be the approach used for this present study because I am interested in how student athletes are perceived by faculty According to Paternoster and Iovanni (1989) the theoretical origins of labeling theory come from symbolic interactionism and conflict theory. The deviance amplification hypothesis or secondary deviance hypothesis is drawn from symbolic interactionism and the sociological idea of the looking glass self, where the experience of being labeled by agencies of social control may concept. Alternatively, the differential enforcement or status characteristics hypothesis comes from the conflict tradition. Conflict theory traditionally explains how political or economic powers create deviant label s or statuses concerning certain behaviors or actions. Therefore, these delinquency statuses are based on extra legal attributes of groups not in political or economic control. Labeling theory as originally presented by Tannenbaum (1938), Becker (1963), a nd Lemert (1972) and was one of the most popular perspectives in the 1960s and 1970s. Over the years, labeling theory has been criticized for being vague, ambiguous,

PAGE 32

32 and failing to provide empirically testable propositions (Wellford, 1975 1993 ; Gove, 1980 ; Paternoster & Iovanni, 1989). Today, labeling is rarely considered a theory of criminal and deviant behavior; instead, it is considered a perspective that is meant to offer only sensitizing concepts (Cullen & Agnew, 2003). In the present study, I will us e concepts from the labeling perspective to examine faculty perceptions of student athletes. Labeling Perspective Applied to the Present Study This study will look at the label as the dependent variable rather than the label as an independent variable. T herefore, this study will use the differential enforcement/status characteristics hypothesis. Research that has been conducted in this area comes out of the conflict tradition, where the powerlessness of particular groups (minorities, poor, etc.) increases and being punished than those from more powerful groups (Paternoster & Iovanni, 1989). In the present study, faculty are the group in power and student athletes are the group not in power. Actual deviant beha vior by student athletes is of secondary importance from this perspective. According to labeling theorists, variables like individual and community characteristics, social distance, and visibility predict the formation of the deviant label.

PAGE 33

33 Figure 2 1. P erspectives in labeling Although most labeling theorists do not believe that status attributes are the only factors that attribute to a label, they do believe they are influential and have some effect on labeling outcomes (Bernstein, Kelly, & Doyle, 1977). Tittle (1980) argues that social characteristics should be the most important factor in determining the outcomes of deviant labels, more important than the actual rule breaking. A majority of research that examines the impact of social attributes is withi n the formal setting of the criminal justice system because many believe that criminal justice decisions (i.e., decision to arrest, sentencing outcomes, etc.) are labels (Bernstein et al., 1977). For example, several studies have found that offenders who a re male, black, and young are sentenced more harshly than offenders who are female, non black, and older (Steffensmeier, Ullmer, & Kramer, 1998; Mitchell, 2005). T he labeling perspective, and the differential enforcement/status characteristics hypothesis, in particular, will help inform the analysis of the attributes of faculty members or their institutions that influence professors to label all or some student athletes as deviant. Therefore, in the present study, I am interested in what faculty and univer sity factors are related to faculty holding negative labels of student athletes. Although this study only focuses on one aspect of the labeling perspective, it is important to understand the status attributes of people (in this case faculty) who are or are not labeling student athletes as deviants. Once there is a better understanding of how the label is formed, this research could be expanded to explore the deviance amplification hypothesis or stereotype threat.

PAGE 34

34 Unique Contributions of the Current Study Th is study contributes to the understanding of faculty perceptions of student athlete deviance by expanding on prior literature in at least three important ways. First, this is one of the first quantitative studies to examine status attributes of faculty bot h athlete deviance specifically. Second, this study is the first to examine faculty perceptions of student ormative deviance. Third, this study expands the labeling perspective by looking at an informal social control networks (faculty) formation of deviant labels for student athletes. More specifically, the differential enforcement hypothesis of labeling persp ective is used as a framework to explore whether there is a development of a deviant label for student athletes by faculty. Research Questions Given that there is limited prior research on faculty perceptions of student pective, this study is exploratory. Therefore, the following research questions are examined : Research Questions Regarding Academic Deviance Label of Student Athletes 1. What, if any, individual s tatus attributes of faculty (i.e., age, race, gender, academi c discipline, academic rank, tenure status, administrative status, sports fandom, and contac t with student athletes) relate to an increased likelihood that faculty will view student athletes as more academically deviant? 2. Are there university status attribu tes (i.e., university, region, faculty population, and student population) related to an increased likelihood that faculty will view student athletes as more academically deviant? 3. Are there individual status attributes of student athletes (i.e., gender, ra ce, and sport) related to an increased likelihood that faculty will view student athletes as more academically deviant?

PAGE 35

35 4. Are certain athletic status attributes related to the university (i.e., NCAA ding, number of varsity teams, total athletic revenue, and major NCAA infractions) related to an increased likelihood that faculty will view student athletes as more academically deviant? Research Questions Regarding Normative Deviance Label of Student Ath letes 5. What, if any, i ndividual status attributes of faculty (i.e., age, race, gender, academic discipline, academic rank, tenure status, administrative status, sports fandom, and contact with student athletes) related to an increased likelihood that facult y will view student athletes as more normatively deviant? 6. Are there university status attributes (i.e., university, region, faculty population, and student population) related to an increased likelihood that faculty will view student athletes as more norma tively deviant? 7. Are there individual status attributes of student athletes (i.e., gender, race, and sport) related to an increased likelihood that faculty will view student athletes as more normatively deviant? 8. Are certain athletic status attributes relate d to the university (i.e., NCAA teams, total athletic revenue, and major NCAA infractions) related to an increased likelihood that faculty will view student athletes as more normatively deviant? Research Hypotheses Although this study is exploratory, I have some hypotheses based on the idea of intergroup contact in social psychology. The intergroup contact hypothesis is the more contact different groups of people have with ea ch other the less prejudice and better social relations there will be between the groups (Allport, 1954). According to Allport (1954) there are four conditions that are ideal for contact to occur between groups: equal status, intergroup cooperation, common goals, and support by social and institutional authorities. Faculty and student athlete contact involves cooperation, a common goal of education, and support of positive contact by social and institutional authorities. However, faculty and student athlete s contact situation does not have equal

PAGE 36

36 status, because there is a hierarchical relationship between instructors and students. According to research, the three conditions that are met in this situation are all strongly associated with reducing prejudice (P ettigrew & Tropp, 2006). As applied to this study, I believe any opportunity faculty and student athletes have to create more contact will lessen the perception of student athlete deviance. Ther efore, this study has the following hypotheses: Older faculty will have lower perceptions of student athlete academic deviance. This is based on the idea that they have had more time to interact with student athletes throughout their careers. Additionally, older faculty will have more opportunity to become fans of th eir university sports program. Faculty affiliated with Science, Technology, Engineering, and Math ( STEM ) disciplines will have increased perceptions of student athlete deviance compared to faculty in other disciplines. This hypothesis is based on research regarding academic clustering of student athletes, which shows student athletes are more Iikely to be overrepresented in non STEM majors (Fountain & Finley, 2009). Faculty with more interaction with student athletes decreases perceptions of deviance. Facu lty involved in service to athletics will have lower perceptions of student athlete deviance than faculty not involved in service to athletics. Faculty with higher levels of fandom will have lower perceptions of student athlete deviance. Faculty that atte nded more university sporting events will have lower perceptions of student athlete deviance I have two hypotheses based on the sporting group for which faculty were asked questions. These hypotheses are based on perceptions or stereotypes related to devia nce by the general public, which may also extend to faculty. athletes more negatively than that receives a lot of fan support an users has been reinforced by media stories of high profile athletes being involved with these types of situations (Kluger, 2014 ; W adhwani, 2016 ) This can lead to

PAGE 37

37 misconceptions or stereotypes by faculty involved in deviance generally. Additionally, f aculty may use representations of (Skolnick, 1993). T he concept of a criminal in our culture is disproportionately portrayed as a young, black male (Drummond, 1990; Barlow, 1998; Russell, 2002). This typification of crime is compounded for football student athletes because the stereotype of a talented athlete is also a young, black male (Sailes, 1998) Faculty will perceive w student athletes more positively than revenue producing sport that receives t he least fan support and media attention of the three groups (Adams & Tuggle, 2004) More importantly, it is a female sport. An undisputed fact in criminology is that males commit more crime than females (Lauritsen, Heimer, & Lynch, 2009). Additionally, th e public perceives most crime to be committed by black males, not females (Drummond, 1990; Barlow, 1998; Russell, 2002). Therefore, I believe faculty would have similar perceptions as the general public, where a female sporting group would engage in less d eviance than a male sporting group.

PAGE 38

38 CHAPTER 3 METHODOLOGY Research Design Faculty at four universities were sampled for this project. The universities were selected from two regions of the country, the Midwest and South. This section will describe the h ow I selected two universities from each of these regions. The sampling frame of universities is based on their ranking in the 2017 U.S. News top 25 public schools for national universities (U.S. News, 2016). Table 3 1 shows each of the top 25 universitie s rank, region, NCAA Division I status, Football Bowl Subdivision participation, NCAA athletic conference, and undergraduate and faculty populations. These criteria help to narrow down the four universities ultimately selected for administration of faculty surveys. Selection criteria: US News Top 25 Public Universities Ranking University in the Midwest or South Regions of the country NCAA Division I status Football Bowl Subdivision (FBS) status NCAA conference in the Big 10 or SEC Comparable undergraduat e and faculty populations From the list of top 25 public universities, two institutions were automatically removed from the sampling frame because they have special status that would skew results (U.S. News, 2016). The University of North Carolina Chapel Hill and Pennsylvania State University were removed due to highly publicized NCAA violations

PAGE 39

39 those two schools from the sampling frame, institutions were first split b y region, focusing on the Midwest and South regions 1 Second, only those universities with NCAA Division I athletic programs that participate in the Football Bowl Subdivision (FBS) were considered. This is because FBS Division I athletic departments have larger student athlete populations and more publicized athletic events, increasing the probability of faculty interaction and awareness of student athletes on campus and in their classrooms (Lawrence et al., 2007). Third, only those universities in the Bi g Ten and Southeastern Conference (SEC) were selected. The only conference remaining in the Midwest region is the Big Ten. The South region still contains both the Atlantic Coast Conference (ACC) and SEC. However, a majority of the institutions in the samp ling frame from the North region are included in the ACC conference, limiting the number of institutions from the South. Therefore, institutions in the SEC were selected to compare in the South region. Fourth, institutions with the largest undergraduate student populations in each of the regions (South and Midwest) are selected from the institutions that remained in the sampling frame based on the above criteria. This information comes from the Integrated Postsecondary Education Data System (IPEDS) of 201 4 fall enrollment data of total degree seeking undergraduate students ( National Center for Educational Statistics 2015). This leaves the final four universities to include: University of Illinois (IL), 1 The South regio n was selected for convenience to the researcher and gaining IRB approval. The Midwest region is selected as a comparison group because the undergraduate student populations are most similar in size compared to the institutions in the North and West. Addit ionally, very few institutions in the West region are NCAA Division I athletic programs that participate in the Football Bowl Subdivision (FBS). The ACC is the only conference in the North region with more than one institution to compare. However, the ACC is spread throughout the entire east coast ranging from Florida to New York. There could be some bias with including these schools due to the spread across different regions of the country.

PAGE 40

40 University of Florida (UF), Ohio State University (OS U), and University of Georgia (UGA). These universities are also comparable on a number of other factors, making it reasonable to compare faculty perceptions of student athletes. These factors include US News Ranking, campus setting, undergraduate student populations, graduation rates, faculty populations, student athlete populations, Directors Cup standing, number of varsity teams, athletic revenue, and NCAA infractions in the last 10 years. See Table 3 2 for a more detailed comparison among the final fou r institutions selected. Institutional data come from the Integrated Postsecondary Education Data System (IPEDS) reporting system, which includes campus setting, undergraduate student population, graduation rates, and faculty population ( National Center f or Educational Statistics, 2015) UF and UGA are located in midsized cities, UI is located in a small city, and OSU is located in a large city. The undergraduate student populations are 32,959 (UI), 32,829 (UF), 44,741 (OSU), and 26,882 (UGA). Table 3 2 al so includes a breakdown of the undergraduate student populations by race and gender. The graduation rates are 84% (UI), 88% (UF), 84% (OSU), and 85% (UGA). Finally, the instructional faculty populations are 2,224 (UI), 2,472 (UF), 3,587 (OSU), and 1,908 (U GA). The data for athletics participation come from the Department of Education Equity in Athletics Data Analysis Tool for the reporting period of July 2014 June 2015 (U.S. Department of Education, 2015). Each institution is required to report athletic p equality for Title IX. The student athlete populations are 456 (UI), 512 (UF), 1050

PAGE 41

41 (OSU), and 552 (UGA). Table 3 2 also includes a breakdown of the student athlete populatio ns by gender. The total athletic revenue of each of the institutions is $74,469,976 (UI), $130,772,424 (UF), $170,903,135 (OSU), and $116,151,279 (UGA). 15 report for Division I athletics. It is sponsored by the National Association of Collegiate Directors of Athletics (NACDA) and Learfield Sports (Learfield Sports, 2015). Each institution is awarded in the Director institutions included in this study are ranked highly in these standings, with UF ranked 4 th OSU 7 th UGA 14th, and UI 31 st but are in a variety of spots on the ranking list. NCAA infr actions were obtained from the NCAA legislative services database, which I searched for major infractions during the last 10 years (NCAA, 2015a). UF has had one major violation in 2015 for impermissible recruiting by its football team, which resulted in a 30 day suspension of the assistant coach responsible for the action (NCAA, 2015b). UI had one violation in 2005 for improper benefits in their football program, which resulted in a one year probation sanction by the NCAA (NCAA, 2005). UGA has also had one violation in the past 10 years. In this case, the head swimming and diving coach arranged impermissible benefits by organizing an independent study with a professor to keep the student athlete eligible (NCAA, 2014d). The penalty for the violation was a fin e, nine competition suspension, and one year recruiting suspension for the head coach. OSU had two major infractions in the last 10 years in 2006 and 2011. The first (in

PAGE 42

42 im proper recruiting, extra benefits, academic fraud, unethical conduct, and failure to monitor students (NCAA, 2006). The penalty associated with this infraction was three NCAA probation, vacation of wins, reimbursement of revenue associated with cham pionships, and a recruiting ban of an assistant coach. The most recent infraction (in 2011) was for preferential treatment and extra benefits for work that was not performed by a football student athlete (NCAA, 2011). The penalties associated with this inf season bans, and vacation of wins. Sample Procedure and Recruitment Target P opulation The target population of the study is faculty members at four large Division I insti tutions in the South and Midwest results cannot be generalized to smaller NCAA Division I, II, or III institutions, because faculty experience with athletic programs and students athletes may be very different. Sam pling F rame The sampling frame includes a list of all faculty listed on department directory websites at the four institutions selected (UI, UF, OSU, and UGA). The sampling frame involved in this study includes faculty members at four NCAA Division I insti tutions (N = 7,680). More specifically, 1,649 faculty from UF, 1,712 from UGA, 2,314 from OSU, and email address were contacted to participate in the study. Additional ly, faculty listed as emeritus/retired or adjunct were not included in the sampling frame.

PAGE 43

43 Recruitment and Data Collection Procedures Recruitment of participants follows the guidelines of Dillman, Smyth, and Christian (2009) for Internet surveys. See Tabl e 3 3 for a timeline of recruitment emails. After receiving Instituti onal Review Board Approval ( Appendix B for IRB Protocol), faculty participants were sent a series of emails as discussed below. I chose to send the email near the beginning of the Fall se mester, because I expected it would be more likely that faculty would be available (compared to the summer months, for example). On Wednesday, September 14 th 2016 at 11:00 AM, I sent an introduction recruitment email to the sampling frame of faculty at th e four institutions (N = 7,680). This email introduced the researcher, informed faculty members about the study, and indicated that participation is voluntary and anonymous. This email also let them know that the link to the survey will be provided in an a dditional email one week later. See Appendix A for the Introduction Email Template. After sending the introduction recruitment email, I received 61 emails from faculty indicating that they did not want to participate in the study either because they belie ved they were not relevant for the study or they were not interested Reasons faculty declined to participate that fall under the not relevant theme are: retired, on sabbatical, not a faculty member, not teaching, teaching at a satellite/remote campus, jus t hired at the institution, or no longer at the institution. Reasons faculty listed under the not interested theme were: too busy, do not like/take surveys, no opinion, and that they athletes. Several faculty indicated that time was an issue for retired and too busy with other matters to think about student athle

PAGE 44

44 lergic to 4 for more details regarding faculty reasons for declining to participate themes and frequencies. Those 61 faculty were remov ed from the email listing and did not receive the survey link email. One week after the Introduction Email, the researcher sent a Survey Link Email to faculty at the four institutions (N = 7 6 19 ). This email reminded the faculty about the study and provide d the link to an anonymous online survey through www. qualitrics.com. See Appendix A for the Survey Link Email Template. After sending the Survey Link email, the researcher received 79 emails from faculty providing additional feedback about the survey, ind icating they completed the survey (n = 38) or declining to participate (n = 41). The researcher also received one not listed anywhere in the recruitment emails. Therefore, it shows the faculty member had to look it up at some point before or after taking the survey. Several themes emerged from faculty responses to not participate after receiving the Survey Link email ( Table 3 4 for specific details). The themes for declining to participate include: not relevant, no t interested, fear, and being a source of potential bias. Again, several said they did not interact with student athletes enough to participate. One notewor Student athletes have a very poor record of performance in my courses, and I am not willing to speak freely on th e subject for fear Both faculty that did not want to participate and who indicated completion of

PAGE 45

45 the survey wer e removed from the email list and did not receive the Follow Up email reminding them to complete the survey Two weeks after the Survey Link Email was sent, a Follow Up Email was sent to all faculty reminding them to take the survey if they had not (N = 7 540). According to Dillman et al. (2009), follow up emails are common practice to help increase response rates. If participants had taken the survey, they were asked to ignore the message. This vided a link to the survey for their convenience. See Appendix A for the Follow Up Email Template. The survey closed one week following the Follow Up Email. The researcher received 37 email responses from faculty after the Follow Up Email was sent. Simila r to the response after the Survey Link Email, faculty responses indicated they completed the survey (n = 8) or declined to participate (n = 29). See Table 3.4 for reasons they did not want to participate. In both the second email providing the survey lin k and follow up emails participants were provided a link to an anonymous Qualtrics survey. Upon opening the email, participants were brought to an informed consent screen that they read and sign ed electronically ( Appendix C). If they consented to the surve y Qualtrics directed the respondents to the survey questionnaire to complete. Survey Design The internet survey was administered through the internet host, Qualtrics. The questions were divided in 13 blocks or pages. Questions mostly consisted of closed ended questions with ordered response categories in a matrix format. However, some blocks had open ended questions to provide respondents opportunities to explain more. The average time to complete the survey was 7.97 minutes (median time).

PAGE 46

46 The first block of the survey asked respondents demographic questions, which included their age, sex, and race/ethnicity (Appendix D, questions 1 3). The second and third blocks asked questions regarding faculty status attributes. Specifically, the second block included questions regarding the university with which respondents were affiliated, academic rank, tenure status, whether they held an administrative position, their academic discipline, the undergraduate majors their department served, and their time at their curr ent institution (Appendix D, questions 4 10). The university measure asked UGA, UI, and OSU. Respondents were asked about their university affiliation because one anonymou s link was sent to all faculty in the sampling frame (Appendix D, question 4). The academic rank measure asked each faculty respondent to indicate their rank: lecturer, assistant professor, associate professor, full professor, or other (Appendix D, questio with an open ended response. The tenure status measure asked faculty respondents if they were tenured, not yet tenured, or not in tenure track (Appendix D, question 6). The admin istrative position measure asked faculty to indicate whether or not they held an administrative position. If they did, they were asked to indicate the type of administrative position (department/program head, assistant dean, associate dean, and other) (App endix D, question 7). The academic discipline measure asked faculty whether they were in architecture, arts and humanities, physical sciences, physical sciences and mathematics, social and behavioral sciences, or other (Appendix D, question 8). These disci plines were selected as answer options based on the disciplines listed in Digital

PAGE 47

47 Commons (2015). Time at current institution was measured by asking faculty how many years they had been at their current institution (Appendix D, question 10). The third blo ck included questions regarding participation in service involving athletics and service in an institutional governance role with responsibilities for athletics (Appendix D, questions 11 and 13). Respondents could either answer yes or no. Each of these que stions had skip patterns associated with them. If respondents selected no, they were brought to the next block of questions. If they selected yes, they were brought to an open ended question where they were asked to specify their involvement (Appendix D, q uestion 12 and 14). The fourth block asked respondents about their fandom and attendance at events for their university sports programs (Appendix D, questions 15 17). Respondents were displayed questions based on their university affiliation. For example, faculty from by Wann (2002) to measure fandom. These items included: I consider myself to b e a (Gator/Buckeye/Bulldog/Illini) fan, my friends see me as a (Gator/Buckeye/Bulldog/Illini) fan, I believe that following (Gator/Buckeye/Bulldog/Illini) sports is the most enjoyable form of entertainment (Appendix D, question set 15). Response options in cluded: strongly disagree (1), disagree (2), somewhat disagree (3), neither agree or disagree (4), somewhat agree (5), agree (6), and strongly agree (7). Faculty were also asked how often they attended (Gator/Buckeye/Bulldog/Illini) football, baseball, or basketball sporting events in the 2015 2016 academic year (Appendix D, question set 16). Respondents were asked about each of these sport groups. Only three sports were

PAGE 48

48 chosen to keep the survey length reasonable 2 Answer options included: never (1 ), rarely (2), sometimes (3), often (4), and very often (5). Additionally, faculty were asked an open ended question about whether there were any other (Gator/Buckeye/Bulldog/Illini) sporting events that they went to in the 2015 2016 academic year and to e xplain (Appendix D, question 17). The fifth block asked respondents about their interaction with student athletes on their campus during the 2015 2016 academic year (Appendix D, question 18 and 19). First, respondents were asked a group of close ended que stions with ordered response categories that were adapted from the Knight Commission Survey (Lawrence et al., 2007) (Appendix D, question set 18). The items included: student athletes are in you courses, student athlete communicate with you by email or in person, and student athletes interact with you during your class sessions. Response options included: never (1), rarely (2), sometimes (3), often (4), and very often (5). Additionally, respondents were provided an open ended question where they could expla in any other interaction they had with student athletes on their campus during the 2015 2016 academic year (Appendix D, question 19) The sixth and seventh blocks asked respondents their perceptions of the percentage of student athletes on their campus over all and by gender, race, and sport (Appendix D, questions 20 2 ue production and large fan support at all revenue producing male sport because each of the more i nteraction with student

PAGE 49

49 athletes who are: male/female, different racial categorie sliders from 0 to 100% so they could indicate the percentage of their liking. T he eighth block asked respondents their perception of the percentage of student one of three te ams by Qualtrics to determine if faculty hold different attitudes about student athletes based on the sport they play. Again, only three sports were chosen to keep the survey length reasonable. Additionally, I wanted to prevent an ordering effect of presen ting questions regarding more than one sport at a time. Qualtrics random assignment of sport groups to respondents is beneficial to eliminate any systematic bias 3 The ninth block asked respondents their knowledge about NCAA violations at their university (Appendix D, questions 25 26). Respondents that selected yes, were skipped to an open ended question where they were asked to describe what they know. Respondents that selected no, were skipped to the next block of questions. The tenth block asked responde nts if they were a varsity student athlete in college (Appendix D, questions 27 28). Again, if the respondent selected yes, they were asked to specify. If they selected no, they were skipped to the next block of questions. 3 Qualtrics equally distributed sport groups to participants through random assignment. 33.1% (n = 326) of respondents were a ssigned MFB, 33.3% (n = 328) MBA, and 33.7% (n = 332) WBB.

PAGE 50

50 The eleventh block of questions a sked respondents a group of close ended questions with ordered response categories about general perceptions of student athletes (Appendix D, question 29). These four items were adapted from Lawrence (2009). These were not sport coded; respondents were ask ed to indicate how strongly they agree with four statements about all student athletes on their campus. Response options include: not at all (1), slightly (2), moderately (3), and very much (4). Additionally, respondents were asked an open ended question w here they could provide any other comments they had about student athletes on their campus (Appendix D, question 30). The thirteenth block of questions asked about academic deviance for the specific sport group randomly assigned by Qualtrics earlier in the survey (either MFB, MBA, or WBB) at the eighth block, where they were asked about specific sport graduation rates questions at the eighth block will only answer questions the rest of the survey. Again, respondents were only asked academic and deviance questions about one sport group to keep the length of the survey short and prevent any ordering effect. The academic deviance items were close ended with ordered response cat egories in a matrix format ( Appendix D, question set 31). The items wer e adapted from Lin and Wen (2007 ). Participants were asked how many times in the last year athletes on their campus engaged in a variety of academic deviance behaviors. These academic deviance items included copying from other students, passing answers to other students during a test, using prohibited notes, obtaining tes t questions illegally,

PAGE 51

51 assignments, working on assignments with others when asked for individual work, getting extra help on an assignment from a tutor, providing a paper or a ssignment for another student, giving forbidden help to others on their assignments, doing less of their share of work in a group project, copying materials without citing them, and falsifying athletic travel letters to postpone exams or assignments. Respo nse options include: never (1), rarely (2), sometimes (3), often (4), all of the time (5). Finally, the fourteenth block asked about normative deviance for the specific sport group randomly assigned by Qualtrics earlier in the survey (either MFB, MBA, or WBB) at the eighth block, where they were asked about specific sport graduation rates (Appendix D, question 24). The student athlete normative deviance scale was adapted based on several criminological studies and the deviance scale in the National Youth S urvey (Agnew, 1991; Osgood, McMorris, & Potenza, 2002) ( Appendix D, question set 32 ) Participants were asked how many times in the last year their randomly assigned sport group on their campus engaged in a range of deviant behaviors. These normative devia nce items included: purposely damaging or destroying property belonging to others, st ealing somet hing worth more than $50, throwing objects at cars or people, stealing things worth $5 0 or less stealing money or other things from their friends, neighbors, or roommates, breaking into a building or vehicle to steal something or just look around being involved in a group fight, hit ting (or threatening to hit) other people, having (or trying to have) sexual relations with someone against their will drinking a lcohol, drinking more than 5 alcoholic drinks at once, selling/using marijuana or

PAGE 52

52 LSD, lying about their age to gain entrance or to purchase something, having sexual r elations with someone other than their significant other, being loud rowdy or unruly in a public place (disorderly conduct), buying or providing liquor for a minor, avoiding paying for such things as movies, clothing, and food, and being drunk in a publi c place Response opt ions include: never (1), rarely (2), sometimes (3), often (4), and all of the time (5) After participants answered these questions, they were thanked for completing the survey. Sample Characteristics Sample Size The total sample size for the study is 1,100 responses. There were 266 respondents from OSU (29.9%) 212 from UF (23.8%) 223 from UGA (25.1%) and 189 from UI (21.2%) An a priori power analysis was performed for sample size estimation using G Power 3.1 software A small eff ect size was used ( ES = .02 for regression f test) because this is an exploratory study and there are no similar studies to date (Cohen, 1988). With an alpha = 0.05 and power = 0.80, the projected sample size needed for each institution with this effect si ze is approximately N = 311. This sample size is needed at each institution because the researcher is examining the effect of variables by each university and region. Therefore, the ideal total sample would be 1244 (311 x 4 institutions) or a general respo nse rate of 16.2% of the sampling frame. I did not achieve this goal for the response rate, which is explained in the section below.

PAGE 53

53 Response Rate The general response rate of the survey was 14.3% (total of 1,100 responses divided by the sampling frame of 7,680). The response rate by university was 11.5% for OSU, 12.9% for UF, 13% for UGA, and 9.4% for UI. 4 Additionally, the researcher calculated the response rate using the American Association for Public Opinion (AAPOR) Research Response Rate Calculator, which is a stronger measure of survey quality when response rates are low (AAPOR, 2016). This measure uses a formula that standardizes response and non response rates for a variety of survey types, in particular for internet surveys of specifically named persons. cooperation, refusal, and contact rates for the survey (AAPOR, 2016). These rates are calculated using formulas based on four cate gories of outcomes ( Table 3 5). There are tw o values for response rate. Response Rate 1 (RR1) or the minimum resp onse rate is 11.7% (Table 3 6). This is the most conservative estimate of response rate for the sample because it does not take into account partially completed surveys. RR1 is the number of completed surveys (I) divided by the number of total surveys (both partial (P) and complete (I)) plus the number of refusals (R) plus the cases of unknown eligibility (UH and UO) (nothing returned or undeliverable) (AAPOR, 2016). See Equation 3 1 below (3 1) 4 Respondents were not forced to respond to any question in the survey, including the university they belonged to. Therefore, there was some item non response on that measure, which is why the uni versity response rates are lower than the general response rate.

PAGE 54

54 Response Rate 2 (RR2) is 14.3%. This is a less conservative estimate of response rate because it includes partially completed surveys as respondents. RR2 includes partially completed surveys (P) as respondents in the nu merator of its formula (AAPOR, 2016). See Equation 3 2 below. (3 2 ) The cooperation rate is the proportion of all faculty surveyed of all eligible faculty contacted (AAPOR, 2016). Cooperation Rate 1 (COOP1) or the minimum cooper ation rate is 72.5%. COOP1 is the number of completed surveys (I) divided by the number of surveys (both partial (P) and complete (I)) plus the number of refusals (R) plus other (O). This is different from refusal rate because it does not include unknown e ligibility, meaning faculty who never clicked the survey link and took the survey and faculty who never received the survey link because their emails bounced. See Equation 3 3 below. (3 3 ) Cooperation Rate 2 (COOP2) is 88.9%, which includ es partially (P) completed surveys in its formula. See Equation 3 4 below. (3 4 ) The refusal rate is the proportion of faculty who refused to participate in the survey (AAPOR, 2016). Refusal Rate 1 (REF1) is 1.8%. RR1 is calculated with the number of refusals (R) divided by the number of surveys (both partial (P) and complete

PAGE 55

55 (I)) plus the cases of unknown eligibility (UH and UO). Therefore, this is the refusal rate for the entire sampling frame. See Equation 3 5 below. (3 5 ) Refusal Rate 3 (REF3) is 11.1%. This calculation does not include the cases of unknown eligibility (UH and UO), meaning faculty who never clicked the survey link and those that bounced back as undeliverable. Therefore, this is the refusal rate am ong respondents with which I had some contact, either by them clicking the survey link or emailing me directly. See Equation 3 6 below. (3 6 ) Finally, the contact rate is the proportion of all faculty that was reached by the survey (AAPOR, 201 6). This calculation takes into account the entire sampling frame. More faculty may have been reached by the survey link email, but it is impossible to know without respondents actually clicking the survey link or responding back to me via email. The Conta ct Rate 1 (CON1) is 16.1%. See Equation 3 7 below. (3 7 ) Respondents versus S a mpling F rame In order to examine whether individuals with certain characteristics responded to the survey more frequently than others or non response bias, I compared respondents in the sample to the sampling frame. I am not able to compare non respondents to respondents because the survey was anonymous and I am not able to tell who did not participate from those who did. However, it is important to de termine if these differences exist to include the variables related to nonresponse as controls in the outcome models.

PAGE 56

56 The characteristics known about faculty from both the sample and sampling frame are university, academic rank, and discipline 5 The sampl ing frame and sample data were significantly different on all three of these characteristics (Table 3 7). Fewer faculty from Ohio State University and University of Illinois, institutions in the Midwest region and Big Ten conference, participated in the st udy compared to those in the sampling frame data ( X = 12.13, p < .01) Additionally, more faculty from University of Florida and University of Georgia, institutions in the south and Southeastern Conference, participated in the study compared to the sampli ng frame data. A possible explanation for the difference is because faculty from the south and southeastern conference felt more connected to the researcher because I am affiliated with University of Florida, a southern university. The sampling frame data were also significantly different from the sample data for the academic rank characteristics ( X = 134.91, p < .001). There were fewer lecturer, assistant professor, associate professor, and full professor faculty that participated in the survey than in t category in the sample data than in the sampling frame data. Although the overall chi square is significant, the percentages were within a point or two difference. A possible explanation for th e larger other category is that different titles that faculty have may not have been listed on the department websites from which the sampling frame was 5 Unfortunately, I do not have demographic information (i.e., race or sex) of the sampling frame. This is because I created my sampling frame of faculty based on department directories that do not list that type of demographic information, which would likely be biased.

PAGE 57

57 survey. Many o f them indicated they were emeritus (n = 7), clinical assistant or associate professors (n = 4), or adjunct/visiting professor (n = 4). There is no way to determine the actual reason for the discrepancy between the number of people in different ranks in th e sampling frame and the final sample, because the survey was anonymous. One possibility is that their ranks/titles may not have been updated in the department directory on the date the sampling frame was constructed. However, it is also possible that peop le of different ranks differentially participated. The sample also differed significantly overall from the sampling frame on academic discipline ( X = 170.98, p < .001). Fewer faculty in the academic disciplines of architecture, arts and humanities, busin ess, engineering, and life sciences participated in the survey than in the sampling frame. More faculty in the academic disciplines of education, law, medicine and health sciences, physical sciences and mathematics, social and behavioral sciences, and othe r participated in the survey than were in the sampling frame data. A possible explanation for this is that faculty in certain disciplines may be more willing to participate in survey research than others (i.e., social and behavioral sciences). Additionally I revealed my department affiliation, which is in the social and behavioral sciences, so faculty in that same discipline may have felt more compelled to participate than other disciplines. Early versus Late Respondents In order to examine whether the da ta from respondents represents the opinions of the sampling frame or population of faculty in general, I also compared early respondents to late respondents. There is research that supports that late respondents are similar to non respondents in surveys (M iller & Smith, 1983). Therefore, subjects who participated in the survey after the Survey Link Email was sent were identified as

PAGE 58

58 (n = 789) and those who participated after the Follow Up Email was sent (n = 311). Early and late respondents were compared on key independent and dependent variables ( Table 3 8) There was only one variable that early and late respondents were significant ly different on, which was race. More white faculty participated in the study early compared to non white faculty ( X = 6.05, p < .01) ( Table 3 8) These results, where there was only one significant difference between early and late respondents, statistically conclude that non respondents may be similar to late respondent s, which increases the external validity and generalizability of the study ( Radhakrishna & Doamekpor, 2008; Miller & Smith, 1983). Measures Independent Variables The independent variables used in this study can be grouped into four broad constructs : fac ulty status attributes, university status attributes, student athlete attributes, and university athletic status attributes. Faculty status attributes Faculty status a ttributes include respondent age, sex, race, academic rank, tenure status, administrati ve position, academic discipline, time at current institution, service involving athletics, sports fandom, and contact with student athletes. The age variable is a scale measure based on the number the respondent indicated as their age. However, two respon dents age responses were coded as missing because they entered 0 and 16 years old for their age, which do not appear to be plausible ages for faculty. The sex variable was dummy coded (male = 1, female = 0). The race variable was also dummy coded (white = 1, non white = 0). I chose to put all the non white race/ethnicities together because there were so few in each of the

PAGE 59

59 categories. Academic rank is measured in several dummy variables based on the answer option categories from the instrument (lecturer, as sistant professor, associate professor, full professor, and other). The tenure status, administrative position, and service for athletics variables were made into dummy variables (tenure = 1, non tenure = 0; administrator = 1, non administrator = 0; athlet ic service = 1, no athletic service = 0). The academic discipline measure was also made into a series of dummies based on the answer option categories (architecture, arts and humanities, business, education, engineering, law, life sciences, medicine and he alth sciences, physical sciences and mathematics, social and behavioral sciences, and other). Years at institution is a scale variable created by using the number faculty provided for the years they had been at their current institution. The sports fan var iable was created after running a Principal Component Analysis (PCA) using varimax rotation of the three fandom items. These items represented one component (one Eigenvalue greater than 1.0). This construct was used to create the index measure, where the t hree items were added up then divided by three. The fandom scale consists of 3 items, with components ranging from 0.94 to 0.84. Additionally, the scale had high internal consistency with a Cronba 0.89. See Table 3 9 for scale items and loadin gs. The scale ranges from 1 to 6, with higher scores indicating stronger fandom. variables were created by using the numeric response option selected from the Likert scale in the instrument. Answer options included: never (1), rarely (2), sometimes (3),

PAGE 60

60 often (4), very often (5). Therefore, higher scores indicate more attendance at sporting events. Finally, the student athlete interaction variable was created after running a P rincipal Component Analysis (PCA) using varimax rotation. The items indicated that these three items represented one construct (one Eigenvalue greater than 1.0), with components ranging from 0.94 to 0.93. This construct was used to create the index measure where the three items were added up then divided by three. Additionally, the scale had high internal consistency with a Cronba 10 for scale items and loadings. The scale ranges from 1 to 5, with higher scores indicating mor e interaction with student athletes. University status attributes University status attributes include the university, region, size of undergraduate student population, and size of faculty population. There are four dummy coded university related variabl es (UF, UGA, UI, and OSU). The regions variable is dummy coded based on where the university is in the country respondents are affiliated with, which are the South and Midwest. UF and UGA are in the south (coded 0). OSU and UI are in the Midwest (coded 1). The sizes of the undergraduate student populations and faculty populations were drawn from the IPEDS reporting system for each individual u niversity in (IPEDS, 2014) ( Table 3 2). Perceptions of student athlete a ttributes. The variables for student athlet e a ttributes include faculty perceptions of the gender, race, and sport of student athletes on their campus. Scale variables were created based on the numeric percentage faculty estimated of student athlete characteristics. Therefore, there are 12 variable s for faculty perceptions ( percentage of male, female, black/African American, white/Caucasian,

PAGE 61

61 Latino /Hispanic, Asian/pacific islander, other racial group, MFB, MBA, and WBB student athletes ). University Athletic Status Attributes Variables reflecting u niversity athletic status attributes are based on official measures of student athlete population, number of infractions for each of the four included universities. Student athlete population data, number of varsity teams, and athletic revenue data come from the Department of Education Equity in Athletics Data Analysis Tool for the reporting period of July 2014 June 2015, which are the latest available data ( Table 3 2). The D 15 report for Division I athletics, which is sponsored by the National Association of Collegiate Directors of Athletics (NACDA) and Learfield Sports (Learfield, 2015). Each institution is awarded points for the 3 2). NCAA infractions are obtained from the NCAA legislative services database. I searched fo r major infrac tions in the last 10 years ( Table 3 2). Dependent Variables There are two dependent variables utilized in this study based on the research questions. These include faculty perceptions of student athlete academic d eviance (e.g., cheating) an d student athlete normative d eviance (e.g., theft). Specifically, each faculty respondent was asked questions about only one sport baseball Academic deviance. I constructed the academic deviance scale using i tems from questio n set 31 in the instrument ( Appendix D). Principal components analysis

PAGE 62

62 (PCA) using varimax rotation was conducted to determine emergent constructs. PCA was run for the combined sport sample and for three sport subsamples, which was based o n the individual sporting team for which each respondent answered perceptions of deviance questions 6 Only components with eigenvalues above 1 were considered, using Kaiser (1960) criterion. Scales were determined using the .60/.40 rule, where loadings ab ove .60 were kept and below .40 were not (Costello & Osborne, 2005). This is the most conservative criterion to ensure there is not extraneous error in the scales created. Additionally, cross loadings (items that have loadings of .40 or higher on more than one components or factors) were dropped from consideration on the scales (Costello & Osborne, 2005). Table 3 11 shows how all the academic deviance items loaded using PCA for the combined and sport subsamples. Factors above .60 are highlighted and any it em with a cross loading has a strikethrough. There are consistent loading patterns across the combined sample and sport subsamples. There were 5 items that loaded on Component 1 across all groups: passed answers to other students during a test, obtained th e test questions illegally, used unauthorized electronic equipment on a test or assignment, provided a paper or assignment for another student, and falsified athletic travel letters to postpone exams or assignments. The WBB sport subsample also had one ite m (used prohibited notes) load on Component 1. There were 2 items that loaded on Component 2 across almost all groups: got extra help on an assignment from a tutor and did less of their share of work in a group project. Additionally, the MFB subsample 6 PCA was run for both the combined sample and individual sporting groups because faculty were randomly assigned one of three sports to answer a set of deviance questions about. This approach was to be sure the components and their variance did not differ across sports.

PAGE 63

63 also had one item (worked on assignments with others when asked for individual work) load on Component 2. Additional PCAs were run using each of the items that loaded on the two components individually to ensure they stick together and have strong scale reliab ility (Table 3 12 Component 1 were analyzed together and the 3 items that loaded on Component 2 were analyzed together. For each of these, the items loaded on one component with an Eigenvalue above 1 and had strong scale reliability. From this PCA, I created two academic deviance scales representing two areas of academic dishonesty: general cheating and relying on others for work. The general cheating scale included six items fro m the combined sample, with factor loadings ranging from .75 (falsified athletic travel letters to postpone exams or assignments) to .91 (passed answers to other students during a test). More specifically, the general cheating items included: passed answer s to other students during a test, used prohibited notes, obtained the test questions illegally, used unauthorized equipment on a test or assignment, provided a paper or assignment for another student, falsified athletic travel letters to postpone exams or assignments. The component loadings for the three sporting teams were also consistent with the combined sample 7 The scale values range from 1 to 5, with higher scores indicating higher perceptions of general cheating. 7 The WBB subgroup had an additional item that loaded on Componen t 1 different from the other subgroups. I chose not to include this item in the scale for the main analysis. There are additional analyses in Appendix E, which focus on the individual subgroups. Additionally, the factor loadings did vary in strength betwee n the sporting groups and combined sample. However, across all subgroups, the

PAGE 64

64 The second academic deviance scale, relying on others for work, was made up of three items, with loadings ranging from .89 (did less of their share of work in a group project) to .83 (got extra help on an assignment from a tutor) for the combined sample. The three items for the relying on ot hers for work scale include: got extra help on an assignment from a tutor, did less of their share of work in a group project, and worked on an assignment with others when asked for individual work. The component loadings for the three sporting teams were also consistent with the combined sample. The scale values range from 1 to 5, with higher scores indicating higher perceptions of relying on others. Normative deviance. A principal component analysis (PCA) using varimax rotation was conducted to determine emergent constructs on the normative deviance items (Appendix D, question set 32). Again, I ran PCA for the combined sample and each individu al sporting team. See Table 3 13 for how all the normative deviance items in the survey loaded using PCA. I conside red only components with eigenvalues above 1, using Kaiser (1960) criterion. I created scales using the .60/.40 rule, where loadings above .60 were kept and below .40 were not (Costello & Osborne, 2005). Additionally, cross loadings were dropped from consi deration on the scales. Table 3.13 shows how each of the normative deviance items load using PCA for the combined and sport subsamples. Factors above .60 are highlighted and any item with a cross loading has a strikethrough. There are some consistent load ing patterns across the combined sample and sport subsamples. There were 9 items that loaded on Component 1 similarly across several groups: purposefully damaged or destroyed property belonging to others, stolen something worth more than $50, thrown object s at

PAGE 65

65 cars or people, stolen things worth $50 or less, sold marijuana or hashish, stolen money or other things from their friends, neighbors, or roommates, sold harsh drugs such as heroin, cocaine, and LSD, used harsh drugs such as heroin, cocaine, and LSD, and broken into a building or vehicle to steal something or just look around. There were 5 items that loaded similarly on Component 2 across all groups: drank alcohol, drank more than 5 alcohol drinks at once, h ad sexual relations with a person other than their significant other b ought or provided liquor for a minor and been drunk in a public place. 8 Component 3 only emerged for the MFB subgroup with three items: s old marijuana or hashish s old harsh drugs such as heroin, cocaine, and LSD and used harsh drugs such as heroin, cocaine, and LSD It is reasonable for there to be differences between the three sporting groups and the items that are correlated because respondents are answering them thinking about the group they are randomly assigned. Because t here are not clear cut constructs emerging similarly from each group, I will use the combined group loadings to create the scales of normative deviance. Therefore, I will present and use the scales developed based on the loadings for the combined group due to differences across the three groups, so that I can compare apples to apples. However, I will present the models for the scales developed on individual loadings for each sport in Appendix E for the interested reader. PCAs were run using the items that l oaded on the two components individually for the combined group to ensure they stick together and have strong scale reliability ( Table 3 8 There were some differences in the factor loadings for Component 2 between the sporting group and combined sa mple. For the main analyses, I will use the combined scale loadings, but I also ran additional analysis in Appendi x E for each sporting group.

PAGE 66

66 presented in the main text, all 9 items that load ed on Component 1 for the combined group were analyzed together and the 6 items that loaded on Component 2 for the combined group were analyzed together. For each of these, the items loaded on one component with an Eigenvalue above 1 and had strong scale r eliability. Based on this information, normative deviance items represented two areas of deviance: criminal deviance and deviance related to alcohol behaviors. I believe the items that loaded on Component 1 represent a criminal deviance construct. Each of the items represent behaviors that are more serious and would be categorized as a crime. The items that loaded on Component 2 represent behaviors related to the college party environment, high risk activities, and drinking. 9 The first normative deviance s cale, criminal deviance, consists of 9 items, with loadings ranging from .82 (sold harsh drugs such as heroin, cocaine, and LSD) to .90 (stolen something worth more than $50) for the combined sample ( Tabl e 3 14 ). Other items include: purposely damaging or destroying property belonging to others, throwing objects at cars or people, stealing things worth $50 or less, selling marijuana or hashish, stealing money or things from friends, neighbors or roommates, using harsh drugs such as heroin, cocaine, and LSD, and breaking into a building or vehicle to steal something or just look around. The scale was created by adding each item up and dividing by the total number of items (9). The scale values range from 1 to 5, with higher scores indicating higher perception s of criminal deviance. 9 There are items similar to this in scales from studies looking at drinking and high risk behavior in Greek orga nizations on college campuses (Larimer et al., 2004).

PAGE 67

67 The second normative deviance scale, drinking related deviance, consisted of 6 items, with loadings ranging from .72 (had sexual relations with a person other than their significant other) to .89 (been drunk in a public place) for t he combined sample (refer to Table 3 13). Other items include: lying about their age to gain entrance or to purchase something, drinking alcohol, drinking more than 5 alcohol drinks at once, and buying or providing liquor for a minor. The scale was created by adding each item up and dividing by the total number of items (6). The scale values range from 1 to 5, with higher scores indicating higher perceptions of drinking related deviance. Analysis I first ran descriptive analyses to retrieve information rega rding the distributions of each variable. The design of the study is a 4 x 3 factorial, where university has four levels (UI, UF, UGA, and UF) and student athlete sport group has three levels (MFB, MBA, and WBB). Therefore, I examined the main effects of u niversity and sport group and whether there is an interaction effect by running a two way factorial ANOVA. I then obtained bivariate correlations for each of the variables to determine direction and strength of relationships. Additionally, significant v ariables at the bivariate level were used to inform the multivariate regression models. O rdinary Least Squares (OLS) regression models are run for each of the research questions for the entire sample. A key assumption of traditional OLS regression models i s the independence of observations, which is often a concern with nested data (Osborne, 2000; Johnson, 2010). In this case, faculty within university clusters may share unaccounted similarities, making the residual errors correlated. This could increase th e risk of making a Type I error, which shows a significant difference when there is not one. To control for the variance due to the university across analyses I used a model based approach by

PAGE 68

68 adding fixed effects of clusters in the regression models using dummy coding. That is, for the four universities, I included three university dummy coded variables into each of the regression models 10 After the research questions are considered for the entire sample, they are then re run by university to determine wh ether there are other important predictor variables of academic and normative deviance by university Bivariate correlations and traditional OLS regressions are run for each of the research questions. To determine statistically significant differences by u niversity, I ran a regression coefficient comparison tests developed by Clogg, Petkova, and Haritou (1995). Next, the research questions are examined by student athlete sporting group (MFB, MBA, and WBB) to determine whether there are differences in facul ty perceptions by sport. The Clogg et al. (1995) analysis is also used to determine significant differences exist between the coefficients. 10 Additionally, the intra class correlation coefficient (ICC) determines the proportion of variance in the outcomes that can be attributed to clustering. ICCs for each dependent variable determined the clustering effect of university is small, meaning faculty at the four universities are heterogeneous within and similar between them. The variability in the academic deviance dependent variables that can be attributed to between group differences was 2.2% for the general cheating scale and 1.7% for the relying on other scale. Additionally, the ICCs for the normative deviance scales was 0.0% for the criminal deviance scale and 1.2% for the drinking related deviance scale.

PAGE 69

69 Table 3 1. Demographics of Top 25 Public Universities 2017 US News Ranking 11 University Region NCAA Division I Foot ball Bowl Subdivision NCAA Conference Fall 2014 Undergraduate Student Enrollment 12 Fall 2014 Total Faculty 13 1 University of California Berkeley West Yes Yes Pac 12 27,126 1,672 2 University of California Los Angeles West Yes Yes Pac 12 29,633 3,137 2 U niversity of Virginia South Yes Yes ACC 16,483 2,370 4 University of Michigan Ann Arbor Midwest Yes Yes Big Ten 28,395 6,068 5 University of North Carolina Chapel Hill South Yes Yes ACC 18,350 3,318 6 College of William and Mary South No Yes Independe nt* 6,299 674 7 Georgia Institute of Technology South Yes Yes ACC 14,682 1,035 8 University of California Santa Barbara West Yes No Big West 20,238 909 9 University of California Irvine West Yes No Big West 24,489 1,664 10 University of California D avis West No No Big West 27,565 2,148 10 University of California San Diego West No No California Collegiate 24,810 2,035 10 University of Illinois Urbana Champaign Midwest Yes Yes Big Ten 32,959 2,317 10 University of Wisconsin Madison Midwest Yes Yes Big Ten 30,694 3,324 14 Pennsylvania State University University Park North Yes Yes Big Ten 40,541 3,351 14 University of Florida South Yes Yes SEC 32,829 4,229 16 Ohio State University Columbus Midwest Yes Yes Big Ten 44,741 3,688 16 Universit y of Washington West Yes Yes Pac 12 30,672 4,138 18 University of Georgia South Yes Yes SEC 26,882 2,583 18 University of Texas Austin West Yes Yes Big 12 39,523 2,745 20 Purdue University West Lafayette Midwest Yes Yes Big Ten 30,237 1,889 20 Unive rsity of Connecticut North Yes Yes American 18,395 2,007 20 University of Maryland College Park North Yes Yes Big Ten 27,056 3,262 23 Clemson University South Yes Yes ACC 17,260 1,147 24 University of Pittsburg North Yes Yes ACC 18,757 4,035 25 Rutger s University New Brunswick North Yes Yes Big Ten 34,544 3,822 11 U.S. News (2016) 12 National Cente r for Educational Statistics (2015) 13 National Center for Educational Statistics (2015)

PAGE 70

70 Table 3 2. Demographic of final four institutions University of Illinois University of Florida Ohio State University University of Georgia 2016 US News Ranking 14 10 14 16 18 AAU Member Institution Yes Yes Yes No Region Midwest South Midwest South Campus setting 15 Small city Midsize city Large city Midsize city 2014 Undergraduate Student Population 16 32,959 32,829 44,741 26,882 Gender: % Male 0.56 0.45 0.52 0.43 % Female 0.44 0 .55 0.48 0.57 Race: % Asian 0.14 0.07 0.06 0.08 % Black/African American 0.05 0.06 0.05 0.08 % Hispanic 0.07 0.16 0.04 0.05 % White 0.48 0.55 0.69 0.70 % Other 0.26 0.16 0.16 0.09 2014 Overall graduation rates 17 0.84 0.88 0.84 0.85 2014 Fa culty population 18 2,317 4,229 3,688 2,583 2014 Instructional faculty population 19 2,224 2,472 3,587 1,908 NCAA Conference Big Ten SEC Big Ten SEC Student Athlete Population 20 456 512 1050 552 Gender: % Male 0.60 0.56 0.55 0.47 % Female 0.40 0.44 0.45 0.53 14 U.S. News (2016) 15 National Center for Educational Statistics (2015) 16 National Center for Educational Statistics (2015) 17 National Center for Educational Statistics (2015) 18 National Center for Educational Statistics (2015) 19 National Center for Educational Statistics (2015) 20 U.S. Department of Education (2015)

PAGE 71

71 Table 3 2. Continued University of Illinois University of Florida Ohio State University University of Georgia 2014 2015 Director's Cup Standing 21 31 4 7 14 2015 AP College Football Rank Not ranked 25 4 Not ranked 2015 number of va rsity teams 22 17 17 32 17 FY 2015 total athletic revenue 23 $74,469,976 $130,772,424 $170,903,135 $116,151,279 Major NCAA infractions in last 10 years 24 Yes (1) Yes (1) Yes (2) Yes (1) 21 Learfield Sports (2015) 22 U.S. Department of Education (2015) 23 U.S. Department of Education (2015) 24 NCA A (2015)

PAGE 72

72 Table 3 3 Timeline of recruitment emails Initial email 9/14/2016 Survey link email 9/21/2016 Follow up email 10/5/2016 Survey closed 10/12/2016 Table 3 4 Faculty reasons for declining to participate themes Theme After intro email After survey link email After follow up email Total Frequency Not relevant Retired 8 1 1 10 On sabbatical 2 0 1 3 Not faculty 1 0 0 1 Not teaching 2 0 0 2 Teach at satellite/remote campus 2 2 1 5 Just hired at institution 1 3 0 4 No longer at institution 3 0 0 3 Not interested Decline/unsubscribe (no reason) 13 10 15 38 Too busy 15 5 4 24 Do not like surveys 3 2 1 6 No opinion 4 3 0 7 athletes 7 11 6 24 Fear Reprisal 0 1 0 1 Potential bias Conflict of interest 0 1 0 1 Dissertation committee member 0 2 0 2 Tot al 61 41 29 131

PAGE 73

73 Table 3 5 Outcome Rate Categories Categories Number Category 1: Interview Complete (I) 898 Partial (P) 202 Category 2: Eligible, non interview Refusal (i.e., responded to researcher declining participation) 131 Known respondent refusal (i.e., does not sign informed consent) 7 Total Refusal (R) 138 Category 3: Unknown eligibility, non interview Nothing returned (UH) 6420 Undeliverable (i.e., bounced emails) (UO) 42 Total 7 700 Table 3 6 Rate Estimate s Estimates Percent Response rate Response rate 1 (RR1) 11.7% Response rate 2 (RR2) 14.3% Cooperation rate Cooperation rate 1 (COOP1) 72.5% Cooperation rate 2 (COOP2) 88.9% Refusal rate Refusal rate 1 (REF1) 1.8% Refusal rate 3 (REF3) 1 1.1% Contact rate Contact rate 1 (CON1) 16.1%

PAGE 74

74 Table 3 7 Sampling frame data versus sample data Sampling frame data Sample data (N = 7680) (N = 1100) N % N % X University Ohio State University 2314 30.1 2 66 29.9 12.13** University of Florida 1649 21.5 212 23.8 University of Georgia 1712 22.3 223 25.1 University of Illinois 2005 26.1 189 21.2 Academic Rank Lecturer 923 12.0 122 11.7 134.91*** Assistant Professor 1570 20.4 206 19.8 Asso ciate Professor 2045 26.6 262 25.1 Full Professor 3098 40.3 407 39.0 Other 43 0.6 46 4.4 Discipline Architecture 149 1.9 10 1 170.98*** Arts and Humanities 1719 22.4 171 16.5 Business 537 7.0 57 5.5 Education 524 6.8 81 7.8 Engine ering 1187 15.5 92 8.9 Law 0 0 4 0.4 Life Sciences 1047 13.6 124 12.0 Medicine and Health Sciences 216 2.8 37 3.6 Physical Sciences and Mathematics 740 9.6 108 10.4 Social and Behavioral Sciences 999 13.0 100 19.3 Other 562 7.3 152 14.7

PAGE 75

75 Table 3 8. Early versus late respondents Early Late n % n % X Chi square analysis for nominal and ordinal variables (N = 789) (N = 311) Sex Female 306 70.00% 131 30.00% 1.54 Male 447 73.50% 161 26.50% Ra ce Non white 87 63.50% 50 36.50% 6.05** White 677 73.60% 243 26.40% University OSU 191 71.80% 75 28.20% 3.71 UF 159 75% 53 25% UGA 153 68.60% 70 31.40% UI 144 76.20% 45 23.80% Region South/SEC 312 71.70% 123 28.3 0% 0.41 Midwest/Big 10 335 73.63% 120 26.40% Academic Rank Lecturer 90 73.80% 32 26.20% 2.24 Assistant Professor 144 69.90% 62 30.10% Associate Professor 192 73.30% 70 26.70% Full Professor 305 74.90% 102 25.10% Other 30 65.20% 16 34.80% Tenure Status Non tenure 260 69.70% 113 30.30% 2.94 Tenure 500 74.60% 170 25.40% Administrative position Non administrator 581 72.60% 219 27.40% 0.12 Administrator 177 73.80% 63 26.30% Discipline Architectu re 11 73.30% 4 26.70% 8.77 Arts and Humanities 135 69.60% 59 30.40% Business 48 78.70% 13 21.30% Education 58 70.70% 24 29.30% Engineering 75 80.60% 18 19.40% Law 4 100% 0 0% Life Sciences 92 68.10% 43 31.90% Medicine and Health Sciences 34 75.60% 11 24.40% Physical Sciences and Mathematics 79 71.20% 32 28.80% Social and Behavioral Sciences 158 74.50% 54 25.50% Note. *p < .05, **p < .01, ***p < .001

PAGE 76

76 Table 3 8. Continued Early Late n % n % X Chi square analysis for nominal and ordinal variables (N = 789) (N = 311) Other 61 72.60% 23 27.40% Service to athletics No 487 71.50% 194 28.50% 1.25 Yes 264 74.80% 89 25.20% Athletic governance No 727 73% 269 27% 0.03 Yes 28 71.80% 11 28.20% T test for scale variables n x bar (sd) n x bar (sd) t Age 750 49.95 (11.74) 280 49.95 (11.98) 0.00 Years at institution 748 14.12 (10.47) 275 14.04 (10.49) 0.11 Negative perceptions of student athletes 540 2.13 (0.67) 198 2.22 (0.65) 1.6 5 Sport fandom 639 3.84 (1.61) 236 3.93 (1.57) 0.72 Interaction with student athletes 717 2.48 (1.14) 265 2.44 (1.11) 0.55 Attendance at MFB events 639 1.85 (1.29) 238 1.89 (1.20) 0.42 Attendance at MBA events 632 1.36 (0.78) 238 1.31 (0.70) 0.91 Attendance at WBB events 634 1.25 (0.65) 238 1.28 (0.76) 0.58 % estimate student athletes 499 6.16 (6.89) 187 6.14 (5.25) 0.04 % estimate male student athletes 467 42.93 (21.53) 162 41.17 (22.71) 0.89 % estimate female student athletes 466 35.94 (18.43) 161 34.10 (18.72) 1.09 % estimate black student athletes 408 34.87 (17.79) 155 34.30 (16.14) 0.35 % estimate white student athletes 414 46.10 (20.35) 157 44.83 (17.74) 0.69 % estimate Hispanic student athletes 372 9.26 (6.40) 143 8.69 (5.44) 0.95 % estimate Asian student athletes 317 4.96 (3.94) 120 4.74 (4.01) 0.51 % estimate other race student athletes 176 4.75 (4.33) 61 4.74 (3.72) 0.02 % estimate MFB student athletes 429 18.40 (17.47) 157 20.68 (18.38) 1.38 % estimate MBA student athletes 425 8.71 (8.71) 156 10.39 (10.92) 1.92 % estimate WBB student athletes 423 7.30 (7.88) 155 8.57 (9.60) 1.62 General cheating AD 343 1.95 (0.68) 125 1.90 (0.62) 0.70 Relying on others AD 360 2.73 (0.81) 128 2.87 (0.80) 1.69 Criminal devi ance 333 1.67 (0.55) 120 1.61 (0.47) 1.12 Drinking related deviance 325 2.65 (0.80) 118 2.66 (0.70) 0.07 Note. *p < .05, **p < .01, ***p < .001

PAGE 77

77 Table 3 9 Principal components analysis for fandom variable Item Loading Item 1. I consider myself t o be a (Gator/Buckeye/Bulldog/Illini) fan 0.92 Item 2. My friends see me as a (Gator/Buckeye/Bulldog/Illini) fan 0.94 Item 3. I believe that following (Gator/Buckeye/Bulldog/Illini) sports is the most enjoyable form of entertainment 0.84 Cronbach's alph a 0.89 Table 3 10 Principal component analysis for student athlete interaction variable Item Loading Item 1. Student athletes are in your courses 0.94 Item 2. Student athletes communicate with you by email or in person 0.93 Item 3. Student athletes interact with you during class sessions 0.94 Cronbach's alpha 0.93

PAGE 78

78 Table 3 11 Principal Components Analysis for all academic deviance items in instrument Item Factor loading Combined MFB MBA WBB 1 2 1 2 1 2 1 2 Copi ed from other students 0.67 0.60 0.60 0.67 0.62 0.67 0.78 0.43 Passed answers to other students during a test 0.85 0.31 0.85 0.28 0.86 0.30 0.87 0.30 Used prohibited notes 0.76 0.48 0.73 0.48 0.72 0.54 0.82 0.37 Obtained the test questions illegall y 0.79 0.35 0.76 0.37 0.76 0.38 0.85 0.25 Used unauthorized electronic equipment on a test or assignment 0.80 0.36 0.75 0.38 0.80 0.37 0.87 0.25 0.57 0.70 0.52 0.74 0.58 0.69 0.61 0.67 Worked on assignments with others when asked for individual work 0.48 0.77 0.39 0.82 0.54 0.74 0.50 0.75 Got extra help on an assignment from a tutor 0.01 0.86 0.02 0.86 0.02 0.87 0.01 0.82 Provided a paper or assignment for another student 0.81 0.35 0.79 0.32 0.83 0.32 0.81 0 .40 Gave forbidden help to others on their assignments 0.76 0.43 0.75 0.40 0.78 0.42 0.74 0.46 Did less of their share of work in group project 0.39 0.75 0.40 0.71 0.33 0.75 0.41 0.77 Copied materials without citing them 0.47 0.69 0.55 0.62 0.37 0.73 0.48 0.74 Falsified athletic travel letters to postpone exams or assignments 0.80 0.05 0.81 0.07 0.75 0.01 0.80 0.10

PAGE 79

79 Table 3 12 Principal Components Analysis for individual components of academic deviance Factor loading Combi ned MFB MBA WBB Component 1: General cheating Item 1: Passed answers to other students during a test 0.91 0.90 0.91 0.92 Item 2: Used prohibited notes 0.90 0.89 0.88 0.91 Item 3: Obtained the test questions illegally 0.89 0.87 0.88 0.90 It em 4: Used unauthorized electronic equipment on a test or assignment 0.90 0.87 0.89 0.92 Item 5: Provided a paper or assignment for another student 0.87 0.84 0.88 0.90 Item 6: Falsified athletic travel letters to postpone exams or assignments 0.75 0.76 0.67 0.81 Component 2: Relying on others Item 1: Got extra help on an assignment from a tutor 0.83 0.88 0.89 0.89 Item 2: Did less of their share of work in group project 0.89 0.85 0.84 0.78 Item 3: Worked on assignments with others when asked for individual work 0.87 0.84 0.86 0.90

PAGE 80

80 Table 3 13 Principal Components Analysis for all normative deviance items in instrument deviance variable Loading Combin ed MFB MBA WBB 1 2 1 2 3 1 2 1 2 3 Purposely damaged or destroyed property belonging to others? 0.75 0.39 0.83 0.28 0.17 0.68 0.41 0.60 0.27 0.64 Stolen (or tried to steal) something worth more than $50? 0.84 0.27 0.87 0.16 0.30 0.85 0.23 0.55 0. 19 0.70 Thrown objects (such as rocks, bottles, etc.) at cars or people? 0.77 0.30 0.67 0.31 0.28 0.81 0.17 0.69 0.22 0.47 Lied about their age to gain entrance or to purchase something: for example, lying about their age to buy liquor? 0.32 0.75 0. 25 0.74 0.12 0.33 0.72 0.10 0.68 0.45 Drank alcohol? 0.15 0.86 0.20 0.82 0.07 0.08 0.87 0.15 0.88 0.07 Drank more than 5 alcoholic drinks at once? 0.22 0.83 0.16 0.83 0.08 0.14 0.82 0.25 0.83 0.07 Stolen (or tried to steal) things worth $50 or less? 0.80 0.33 0.82 0.23 0.33 0.80 0.26 0.41 0.22 0.79 Had sexual relations with a person other than their significant other? 0.26 0.78 0.28 0.70 0.20 0.16 0.78 0.02 0.74 0.34 Been involved in a group fight? 0.63 0.49 0.53 0.41 0.29 0.57 0.49 0. 51 0.31 0.51 Sold marijuana or hashish ("pot", "grass", "hash")? 0.73 0.35 0.31 0.36 0.64 0.76 0.28 0.81 0.26 0.21 Used marijuana or hashish ("pot", "grass", "hash")? 0.42 0.68 0.27 0.67 0.24 0.37 0.68 0.39 0.61 0.26 Stolen money or other things from their friends, neighbors, or roommates? 0.81 0.31 0.68 0.21 0.50 0.81 0.25 0.55 0.32 0.61 Taken money or gifts from alumni 0.52 0.51 0.14 0.49 0.58 0.56 0.47 0.19 0.40 0.54 Hit (or threatened to hit) other people? 0.63 0.51 0.37 0.48 0.45 0.58 0.4 9 0.34 0.28 0.74 Been loud, rowdy, or unruly in a public place (disorderly conduct)? 0.46 0.68 0.35 0.61 0.27 0.32 0.71 0.31 0.57 0.49

PAGE 81

81 Table 3 13 C ontinued Loading Combined MFB MBA WBB 1 2 1 2 3 1 2 1 2 3 Sold harsh drugs such as heroin, cocai ne, and LSD? 0.81 0.12 0.26 0.08 0.86 0.81 0.06 0.89 0.13 0.16 Used hard drugs such as heroin, cocaine, and LSD? 0.77 0.28 0.35 0.19 0.72 0.73 0.30 0.82 0.22 0.25 Bought or provided liquor for a minor? 0.39 0.72 0.07 0.73 0.35 0.39 0.66 0.36 0 .69 0.37 Had (or tried to have) sexual relations with someone against their will? 0.61 0.49 0.42 0.42 0.47 0.42 0.58 0.63 0.19 0.41 Avoided paying for such things as movies, clothing, and food? 0.54 0.54 0.28 0.54 0.40 0.57 0.42 0.27 0.47 0.67 Been dru nk in a public place? 0.31 0.82 0.16 0.78 0.26 0.18 0.84 0.27 0.81 0.31 Broken into a building or vehicle (or tried to break in) to steal something or just look around? 0.81 0.27 0.43 0.23 0.64 0.83 0.19 0.80 0.18 0.33

PAGE 82

82 Table 3 14 Principal Components Analysis for individual components of normative deviance Loading Component 1: Criminal deviance Item 1: Purposely damaged or destroyed property belonging to others? 0.86 Item 2: Stolen (or tried to steal) something wor th more than $50? 0.90 Item 3: Thrown objects (such as rocks, bottles, etc.) at cars or people? 0.84 Item 4: Stolen (or tried to steal) things worth $50 or less? 0.87 Item 5: Sold marijuana or hashish ("pot", "grass", "hash")? 0.83 Item 6: Stolen m oney or other things from their friends, neighbors, or roommates? 0.89 Item 7: Sold harsh drugs such as heroin, cocaine, and LSD? 0.82 Item 8: Used hard drugs such as heroin, cocaine, and LSD? 0.83 Item 9: Broken into a building or vehicle (or tried to break in) to steal something or just look around? 0.86 Component 2: Minor deviance Item 1: Lied about their age to gain entrance or to purchase something: for example, lying about their age to buy liquor? 0.84 Item 2: Drank alcohol? 0.84 Item 3: Drank more than 5 alcoholic drinks at once? 0.88 Item 4: Had sexual relations with a person other than their significant other? 0.72 Item 5: Bought or provided liquor for a minor? 0.84 Item 6: Been drunk in a public place? 0.89

PAGE 83

83 CHAPTER 4 DESCRIPTIVE STATISTICS Independen t Variables Faculty Status A ttributes The sample cons isted of 1,100 faculty (Table 4 1 ). Faculty respondents ages ranged from 26 to 85 25 (mean = 49.95; median = 50). A majority of the sample was male (n = 608; 58.2%). The racial composition of the sample w as a majority White (n = 920; 87 .0 %), 4.4% Asian/Pacific Islander (n = 46), 3.2% Black/African American (n = 34), 2.6% Hispanic/Latino (n = 27), 1.3% Mixed Race/Biracial (n = 14), and 1.5% Other (n = 16). However, to allow for comparison throughout the stu dy between whites and non whites the group non white was comprised of Asian/Pacific Islander, Black/African American, Hispanic/Latino, Mixed Race/Biracial, and Other, which represented 13.0 % of respondents (n = 137 ). Faculty respondents average years at their institution was 14.1 (sd = 10.47). The academic rank composition of faculty that participated in the study was a majority full professor (n = 407; 39 .0 %), 25.1% associate professor (n = 262), 19.8% assistant professor (n = 206), 11.7% lecturer (n = 122), and 4.4% other (n = 46). A majority of the sample also were tenured (n = 670; 66.2%), 19.7% were not yet tenured (n = 199), and 14.1% were not in tenure track (n = 143). A majority of faculty surveyed were not in administrative positions (n = 800; 7 6.9%), However, 10.1% were department or program head, 0.4% were assistant deans (n = 4), 1.7% were associate deans (n = 18), and 10.9% said they had other 25 their age.

PAGE 84

84 administrative positions (n = 113). Faculty specified other administrative positions as associate director, director of center, associate chair, chair of undergraduate studies, teaching coordinator, graduate coordinator, and program coordinator. Social and behavioral sciences (20.5%) and arts and humanities (18.7%) were the most common academic discip lines reported by faculty. After that, faculty reported the following disciplines: 13.0% life sciences, 10.7% physical sciences and mathematics, 9.0% engineering, 8.1% other, 7.9% education, 5.9% business, 4.3% medicine and health sciences, 1.4% architectu re, and 0.4% law. Faculty specified the other disciplines as advertising, public affairs, agriculture, crop science, food science, environmental studies, horticulture, construction management, textiles, city and regional planning, urban planning, and sport management. A majority of faculty had not participated in service for athletics (65.9%, n = 681). However, faculty that did participate in service for athletics specified the type of service, with most faculty claiming teaching student athletes and fillin g out progress reports for at hletics as service ( Table 4 2 for types of responses). Additionally, only 3.8% (n = 39) of faculty indicated they served in an institutional governance role for athletics. There were 4 faculty that said they were faculty athlet ic representatives, 16 on the campus advisory board, 5 on the NCAA certification team, and 18 in some othe r institutional governance role The mean of the sport fandom scale was a 3.86 (s.d. = 1.60). The fandom items included: I consider myself to be a (G ator/Buckeye/Bulldog/Illini) fan, my friends see me as a (Gator/Buckeye/Bulldog/Illini) fan, and I believe that following (Gator/Buckeye/Bulldog/Illini) sports is the most enjoyable form of entertainment. The

PAGE 85

85 scale ranges from 1 to 6, with higher scores in dicating stronger fandom. Therefore, most faculty were between the somewhat disagree or neither agree or disagree range The mean for attendance at MFB ) events was 1.86 (s.d. = 1.26), MBA ) events 1.35 (s.d. = 0.76), and WBB ) events 1.26 (s.d. = 0.68). For this measure faculty were asked how often they attended in the 2015 2016 ac ademic year. Answer options ranged from 1 to 5, which means most faculty indicated that they rarely or never attended university athletic events. However, MFB had the most self reported attendance by faculty. Additionally, faculty indicated that they had between rarely and some interaction with student athletes on their campuses during the 2015 2016 academic year with a mean of 2.47 (s.d. = 1.31). The items for this scale included: student athletes are in you courses, student athlete communicate with you b y email or in person, and student athletes interact with you during your class sessions. The response options ranged from 1 to 5, with higher score indicating more interaction. University Status A ttributes The composition of university faculty surveyed wer e 29.9% from OSU (n = 266), 23.8% UF (n = 212), 25.1% UGA (n = 223), and 21.2% UI (n = 189). Therefore, 48.9% (n = 435) of faculty in the sample work for universities in the south region or a university in the Southeastern Conference and 51.1% (n = 455) wo rk in the Midwest region or university in the Big Ten Conference The undergraduate student populations are 32,959 (UI), 32,829 (UF), 44,741 (OSU), and 26,882 (UGA) ( Table 3 2) Table 3 2 also includes a breakdown of the

PAGE 86

86 undergraduate student populations by race and gender. Finally, the instructional faculty populations are 2,224 (UI), 2,472 (UF), 3,587 (OSU), and 1,908 (UGA). Perceptions of Student Athlete A ttributes The average percentage respondents estimated of student athletes on campus was 6.16 % (T able 4 3). The average percentage of male student athletes estimated by respondents was 42.48 % and females was 35.47 % 26 The average percentage estimated by respondents of student athletes racial groups was 32.71 % black/African American, 45.75 % white/Caucas ian, 9.10 % Latino/Hispanic, 4.90 % Asian/pacific islander, and 4.75 % other. Additionally, the average percentage respondents estimated of student athletes by sport that play football was 19.01 % 9.38 % 8.39 % University Athletic Status A ttributes Recall that t he actual student athlete populations are 456 (UI), 512 (UF), 1050 (OSU), and 552 (UGA) (Table 3 2) UI, UF, and UGA all have 17 varsity athletic teams, where OSU has 32. Each of the institutions has had at least o ne NCAA infraction in the last 10 years, OSU has had two infractions. The total athletic revenue of each of the institutions is $74,469,976 (UI), $130,772,424 (UF), $170,903,135 (OSU), and $116,151,279 (UGA). For UF ranked 4 th OS U 7 th UGA 14th, and UI 31 st 26 An issue with this measure using the Qualtrics software was that it did not force the respondents answers to add up to 100%.

PAGE 87

87 Dependent Variables Academic Deviance Results indicated that the mean of the general cheating academic deviance scale for the combined sample was 1.93 (s.d. = 0.66), meaning faculty rarely believe student athletes cheat gene ra lly. Recall, the general cheating items included: passed answers to other students during a test, used prohibited notes, obtained the test questions illegally, used unauthorized equipment on a test or assignment, provided a paper or assignment for another student, falsified athletic travel letters to postpone exams or assignments. The scale ranges from 1 to 5. The mean for the relying on others academic deviance scale for the combined sample was 2.76 (s.d. = 0.81), meaning faculty sometimes believed studen t athletes rely on others to cheat. See Table 4 4 for scale items descriptive statistics for the combined sample. Recall, the general cheating items included: passed answers to other students during a test, used prohibited notes, obtained the test question s illegally, used unauthorized equipment on a test or assignment, provided a paper or assignment for another student, falsified athletic travel letters to postpone exams or assignments. The scale ranges from 1 to 5. Normative Deviance Results indicated tha t the mean of the criminal deviance scale for the combined sample was 1.66 (s.d. = 0.53), meaning faculty rarely believe student athletes engage in criminal deviance. Recall, the criminal deviance items included: sold harsh drugs such as heroin, cocaine, a nd LSD, stolen something worth more than $50, purposely damaging or destroying property belonging to others, throwing objects at cars or people, stealing things worth $50 or less, selling marijuana or hashish, stealing money or things

PAGE 88

88 from friends, neighbo rs or roommates, using harsh drugs such as heroin, cocaine, and LSD, and breaking into a building or vehicle to steal something or just look around The scale ranges from 1 to 5. The mean of the drinking related deviance scale for the combined sample was 2 .65 (s.d. = 0.77), meaning faculty rarely believe stude nt athletes engage in drinking related deviance. Recall, the drinking related deviance items included: had sexual relations with a person other than their significant other, been drunk in a public plac e, lying about their age to gain entrance or to purchase something, drinking alcohol, drinking more than 5 alcohol drinks at once, and buying or providing liquor for a minor. The scale ranges from 1 to 5. See Table 4 5 for scale items descriptive statistic s for the combined sample. Diagnosing Missing Data For t he dependent variables, there are significant missing data due to item non respondents on most items) or did n ot respond to items at all (about 30 % of respondents) ( Table s 4 4 and 4 5 provide more details in regards to the dependent variable items and missing data). Therefore, it is important to determine the type of missing data and possible strategies to address it. The three types of missing data are: missing at random (MAR), missing completely at random (MCAR), or missing not at random (MNAR) (Rubin, 1976). Missing data methods can be used for data that are MCAR and MAR. However, there are limited solutions for data that are MNAR. Data that are MCAR have missing values that are unrelated to other variables in the data set as well as variables that are not measured in the dataset (Peugh & Enders, 2004). The cause of missingness is completely at random, which is t he ideal type of

PAGE 89

89 missing data to have because it is the least harmful to estimates. This is the only type of present study, Litt l MCAR test rules out the possibility that data are MCAR ( X = 86.47, p < .01). 27 Data are MAR when the missing values on a variable can be related to another measured variable or variables in the data set (Peugh & Enders, 2004). Additionally, the missing values must be unrelated to the value s of the variable with missingness itself to be MAR. There is no way to empirically confirm if the MAR mechanism is occurring in the data solely because of a variable that is measured in the data set because the data are missing. There were two codes of m r not to of answer options. Missing refers to the group of respondents that did not select anything. Chi square analyses were run to compare these three group differences on categorical variables. ANOVA were run to compares these three group differences on quantitative variables. There are several variables related to whether respondents h e dataset as shown in Tables 4 6 through 4 29 Variables that were consistently related to missingness for both academic and normative deviance were: academic discipline, service to athletics, sports fandom, interact ion with student athletes, and attendance at 27 has the null hypotheses that data are MCAR. Since the chi square has a p value that is statistically significant, I reject the null hypothesis that data are MCAR.

PAGE 90

90 MFB and WBB events were significantly related to missingness on the academic deviance items. More specifically, lecturers and non tenure track faculty were more likely to respond to the academic deviance items than faculty in other academic ranks. Additionally, faculty in the disciplines of business and medicine and health sciences were more likely to respond to the academic and normative deviance items than select prefer not to answer or not respond at all. In general, faculty who indicated they had more involvement and experience with athletics were more likely to respond to the deviance items than those who did not. For example faculty involved in service for athletics were more likely to respond to the deviance items than those who did not. had significantly more negative views of student athletes generally compared to those who respo n ded to the i tems ( Tables 4 6 to 4 29 ). Also, faculty that responded to the items had significantly higher levels of fandom, interaction with student athletes, and attendance at MFB events compared to those th were missing (Tables 4 6 to 4 29 ). There is a strong possibility that the data are missing not at random (MNAR), meaning there is a relationship between the tendency of a value to be missing and its values (Peugh & End ers, 2004). The issue with MNAR is that missing cases are produced by factors that are unknown, so the researcher cannot effectively control for them. In this particular study, data may be MNAR due to the sensitivity of the questions, which is common in so cial science research (Allison, 2002; Schwartz & Beaver, 2014). are systematically different from faculty that did answer beyond the variables that are

PAGE 91

91 already measured in th e data set. For most of the deviance items, more than 25% of Table 4 4 and 4 5 ). For example, one respondent emailed me after taking the survey saying unfortunately, I have a more negative view of what I see some of the players in other sports getting away with because some of them have been arrested. I also do think that it is very likely that athletes have access to old exams, old p apers and do share that information in the same way that we are aware that some of the fraternities and sororities do. I have no evidence of this, but many think that the Athletic Department itself provides that ma terial like this convey that for some respondents they felt uncomfortable or bad responding to the deviance items, meaning there was a social desirability response bias. Therefore, it was the items themselves that caused respondents to not respond, providing str ong evidence that the data are MNAR. Additionally, there were a few faculty that emailed me after taking the survey saying they felt uncomfortable selecting a response about student athlete deviance kn ow anything at all about [student athletes] collectively or individually the high percentage of missing ness for the dependent variables items. Participants may present r ather than not answering at all t actually know, but also makes it harder to analyze the construct of interest. In addition, the respondents are researchers and analytical, so they may be less like ly than the general public might be to make opinions without data.

PAGE 92

92 The most common way to deal with missing data is list wise deletion, where the entire case is deleted from the analysis if a single value is missing (Peugh & Enders, 2004). The advantage of list wise deletion is the ease of its use and that it provides a solid basis for estimating standard errors because the sample size is known. A disadvantage of list wise deletion is that it could result in biased parameter estimates when the data are not MCAR and the fraction of missing data are more than 5% MCAR test. Another disadvantage of listwise deletion is a loss of statistical power because the sample size is subst antially reduced by eliminating cases with one or more missing values (Peugh & Enders, 2004). According to Enders (2010), there ad hoc methods to account for MNAR data, which are the selection model and pattern mixture model. However, these methods are pri marily used with longitudinal data and clinical trials wit h attrition issues (Enders, 2010 ). Additionally, the assumptions to these methods require the researcher the specify values for the imputed scores. For this study, the reasons the data are missing a re uncertain and believed to be because of social desirability. Due to these concerns, I will use the traditional list wise deletion method for non response and caution the results of the study due to missing data.

PAGE 93

93 Table 4 1. Independent variables descr iptive statistics (N = 1,100) Age Mean = 49.95 SD = 11.80 Administrative Position Sex No 76.9% 800 Female 41.8% 437 Department/Program Head 10.1% 105 Male 58.2% 608 Assistant Dean 0.4% 4 Race Associate Dean 1.7% 18 White 87.0% 920 Other 10.9% 113 Non White 13.0% 137 Discipline Black/African American 3.2% 34 Architecture 1.4% 15 Latino/Hispanic 2.6% 27 Arts and Humanities 18.7% 194 Asian/Pacific Islander 4.4% 46 Business 5.9% 61 Mixed Race/Biracial 1.3% 1 4 Education 7.9% 82 Other 1.5% 16 Engineering 9.0% 93 University Law 0.4% 4 Ohio State University 29.9% 266 Life Sciences 13.0% 135 University of Florida 23.8% 212 Medicine and Health Sciences 4.3% 45 University of Georgia 25.1% 223 Physical Sciences and Mathematics 10.7% 111 University of Illinois 21.2% 189 Social and Behavioral Sciences 20.5% 212 Region/Conference Other 8.1% 84 South/SEC 48.9% 435 Years at Institution Mean= 14.10 SD = 10.47 Midwest/Big 10 51.1% 455 S ervice for athletics Academic Rank No 65.9% 681 Lecturer 11.7% 122 Yes 34.1% 353 Assistant Professor 19.8% 206 Institutional Governance Role for athletics Associate Professor 25.1% 262 No 96.2% 996 Full Professor 39.0% 407 Yes 3.8% 3 9 Other 4.4% 46 Faculty Athletic Representative 0.4% 4 Tenure Status Campus Advisory Board 1.4% 16 Tenured 66.2% 670 NCAA Certification Team 0.5% 5

PAGE 94

94 Table 4 1. Continued Not yet tenured 19.7% 199 Other 1.6% 18 Not in tenur e track 14.1% 143 Sport fandom Mean = 3.86 SD = 1.60 Attendance at MFB events Mean = 1.86 SD = 1.26 Attendance at MBA events Mean = 1.35 SD = 0.76 Attendance at WBB events Mean = 1.26 SD = 0.68 Student athlete interaction Mean = 2.47 SD = 1.31

PAGE 95

95 Table 4 2. Faculty athletic service themes (N = 353) Type of service n % Teaching student athletes 96 27.2% Filling out progress reports about student athletes 51 14.4% Serving on various committees (IAC, Title IX, audit, admissions review, eligibility, facility, coach search) 34 9.6% Involved with sports in college or high school 22 6.2% Tutoring 21 5.9% Attending athletic events 19 5.4% Leadership/professional development for athletes 18 5.1% Athletic board member 1 5 4.2% Consulting (Research, psychology, nutrition, medical) 11 3.1% Working with band 11 3.1% Recruiting 9 2.5% Handling absences/accommodating travel 8 2.3% Advising 7 2.0% Mentoring 7 2.0% Recommending tutors/Grad students tutors 5 1.4% Student conduct cases 2 0.1% Support student athletes with disabilities 2 0.1% Volunteering in community with athletes 2 0.1% NCAA certification team 1 0.0% No response 12 3.4%

PAGE 96

96 Table 4 3. Individual status attributes of student athl etes descriptive statistics Mean s.d. n Estimate % of total student athletes on campus 6.16 6.48 686 Estimate % of student athlete gender Male 42.48 21.83 629 Female 35.47 18.50 627 Estimate % of student athlete race Black/African Ameri can 34.71 17.34 563 White/Caucasian 45.75 19.66 571 Latino/Hispanic 9.10 6.15 515 Asian/Pacific Islander 4.90 3.96 437 Other 4.75 4.17 237 Estimate % of student athlete sport MFB 19.01 17.73 586 MBA 9.38 9.38 581 WBB 8.39 8.39 578

PAGE 97

97 Tab le 4 4. Descriptives of academic deviance scale s and i tems for the entire sample Never (1) Rarely (2) Sometimes (3) Often (4) All of the time (5) mean SD n Prefer not to answer Missing General cheating Passed answers to other students dur ing a test 26.6% 53.7% 17.9% 1.2% 0.6% 1.96 0.74 492 319 289 131 264 88 6 3 44.7% 29.0% 26.3% Used prohibited notes 21.7% 51.2% 22.4% 3.7% 1.0% 2.11 0.82 492 316 292 107 252 110 18 5 44.7% 28.7% 26.5% Obtained the test questions illegally 29. 9% 50.0% 16.8% 2.5% 0.8% 1.94 0.80 488 321 291 146 244 82 12 4 44.4% 29.2% 26.5% Used unauthorized electronic equipment on a test or assignment 27.0% 51.6% 18.3% 2.4% 0.6% 1.98 0.78 492 319 289 133 254 90 12 3 44.7% 29.0% 26.3% Provided a pap er or assignment for another student 26.5% 53.0% 17.9% 1.8% 80.0% 1.98 0.77 491 310 299 130 260 88 9 4 44.6% 28.2% 27.2% Falsified athletic travel letters to postpone exams or assignments 49.6% 42.1% 7.1% 0.6% 0.6% 1.60 0.70 496 310 294 246 209 3 5 3 3 45.1% 28.2% 26.7% Relying on others Got extra help on an assignment from a tutor 6.6% 13.2% 36.4% 34.5% 9.3% 3.27 1.02 516 288 296 34 68 188 178 48 46.9% 26.2% 26.9% Did less of their share of work in a group project 11.3% 32.0% 45.5% 9.5% 1.6% 2.58 0.87 503 303 294 57 161 229 48 8 45.7% 27.5% 26.7% Worked on assignments with others when asked for individual work 14.9% 35.9% 37.9% 9.3% 2.0% 2.48 0.93 496 310 294 74 178 188 46 10 45.1% 28.2% 26.7%

PAGE 98

98 Table 4 5 Descript ives of normative deviance scales and items for the entire sample Never (1) Rarely (2) Sometimes (3) Often (4) All of the time (5) mean SD n Prefer not to answer Missing Criminal deviance Purposely damaged or destroyed property belongin g to others? 34.7% 53.5% 10.7% 0.8% 0.4% 1.79 0.69 507 271 322 176 271 54 4 2 46.1% 24.6% 29.3% Stolen (or tried to steal) something worth more than $50? 36.2% 55.7% 7.4% 0.4% 0.4% 1.73 0.64 503 272 325 182 280 37 2 2 45.7% 24.7% 29.5% Thrown objects (such as rocks, bottles, etc.) at cars or people? 44.5% 48.7% 6.0% 0.4% 40.0% 1.64 0.65 497 279 324 221 242 30 2 2 45.2% 25.4% 29.5% Stolen (or tried to steal) things worth $50 or less? 29.8% 61.3% 8.1% 0.4% 0.4% 1.80 0.63 493 283 324 14 7 302 40 2 2 44.8% 25.7% 29.5% Sold marijuana or hashish ("pot", "grass", "hash")? 40.0% 54.0% 4.7% 0.9% 0.4% 1.68 0.64 470 303 327 188 254 22 4 2 42.7% 27.5% 29.7% Stolen money or other things from their friends, neighbors, or roommates? 35.3% 58.8% 5.3% 0.2% 0.4% 1.72 0.61 476 296 328 168 280 25 1 2 43.3% 26.9% 29.8% Sold harsh drugs such as heroin, cocaine, and LSD? 56.9% 41.6% 1.1% 0.0% 0.4% 1.45 0.57 471 302 327 268 196 5 0 2 42.8% 27.5% 29.7% Used hard drugs such as heroin, co caine, and LSD? 45.3% 50.4% 3.8% 0.0% 0.4% 1.60 0.61 472 299 329 214 238 18 0 2 42.9% 27.2% 29.9% Broken into a building or vehicle (or tried to break in) to steal something or just look around? 42.9% 53.2% 3.3% 0.2% 0.4% 1.62 0.60 483 291 326 20 7 257 16 1 2 43.9% 26.5% 29.6% Drinking related deviance Lied about their age to gain entrance or to purchase something: for example, lying about their age to buy liquor? 17.4% 35.4% 34.1% 11.1% 2.0% 2.45 0.97 505 274 321 88 179 172 56 10 45.9% 24.9% 29.2%

PAGE 99

99 Table 4 5 Continued Drank alcohol? 6.6% 9.3% 39.0% 41.5% 3.7% 3.26 0.92 516 262 322 34 48 201 214 19 46.9% 23.8% 29.3% Drank more than 5 alcoholic drinks at once? 10.0% 22.8% 42.3% 22.8% 2.2% 2.85 0.96 492 284 324 49 112 208 112 11 44.7% 25.8% 29.5% Had sexual relations with a person other than their significant other? 9.2% 24.1% 54.5% 10.5% 1.7% 2.71 0.84 468 306 326 43 113 255 49 8 42.5% 27.8% 29.6% Bought or provided liquor for a minor? 20.9% 42.6% 29.2% 6.1 % 1.3% 2.24 0.89 479 292 329 100 204 140 29 6 43.5% 26.5% 29.9% Been drunk in a public place? 12.8% 33.7% 44.4% 8.1% 1.0% 2.51 0.85 507 271 322 65 171 225 41 5 46.1% 24.6% 29.3%

PAGE 100

100 Table 4 6 Missing data analysis for item 1 of academi c deviance scale (N = 492) Item 1: Passed answers to other students during a test Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 492) (N = 319) (N = 289) Sex Female 202 46. 2 % 124 28.4 % 111 25.4 % 1.62 Male 283 46.5 % 189 31.1 % 136 22.4 % Race Non white 47 41.6 % 46 33.6 % 34 24.8 % 1.47 White 432 47.0 % 274 29.6 % 216 23.5 % University OSU 133 50 .0 % 71 26.7 % 62 23.3 % 5.31 UF 106 50 .0 % 55 25.9 % 51 24 .1 % UGA 107 48 .0 % 72 32.3 % 44 19.7 % UI 86 45.5 % 64 33.9 % 39 20.6 % Region South/SEC 213 49 .0 % 127 29.2 % 95 21.8 % 0.06 Midwest/Big 10 219 48.1 % 135 29.7 % 101 22.2 % Academic Rank Lecturer 73 59.8 % 27 22.1 % 22 18 .0 % 19.65* Assistant Professor 102 49.5 % 66 32 .0 % 38 18.4 % Associate Professor 105 40.1 % 82 31.3 % 75 28.6 % Full Professor 192 47.2 % 130 31.9 % 85 20.9 % Other 20 43.5 % 12 26.1 % 14 30.4 % Tenure Status Non tenure 196 52.5 % 105 28.2 % 72 19.3 % 7 .34* Tenure 295 44 .0 % 212 31.6 % 163 24.3 % Administrative position Non administrator 373 46.6 % 240 30 .0 % 187 23.40% 1.56 Administrator 116 48.3 % 77 32.1 % 47 19.60% Discipline Architecture 3 20 .0 % 9 60 .0 % 3 20 .0 % 34.21* Art s and Humanities 88 45.4 % 62 32 .0 % 44 22.7 % Business 40 65.6 % 13 21.3 % 8 13.1 % Education 37 45.1 % 23 28 .0 % 22 26.8 % Engineering 40 43 .0 % 25 26.9 % 28 30.1 % Law 4 100% 0 0 .0 % 0 0 .0 % Life Sciences 62 45.9 % 48 35.6 % 25 18.5 % Medicine an d Health Sciences 25 55.6 % 14 31.1 % 6 13.3 % Note. *p < .05, **p < .01, ***p < .001

PAGE 101

101 Table 4 6 Continued Item 1: Passed answers to other students during a test Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 492) (N = 319) (N = 289) Physical Sciences and Mathematics 45 40.5 % 38 34.2 % 28 25.2 % Social and Behavioral Sciences 104 49.1 % 65 30.7 % 43 20.3 % Other 39 46.4 % 20 23.8 % 25 29.8 % Service to athletics No 302 4 4.3 % 220 32.3 % 159 23.3 % 7.03* Yes 187 53 .0 % 94 26.6 % 72 20.4 % Athletic governance No 470 47.2 % 309 31 .0 % 217 21.8 % 1.28 Yes 22 56.4 % 10 25.6 % 7 17.9 % ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 485 50.02 (11.88) 309 50.21 (11.67) 236 49.48 (11.82) 0.27 Years at institution 488 13.89 (10.48) 314 14.37 (10.78) 221 14.19 (10.04) 0.21 Negative perceptions of student athletes 434 2.12 (0.66) 199 2.25 (0.68) 105 2.10 (0.65) 2.66 Sport fandom 429 4.19 (1.58) 262 3.61 (1.55) 184 2.66 (1.64) 9.35*** Interaction with student athletes 482 2.63 (1.14) 298 2.31 (1.13) 202 2.32 (1.07) 9.92*** Attendance at MFB events 431 2.06 (1.39) 262 1.70 (1.10) 184 1.61 (1.10) 11.44*** Attendance at MBA events 4 28 1.42 (0.80) 259 1.27 (0.64) 183 1.31 (0.81) 3.59* Attendance at WBB events 429 1.30 (0.71) 260 1.27 (0.72) 183 1.14 (0.52) 3.47* Note. *p < .05, **p < .01, ***p < .001

PAGE 102

102 Table 4 7 Missing data analysis for item 2 of academic deviance scale (N = 492) Item 2: Used prohibited notes Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 492) (N = 316) (N = 292) Sex Female 202 46.2 % 123 28.1 % 112 25.6 % 1.50 Male 283 46.5 % 187 30.8 % 138 22.7 % Race Non white 59 43.1 % 43 31.4 % 35 25.5 % 0.65 White 430 46.7 % 272 29.6 % 218 23.7 % University OSU 132 50 .0 % 72 27.1 % 62 23.3 % 3.88 UF 107 51 .0 % 55 25.9 % 50 23.6 % UGA 107 48 .0 % 71 31.8 % 45 20 .2 % UI 88 46.6 % 62 32.8 % 39 20.6 % Region South/SEC 214 49 .0 % 126 29 .0 % 95 21.8 % 0.06 Midwest/Big 10 220 48.4 % 134 29.5 % 101 22.2 % Academic Rank Lecturer 74 60.7 % 26 21.3 % 22 18 .0 % 18.41* Assistant Professor 101 49.0 % 65 31.6 % 40 19.4 % Associate Professor 107 40.8 % 80 30.5 % 75 28.6 % Full Professor 189 46.4 % 131 32.2 % 87 21.4 % Other 21 45.7 % 12 26.1 % 13 28.3 % Tenure Status Non tenure 196 52.5 % 103 27.6 % 74 19.8 % 7.15* Tenure 295 44 .0 % 211 31.5 % 1 64 24.5 % Administrative position Non administrator 371 46.4 % 238 29.8 % 191 23.9 % 2.33 Administrator 118 49.2 % 76 31.7 % 46 19.2 % Discipline Architecture 3 20 .0 % 9 60 .0 % 3 20 .0 % 36.08* Arts and Humanities 90 46.4 % 60 30.9 % 44 22.7 % Business 40 65.6 % 13 21.3 % 8 13.1 % Education 35 42.7 % 23 28 .0 % 24 29.3 % Engineering 38 41 .0 % 26 28 .0 % 29 31.2 % Law 4 100% 0 0 .0 % 0 0 .0 % Life Sciences 62 45.9 % 48 35.6 % 25 18.5 % Medicine and Health Sciences 26 57.8 % 13 28.9 % 6 13.3 % Note. *p < .05, **p < .01, ***p < .001

PAGE 103

103 Table 4 7 Continued Item 2: Used prohibited notes Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 492) (N = 316 ) (N = 292 ) Physical Sciences and Mathematics 46 41.4 % 37 33.3 % 28 25.2 % Social and Behavioral Sciences 104 49.1 % 65 30.7 % 43 20.3 % Other 39 46.4 % 20 23.8 % 25 29.8 % Service to athletics No 302 44.3 % 217 31.9 % 162 23.8 % 6.96* Yes 187 53 .0 % 94 2 6.6 % 72 20.4 % Athletic governance No 469 47.1 % 306 30.7 % 221 22.2 % 2.23 Yes 23 59.0 % 10 25.6 % 6 15.4 % ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 485 49.89 (11.90) 306 50.39 (11.63) 239 49.54 (11.82 ) 0.37 Years at institution 489 13.93 (10.52) 311 14.46 (10.77) 223 13.99 (9.96) 0.26 Negative perceptions of student athletes 432 2.11 (0.65) 198 2.27 (0.69) 108 2.10 (0.64) 3.87* Sport fandom 431 4.11 (1.58) 260 3.61 (1.57) 184 3.65 (1.62) 10.08*** Interaction with student athletes 482 2.64 (1.14) 295 2.30 (1.13) 205 2.30 (1.07) 11.53*** Attendance at MFB events 433 2.06 (1.39) 260 1.70 (1.10) 184 1.61 (1.10) 11.50*** Attendance at MBA events 430 1.42 (0.80) 257 1.26 (0.64) 183 1.31 (0.81) 3.6 8* Attendance at WBB events 431 1.30 (0.71) 258 1.26 (0.72) 183 1.14 (0.52) 3.50* Note. *p < .05, **p < .01, ***p < .001

PAGE 104

104 Table 4 8 Missing data analysis for item 3 of academic deviance scale (N = 488) Item 3: Obtained the test quest ions illegally Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 488) (N = 321) (N = 291) Sex Female 200 45.8 % 125 28.6 % 112 25.6 % 1.63 Male 281 46.2 % 190 31.3 % 137 22.5 % Race Non white 55 40.9 % 46 41.5 % 35 32.7 % 1.62 White 429 46.6 % 274 29.8 % 217 23.6 % University OSU 130 48.9 % 74 27.8 % 62 23.3 % 3.68 UF 105 49.5 % 56 26.4 % 51 24.1 % UGA 107 48 .0 % 72 32.3 % 44 19.7 % UI 87 46.0% 62 32.8 % 40 21.2 % Region South/SEC 212 48.7 % 128 29.4 % 95 21.8 % 0.10 Midwest/Big 10 217 47.7 % 136 29.9 % 102 22.4 % Academic Rank Lecturer 73 59.8 % 28 23 .0 % 21 17.2 % 19.28* Assistant Professor 102 49.5 % 65 31.6 % 39 18.9 % Associate Pro fessor 104 39.7 % 82 31.3 % 76 29 .0 % Full Professor 188 46. 2 % 132 32.4 % 87 21.4 % Other 21 45.7 % 12 26.1 % 13 28.3 % Tenure Status Non tenure 196 52.5 % 105 28.2 % 72 19.3 % 8.15* Tenure 292 43.6 % 213 31.8 % 165 24.6 % Administrative posit ion Non administrator 370 46.3 % 240 30 .0 % 190 23.8 % 2.33 Administrator 115 47.9 % 79 32.9 % 46 19.2 % Discipline Architecture 2 13.3 % 9 60.0 % 4 26.7 % 30.37** Arts and Humanities 88 45.4 % 62 32 .0 % 44 22.7 % Business 41 67.2 % 13 21.3 % 7 11.5 % Education 37 45.1 % 23 28 .0 % 22 26.8 % Engineering 40 43 .0 % 25 26.9 % 28 30.1 % Law 4 100 % 0 0 .0 % 0 0 .0 % Life Sciences 62 45.9 % 48 35.6 % 25 18. % Medicine and Health Sciences 26 57.8 % 14 31.1 % 5 11.1 % Note. *p < .05, **p < .0 1, ***p < .001

PAGE 105

105 Table 4 8 Continued Item 3: Obtained the test questions illegally Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 488 ) (N = 321 ) (N = 291 ) Physical Scienc es and Mathematics 44 39.6 % 39 35.1 % 28 25.2 % Social and Behavioral Sciences 101 47.6 % 66 31.1 % 45 21.2 % Other 38 45.2 % 20 23.8 % 26 31.0% Service to athletics No 299 43.9 % 219 32.2 % 163 23.9 % 7.23* Yes 186 52.7 % 97 27.5 % 70 19.8 % Athletic governance No 466 46.8 % 311 31.2 % 219 22 .0 % 1.40 Yes 22 56.4 % 10 25.6 % 7 17.9 % ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 481 49.98 (11.94) 311 50.32 (11.61) 238 49.44 (11.78) 0.38 Years a t institution 484 13.85 (10.53) 316 14.47 (10.78) 223 14.13 (9.91) 0.34 Negative perceptions of student athletes 428 2.12 (0.65) 201 2.26 (0.68) 109 2.10 (0.66) 2.79* Sport fandom 426 4.10 (1.59) 264 3.63 (1.57) 185 2.65 (1.62) 9.30*** Interaction wi th student athletes 478 2.64 (1.13) 300 2.32 (1.13) 204 2.30 (1.07) 10.25*** Attendance at MFB events 428 2.06 (1.39) 264 1.70 (1.10) 185 1.61 (1.09) 11.40*** Attendance at MBA events 425 1.42 (0.80) 261 1.27 (0.66) 184 1.30 (0.81) 3.32* Attendance a t WBB events 426 1.30 (0.71) 262 1.26 (0.72) 184 1.14 (0.51) 3.59* Note. *p < .05, **p < .01, ***p < .001

PAGE 106

106 Table 4 9 Missing data analysis for item 4 of academic deviance scale (N = 492) Item 4: Used unauthorized electronic equipment on a test or assignment Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 492) (N = 319) (N = 289) Sex Female 203 46.5 % 123 28.1 % 111 25.4 % 1.81 Male 282 46.4 % 190 31.3 % 136 2 2.4 % Race Non white 59 43.1 % 44 32.1 % 34 24.8 % 0.65 White 430 46.7 % 274 29.8 % 216 23.5 % University OSU 130 48.9 % 73 27.4 % 63 23.7 % 4.40 UF 107 50.5 % 55 25.9 % 50 23.9 % UGA 108 48.4 % 72 32.3 % 43 19.3 % UI 88 46.6 % 62 32 .8 % 39 20.6 % Region South/SEC 215 49.4 % 127 29.2 % 93 21.4 % 0.23 Midwest/Big 10 218 47.9 % 135 29.7 % 102 22.4 % Academic Rank Lecturer 75 61.5 % 26 21.3 % 21 17.2 % 21.94** Assistant Professor 103 50 .0 % 65 31.6 % 38 18.4 % Assoc iate Professor 105 40.1 % 81 30.9 % 76 29 .0 % Full Professor 188 46.2 % 133 32.7 % 86 21.1 % Other 21 45.7 % 12 26.1 % 13 28.3 % Tenure Status Non tenure 199 53.4 % 103 27.6 % 71 19.0 % 9.48** Tenure 292 43.6 % 214 31.9 % 164 24.5 % Administrat ive position Non administrator 373 46.6 % 240 30 .0 % 187 23.4 % 1.56 Administrator 116 48.3 % 77 32.1 % 47 19.6 % Discipline Architecture 3 20 .0 % 9 60 .0 % 3 20 .0 % 37.84** Arts and Humanities 88 45.5 % 60 30.9 % 46 23.7 % Business 41 67.2 % 13 21.3 % 7 11.5 % Education 37 45.1 % 23 28 .0 % 22 26.8 % Engineering 39 41.9 % 26 28.0 % 28 30.1 % Law 4 100 % 0 0 .0 % 0 0 .0 % Life Sciences 62 45.9% 48 35.6 % 25 18.5 % Medicine and Health Sciences 26 57.8 % 14 31.1 % 5 11.1 % Note. *p < .05 **p < .01, ***p < .001

PAGE 107

107 Table 4 9 Continued Item 4: Used unauthorized electronic equipment on a test or assignment Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 492) (N = 319) (N = 289) Physical Sciences and Mathematics 44 39.6 % 39 35.1 % 28 25.2 % Social and Behavioral Sciences 104 49.1 % 65 30.7 % 43 20.3 % Other 39 46.4 % 20 23.8 % 25 29.8 % Service to athletics No 301 44.2 % 219 32.2 % 161 23.6 % 7.65* Yes 188 53 .3 % 95 26.9 % 70 19.8 % Athletic governance No 468 47.1 % 309 31 .0 % 218 21.9 % 2.20 Yes 23 59 .0% 10 25.6 % 6 15.4 % ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 485 49.83 (11.91) 309 50.43 (11.61) 236 49.6 0 (11.84) 0.38 Years at institution 488 13.79 (10.49) 314 14.48 (10.80) 221 14.24 (9.98) 0.44 Negative perceptions of student athletes 432 2.11 (0.65) 199 2.27 (0.69) 107 2.12 (0.65) 3.92* Sport fandom 430 4.10 (1.58) 262 3.61 (1.58) 183 3.65 (1.62) 9.80*** Interaction with student athletes 482 2.64 (1.14) 298 2.30 (1.12) 202 2.30 (1.07) 11.53*** Attendance at MFB events 432 2.05 (1.29) 262 1.70 (1.10) 183 1.62 (1.10) 10.62*** Attendance at MBA events 429 1.42 (0.80) 259 1.26 (0.64) 182 1.31 (0. 81) 3.79* Attendance at WBB events 430 1.30 (0.71) 260 1.26 (0.72) 182 1.15 (0.52) 3.13* Note. *p < .05, **p < .01, ***p < .001

PAGE 108

108 Table 4 10 Missing data analysis for item 5 of academic deviance scale (N = 491) Item 5: Provided a paper or assignment for another student Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 491) (N = 310) (N = 299) Sex Female 204 46.7 % 118 27 .0 % 115 26.3 % 2.17 Male 279 45.9 % 187 30.8 % 142 23.4 % Race Non white 59 43.1 % 43 31.4 % 35 25.5 % 0.64 White 429 46.6 % 266 28.9 % 225 24.5 % University OSU 132 49.6 % 71 26.7 % 63 23.7 % 4.38 UF 104 49.1 % 54 25.5 % 54 25.5 % UGA 106 47.5 % 71 31.8 % 46 20.6 % UI 87 46 .0 % 61 32.3 % 41 21.7 % Region South/SEC 210 48.3 % 125 28.7 % 100 23 .0 % 0.01 Midwest/Big 10 219 48.1 % 132 29 .0 % 104 22.9 % Academic Rank Lecturer 74 60.7 % 25 20.5 % 23 18.9 % 18.59* Assistant Professor 100 48.5 % 65 31.6 % 41 19. 9 % Associate Professor 106 40.5 % 80 30.5 % 76 29 .0 % Full Professor 191 46.9 % 126 31 .0 % 90 22.1 % Other 20 43.5 % 12 26.1 % 14 30.4 % Tenure Status Non tenure 195 52.3 % 102 27.3 % 76 20.4 % 6.81* Tenure 295 44 .0 % 206 30.7 % 169 25.2 % Administrative position Non administrator 371 46.4 % 232 29 .0 % 197 24.6 % 2.67 Administrator 117 48.8 % 76 31.7 % 47 19.6 % Discipline Architecture 3 20 .0 % 9 60 .0 % 3 20 .0 % 32.78* Arts and Humanities 90 46.4 % 59 30.4 % 45 23.3 % Bu siness 40 65.6 % 13 21.3 % 8 13.1 % Education 37 45.1 % 23 28 .0 % 22 26.8 % Engineering 39 41.9 % 25 26.9 % 29 31.2 % Law 4 100 % 0 0 .0 % 0 0 .0 % Life Sciences 60 44.4 % 47 34.8 % 28 20.7 % Medicine and Health Sciences 24 53.3 % 13 28.9 % 8 17.8 % Note *p < .05, **p < .01, ***p < .001

PAGE 109

109 Table 4 10 Continued Item 5: Provided a paper or assignment for another student Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 491) (N = 310) (N = 2 99) Physical Sciences and Mathematics 45 40.5 % 37 33.3 % 29 26.1 % Social and Behavioral Sciences 106 50 .0 % 62 29.2 % 44 20.8 % Other 38 45.2 % 20 23.8 % 26 31.0 % Service to athletics No 301 44.2 % 213 31.3 % 167 24.5 % 7.20* Yes 187 53 0 % 92 26.1 % 74 21 .0 % Athletic governance No 468 47.1 % 300 30.1 % 227 22.8 % 1.33 Yes 22 56.4 % 10 25.6 % 7 17.9 % ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 484 49.92 (11.92) 300 50.26 (11.63) 246 49.64 (11.80) 0.19 Years at institution 485 13.93 (10.47) 308 14.22 (10.69) 230 14.31 (10.22) 0.13 Negative perceptions of student athletes 433 2.12 (0.66) 194 2.26 (0.69) 111 1.09 (0.64) 3.52* Sport fandom 426 4.10 (1.57) 257 3.60 (1.57) 192 3.68 (1.64) 9.69*** Interaction with student athletes 481 2.65 (1.14) 290 2.28 (1.11) 211 2.31 (1.08) 12.56*** Attendance at MFB events 428 2.06 (1.38) 257 1.69 (1.09) 192 1.64 (1.13) 10.62*** Attendance at MBA events 425 1.41 (0.80) 254 1.26 (0.63) 191 1.33 (0. 82) 3.52* Attendance at WBB events 426 1.29 (0.70) 255 1.25 (0.69) 191 1.18 (0.60) 1.93 Note. *p < .05, **p < .01, ***p < .001

PAGE 110

110 Table 4 11 Missing data analysis for item 6 of academic deviance scale (N = 496) Item 6: Falsified athleti c travel letters to postpone exams or assignments Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 496) (N = 310) (N = 294) Sex Female 206 47.1 % 118 27.0 % 113 25.9 % 2.21 Ma le 282 46.4 % 187 30.8 % 139 22.9 % Race Non white 60 43.8 % 43 31.4 % 34 34.8 % 0.55 White 433 47.1 % 266 28.9 % 221 24.0 % University OSU 132 49.6 % 70 26.3 % 64 2 4.1 % 4.89 UF 107 50.5 % 53 25 .0 % 53 24.5 % UGA 109 48.9 % 70 31.4 % 44 19.7 % UI 87 46.0 % 61 32.3 % 41 21.7 % Region South/SEC 216 49.7 % 123 28.3 % 96 22.1 % 0.23 Midwest/Big 10 219 48.1 % 131 28.8 % 105 23.1 % Academic Rank Lecturer 75 61.5 % 24 19.7 % 23 18.9 % 20.47** Assistant Professor 105 51 .0 % 61 30.1 % 39 18.9 % Associate Professor 105 40.1 % 82 31.3 % 75 28.6 % Full Professor 190 46.7 % 128 31.4 % 89 21.9 % Other 21 45.7 % 12 26.1 % 13 28.3 % Tenure Status Non tenure 201 53.9 % 98 26.3 % 74 19.8 % 9.68** Tenure 294 43.9 % 210 31. 3 % 166 24.8 % Administrative position Non administrator 376 47 .0 % 232 29 .0 % 192 24 .0 % 2.13 Administrator 117 48.8 % 76 31.7 % 47 19.6 % Discipline Architecture 3 20.0 % 9 60.0 % 3 20 .0 % 40.30** Arts and Humanities 92 47.4 % 57 29.4 % 45 23.2 % Business 41 67.2 % 12 19.7 % 8 13.1 % Education 36 43.9 % 23 28 .0 % 23 28 .0 % Engineering 39 41.9 % 25 26.9 % 29 31.2 % Law 4 100% 0 0 .0 % 0 0 .0 % Life Sciences 60 44.4 % 49 36.3 % 26 19.3 % Medicine and Health Sciences 27 60 .0 % 13 28. 9 % 5 11.1 % Note. *p < .05, **p < .01, ***p < .001

PAGE 111

111 Table 4 11 Continued (N = 496) Item 6: Falsified athletic travel letters to postpone exams or assignments Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and o rdinal variables (N = 496) (N = 310) (N = 294) Physical Sciences and Mathematics 43 38.7 % 39 35.1 % 29 26.1 % Social and Behavioral Sciences 107 50.5 % 61 28.8 % 44 20.8 % Other 39 46.4 % 20 23.8 % 25 29.8 % Service to athletics No 304 44. 6 % 213 31.3 % 164 24.1 % 7.39* Yes 189 53.5 % 92 26.1 % 72 20.4 % Athletic governance No 473 47.5 % 300 30.1 % 223 22.4 % 2.12 Yes 23 59.0% 10 25.6 % 6 15.4 % ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 48 9 49.71 (11.88) 300 50.73 (11.69) 241 49.50 (11.77) 0.93 Years at institution 491 13.77 (10.46) 307 14.49 (10.79) 225 14.30 (10.07) 0.50 Negative perceptions of student athletes 438 2.12 (0.66) 191 2.28 (0.68) 109 2.09 (0.64) 4.56* Sport fandom 432 4 .10 (1.58) 254 3.59 (1.57) 189 3.68 (1.61) 9.98*** Interaction with student athletes 485 2.67 (1.14) 290 2.24 (1.10) 207 2.31 (1.07) 15.70*** Attendance at MFB events 434 2.06 (1.38) 254 1.67 (1.09) 189 1.65 (1.13) 11.03*** Attendance at MBA events 4 31 1.42 (0.80) 251 1.25 (0.63) 188 1.32 (0.82) 2.94* Attendance at WBB events 432 1.30 (0.71) 252 1.25 (0.69) 188 1.16 (0.58) 2.57 Note. *p < .05, **p < .01, ***p < .001

PAGE 112

112 Table 4 12 Missing data analysis for item 1 of relying on others scale (N = 516) Item 1: Got extra help on an assignment from a tutor Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 516) (N = 288) (N = 296) Sex Female 218 49.9 % 108 24.7 % 111 25.4 % 2.34 Male 289 47.5 % 176 28.9 % 143 23.5 % Race Non white 61 44.5 % 42 30.7 % 34 24.8 % 1.25 White 452 49.1 % 245 26.6 % 223 24.4 % University OSU 136 51.1 % 67 25.2 % 63 23.7 % 3.60 UF 109 51.4 % 50 23.6 % 53 25 .0 % UGA 117 52.5 % 61 27.4 % 45 20.2 % UI 92 48.7 % 57 30.2 % 40 21.2 % Region South/SEC 226 52 .0 % 111 25.5 % 98 22.5 % 0.40 Midwest/Big 10 228 50.1 % 124 27.3 % 103 22.6 % Academic Rank Lecturer 77 63.1 % 22 18 .0 % 23 18.9 % 19.49* Assistant Professor 104 50.5 % 61 30.1 % 40 19.4 % Associate Professor 110 42 .0 % 76 29 .0 % 76 29 .0 % Full Professor 203 49.9 % 115 28.3 % 89 21.9 % Other 21 45.7 % 12 26.1 % 13 28.3 % Tenure Status Non tenure 202 54.2 % 96 25.7 % 75 20.1 % 5.87 Tenure 312 46.6 % 191 28.5 % 167 24.9 % Administrative position Non administrator 388 48.5 % 217 27.1 % 195 24.4 % 2.82 Administrator 124 51.7 % 70 29.2 % 46 19.2 % Discipline Architecture 3 20 .0 % 9 60 .0 % 3 20 .0 % 37.09* Arts and Humanities 96 49.5 % 53 27.3 % 45 23.2 % Business 42 68.9 % 11 18 .0 % 8 13.1 % Education 38 46.3 % 22 26.8 % 22 26.8 % Engineering 40 43 .0 % 24 25.8 % 29 31.2 % Law 4 100% 0 0 .0 % 0 0 .0 % Life Sciences 63 46.7 % 44 32.6 % 28 20.7 % Medicine and Health Science s 27 60 .0 % 12 26.7 % 6 13.3 % Note. *p < .05, **p < .01, ***p < .001

PAGE 113

113 Table 4 12 Continued Item 1: Got extra help on an assignment from a tutor Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variable s (N = 516) (N = 288) (N = 296) Physical Sciences and Mathematics 47 42.3 % 35 31.5 % 29 26.1 % Social and Behavioral Sciences 111 52.4 % 58 27.4 % 43 20.3 % Other 39 46.4 % 19 22.6 % 26 31 .0 % Service to athletics No 317 46.5 % 199 29.2 % 16 5 24.2 % 7.06* Yes 195 55.2 % 85 24.1 % 73 20.7 % Athletic governance No 492 49.4 % 279 28 .0 % 225 22.6 % 2.31 Yes 24 61.5 % 9 23.1 % 6 15.4 % ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 506 50.07 (11.97) 281 50.16 (11.55) 243 49.47 (11.75) 0.27 Years at institution 510 14.15 (10.65) 286 13.97 (10.51) 227 14.16 (10.05) 0.03 Negative perceptions of student athletes 450 2.13 (0.66) 179 2.26 (0.69) 109 2.09 (0.64) 3.21* Sport fandom 451 4.08 (1.58) 235 3 .60 (1.56) 189 3.67 (1.63) 8.73*** Interaction with student athletes 505 2.66 (1.14) 268 2.25 (1.10) 209 2.29 (1.08) 14.78*** Attendance at MFB events 453 2.03 (1.37) 235 1.69 (1.09) 189 1.65 (1.13) 9.02*** Attendance at MBA events 450 1.40 (0.79) 23 2 1.26 (0.65) 188 1.34 (0.83) 2.52 Attendance at WBB events 451 1.28 (0.69) 233 1.27 (0.71) 188 1.18 (0.60) 1.48 Note. *p < .05, **p < .01, ***p < .001

PAGE 114

114 Table 4 13 Missing data analysis for item 2 of relying on others scale (N = 503) Item 2: Did less of their share of work in a group project Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 503) (N = 303) (N = 294) Sex Female 208 47.6 % 117 26.8 % 112 25.6 % 1 .52 Male 287 47.2 % 181 29.8 % 140 23.0 % Race Non white 58 42.3 % 44 32.1 % 35 25.5 % 1.65 White 442 48 .0 % 258 28.0 % 220 23.9 % University OSU 134 50.4 % 69 25.9 % 63 23.7 % 5.09 UF 109 51.4 % 51 24.1 % 52 24.5 % UGA 107 48.0% 71 31.8 % 45 20.2 % UI 89 47.1 % 59 31.2 % 41 21.7 % Region South/SEC 216 49.7 % 122 28 .0 % 97 22.3 % 0.05 Midwest/Big 10 223 49 .0 % 128 28.1 % 104 22.9 % Academic Rank Lecturer 74 60.7 % 25 20.5 % 23 18.9 % 19.30* Assistant Professor 1 06 51.5 % 62 30.1 % 38 18.4 % Associate Professor 106 40.5 % 80 30.5 % 76 29 .0 % Full Professor 195 47.9 % 123 30.2 % 89 21.9 % Other 22 47.8 % 11 23.9 % 13 28.3 % Tenure Status Non tenure 201 53.9 % 99 26.5 % 73 19.6 % 8.06* Tenure 301 44.9 % 202 30.1 % 167 24.9 % Administrative position Non administrator 381 47.6 % 227 28.4 % 192 24 .0 % 2.10 Administrator 119 49.6 % 74 30.8 % 47 19.6 % Discipline Architecture 3 20 .0 % 9 60 .0 % 3 20 .0 % 39.05** Arts and Humanities 94 48.5 % 55 28.4 % 45 23.2 % Business 40 65.6 % 13 21.3 % 8 13.1 % Education 37 45.1 % 22 26.8 % 23 28.0 % Engineering 40 43 .0 % 24 25.8 % 29 31.2 % Law 4 100 % 0 0 .0 % 0 0 .0 % Life Sciences 63 46.7 % 46 34.1 % 26 19.3 % Medicine and Health Sciences 27 60 .0 % 13 28.9 % 5 11.1 % Note. *p < .05, **p < .01, ***p < .001

PAGE 115

115 Table 4 13 Continued Item 2: Did less of their share of work in a group project Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 503) (N = 303) (N = 294) Physical Sciences and Mathematics 42 37.8 % 39 35.1 % 30 27 .0 % Social and Behavioral Sciences 108 50.9 % 61 28.8 % 43 20.3 % Other 40 47.6 % 19 22.6 % 25 29.8 % Service to athletics No 307 45.1 % 210 30.8 % 164 24 .1 % 8.62* Yes 193 54.7 % 88 24.9 % 72 20.4 % Athletic governance No 480 48.2 % 293 29.4 % 223 22.4 % 1.91 Yes 23 59 .0 % 10 25.6 % 6 15.4 % ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 495 49.95 (11.87) 294 50.31 (11.73) 241 49.52 (11.75) 0.30 Years at institution 497 13.90 (10.45) 301 14.29 (10.69) 225 14.29 (10.07) 0.18 Negative perceptions of student athletes 443 2.14 (0.67) 187 2.23 (0.67) 108 2.09 (0.64) 1.83 Sport fandom 436 4.08 (1.59) 250 3.63 (1.57) 189 3.66 (1.62) 8.34*** Interaction with student athletes 493 2.65 (1.14) 282 2.27 (1.11) 207 2.31 (1.08) 13.09*** Attendance at MFB events 438 2.04 (1.37) 250 1.71 (1.12) 189 1.64 (1.13) 9.09*** Attendance at MBA events 436 1.42 (0.80) 246 1. 25 (0.63) 188 1.32 (0.82) 3.75* Attendance at WBB events 436 1.29 (0.69) 248 1.27 (0.72) 188 1.16 (0.58) 2.25 Note. *p < .05, **p < .01, ***p < .001

PAGE 116

116 Table 4 14 Missing data analysis for item 3 of relying on others scale (N = 496) Item 3: Worked on an assignment with others when asked for individual work Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 496) (N = 310) (N = 294) Sex Female 206 47.1% 120 27.5% 111 25.4% 1.31 Male 282 46.6% 185 30.4% 141 23.3% Race Non white 60 43.8% 43 31.4% 34 24.8% 0.55 White 433 47.1% 266 28.9% 221 24.0% University OSU 134 50.4% 69 25.9% 63 23.7% 5.11 UF 105 49.5% 54 25.5% 53 25.0% UGA 108 48.4% 71 31.8% 44 19.7% UI 88 46.6% 61 32.3% 40 21.2% Region South/SEC 213 49.0% 125 28.7% 97 22.3% 0.02 Midwest/Big 10 222 48.8% 130 28.6% 103 22.6% Academic Rank Lecturer 74 60.7% 25 20.5% 23 18.9% 18.39* Assistant Professor 102 49.5% 65 31.6% 39 18.9% Associate Professor 106 40.5% 81 30.9% 75 28.6% Full Professor 193 47.4% 125 30.7% 89 21.9% Other 21 45.7% 12 26.1% 13 28.3% Tenure Status Non tenure 197 52.8% 102 27.3% 74 19.8% 6.99* Tenu re 298 44.5% 206 30.7% 166 24.8% Administrative position Non administrator 375 46.9% 232 29.0% 193 24.1% 2.62 Administrator 118 49.2% 76 31.7% 46 19.2% Discipline Architecture 3 20.0% 9 60.0% 3 20.0% 35.83* Arts and Humaniti es 92 47.4% 57 29.4% 45 23.2% Business 40 65.6% 13 21.3% 8 13.1% Education 37 45.1% 23 28.0% 22 26.8% Engineering 40 43.0% 24 25.8% 29 31.2% Law 4 100.0% 0 0.0% 0 0.0% Life Sciences 61 45.2% 48 35.6% 26 19.3% Medicine and Health Sci ences 26 57.8% 13 28.9% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 117

117 Table 4 14 Continued Item 3: Worked on an assignment with others when asked for individual work Response Prefer not to answer Missing n % n % n % X Chi square analysis for no minal and ordinal variables (N = 496) (N = 310) (N = 294) Physical Sciences and Mathematics 45 40.50 % 37 33.3% 29 26.1% Social and Behavioral Sciences 105 49.5% 64 30.2% 43 20.3% Other 38 45.2% 20 23.8% 26 31.0% Service to athletics No 304 44.6% 213 31.3% 164 24.1% 7.39* Yes 189 53.5% 92 26.1% 72 20.4% Athletic governance No 473 47.5% 300 30.1% 223 22.4% 2.12 Yes 23 59.0% 10 25.6% 6 15.4% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd ) F Age 489 49.97 (11.89) 300 50.29 (11.69) 241 29.50 (11.78) 0.30 Years at institution 491 13.89 (10.48) 307 14.34 (10.77) 225 14.24 (10.06) 0.20 Negative perceptions of student athletes 437 2.12 (0.66) 193 2.26 (0.69) 108 2.10 (0.64) 3.45* Sport fandom 432 4.09 (1.59) 255 3.61 (1.56) 188 3.68 (1.62) 8.89*** Interaction with student athletes 486 2.65 (1.14) 289 2.29 (1.12) 207 2.30 (1.08) 12.23*** Attendance at MFB events 434 2.04 (1.38) 255 1.69 (1.10) 188 1.65 (1.13) 9.55*** Attendance at M BA events 431 1.41 (0.80) 252 1.25 (0.63) 187 1.34 (0.83) 3.41* Attendance at WBB events 432 1.29 (0.70) 253 1.25 (0.69) 187 1.17 (0.59) 2.15 Note. *p < .05, **p < .01, ***p < .001

PAGE 118

118 Table 4 15 Missing data analysis for Item 1 of crimina l deviance scale (N = 507) Item 1: Purposely damaged or destroyed property belonging to others Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 507) (N = 271) (N = 322) Sex 209 47.8 % 1 05 24 .0 % 123 26.1 % 1.04 Female 291 47.9% 160 26.3% 157 25.8% Male Race Non white 58 42.3% 40 29.2% 39 28.5% 1.93 White 446 48.5% 230 25.0% 244 26.5% University OSU 130 48.9% 64 24.1% 72 27.1% 3.81 UF 106 50.0% 4 7 22.2% 59 27.8% UGA 116 52.0% 57 25.6% 50 22.4% UI 95 50.3% 52 27.5% 42 22.2% Region South/SEC 222 51.0% 104 23.9% 109 25.1% 0.34 Midwest/Big 10 225 49.5% 116 25.5% 114 25.1% Academic Rank Lecturer 72 59.0% 21 17.2% 29 23.8% 15.44 Assistant Professor 95 46.1% 63 30.6% 48 23.3% Associate Professor 115 43.9% 68 26.0% 79 30.2% Full Professor 206 50.6% 106 26.0% 95 23.3% Other 19 41.3% 11 23.9% 16 34.8% Tenure Status Non tenure 185 49.6% 95 25.5% 93 24.9% 0.30 Tenure 321 47.9% 174 26.0% 175 26.1% Administrative position Non administrator 387 48.4% 205 25.6% 208 26.0% 0.34 Administrator 118 40.2% 64 26.7% 58 24.2% Discipline Architecture 2 13.3% 6 40.0% 7 46.7% 34.66 Arts and Humanities 94 48.5% 53 27.3% 47 24.2% Business 39 63.9% 11 18.0% 11 18.0% Education 39 47.6% 20 24.4% 23 28.0% Engineering 43 46.2% 21 22.6% 29 31.2% Law 4 100.0% 0 0.0% 0 0.0% Life Sciences 68 50.4% 37 27.4% 30 22.2% Medicine and Health Sciences 30 66.7% 9 20.0% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 119

119 Table 4 15 Continued Item 1: Purposely damaged or destroyed property belonging to others Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 507) (N = 271) (N = 322) Physical Sciences and Mathematics 43 38.7% 32 28.8% 36 32.4% Social and Behavioral Sciences 101 47.6% 62 29.2% 49 23.1% Other 42 50.0% 17 20.2% 25 29.8% Service to athl etics No 312 45.8% 186 27.3% 183 26.9% 5.98 Yes 190 53.8% 83 23.5% 80 22.7% Athletic governance No 482 48.4% 264 26.5% 250 25.1% 3.71 Yes 25 64.1% 7 17.9% 7 17.9% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 500 50.45 (11.80) 260 49.99 (11.97) 270 49.00 (11.61) 1.33 Years at institution 503 14.68 (10.82) 267 13.39 (10.16) 253 13.71 (10.05) 1.58 Negative perceptions of student athletes 439 2.14 (0.65) 171 2.26 (0.71) 128 2.08 (0.63) 3. 03* Sport fandom 444 4.06 (1.56) 220 3.61 (1.62) 211 3.70 (1.62) 7.39** Interaction with student athletes 496 2.61 (1.13) 256 2.30 (1.12) 230 2.36 (1.10) 8.23*** Attendance at MFB events 446 2.02 (1.35) 220 1.70 (1.13) 211 1.67 (1.16) 8.09*** Atten dance at MBA events 443 1.40 (0.80) 218 1.28 (0.64) 209 1.31 (0.79) 2.22 Attendance at WBB events 444 1.29 (0.71) 219 1.24 (0.67) 209 1.20 (0.61) 1.44 Note. *p < .05, **p < .01, ***p < .001

PAGE 120

120 Table 4 16 Missing data analysis for Item 2 o f criminal deviance scale (N = 503) Item 2: Stolen (or tried to steal) something worth more than $50 Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 503) (N = 272) (N = 325) Sex Female 208 47.6% 103 23.6% 126 28.8% 1.90 Male 288 47.4% 163 26.8% 157 25.8% Race Non white 57 41.6% 41 29.9% 39 28.5% 2.41 White 444 48.3% 229 24.9% 247 26.8% University OSU 129 48.5% 65 24.4% 72 27.1% 3.57 UF 105 4 9.5% 47 22.2% 60 28.3% UGA 116 52.0% 56 25.1% 51 22.9% UI 94 49.7% 52 27.5% 43 22.8% Region South/SEC 221 50.8% 103 23.7% 111 25.5% 0.52 Midwest/Big 10 223 49.0% 117 25.7% 115 25.3% Academic Rank Lecturer 70 57.4% 24 19 .7% 28 23.0% 12.59 Assistant Professor 95 46.1% 63 30.6% 48 23.3% Associate Professor 116 44.3% 66 25.2% 80 30.5% Full Professor 203 49.9% 106 26.0% 98 24.1% Other 19 41.3% 11 23.9% 16 34.8% Tenure Status Non tenure 183 49.1% 98 26.3% 92 24.7% 0.53 Tenure 319 47.6% 172 25.7% 179 26.7% Administrative position Non administrator 383 47.9% 206 25.8% 211 26.4% 0.47 Administrator 118 49.2% 64 26.7% 58 24.2% Discipline Architecture 2 13.3% 6 40.0% 7 46.7% 30.88 Arts and Humanities 92 47.4% 54 27.8% 48 24.7% Business 38 62.3% 11 18.0% 12 19.7% Education 39 47.6% 20 24.4% 23 28.0% Engineering 43 46.2% 21 22.6% 29 31.2% Law 4 100.0% 0 0.0% 0 0.0% Life Sciences 68 50.4% 36 26.7% 31 23.0% Medicine and Health Sciences 29 64.4% 10 22.2% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 121

121 Table 4 16 Continued Item 2: Stolen (or tried to steal) something worth more than $50 Response Prefer not to answer Missing n % n % n % X Chi squa re analysis for nominal and ordinal variables (N = 503) (N = 272) (N = 325) Physical Sciences and Mathematics 43 38.7% 33 29.7% 35 31.5% Social and Behavioral Sciences 102 48.1% 60 28.3% 50 23.6% Other 41 48.8% 18 21.4% 25 29.8% Service to a thletics No 308 45.2% 187 27.5% 186 27.3% 6.91* Yes 190 53.8% 83 23.5% 80 22.7% Athletic governance No 479 48.0% 266 26.7% 252 25.3% 4.18 Yes 25 64.1% 6 15.4% 8 20.5% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 496 50.50 (11.79) 261 49.89 (12.02) 273 49.02 (11.57) 1.39 Years at institution 499 14.67 (10.79) 268 13.33 (10.21) 256 13.82 (10.09) 1.55 Negative perceptions of student athletes 434 2.14 (0.65) 174 2.26 (0.71) 130 2.07 (0.63 ) 3.70* Sport fandom 441 4.07 (1.55) 220 3.61 (1.63) 214 3.69 (1.62) 8.01*** Interaction with student athletes 492 2.60 (1.13) 259 2.31 (1.13) 231 2.37 (1.10) 6.92** Attendance at MFB events 443 2.03 (1.36) 220 1.71 (1.14) 214 1.65 (1.14) 8.55*** A ttendance at MBA events 439 1.40 (0.80) 219 1.29 (0.65) 212 1.30 (0.77) 2.21 Attendance at WBB events 441 1.30 (0.73) 219 1.22 (0.63) 212 1.19 (0.60) 2.33

PAGE 122

122 Table 4 17 Missing data analysis for Item 3 of criminal deviance scale (N = 49 7) Item 3: Thrown objects (such as rocks, bottles, etc.) at cars or people Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 497) (N = 279) (N = 324) Sex Female 206 47.1% 108 24.7% 123 28.1% 0.96 Male 284 46.7% 165 27.1% 159 26.2% Race Non white 57 41.6% 41 29.9% 39 28.5% 1.88 White 438 47.6% 236 26.7% 246 26.7% University OSU 126 47.4% 67 25.2% 73 27.4% 4.01 UF 105 49.5% 48 22.6% 59 27.8% UGA 114 51.1% 58 26.0% 51 22.9% UI 93 49.2% 54 28.6% 42 22.2% Region South/SEC 219 50.3% 106 24.4% 110 25.3% 0.65 Midwest/Big 10 219 48.1% 121 26.6% 115 25.3% Academic Rank Lecturer 71 58.2% 22 18.0% 29 23.8% 13.90 Assi stant Professor 95 46.1% 63 30.6% 48 23.3% Associate Professor 114 43.5% 69 26.3% 79 30.2% Full Professor 198 48.6% 112 27.5% 97 23.8% Other 19 41.3% 11 23.9% 16 34.8% Tenure Status Non tenure 184 49.3% 96 25.7% 93 24.9% 0.74 Te nure 312 46.6% 181 27.0% 177 26.4% Administrative position Non administrator 379 47.4% 212 26.5% 209 26.1% 0.23 Administrator 116 48.3% 65 27.1% 59 24.6% Discipline Architecture 2 13.3% 6 40.0% 7 46.7% 37.13* Arts and Humani ties 92 47.4% 55 28.4% 47 24.2% Business 39 63.9% 11 18.0% 11 18.0% Education 38 46.3% 20 24.4% 24 29.3% Engineering 42 45.2% 21 22.6% 30 32.3% Law 4 100.0% 0 0.0% 0 0.0% Life Sciences 65 48.1% 40 29.6% 30 22.2% Medicine and Health Sciences 30 66.7% 9 20.0% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 123

123 Table 4 17 Continued Item 3: Thrown objects (such as rocks, bottles, etc.) at cars or people Response Prefer not to answer Missing n % n % n % X Chi square analysis for nom inal and ordinal variables (N = 497) (N = 279) (N = 324) Physical Sciences and Mathematics 41 36.9% 34 30.6% 36 32.4% Social and Behavioral Sciences 101 47.6% 62 29.2% 49 23.1% Other 41 48.8% 18 21.4% 25 29.8% Service to athletics No 303 44.5% 193 28.3% 185 27.2% 7.63* Yes 189 53.5% 84 23.8% 80 22.7% Athletic governance No 472 47.4% 272 27.3% 252 25.3% 4.21 Yes 25 64.1% 7 17.9% 7 17.9% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 491 50.41 (11.81) 267 50.11 (11.98) 272 48.97 (11.58) 1.34 Years at institution 493 14.46 (10.66) 275 13.80 (10.55) 255 13.74 (10.03) 0.55 Negative perceptions of student athletes 431 2.13 (0.65) 177 2.26 (0.72) 130 2.08 (0.62) 3.49* Sport fan dom 435 4.07 (1.58) 227 3.61 (1.63) 213 3.71 (1.61) 7.49** Interaction with student athletes 486 2.62 (1.14) 264 2.30 (1.12) 232 2.35 (1.10) 8.64*** Attendance at MFB events 437 2.03 (1.36) 227 1.70 (1.12) 213 1.67 (1.16) 8.46*** Attendance at MBA ev ents 434 1.41 (0.81) 225 1.27 (0.64) 211 1.32 (0.79) 2.55 Attendance at WBB events 435 1.30 (0.72) 226 1.23 (0.66) 211 1.19 (0.61) 1.77 Note. *p < .05, **p < .01, ***p < .001

PAGE 124

124 Table 4 18 Missing data analysis for item 4 of criminal dev iance scale (N = 493) Item 4: Stolen (or tried to steal) things worth $50 or less Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 493) (N = 283) (N = 324) Sex Female 204 46. 7% 110 25.2% 123 28.1% 0.89 Male 282 46.4% 167 27.5% 159 26.2% Race Non white 57 41.6% 41 29.9% 39 28.5% 1.59 White 434 47.2% 240 26.1% 246 26.7% University OSU 125 47.0% 69 25.9% 72 27.1% 4.00 UF 104 49.1% 49 23.1% 59 27 .8% UGA 116 52.0% 57 25.6% 50 22.4% UI 90 47.6% 55 29.1% 44 23.3% Region South/SEC 220 50.6% 106 24.4% 109 25.1% 1.24 Midwest/Big 10 215 47.3% 124 27.3% 116 25.5% Academic Rank Lecturer 66 54.1% 27 22.1% 29 23.8% 8.97 Assistant Professor 94 45.6% 63 30.6% 49 23.8% Associate Professor 115 43.9% 69 26.3% 78 29.8% Full Professor 199 48.9% 111 27.3% 97 23.8% Other 19 41.3% 11 23.9% 16 32.8% Tenure Status Non tenure 178 47.7% 101 27.1% 94 25.2% 0.1 5 Tenure 314 46.9% 180 26.9% 176 26.3% Administrative position Non administrator 373 46.6% 216 27.0% 211 26.4% 0.75 Administrator 118 49.2% 65 27.1% 57 23.8% Discipline Architecture 2 13.3% 6 40.0% 7 26.7% 31.97* Arts and Humanities 90 46.6% 56 28.9% 48 24.7% Business 39 63.9% 11 18.0% 11 18.0% Education 39 47.6% 20 24.4% 23 28.0% Engineering 42 45.2% 23 24.7% 28 30.1% Law 3 75.0% 1 25.0% 0 0.0% Life Sciences 65 48.1% 40 29.6% 30 22.2% Medicine and Health Sciences 29 64.4% 10 22.2% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 125

125 Table 4 18 Continued Item 4: Stolen (or tried to steal) things worth $50 or less Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 493) (N = 283) (N = 324) Physical Sciences and Mathematics 42 37.8% 32 28.8% 37 33.3% Social and Behavioral Sciences 100 37.2% 63 29.7% 49 23.1% Other 40 47.6% 18 21.4% 26 31.0% Service to athletics No 3 01 44.2% 195 28.6% 185 27.2% 7.68* Yes 188 79.0% 86 24.4% 79 22.4% Athletic governance No 468 47.0% 276 27.7% 252 25.3% 4.43 Yes 25 64.1% 7 17.9% 7 17.9% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F A ge 486 50.42 (11.82) 272 50.01 (11.94) 272 49.06 (11.80) 1.16 Years at institution 490 14.55 (10.73) 278 13.57 (10.32) 255 13.82 (10.12) 0.91 Negative perceptions of student athletes 428 2.13 (0.65) 181 2.26 (0.71) 129 2.09 (0.63) 3.24* Sport fandom 432 4.09 (1.55) 230 3.60 (1.62) 213 3.68 (1.62) 9.03*** Interaction with student athletes 482 2.61 (1.13) 268 2.32 (1.14) 232 2.35 (1.10) 7.17** Attendance at MFB events 434 2.04 (1.36) 230 1.69 (1.13) 213 1.66 (1.16) 9.30*** Attendance at MBA events 430 1.40 (0.80) 229 1.30 (0.69) 211 1.30 (0.75) 1.05 Attendance at WBB events 432 1.30 (0.73) 229 1.23 (0.64) 211 1.19 (0.61) 2.04 Note. *p < .05, **p < .01, ***p < .001

PAGE 126

126 Table 4 19 Missing data analysis for item 5 of criminal devianc e scale (N = 470) Item 5: Sold marijuana or hashish ("pot", "grass", "hash") Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 470) (N = 303) (N = 327) Sex Female 195 44.6% 11 8 27.0% 124 28.4% 0.81 Male 269 44.2% 178 29.3% 161 26.5% Race Non white 53 38.7% 44 32.1% 40 29.2% 2.06 White 415 45.1% 257 27.9% 248 27.0% University OSU 122 45.9% 72 27.1% 72 27.1% 4.38 UF 101 47.6% 51 24.1% 60 28.3% UGA 106 47.5% 66 29.6% 51 22.9% UI 87 46.0% 59 31.2% 43 22.8% Region South/SEC 207 47.6% 117 26.9% 111 25.5% 0.42 Midwest/Big 10 209 45.9% 131 28.8% 115 25.3% Academic Rank Lecturer 67 54.9% 26 21.3% 29 23.8% 11.96 As sistant Professor 93 45.1% 64 31.1% 49 23.8% Associate Professor 109 41.6% 73 27.9% 80 30.5% Full Professor 184 45.2% 125 30.7% 98 24.1% Other 17 37.0% 13 28.3% 16 34.8% Tenure Status Non tenure 177 47.5% 102 27.3% 94 25.2% 1.47 Tenure 292 43.6% 199 29.7% 179 26.7% Administrative position Non administrator 358 44.8% 229 28.6% 213 26.6% 0.60 Administrator 110 45.8% 72 30.0% 58 24.2% Discipline Architecture 2 13.3% 6 40.0% 7 46.7% 35.95* Arts and Hum anities 89 45.9% 57 29.4% 48 24.7% Business 39 63.9% 11 18.0% 11 18.0% Education 35 42.7% 24 29.3% 23 28.0% Engineering 41 44.1% 23 24.7% 29 31.2% Law 3 75.0% 1 25.0% 0 0.0% Life Sciences 62 45.9% 42 31.1% 31 23.0% Medicine and Heal th Sciences 28 62.2% 11 24.4% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 127

127 Table 4 19 Continued Item 5: Sold marijuana or hashish ("pot", "grass", "hash") Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 470) (N = 303) (N = 327) Physical Sciences and Mathematics 38 34.2% 37 33.3% 36 32.4% Social and Behavioral Sciences 92 43.4% 70 33.0% 50 23.6% Other 39 46.4% 18 21.4% 27 32.1% Service to athletics No 287 42. 1% 207 30.4% 187 27.5% 6.99* Yes 179 50.7% 94 26.6% 80 22.7% Athletic governance No 449 45.1% 292 29.3% 255 25.6% 1.52 Yes 21 53.8% 11 28.2% 7 17.9% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 46 4 50.03 (11.81) 291 50.65 (11.94) 275 49.09 (11.61) 1.26 Years at institution 467 14.27 (10.64) 298 14.08 (10.57) 258 13.82 (10.07) 0.16 Negative perceptions of student athletes 409 2.13 (0.65) 196 2.25 (0.71) 133 2.09 (0.63) 2.99 Sport fandom 413 4. 09 (1.55) 248 3.61 (1.62) 214 3.70 (1.62) 8.55*** Interaction with student athletes 459 2.61 (1.13) 288 2.34 (1.14) 235 2.35 (1.10) 6.98** Attendance at MFB events 415 2.04 (1.35) 248 1.72 (1.16) 214 1.67 (1.16) 8.09*** Attendance at MBA events 412 1 .40 (0.79) 246 1.30 (0.69) 212 1.32 (0.79) 1.63 Attendance at WBB events 414 1.29 (0.70) 246 1.25 (0.69) 212 1.20 (0.61) 1.31 Note. *p < .05, **p < .01, ***p < .001

PAGE 128

128 Table 4 20 Missing data analysis for item 6 of criminal deviance scal e (N = 476) Item 6: Stolen money or other things from their friends, neighbors, or roommates Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 476) (N = 296) (N = 328) Sex Fem ale 195 44.6% 118 27.0% 124 28.4% 0.42 Male 275 45.2% 171 28.1% 162 26.6% Race Non white 53 38.7% 45 32.8% 39 28.5% 2.79 White 421 45.8% 250 27.2% 249 27.1% University OSU 122 45.9% 70 26.3% 74 27.8% 5.21 UF 104 49.1% 48 22.6% 60 28.3% UGA 109 48.9% 63 28.3% 51 22.9% UI 87 46.0% 58 30.7% 44 23.3% Region South/SEC 213 49.0% 111 25.5% 111 25.5% 1.01 Midwest/Big 10 209 45.9% 128 28.1% 118 25.9% Academic Rank Lecturer 65 53.3% 27 22.1% 30 2 4.6% 9.72 Assistant Professor 94 45.6% 63 30.6% 49 23.8% Associate Professor 109 41.6% 73 27.9% 80 30.5% Full Professor 190 46.7% 119 29.2% 98 24.1% Other 18 39.1% 12 26.1% 16 34.8% Tenure Status Non tenure 176 47.2% 102 27.3% 9 5 25.5% 0.63 Tenure 299 44.6% 192 28.7% 179 26.7% Administrative position Non administrator 363 45.4% 223 27.9% 214 26.8% 0.69 Administrator 111 46.3% 71 29.6% 58 24.2% Discipline Architecture 2 13.3% 6 40.0% 7 46.7% 35.08* Arts and Humanities 89 45.9% 57 29.4% 48 24.7% Business 39 63.9% 11 18.0% 11 18.0% Education 37 45.1% 22 26.8% 23 28.0% Engineering 42 45.2% 22 23.7% 29 31.2% Law 3 75.0% 1 25.0% 0 0.0% Life Sciences 63 46.7% 41 30.4% 31 23.0% Me dicine and Health Sciences 28 62.2% 11 24.4% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 129

129 Table 4 20 Continued Item 6: Stolen money or other things from their friends, neighbors, or roommates Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 476) (N = 296) (N = 328) Physical Sciences and Mathematics 38 34.2% 36 32.4% 37 33.3% Social and Behavioral Sciences 93 43.9% 68 32.1% 51 24.1% Other 40 47.6% 18 21.4% 26 31.0% Serv ice to athletics No 290 42.6% 204 30.0% 187 27.5% 7.56* Yes 182 51.6% 90 25.5% 81 22.9% Athletic governance No 454 45.6% 286 28.7% 256 25.7% 1.97 Yes 22 56.4% 10 25.6% 7 17.9% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 470 50.15 (11.76) 284 50.46 (11.98) 276 49.10 (11.66) 1.05 Years at institution 473 14.33 (10.64) 291 13.99 (10.59) 259 13.83 (10.06) 0.22 Negative perceptions of student athletes 413 2.12 (0.65) 192 2.25 (0.71) 133 2 .10 (0.63) 2.97 Sport fandom 419 4.10 (1.56) 239 3.60 (1.59) 217 3.69 (1.63) 9.11*** Interaction with student athletes 465 2.60 (1.13) 281 2.35 (1.14) 236 2.35 (1.09) 5.97** Attendance at MFB events 421 2.04 (1.36) 239 1.72 (1.14) 217 1.66 (1.16) 8.3 2*** Attendance at MBA events 418 1.41 (0.81) 237 1.27 (0.64) 215 1.31 (0.79) 2.81 Attendance at WBB events 420 1.30 (0.72) 237 1.23 (0.65) 215 1.20 (0.60) 1.90 Note. *p < .05, **p < .01, ***p < .001

PAGE 130

130 Table 4 21 Missing data analysis for item 7 of criminal deviance scale (N = 471) Item 7: Sold harsh drugs such as heroin, cocaine, and LSD Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 471) (N = 302) (N = 327) Sex Female 196 44.9% 118 27.0% 123 28.1% 0.63 Male 269 44.2% 177 29.1% 162 26.6% Race Non white 53 38.7% 45 32.8% 39 28.5% 2.34 White 416 45.2% 255 27.7% 249 27.1% University OSU 123 46.2% 71 26.7% 72 27.1% 4.13 UF 100 47.2 % 52 24.5% 60 28.3% UGA 106 47.5% 66 29.6% 51 22.9% UI 87 46.0% 59 31.2% 43 22.8% Region South/SEC 206 47.4% 118 27.1% 111 25.5% 0.24 Midwest/Big 10 210 46.2% 130 28.6% 115 25.3% Academic Rank Lecturer 67 54.9% 26 21.3% 29 23.8% 13.30 Assistant Professor 93 45.1% 65 31.6% 48 23.3% Associate Professor 110 42.0% 72 27.5% 80 30.5% Full Professor 185 45.5% 124 30.5% 98 24.1% Other 16 34.8% 13 28.3% 17 37.0% Tenure Status Non tenure 177 47. 5% 103 27.6% 93 24.9% 1.35 Tenure 293 43.7% 197 29.4% 180 26.9% Administrative position Non administrator 359 44.9% 228 28.5% 213 26.6% 0.61 Administrator 110 45.8% 72 30.0% 58 24.2% Discipline Architecture 2 13.3% 6 40.0% 7 46.7% 39.32** Arts and Humanities 88 45.4% 58 29.9% 48 24.7% Business 39 63.9% 11 18.0% 11 18.0% Education 35 42.7% 24 29.3% 23 28.0% Engineering 41 44.1% 22 23.7% 30 32.3% Law 4 100.0% 0 0.0% 0 0.0% Life Sciences 61 45.2% 42 31.1% 32 23.7% Medicine and Health Sciences 28 62.2% 11 24.4% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 131

13 1 Table 4 21 Continued Item 7: Sold harsh drugs such as heroin, cocaine, and LSD Response Prefer not to answer Missing n % n % n % X Chi squ are analysis for nominal and ordinal variables (N = 471) (N = 302) (N = 327) Physical Sciences and Mathematics 38 34.2% 37 33.3% 36 32.4% Social and Behavioral Sciences 93 43.9% 70 33.0% 49 23.1% Other 40 47.6% 18 21.4% 26 31.0% Service to a thletics No 288 42.3% 208 30.5% 185 27.2% 6.65* Yes 179 50.7% 92 26.1% 82 23.2% Athletic governance No 449 45.1% 292 29.3% 255 25.6% 2.10 Yes 22 56.4% 10 25.6% 7 17.9% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 465 49.98 (11.71) 291 50.63 (12.01) 274 49.20 (11.72) 1.05 Years at institution 468 14.22 (10.51) 297 14.03 (10.65) 258 13.97 (10.22) 0.06 Negative perceptions of student athletes 410 2.13 (0.65) 196 2.25 (0.71) 132 2.09 (0.6 4) 3.12* Sport fandom 413 4.11 (1.55) 248 3.59 (1.62) 214 3.70 (1.62) 9.49*** Interaction with student athletes 460 2.61 (1.13) 287 2.33 (1.14) 235 2.36 (1.09) 6.99** Attendance at MFB events 415 2.05 (1.36) 248 1.70 (1.13) 214 1.67 (1.17) 9.15*** Attendance at MBA events 412 1.41 (0.79) 246 1.29 (0.68) 212 1.31 (0.78) 2.13 Attendance at WBB events 413 1.30 (0.70) 247 1.24 (0.69) 212 1.19 (0.60) 0.19 Note. *p < .05, **p < .01, ***p < .001

PAGE 132

132 Table 4 22 Missing data analysis for it em 8 of criminal deviance scale (N = 472) Item 8: Used hard drugs such as heroin, cocaine, and LSD Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 472) (N = 299) (N = 329) Sex Female 194 44.4% 118 27.0% 125 28.6% 0.60 Male 272 44.7% 174 28.6% 162 26.6% Race Non white 53 38.7% 44 32.1% 40 29.2% 2.27 White 417 45.3% 253 27.5% 250 27.2% University OSU 123 46.2% 70 26.3% 73 27.4% 3.72 UF 100 47. 2% 52 24.5% 60 28.3% UGA 107 48.0% 65 29.1% 51 22.9% UI 86 45.5% 58 30.7% 45 23.8% Region South/SEC 207 47.6% 117 26.9% 111 25.5% 0.27 Midwest/Big 10 209 45.9% 128 28.1% 118 25.9% Academic Rank Lecturer 67 54.9% 26 21.3 % 29 23.8% 13.25 Assistant Professor 94 45.6% 64 31.1% 48 23.3% Associate Professor 109 41.6% 72 27.5% 81 30.9% Full Professor 186 45.7% 122 30.0% 99 24.3% Other 16 34.8% 13 28.3% 17 37.0% Tenure Status Non tenure 178 47.7% 102 27.3% 93 24.9% 1.56 Tenure 293 43.7% 195 29.1% 182 27.2% Administrative position Non administrator 360 45.0% 226 28.2% 214 26.8% 0.47 Administrator 110 45.8% 71 29.6% 59 24.6% Discipline Architecture 2 13.3% 6 40.0% 7 46.7% 37.85** Arts and Humanities 88 45.4% 58 29.9% 48 24.7% Business 38 62.3% 11 18.0% 12 19.7% Education 35 42.7% 24 29.3% 23 28.0% Engineering 42 45.2% 21 22.6% 30 32.3% Law 4 100.0% 0 0.0% 0 0.0% Life Sciences 61 45.2% 42 31.1% 32 23.7 % Medicine and Health Sciences 28 62.2% 11 24.4% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 133

133 Table 4 22 Continued Item 8: Used hard drugs such as heroin, cocaine, and LSD Response Prefer not to answer Missing n % n % n % X Chi square anal ysis for nominal and ordinal variables (N = 472) (N = 299) (N = 329) Physical Sciences and Mathematics 39 35.1% 35 31.5% 37 33.3% Social and Behavioral Sciences 93 43.9% 70 33.0% 49 23.1% Other 40 47.6% 18 21.4% 26 31.0% Service to athletics No 289 42.4% 206 30.2% 186 27.3% 6.42* Yes 179 50.7% 91 25.8% 83 23.5% Athletic governance No 449 45.1% 290 29.1% 257 25.8% 2.97 Yes 23 59.0% 9 23.1% 7 17.9% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 466 50.03 (11.80) 287 50.53 (11.88) 277 49.22 (11.70) 0.89 Years at institution 469 14.31 (10.67) 294 13.89 (10.37) 260 13.98 (10.26) 0.17 Negative perceptions of student athletes 410 2.12 (0.65) 194 2.26 (0.71) 134 2.09 (0.64) 3.56* Sport fandom 413 4.11 (1.56) 245 3.60 (1.60) 217 3.69 (1.62) 9.86*** Interaction with student athletes 461 2.60 (1.13) 284 2.34 (1.15) 237 2.37 (1.09) 6.13** Attendance at MFB events 415 2.06 (1.36) 245 1.69 (1.12) 217 1.67 (1.17) 9.85*** Attendanc e at MBA events 412 1.41 (0.79) 243 1.30 (0.69) 215 1.30 (0.78) 2.11 Attendance at WBB events 413 1.30 (0.70) 244 1.23 (0.67) 215 1.20 (0.63) 1.45 Note. *p < .05, **p < .01, ***p < .001

PAGE 134

134 Table 4 23 Missing data analysis for item 9 of c riminal deviance scale (N = 483) Item 9: Broken into a building or vehicle to steal something or just look around Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 483) (N = 291) (N = 326) Sex Female 194 44.4% 117 26.8% 126 28.8% 1.05 Male 282 46.4% 168 27.6% 158 26.0% Race Non white 55 40.1% 43 31.4% 39 28.5% 2.04 White 426 46.3% 246 26.7% 248 27.0% University OSU 123 46.2% 69 25.9% 74 27.8% 4.8 8 UF 104 49.1% 49 23.1% 59 27.8% UGA 109 48.9% 63 28.3% 51 22.9% UI 88 46.6% 58 30.7% 43 22.8% Region South/SEC 213 49.0% 112 25.7% 110 25.3% 0.72 Midwest/Big 10 211 46.4% 127 27.9% 117 25.7% Academic Rank Lecturer 6 5 53.3% 27 22.1% 30 24.6% 9.07 Assistant Professor 94 45.6% 63 30.6% 49 23.8% Associate Professor 113 43.1% 70 26.7% 79 30.2% Full Professor 192 47.2% 118 29.0% 97 23.8% Other 19 41.3% 11 23.9% 16 34.8% Tenure Status Non tenure 177 47.5% 101 27.1% 95 25.5% 0.36 Tenure 305 45.5% 188 28.1% 177 26.4% Administrative position Non administrator 367 45.9% 221 27.6% 212 26.5% 0.53 Administrator 114 47.5% 68 28.3% 58 24.2% Discipline Architecture 2 13.3% 6 40.0% 7 46.7% 32.38* Arts and Humanities 90 46.4% 56 28.9% 48 24.7% Business 39 63.9% 11 18.0% 11 18.0% Education 36 43.9% 22 26.8% 24 29.3% Engineering 44 47.3% 21 22.6% 28 30.1% Law 3 75.0% 1 25.0% 0 0.0% Life Sciences 62 45.9% 43 31.9% 30 22.2% Medicine and Health Sciences 28 62.2% 11 24.4% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 135

135 Table 4 23 Continued Item 9: Broken into a building or vehicle to steal something or just look around Response Prefer not to answer Missin g n % n % n % X Chi square analysis for nominal and ordinal variables (N = 483) (N = 291) (N = 326) Physical Sciences and Mathematics 42 37.8% 33 29.7% 36 32.4% Social and Behavioral Sciences 95 44.8% 66 31.1% 51 24.1% Other 40 47.6% 18 2 1.4% 26 31.0% Service to athletics No 297 43.6% 199 29.2% 185 27.2% 5.93 Yes 182 51.6% 90 25.5% 81 22.9% Athletic governance No 460 46.2% 282 28.3% 254 25.5% 2.53 Yes 23 59.0% 9 23.1% 7 17.9% ANOVA for scale var iables n x bar (sd) n x bar (sd) n x bar (sd) F Age 477 50.36 (11.87) 279 50.16 (11.83) 274 49.04 (11.63) 1.16 Years at institution 480 14.52 (10.77) 286 13.70 (10.33) 257 13.76 (10.06) 0.74 Negative perceptions of student athletes 418 2.13 (0.65) 18 7 2.26 (0.71) 133 2.08 (0.63) 3.30* Sport fandom 421 4.11 (1.56) 239 3.57 (1.60) 215 3.69 (1.61) 10.49*** Interaction with student athletes 471 2.59 (1.13) 276 2.36 (1.15) 235 2.36 (1.09) 5.17** Attendance at MFB events 423 2.05 (1.37) 239 1.68 (1.11 ) 215 1.67 (1.16) 10.12*** Attendance at MBA events 419 1.41 (0.81) 238 1.27 (0.63) 213 1.31 (0.78) 2.94 Attendance at WBB events 421 1.31 (0.74) 238 1.22 (0.63) 213 1.19 (0.60) 2.59 Note. *p < .05, **p < .01, ***p < .001

PAGE 136

136 Table 4 24 Missing data analysis for item 1 of drinking related deviance scale (N = 505) Item 1: Lied about their age to gain entrance or to purchase something Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal vari ables (N = 505) (N = 274) (N = 321) Sex Female 209 47.8% 105 24.0% 123 28.1% 1.46 Male 288 47.4% 164 27.0% 156 25.7% Race Non white 61 44.5% 37 27.0% 39 28.5% 0.57 White 441 47.9% 236 25.7% 243 26.4% University OSU 132 49.6% 63 23.7% 71 26.7% 4.38 UF 105 49.5% 48 22.6% 59 27.8% UGA 116 52.0% 57 25.6% 50 22.4% UI 92 48.7% 55 29.1% 42 22.2% Region South/SEC 221 50.8% 105 24.1% 109 25.1% 0.40 Midwest/Big 10 224 29.2% 118 25.9% 113 24.8% Academic Rank Lecturer 71 58.2% 23 18.9% 28 23.0% 13.68 Assistant Professor 95 46.1% 63 30.6% 48 23.3% Associate Professor 116 44.3% 67 25.6% 79 30.2% Full Professor 203 49.9% 109 26.8% 95 23.3% Other 20 43.5% 10 21.7% 16 34.8% Tenure Status Non tenure 185 49.6% 96 25.7% 92 24.7% 0.42 Tenure 319 47.6% 176 26.3% 175 26.1% Administrative position Non administrator 386 48.3% 207 25.9% 207 25.9% 0.32 Administrator 117 48.8% 65 27.1% 58 24.2% Disciplin e Architecture 2 13.3% 6 40.0% 7 46.7% 34.27* Arts and Humanities 94 48.5% 53 27.3% 47 24.2% Business 40 65.6% 11 18.0% 10 16.4% Education 38 46.3% 21 35.6% 23 28.0% Engineering 43 46.2% 21 22.6% 29 31.2% Law 4 100.0% 0 0.0% 0 0.0% Life Sciences 66 48.9% 39 28.9% 30 22.2% Note. *p < .05, **p < .01, ***p < .001

PAGE 137

137 Table 4 24 Continued Item 1: Lied about their age to gain entrance or to purchase something: for example, lying about their age to buy liquor Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 505) (N = 274) (N = 321) Medicine and Health Sciences 29 64.4% 10 22.2% 6 13.3% Physical Sciences and Mathematics 43 38.7% 32 28.8% 36 32.4% Social and Behavioral Sciences 103 48.6% 60 28.3% 49 23.1% Other 41 48.8% 18 21.4% 25 29.8% Service to athletics No 308 45.2% 191 28.0% 182 26.7% 7.85* Yes 192 54.4% 81 22.9% 80 22.7% Athletic governance No 479 48.1% 2 68 26.9% 249 25.0% 5.29 Yes 26 66.7% 6 15.4% 7 17.9% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 499 50.36 (11.79) 262 50.19 (11.98) 269 48.96 (11.62) 1.31 Years at institution 500 14.54 (10.62) 271 13.70 (10. 59) 252 13.68 (10.06) 0.84 Negative perceptions of student athletes 436 2.14 (0.65) 174 2.25 (0.71) 128 2.08 (0.63) 2.53 Sport fandom 442 4.06 (1.56) 223 3.62 (1.62) 210 3.70 (1.62) 7.39** Interaction with student athletes 494 2.61 (1.12) 259 2.30 (1 .14) 229 2.35 (1.10) 8.09*** Attendance at MFB events 444 2.03 (1.36) 223 1.69 (1.12) 210 1.67 (1.17) 8.67*** Attendance at MBA events 440 1.40 (0.80) 222 1.28 (0.66) 208 1.31 (0.79) 2.04 Attendance at WBB events 442 1.30 (0.71) 222 1.23 (0.67) 208 1 .19 (0.61) 1.97 Note. *p < .05, **p < .01, ***p < .001

PAGE 138

138 Table 4 25 Missing data analysis for item 2 of drinking related deviance scale (N = 516) Item 2: Drank alcohol Response Prefer not to answer Missing n % n % n % X Chi square analy sis for nominal and ordinal variables (N = 516) (N = 262) (N = 322) Sex Female 218 49.9% 97 22.2% 122 27.9% 2.36 Male 290 47.7% 160 26.3% 158 26.0% Race Non white 61 44.5% 37 27.0% 39 28.5% 1.03 White 452 49.1% 224 24.3% 244 26.5% University OSU 130 48.9% 64 24.1% 72 27.1% 4.25 UF 110 51.9% 43 20.3% 59 27.8% UGA 118 52.9% 56 25.1% 49 22.0% UI 98 51.9% 48 25.4% 43 22.8% Region South/SEC 228 52.4% 99 22.8% 108 24.8% 0.57 Midwest/Big 10 22 8 50.1% 112 24.6% 115 25.3% Academic Rank Lecturer 73 59.8% 21 17.2% 28 23.0% 14.70 Assistant Professor 97 47.1% 62 30.1% 47 22.8% Associate Professor 119 45.4% 64 24.4% 79 30.2% Full Professor 207 50.9% 103 25.3% 97 23.8% Other 20 43.5% 10 21.7% 16 34.8% Tenure Status Non tenure 189 50.7% 93 24.9% 91 24.4% 0.58 Tenure 326 48.7% 167 24.9% 177 26.4% Administrative position Non administrator 396 49.5% 197 24.6% 207 25.9% 0.32 Administrator 118 49.2% 63 26.3% 59 24.6% Discipline Architecture 2 13.3% 6 40.0% 7 46.7% 32.78* Arts and Humanities 100 51.5% 46 23.7% 48 24.7% Business 41 67.2% 10 16.4% 10 16.4% Education 39 47.6% 20 24.4% 23 28.0% Engineering 43 46.2% 21 22.6% 29 3 1.2% Law 3 75.0% 1 25.0% 0 0.0% Life Sciences 70 51.9% 35 25.9% 30 22.2% Medicine and Health Sciences 29 64.4% 10 22.2% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 139

139 Table 4 25 Continued Item 2: Drank alcohol Response Prefer no t to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 516) (N = 262) (N = 322) Physical Sciences and Mathematics 43 38.7% 33 29.7% 35 31.5% Social and Behavioral Sciences 102 48.1% 60 28.3% 50 23.6% Other 42 50.0% 17 20.2% 25 29.8% Service to athletics No 318 46.7% 179 26.3% 184 27.0% 5.55 Yes 192 54.4% 81 22.9% 80 22.7% Athletic governance No 493 49.5% 253 25.4% 250 25.1% 1.53 Yes 23 59.0% 9 23.1% 7 17.9% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 509 50.43 (11.81) 252 50.00 (11.98) 269 49.01 (11.60) 1.28 Years at institution 511 14.68 (10.72) 259 13.35 (10.31) 253 13.71 (10.09) 1.61 Negative perceptions of student athletes 443 2.14 (0.66) 167 2.25 (0.71) 128 2.09 (0.62) 2.41 Sport fandom 454 4.04 (1.55) 211 3.64 (1.64) 210 3.69 (1.63) 5.99** Interaction with student athletes 504 2.61 (1.14) 248 2.29 (1.12) 230 2.36 (1.10) 7.86*** Attendance at MFB events 455 2.01 (1.3 5) 211 1.72 (1.13) 211 1.68 (1.17) 6.71** Attendance at MBA events 452 1.39 (0.79) 209 1.28 (0.65) 209 1.32 (0.80) 1.73 Attendance at WBB events 453 1.30 (0.72) 210 1.21 (0.63) 209 1.20 (0.62) 2.48 Note. *p < .05, **p < .01, ***p < .001

PAGE 140

140 Table 4 26 Missing data analysis for item 3 of drinking related deviance scale (N = 492) Item 3: Drank more than 5 alcoholic drinks at once Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variabl es (N = 492) (N = 284) (N = 324) Sex Female 204 46.7% 110 25.2% 123 28.1% 1.05 Male 280 46.1% 169 27.8% 159 26.2% Race Non white 60 43.8% 39 28.5% 38 27.7% 0.41 White 429 46.6% 244 26.5% 247 26.8% University OSU 127 47.7% 68 25.6% 71 26.7% 5.57 UF 109 50.9% 45 21.2% 59 27.8% UGA 112 50.2% 60 26.9% 51 22.9% UI 90 47.6% 57 30.2% 42 22.2% Region South/SEC 220 50.6% 105 24.1% 110 25.3% 1.35 Midwest/Big 10 217 47.7% 125 27.5% 113 24.8% Ac ademic Rank Lecturer 69 56.6% 25 20.5% 28 23.0% 11.88 Assistant Professor 96 46.6% 62 30.1% 48 23.3% Associate Professor 113 43.1% 69 26.3% 80 30.5% Full Professor 195 47.9% 115 28.3% 97 23.8% Other 19 41.3% 11 23.9% 16 34.8% Te nure Status Non tenure 183 49.1% 98 26.3% 92 24.7% 0.95 Tenure 308 46.0% 184 27.5% 178 26.6% Administrative position Non administrator 378 47.3% 213 26.6% 209 26.1% 0.49 Administrator 112 46.7% 69 28.7% 59 24.6% Discipline Architecture 0 0.0% 8 53.3% 7 46.7% 41.04** Arts and Humanities 96 49.5% 50 25.8% 48 24.7% Business 40 65.6% 10 16.4% 11 18.0% Education 36 43.9% 23 28.0% 23 28.0% Engineering 43 46.2% 21 22.6% 29 31.2% Law 3 75.0% 1 25.0% 0 0.0 % Life Sciences 66 48.9% 39 28.9% 30 22.2% Medicine and Health Sciences 29 64.4% 10 22.2% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 141

141 Table 4 26 Continued Item 3: Drank more than 5 alcoholic drinks at once Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 492) (N = 284) (N = 324) Physical Sciences and Mathematics 42 37.8% 34 30.6% 35 31.5% Social and Behavioral Sciences 95 44.8% 67 31.6% 50 23.6% Other 40 47.6% 18 21.4% 26 31.0% Service to athletics No 303 44.5% 194 28.5% 184 27.0% 5.85 Yes 185 52.4% 87 24.6% 81 22.9% Athletic governance No 470 47.2% 274 27.5% 252 25.3% 1.53 Yes 22 56.4% 10 25.6% 7 17.9% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 486 50.08 (11.75) 272 50.63 (12.09) 272 49.06 (11.56) 1.27 Years at institution 487 14.25 (10.57) 281 14.24 (10.67) 255 13.67 (10.08) 0.29 Negative perceptions of student athletes 421 2.14 (0.65) 187 2.25 (0.71) 130 2.07 (0.63) 3.03* Sport fandom 434 4.06 (1.57) 230 3.61 (1.59) 211 3.72 (1.62) 7.24** Interaction with student athletes 481 2.59 (1.13) 269 2.34 (1.13) 232 2.36 (1.11) 5.37** Attendance at MFB events 436 2.02 (1.36) 23 0 1.70 (1.11) 211 1.69 (1.19) 7.33** Attendance at MBA events 434 1.39 (0.79) 227 1.30 (0.66) 209 1.32 (0.80) 1.39 Attendance at WBB events 434 1.29 (0.71) 229 1.24 (0.67) 209 1.20 (0.62) 1.64 Note. *p < .05, **p < .01, ***p < .001

PAGE 142

142 T able 4 27 Missing data analysis for item 4 of drinking related deviance scale (N = 468) Item 4: Had sexual relations with a person other than their significant other Response Prefer not to answer Missing n % n % n % X Chi square analysis for nomina l and ordinal variables (N = 468) (N = 306) (N = 326) Sex Female 196 44.9% 117 26.8% 124 28.4% 1.35 Male 266 42.8% 182 29.9% 160 26.3% Race Non white 52 38.0% 46 33.6% 39 28.5% 2.62 White 413 44.9% 259 28.2% 248 27.0% Uni versity OSU 115 43.2% 79 29.7% 72 27.1% 4.63 UF 101 47.6% 51 24.1% 60 28.3% UGA 106 47.5% 66 29.6% 51 22.9% UI 90 47.6% 56 29.6% 43 22.8% Region South/SEC 207 47.6% 117 26.9% 11 25.5% 0.92 Midwest/Big 10 205 45.1% 135 2 9.7% 115 25.3% Academic Rank Lecturer 66 54.1% 27 22.1% 29 23.8% 11.77 Assistant Professor 89 43.2% 68 33.0% 49 23.8% Associate Professor 106 40.5% 77 29.4% 79 30.2% Full Professor 188 46.2% 121 29.7% 98 24.1% Other 19 41.3% 11 23.9% 16 34.8% Tenure Status Non tenure 174 46.6% 105 28.2% 94 25.2% 0.83 Tenure 293 43.7% 199 29.7% 178 26.6% Administrative position Non administrator 355 44.4% 233 29.1% 212 26.5% 0.55 Administrator 111 46.3% 71 29.6% 58 24.2% Discipline Architecture 2 13.3% 6 40.0% 7 46.7% 32.98* Arts and Humanities 87 44.8% 59 30.4% 48 24.7% Business 38 62.3% 12 19.7% 11 18.0% Education 35 42.7% 24 29.3% 23 28.0% Engineering 40 43.0% 24 25.8% 29 31.2% Law 3 75.0% 1 25.0% 0 0.0% Life Sciences 64 47.4% 41 30.4% 30 22.2% Medicine and Health Sciences 27 60.0% 12 26.7% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 143

143 Table 4 27 Continued Item 4: Had sexual relations with a person other than their sign ificant other Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 468) (N = 306) (N = 326) Physical Sciences and Mathematics 39 35.1% 36 32.4% 36 32.4% Social and Behavioral Sciences 92 43.4% 70 33.0% 50 23.6% Other 39 46.4% 18 21.4% 27 32.1% Service to athletics No 291 42.7% 205 30.1% 185 27.2% 3.72 Yes 172 48.7% 100 28.3% 81 22.9% Athletic governance No 447 44.9% 295 29.6% 254 25.5% 1.54 Yes 21 53. 8% 11 28.2% 7 17.9% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 460 50.33 (11.89) 296 50.21 (11.83) 274 49.05 (11.60) 1.11 Years at institution 464 14.50 (10.75) 302 13.78 (10.42) 257 13.78 (10.04) 0.60 Negativ e perceptions of student athletes 406 2.14 (0.65) 200 2.23 (0.71) 132 2.07 (0.63) 2.28 Sport fandom 409 4.10 (1.54) 252 3.61 (1.63) 214 3.71 (1.62) 9.00*** Interaction with student athletes 457 2.61 (1.12) 291 2.35 (1.16) 234 2.35 (1.10) 6.41** Atten dance at MFB events 411 2.01 (1.35) 252 1.75 (1.15) 214 1.69 (1.18) 6.17** Attendance at MBA events 409 1.39 (0.79) 249 1.31 (0.69) 212 1.32 (0.79) 1.31 Attendance at WBB events 410 1.29 (0.70) 250 1.25 (0.70) 212 1.20 (0.61) 1.32 Note. *p < .05, **p < .01, ***p < .001

PAGE 144

144 Table 4 28 Missing data analysis for item 5 of drinking related deviance scale (N = 479) Item 5: Bought or provided liquor for a minor Response Prefer not to answer Missing n % n % n % X Chi square analysis for n ominal and ordinal variables (N = 479) (N = 292) (N = 329) Sex Female 197 45.1% 115 26.3% 125 28.6% 0.66 Male 275 45.2% 171 28.1% 162 26.6% Race Non white 56 40.9% 41 29.9% 40 29.2% 1.11 White 420 45.7% 250 27.2% 250 27.2% University OSU 123 46.2% 70 26.3% 73 27.4% 4.98 UF 102 48.1% 49 23.1% 61 28.8% UGA 109 48.9% 63 28.3% 51 22.9% UI 88 46.6% 58 30.7% 43 22.8% Region South/SEC 211 48.5% 112 25.7% 112 25.7% 0.69 Midwest/Big 10 211 46.4% 128 28.1% 116 25.5% Academic Rank Lecturer 65 53.3% 27 22.1% 30 24.6% 9.72 Assistant Professor 94 45.6% 63 30.6% 49 23.8% Associate Professor 112 42.7% 71 27.1% 29 30.2% Full Professor 190 46.7% 118 29.0% 99 24.3% Other 18 39.1 % 11 23.9% 17 37.0% Tenure Status Non tenure 178 47.7% 100 26.8% 95 25.5% 0.84 Tenure 300 44.8% 190 28.4% 180 26.9% Administrative position Non administrator 364 45.5% 221 27.6% 215 26.9% 0.70 Administrator 113 47.1% 69 28.7 % 58 24.2% Discipline Architecture 2 13.3% 6 40.0% 7 46.7% 34.09* Arts and Humanities 91 46.9% 54 27.8% 49 25.3% Business 39 63.9% 11 18.0% 11 18.0% Education 36 43.9% 23 28.0% 23 28.0% Engineering 42 45.2% 21 22.6% 30 32.3% Law 3 75.0% 1 25.0% 0 0.0% Life Sciences 63 46.7% 40 29.6% 32 23.7% Medicine and Health Sciences 28 62.2% 11 24.4% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 145

145 Ta ble 4 28 Continued Item 5: Bought or provided liquor for a minor Re sponse Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 479) (N = 292) (N = 329) Physical Sciences and Mathematics 40 36.0% 35 31.5% 36 32.4% Social and Behavioral Sciences 93 43.9% 69 32.5 % 50 23.6% Other 40 47.6% 18 21.4% 26 31.0% Service to athletics No 296 43.5% 198 29.1% 187 27.5% 4.67 Yes 178 50.4% 93 26.3% 82 23.2% Athletic governance No 458 46.0% 281 28.2% 257 25.8% 1.41 Yes 21 52.8% 11 28.2% 7 17. 9% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 473 49.94 (11.77) 280 50.73 (11.94) 277 49.20 (11.69) 1.17 Years at institution 475 14.19 (10.52) 288 14.10 (10.62) 260 13.95 (10.25) 0.04 Negative perceptions of student athletes 415 2.12 (0.65) 189 2.26 (0.70) 134 2.10 (0.64) 3.04* Sport fandom 419 4.11 (1.55) 240 3.58 (1.61) 216 3.70 (1.62) 9.98*** Interaction with student athletes 467 2.60 (1.13) 278 2.34 (1.15) 237 2.36 (1.10) 5.87** Attendance at MFB eve nts 421 2.03 (1.35) 240 1.70 (1.14) 216 1.69 (1.19) 7.95*** Attendance at MBA events 418 1.40 (0.79) 238 1.30 (0.69) 214 1.31 (0.78) 1.75 Attendance at WBB events 419 1.29 (0.70) 239 1.25 (0.69) 214 1.19 (0.60) 1.63 Note. *p < .05, **p < .01, ***p < 001

PAGE 146

146 Table 4 29 Missing data analysis for item 6 of drinking related deviance scale (N = 507) Item 6: Been drunk in a public place Response Prefer not to answer Missing n % n % n % X Chi square analysis for nominal and ordinal var iables (N = 507) (N = 271) (N = 322) Sex Female 213 48.7% 102 23.3% 122 27.9% 1.83 Male 286 47.0% 164 27.0% 158 26.0% Race Non white 60 43.8% 39 28.5% 38 27.7% 1.08 White 444 48.3% 231 25.1% 245 26.6% University OSU 131 49.2% 64 24.1% 71 26.7% 4.88 UF 110 51.9% 43 20.3% 59 27.8% UGA 112 50.2% 61 27.4% 50 22.4% UI 95 50.3% 51 27.0% 43 22.8% Region South/SEC 222 51.0% 104 23.9% 109 25.1% 0.25 Midwest/Big 10 226 49.7% 115 25.3% 114 25.1% Academic Rank Lecturer 69 56.6% 24 19.7% 29 23.8% 10.87 Assistant Professor 97 47.1% 62 30.1% 47 22.8% Associate Professor 119 45.4% 65 24.8% 78 29.8% Full Professor 202 49.6% 108 26.5% 97 23.8% Other 20 43.5% 10 21.7% 16 34.8% Tenure Status Non tenure 185 49.6% 96 25.7% 92 24.7% 0.38 Tenure 321 47.9% 173 25.8% 176 26.3% Administrative position Non administrator 386 48.3% 206 25.8% 208 26.0% 0.33 Administrator 119 49.6% 63 26.3% 58 24.2% Discipli ne Architecture 2 13.3% 6 40.0% 7 46.7% 31.55* Arts and Humanities 99 51.0% 48 24.7% 47 24.2% Business 39 63.9% 11 18.0% 11 18.0% Education 37 45.1% 21 25.6% 24 29.3% Engineering 44 47.3% 21 22.6% 28 30.1% Law 3 75.0% 1 25.0% 0 0.0% Life Sciences 70 51.9% 35 25.9% 30 22.2% Medicine and Health Sciences 29 64.4% 10 22.2% 6 13.3% Note. *p < .05, **p < .01, ***p < .001

PAGE 147

147 Table 4 29 Continued Item 6: Been drunk in a public place Response Prefer not to answer Mis sing n % n % n % X Chi square analysis for nominal and ordinal variables (N = 507) (N = 271) (N = 322) Physical Sciences and Mathematics 43 38.7% 33 29.7% 35 31.5% Social and Behavioral Sciences 99 46.7% 64 30.2% 49 23.1% Other 40 47.6% 1 8 21.4% 26 31.0% Service to athletics No 313 46.0% 184 27.0% 184 27.0% 5.15 Yes 188 53.5% 86 24.4% 79 22.4% Athletic governance No 485 48.7% 261 26.2% 250 25.1% 1.23 Yes 22 56.4% 10 25.6% 7 17.9% ANOVA for scale variables n x bar (sd) n x bar (sd) n x bar (sd) F Age 499 50.33 (11.86) 261 50.11 (11.86) 270 49.11 (11.61) 0.97 Years at institution 502 14.54 (10.69) 268 13.57 (10.42) 253 13.80 (10.08) 0.89 Negative perceptions of student athletes 435 2.14 (0.66 ) 174 2.23 (0.70) 129 2.09 (0.63) 1.86 Sport fandom 445 4.06 (1.55) 219 3.61 (1.63) 211 3.70 (1.62) 7.37** Interaction with student athletes 494 2.59 (1.13) 256 2.34 (1.14) 232 2.36 (1.09) 5.48** Attendance at MFB events 447 2.01 (1.35) 219 1.72 (1.1 3) 211 1.67 (1.16) 6.91** Attendance at MBA events 444 1.40 (0.80) 217 1.29 (0.66) 209 1.32 (0.79) 1.81 Attendance at WBB events 445 1.30 (0.72) 218 1.22 (0.64) 209 1.19 (0.61) 2.35 Note. *p < .05, **p < .01, ***p < .001

PAGE 148

148 CHAPTER 5 RESULTS PREDICT ING ACADEMIC DEVIANCE Academic Deviance by University and Sporting Group Factorial ANOVA for General C heating A factorial ANOVA was conducted to compare the effect of university and sporting group and the interaction effect between university and sporting group on perceptions of general cheating academic deviance ( Table 5 1). Uni versity included four levels: University of Illinois (UI) University of Florida ( UF ) University of Georgia ( UGA ), and Ohio State University ( OSU ) and sporting group consisted of t hree levels : football (MFB ) ( MBA ) and WBB). The interaction effect was not significant, but the main effects for university and sport were significant. The main effect for university yielded an F ratio of F(3, 40 1) = 4.47, p < .01, 0.03 indicating a significant difference between UI (M = 2.08, SD = 0.63 ), UF (M = 2.02, SD = 0.64 ), UGA (M = 1.83 SD = 0.68 ), and OSU (M = 1.86 SD = 0.70 ). Significant differences using Tukey HSD found that faculty from UGA wer e less likely to perceive student athletes as general cheater s compared to faculty at UI (Table 5 2) Figure 5 1 shows a visual of the means for perceptions of general cheating by university. The main effect for sporting group yielded an F ratio of F(2, 4 01) = 12.74 p < .001, = 0.06, indicating a significant difference between MFB (M = 2.08 SD = 0.68 ), MBA (M = 2.01, SD = 0.63 ), and WBB (M = 1.72 SD = 0.62 ). Significant differences using Tukey HSD found faculty believed WBB student athletes were less likely to be perceived as cheaters generally compared to MFB and MBA student athletes (Table 5 3) Figure 5 2 shows a visual of the means for perceptions of general cheating by sport.

PAGE 149

149 Factorial ANOVA for Relying on O thers A factorial ANOVA was conducted to compare the effect of university and sporting group and the interaction effect between university and sporting group on perceptions of relying on others academic deviance ( Table 5 4 ). University included four levels (UI, UF, UGA and OSU) and sporting group consisted of three l evels (MFB, MBA, and WBB). The interaction effect was not significant, but the main effects for university and sport were significant. The main effect for university yielded an F ratio of F(3, 415) = 4.31, p < .01, 0.03, indicating a significant diff erence between UI (M = 2.80, SD = 0.71 ), UF (M = 2.97 SD = 0.7 8), UGA (M = 2.72 SD = 0.84 ), and OSU (M = 2.68 SD = 0.83 ). Significant differences using Tukey HSD found faculty at OSU had significantly lower perceptions of student athletes relying on oth ers compared to faculty at UF (Table 5 5) Figure 5 3 shows a visual of the means for perceptions of relying on others by university. The main effect for sporting group yielded an F ratio of F(2, 415) = 13.99, p < .001, indicating a significant difference between MFB (M = 3.01 SD = 0.80 ), MBA (M = 2.75 SD = 0.76 ), and WBB (M = 2.52 SD = 0.80 ). Significant differences using Tukey HSD found faculty believed WBB student athletes were less likely to rely on oth ers as a form of cheating compared to MFB and MBA student athletes (Table 5 6) Additionally, faculty believed MFB student athletes were significantly more likely to rely on others compared to both MBA and WBB. Figure 5 4 shows a visual of the means for pe rceptions of relying on others by sporting group. 28 28 A MANOVA revealed t here was a statistically significant multivariate main effect for university, F(6, 2 = .97. Also, there was a statistically significant main effect for sport 2 = .99.

PAGE 150

150 Bivariate Relationships for the Entire Sample Bivariate correlations were run for the two academic deviance scales (general cheating and relying on others) and the independent variables reflecting individu al status attributes of faculty, student athlete attributes, university attributes, and university athletic attributes ( Tables 5 7 5 8 5 9 and 5 10 ). Following the correlational analysis, ordinary least squares (OLS) regressions were run for any variabl es correlated at the bivariate level for the entire sample. General Cheating There were several independent variables that were significant at th e biva riate level ( Table 5 7 5 8 5 9 and 5 10 ). For the general cheating scale, faculty age, academic rank ( lecturer status), participation in service involving athletics, sports fandom attendance at MFB and MBA events contact with student athletes, UI, MFB sporting group, and WBB sporting group were significantly related to perceptions of student athlete gene ral cheating. As far as faculty status attributes, o lder faculty were less likely to perceive student athletes as general cheaters (r = 0.11, p < .05) (Table 5 7) Lecturers were less likely to perceive student athletes as general cheaters compared to fa culty of other ranks (r = 0.11, p < .05). Faculty involved in service to athletics were also significantly less likely to perceive general cheating among student athletes compared to those who had not participated in service to athletics (r = 0.11, p < 05). Faculty that were bigger fans of their university sports program were less likely to perceive student athletes as general cheaters (r = 0.22, p < .001). Additionally, faculty that attended more MFB events were less likely to perceive student athletes as gene ral cheaters (r = 0.13, p < .01 ). Finally, faculty with higher attendance at MBA events (r = 0.12, p < .05) and that

PAGE 151

151 had more contact with student athletes (r = 0.12, p < .05) were less likely to perceive them as general cheaters. Only one unive rs ity status independent variable was significantly related to the general cheating scale ( Table 5 8 ). Faculty at UI were more likely to perceive student athletes as general cheaters than faculty at the three other universities (r = 0.11, p < .05). As far as student athlete status attributes, two of the manipulation variables were significantly related to general cheating at the bi variate level, MFB and WBB ( Table 5 9 ). Faculty who were randomly assigned MFB student athletes were significantly more likely to perceive general cheating compared to faculty assigned MBA or WBB (r = 0.16, p < .001). Additionally, faculty who were randomly assigned WBB student athletes were significantly less like to perceive general cheating compared to faculty assigned MFB and MBA (r = 0.23, p < .001). N o perceptio ns of student athlete attribute variables were significantly related to the general cheating scale at the bivariate level None of the university athletic status attributes were significantly associated wit h the gene ral cheating scale of academic deviance (Table 5 10 ). Relying on Others There were several variables significantly associated with the relying on others scale (Table 5 7 5 8 5 9 and 5 10 ) These included: faculty age, race, academic rank (assistant and full professor), tenure status, time at current institution, sports fandom, attendance at MFB events UF, MFB sporting group, WBB sporting group, perceptions of female student athletes, and perceptions of black student athletes As far as faculty status attributes, o lder faculty were less likely to perceive student athletes as reliant on others (r = 0.13, p < .01) (Table 5 7 ) White faculty were

PAGE 152

152 significantly more likely to believe student athletes are reliant on others compared to non white faculty (r = 0.09, p < .05). Full professors and those with tenure status were less likely to perceive student athletes as reliant on others compared to faculty of other ranks (r = 0.12, p < .05) and non tenure status (r = 0.10, p < .05). However, assistant professo rs were more likely to perceive student athletes as reliant on others compared to faculty of other ranks (r = 0.09, p < .05). Faculty that had been at their current institution longer were significantly less likely to perceive student athletes as reliant o n others (r = 0.11, p < .05). Faculty that were bigger fans of their university sports program were less likely to perceive student athletes as reliant on others (r = 0.20, p < .001). Additionally, faculty that attended more MFB events were less likely t o perceive student athletes as reliant on others (r = 0.14, p < .01). Only one univers ity status independent variable was significantly related to the re lying on others scale (Table 5 8 ). Faculty at UF were more likely to perceive student athletes as reli ant on others compared to faculty at the three other universities (r = 0.13, p < .01). There were four student athlete status attribute variables significantly associated with the relying on others ac ademic deviance scale (Table 5 9 ). Faculty randomly ass igned the MFB sporting group were more likely to perceive the athletes relying on others compared to faculty assigned MBA or WBB sporting groups (r = 0.22 p < .001). Faculty randomly assigned the WBB sporting group were less likely to perceive relying on others compared to faculty assigned MFB or MBA (r = 0.22, p < .001). Faculty that estimated a smaller percentage of female student athletes on their campus were more likely to perceive student athletes as reliant on others (r = 0.11, p < .05). Faculty th at

PAGE 153

153 also estimated a larger percentage of black student athletes on their campus were more likely to perceive student athletes as reliant on others (r = 0.18, p < .001). None of the university athletic status attributes were significantly associated with th e relying on others scale of academic deviance ( Table 5 10 ). OLS R egression Models for Entire S ample Predicting General C heating OLS regression models were run predicting the general cheating with independent variables that were significant at the bivari ate level. Additionally, the models included the manipulation variables of university and sporting groups, which were dummy coded and with one variable left out. These independent variables include: age, academic rank (lecturer ) service involving athletic s, sports fandom, attendance at MFB events, attendance at MBA events, contact with st udent athletes UI, UF, UGA, MFB and MBA sporting group s ( Table 5 11 ). The overall model predicting general cheating was significant (F = 5.25 p < .0 0 1, R = 0.14 ). Seve ral variable s remained significant after controlling for other variables which include sports fandom attendance at MBA events, UI, U F MFB and MBA The were to perceive s tudent athlete s as general cheaters (b = 0.06 p < .05). The more MBA events faculty attended the less likely they were to perceive student athletes as general cheaters (b = 0.10, p < .05). Faculty at UI and U F were significantly more likely to perceive s tudent athletes as general cheaters compared to faculty at OSU (UI: b = 0.24, p < .01; U F : b = 0.29, p < .01). Additionally, faculty that were assigned MFB and MBA were significantly more likely to perceive student athletes as general cheaters compared to WBB (MFB: b = 0.34, p < .001; MBA: b = 0.29, p < .001).

PAGE 154

154 Predicting Relying on O thers An OLS regression model was run predicting the relying on others with faculty status attributes that were significant at the bivariate level. These independent variables i nclude: age, race (white), academic rank ( assistant professor and full professor ), tenure status, time at current institution, sports fandom, attenda nce at MFB events estimates of black student athletes, estimates of female student athletes, UI, UF, UGA, MFB and MBA sporting group (Table 5 12 ). The overall model predicting relying on others academic de viance was significant (F = 4.39 p < .0 01, R = 0.19 ). Seven variables were significant after controlling for other variables, which included race, sports fandom UF perceptions of female student athletes, perceptions of black student athletes, MFB and MBA White faculty were significantly more likely to perceive student athletes as reliant on others compare d to non white faculty (b = 0.20 p < .05). F acult y that were bigger fans of their university sports program were significantly less likely to perceive student athletes as reliant o n others academically (b = 0.10 p < .01 ). Faculty at U F were significantly more likely to perceive student athletes as reli ant on others compared to faculty at OSU (b = 0.30, p < .01). Faculty that estimated higher percentages of female student athletes on their campus were significantly less likely to perceive student athletes as reliant on others (b = 0.01, p < .05). Howeve r, faculty that estimated higher percentages of black student athletes on their campus were significantly more likely to perceive student athletes as reliant o n others (b = 0.01, p < .001). Additionally, faculty that were assigned MFB and MBA groups, were significantly more likely to perceive student athletes as reliant on others (MFB: b = 0.38, p < .001; MBA: b = 0.25, p < .05).

PAGE 155

155 Academic Deviance by University Bivariate R elationsh ips for Individual U niversities University of Illinois There were five facul ty status attributes significantly related to the general cheating academic deviance scale for UI (Tabl e 5 13 ). Lecturers were significantly less likely to perceive student athletes as general cheaters compared to faculty that were not lecturers (r = 0.22 p < .05). Alternatively, faculty with an associate professor rank at UI were significantly more likely to perceive student athletes as general cheaters compared to faculty of other ranks (r = 0.27, p < .05). Additionally, UI faculty that described themse lves as fans of UI sports (r = 0.32, p < .01), attended MBA events (r = 0.26, p < .05), and had more contact with student athletes (r = 0.23, p < .05) were significantly less likely to perceive student athletes as general cheaters. Of the student athle te status attributes, only two variable s were significantly related to academic deviance at the bivariate level for UI (Table 5 14 ). Faculty that were randomly assigned WBB were less likely to perceive general cheating (b = 0.28, p < .01). Additionally, f aculty at UI that estimated a larger percentage of black student athletes on their campus were significantly more likely to perceive students athletes as general cheaters (b = 0.24, p < .05). There were four faculty status attributes significantly related to the relying on others academic deviance scale for UI faculty race, academic discipline (law), sports fandom, and attendance at MBA events (Table 5 13 ). White faculty at UI were significantly more likely to perceive student athletes as reliant on other s compared to non white faculty (r = 0.22, p < .05). UI faculty that identified law as their discipline were significantly less likely to perceive student athletes as reliant on others (r = 0.28, p < .01). Additionally, UI faculty that described themselve s as fans of UI sports (r = 0.23, p

PAGE 156

156 < .05) and attended MBA events (r = 0.25, p < .05) were significantly less likely to perceive student athletes as reliant on others. There were no stude nt athlete status attributes significantly related to relyin g on o thers for UI (Table 5 14 ) University of Florida Only one faculty status attribute was significantly associated with the general cheating academic deviance scale at U F, which was race (Table 5 13 ). White faculty at UF were significantly more likely to perc eive student athletes as general cheaters compared to non white faculty (r = 0.22, p < .05). Additionally, there were four student athlete status attributes significantly related to general cheating at UF MFB, WBB, and percentage of MFB and WBB ath letes (Table 5 14 ). Faculty at UF that were randomly assigned to the MFB sporting group were significantly more likely to perceive general cheating compared to faculty assigned other sport groups (r = 0.23, p < .05). Additionally, those at UF randomly assigned WBB were significantly less likely to perceive general cheating compared to faculty assigned other groups (r = 0.28, p < .01). Faculty at UF that estimated a higher percentage of MFB and WBB athletes on their campus were significantly less likely to perce ive student athletes as general cheaters (MFB: r = 0.22, p < .05; WBB: r = 0.23, p < .05). There were two faculty status attributes significantly associated with the relying on others scale at UF, race and academic rank (full professor) (Table 5 13 ). Aga in, white faculty at UF were significantly more likely to perceive student athletes as reliant on others compared to non white faculty (r = 0.19, p < .05). Also, UF faculty with the rank of full professor were significantly less likely to perceive student athletes as reliant on others compared to faculty of other ranks (r = 0.21, p < .05).

PAGE 157

157 There was only student athlete status attribute significantly related to relying on others for UF (Table 5 14 ) Similar to the general cheating scale, faculty at UF ran domly assigned to the WBB sporting group were significantly less likely to perceive relying on others compared to faculty assigned other sport groups (r = 0.28, p < .01). University of Georgia There were four faculty status attributes significantly assoc iated with the general cheating academic deviance scale at UGA, which was academic discipline (education and physical sciences and mathematics), service involving athletics, and attenda nce at MBA events (Table 5 13 ). UGA faculty in education were less like ly to perceive student athletes as general cheaters compared to faculty in other disciplines (r = 0.21, p < .05). Alternatively, UGA faculty in physical sciences and mathematics were more likely to perceive student athletes as general cheaters compared to faculty in other disciplines (r = 0.28, p < .01). Additionally, UGA faculty that were involved in service to athletics (r = 0.33, p < .01) and attended more MBA events (r = 0.30, p < .01) were significantly less likely to perceive student athletes as ge neral cheaters compared to faculty that were not as involved in service to athletics and attended less MBA events. There were also two student athlete status attributes significantly related to general cheating at the bivariate level for UGA (Table 5 1 4 ). First, faculty at UGA that were randomly assigned MFB were significantly more likely to perceive general cheating compared to faculty assigned other sport groups (r = 0.20, p < .01). Second, faculty at UGA that were assigned WBB were significantly less li kely to perceive general cheating compared to faculty assigned to other sport groups (r = 0.33 p < .0 0 1).

PAGE 158

158 There were slightly different faculty status attributes significantly associated with the relying on others academic devia nce scale for UGA (Tabl e 5 13 ). First, older faculty were significantly less likely to perceive student athletes as reliant on others (r = 0.21, p < .05). Faculty that had been at UGA longer were also significantly less likely to perceive student athletes as reliant on others comp ared to those who had been at UGA for a shorter period of time (r = 0.23, p < .05). Similar to the general cheating scale, faculty that were involved in service to athletics (r = 0. 21, p < .05) and attended MBA events (r = 0.34, p < .001) were signific antly less likely to perceive student athletes as reliant on others compared to faculty that were not involved in service to athletics and attend less MBA events. As for student athlete attributes, three variable s were significantly related to relying on o thers for UGA ( Table 5 14 ). First, faculty at UGA that were randomly assigned to the MFB sporting group were significantly more likely to perceive relying on others academic deviance compared to faculty assigned other sport groups (r = 0.34, p < .001). Sec ond, faculty at UGA that were randomly assigned to the WBB sporting group were significantly less likely to perceive general cheating compared to faculty assigned to other sport groups (r = 0.32, p < .01). Finally, f aculty at UGA that estimated a higher p ercentage of black student athletes on their campus were significantly more likely to perceive student athletes as reliant on others (r = 0.29, p < .05). O hio S tate U niversity. Only one faculty status attribute had a significant bivariate relationship wit h the general cheating scale at OSU, which was sports fandom ( Table 5 13 ). Sports fandom was negatively related to general cheating, where faculty that are bigger fans of OSU sports were less likely to perceive student athletes as

PAGE 159

159 general cheaters compared to faculty that were not as big of fans of OSU sports (r = 0.26, p < .01). There were no student athlete status attributes significantly related to general cheating at t he bivariate level for OSU ( Table 5 10). There were two faculty status attributes si gnificantly related to the relying on others academic deviance scale, which were sports fandom and attenda nce at MFB events ( Table 5 13 ). Similar to the general cheating scale, faculty that are bigger fans of OSU sports were less likely to perceive student athletes as reliant on others compared to faculty that were not as big of fans of OSU sports (r = 0.32, p < .001). Additionally, faculty that attended more MFB events at OSU were less likely to perceive student athletes as reliant on others compared to f aculty that attended less MFB events at OSU (r = 0.20, p < .05). As for student athlete status attributes, three variable s were significantly related to relying on others for OSU ( Table 5 14 ). OSU f aculty that were randomly assigned MFB were significantl y more likely to perceive student athletes as reliant on others compared to faculty assigned other sporting groups (r = 0.26, p < .01). However, OSU faculty that were randomly assigned WBB were significantly less likely to perceive student athletes as reli ant on others compared to faculty assigned other groups (r = 0.21, p < .05). Additionally, f aculty at OSU that estimated a higher percentage of black student athletes on their campus were significantly more likely to perceive student athletes as reliant o n others (r = 0.24, p < .05). OLS Regression by U niversity and Clogg Coefficient Comparison Test Predicting general c heating Independent variables that were significantly associated with at least one of the university groups were entered into OLS regress ion

PAGE 160

160 models to predict general cheating 29 More specifically, OLS regression models were run for each university using race, academic rank (lecturer and associate professor), academic discipline (education and physical sciences and math), service involving a thletics, sports fandom, attendance at MBA events, contact w ith student athletes, and the man ipulation variables of MFB and MBA Table 5 15 shows OLS regression results predicting general cheating by university. For UI, the ove rall model was significant ( F = 3.13, p < .01, R = 0.34 ) ( Table 5 15 ) Two variable s remained significant for the UI sub group after controlling for other variables, which included the manipulation variables MFB and MBA UI faculty that were randomly assigned MFB student athletes to answer questions about were significant ly more likely to perceive general cheating compared to faculty assigned WBB (b = 0.46 p < .01 ). Similarly, UI faculty that were randomly assigned MBA student athletes were significantly more likely to perceive gener al cheating compared to faculty assigned WBB (b = 0.43, p < .05). For UGA, the overall model was significant (F = 3.11, p < .01, R = 0.28) ( Table 5 15 ). Two variables remained significant for the UGA subgroup after controlling for other variables, which i ncluded service involving athletics and MFB. Faculty at UGA that participated in service involving athletics were significantly less likely to perceive 29 No perceptions of student athlete status attribu te variables were included in the regression analysis, even though they were significant at the bivariate level. Adding these three variables to the model created power issues. When the models were run including all three variables, none of the overall mod els were significant. I suspect this is because of listwise deletion, where cases are thrown out if one of more variables are missing data. There was a high frequency of missingness for the perceptions of student athlete status attribute variables. The sam ple size and number of independent variables included in regression model is important for power. A widely accepted rule of thumb for the minimum number of participants to use in multiple regression analysis is by Harris (1985), where the number of partici pants should exceed the number of independent variables by at least 50. By not including those variables, the n is more appropriate.

PAGE 161

161 student athlete general cheating compared to faculty that have not participated in service involving ath letics (b = 0.29, p < .05). UGA faculty that were randomly assigned MFB student athletes to answer questions about were significantly more likely to perceive general cheating compared to faculty assigned WBB (b = 0 .36 p < .05 ). The overall models for UF and OSU were not significant (UF: F = 0.32, p > .05, R = 0.14 ; OSU: F = 0.92, p > .05, R = 0.09) ( Table 5 15 ). However, for the UF subgroup, MFB was significant controlling for other variables (b = 0.35 p < .05), meaning U F faculty that were assigned MF B student athletes were significantly more likely to perceive general cheating than faculty assigned WBB student athletes The regression coefficient comparison test developed by Clogg, Petkova, and Haritou (1995) determined there are some significant dif ferences concerning some variables by university gr oup ( Table 5 16 for report of z values associated with each independent variable and university). These findings indicate the effect of service involving athletics on perceptions of student athlete general cheating was stronger for resp ondents at UGA than UI (z = 2.10 p < .05) and OSU (z = 2.56 p < .05). Additionally, the effect of the MFB manipulation on general cheating was stronger for UI compared to OSU (z = 1.96, p < .05) Predicting relying on other s. Faculty and student athlete status attributes that were significantly associated with at least one of the university groups were entered into OLS regression models to predict relying on others. More specifically, OLS regression models were run for each university using age, race, academic rank (full professor), time at current institution, service involving athletics, sports fandom, attendance at MFB

PAGE 162

162 and MBA events MFB, and MBA 30 Table 5 17 shows OLS regression results predicting relying on others by un iversity. For UI the overall model w as significant ( F = 2.24 p < .05, R = 0.24). Three variable s remained significant in the model controlling for other variables, which include faculty race, sports fandom and attendance at MBA events. White faculty at UI were significantly more likely to perceive student athletes are reliant on others (b = 0.44, p < .05). F aculty at UI that were larger fans of their university sports program were less likely to perceive student at hletes as reliant on others ( b = 0.15, p < .05). Additionally, faculty at UI that attended more MBA events were significantly less likely to perceive student athletes as reliant on others (b = 0.22, p < .05). For UGA, the overall model was significant (F = 3.32, p < .01, R = 0.27). Two vari ables remained significant in the model controlling for other variables, which include attendance at MBA events and MFB sporting group Faculty at UGA that attended more MBA events were significantly less likely to perceive student athletes as reliant on o thers (b = 0.25, p < .05). Additionally, faculty at UGA that were randomly assigned to the MFB manipulation were significantly more likely to perceive reliance on others compared to those assigned to the WBB sporting group (b = 0.64, p < .01) For OSU, th e overall model was significant (F = 2.40, p < .05, R = 0.18). There were two variables significantly related to relying on others after controlling for other variables, which include sports fandom and MFB sporting group Faculty at OSU that 30 I chose not to include the academic discipline (law) variable in the models ev en though it was significant at the bivar iate level to at least one of the universities because the UF, UGA, and OSU samples did not have any observations. Therefore, the variable would be dropped from their models. Additionally, no perceptions of student athlete status attribute variables inclu ded in the regression analysis, even though they were significant at the bivariate level. Adding these variables to the model created power issues.

PAGE 163

163 were bigger f ans of the sports program were significantly less likely to perceive that student athletes are reliant on others (b = 0.11, p < .05). Also, faculty at OSU that were assigned to the MFB manipulation were significantly more likely to perceive reliance on ot hers compared to those assigned to the WBB sporting group (b = 0.52, p < .01). For UF, the overall model was not significant (F = 1.32, p > .05, R = 0.13). In addition, none of the individual variables were significant predictors of relying on others. Th e regression coefficient comparison test developed by Clogg, Petkova, and Haritou (1995) determined some significant difference s concerning the var iables by university group. Table 5 1 8 shows a report of z values associated with each independent variable a nd universi ty T he effect of race on perceptions of student athletes relying on others was significantly different for faculty at UI compared to UGA (z = 2.01, p < .05). The coefficient for UI was significant and positive, meaning white faculty at UGA were significantly more likely to perceive student athletes relying on others. However, for UGA, the coefficient was negative and non significant. The effect of time faculty have been at their institution on perceptions of student athletes relyin g on others wa s stronger for UGA compared to U I (z = 2.12, p < .05). Finally, the effect of attendance at MBA events on perceptions of student athletes relying on others was significantly different for UGA compared to UF (z = 2.01, p < .05). The coefficient for UGA was significant and negative, meaning the more MBA events UGA faculty attended the less perceptions of relying on others they had regarding student athletes. However, the coefficient for UF was positive and non significant.

PAGE 164

164 Academic Deviance by Student Athlete Sporting Group Bivariate Relationships by Sporting G roup MFB. Six independent variables were significantly associated with the general cheating academic deviance scale for the MFB subgroup ( Table 5 19 5 20 5 21 and 5 22 ). Two of the significant indepen dent variables are faculty status attributes academic rank (associate professor) and sports fandom (Table 5 19 ) A ssociate professor s were significantly more likely to perceive MFB athletes as general cheaters than faculty of other ranks (r = 0.25, p < .0 1). Also, f aculty that were fans of their university sports program were significantly less likely to perceive MFB athletes as general cheaters ( 0.24, p < .01). No student athlete status attributes were significantly related at the bivariate level with ac ademic deviance for the MFB subgroup ( Table 5 20 ). One university status attribute was significantly related for MFB at the bivariate leve l, which was OSU ( Table 5 21 ). Faculty at OSU were significantly less likely to perceive MFB student athletes as gener al cheaters compared to the other universities ( 0.18, p < .05). Finally, three university athletic attributes had significant negative relationships with general cheating for MFB ( Table 5 22 ). These included student athlete population (p = 0.18, p < .05) varsity athletic teams (p = 0.18, p < .05), and NCAA infractions (p = 0.18, p < .05). There were seven independent variables significantly related to the relying on others academic devianc e scale for MFB ( Tables 5 19 5 20 5 21 and 5 22 ). They were a ll faculty status attributes, which included: age, academic rank (full professor), academic discipline (law and other), time at current institution, sports fandom, and attendan ce at MFB events ( Table 5 19 ). Older faculty were less likely to perceive MFB st udent athletes as reliant on others (r = 0.18, p < .05). Faculty in the rank of full

PAGE 165

165 professor were significantly less likely to perceive MFB athletes as reliant on others compared to faculty in other ranks (r = 0.19, p < .05). Faculty in the academic di sciplines of law (r = 0.15, p < .05) and other category (r = 0.17, p < .05) were significantly less likely to perceive MFB athletes as reliant on others compared faculty in other disciplines Finally, faculty that had been at the institution longer (r = 0.17, p < .05), were bigger fans of their university sport program (r = 0.25, p < .01), and attended more MFB events (r = 0.22, p < .01) were also significantly less likely to perceive MFB athletes as reliant on others. MBA. Four variables were signif icantly related to general cheating at the bivariate level for t he MBA subgroup ( Tables 5 19 5 20 5 21 and 5 22 ). Two were faculty status attributes, which included academic discipline (engineering) a nd sports fandom ( Table 5 19 ). Faculty in engineering were significantly less likely to perceive MBA athletes as general cheaters compared to faculty in other disciplines (r = 0.29, p < .001). Additionally, faculty that are bigger fans of their university sports program were significantly less likely to per ceive MBA athletes as general cheaters (r = 0.34, p < .001). There was one student athlete status attribute related at the bivariate level, which was perception of percentage of MBA athletes ( Table 5 20 ). Faculty that estimated a higher percentage of MBA student athletes on their campus, were significantly more likely to perceive MBA athletes as general cheaters (r = 0.19, p < .05). There was also one university status attribute significantly related to general cheating at the bivariate level for the MBA subgroup which was UI ( Table 5 21 ). Faculty at UI were more likely to perceive MBA athletes as general cheaters compared to the three other

PAGE 166

166 institutions (r = 0.22, p < .05) There were no athletic status attributes significantly related to general cheating at the bivariate level for the MBA subgroup ( Table 5 22 ). There were eleven independent variables significantly related to the relying on others academic devianc e scale for MBA ( Tables 5 19 5 20 5 21 and 5 22 ). Three were faculty status attributes, wh ich included academic rank (full professor), academic discipline (business), a nd sports fandom (Table 5 19 ). Full professors were significantly less likely to perceive MBA athletes as reliant on others compared to faculty of other ranks (b = 0.18, p < .05 ). Faculty in business were significantly more likely to perceive MBA athletes as general cheaters compared to other disciplines (r = 0.17, p < .05). Additionally, faculty that are bigger fans of their university sports program were significantly less like ly to perceive MBA athletes as general cheaters (r = 0.31, p < .001). There were three student athlete status attributes significantly related to relying on others at the bivariate level for the MBA subgroup (Table 5 20 ). Faculty that estimated a higher percentage of black student athletes on their campus, were significantly more likely to perceive MBA athletes as reliant on others (r = 0.26, p < .01). Additionally, faculty that estimated a higher percentage of MFB and MBA athletes on their campus were si gnificantly more likely to perceive MBA athletes as reliant on others (MFB: r = 0.21, p < .05; MBA: r = 0.18, p < .05). There were two university status attributes significantly related to relying on others OSU and UF for the M BA subgroup ( Table 5 21 ). Faculty at OSU were significantly less likely to perceive MBA student athletes as reliant on others compared to the three other universities (r = 0.18, p < .05). However, faculty at UF were

PAGE 167

167 significantly more likely to perceive MBA student athletes as rel iant on others compared to facul ty at the three other universities (r = 0.23, p < .01). Finally, there were three university athletic status attributes significantly related to the relying on other academic deviance scale for the MBA subgroup ( Table 5 22 ) These include: student athlete population, varsity athletic teams, and NCAA infractions. Faculty at universities with larger student athlete populations (r = 0.19, p < .05), more varsity athletic teams (r = 0.19, p < .05), and NCAA infractions (r = 0. 19, p < .05) were significantly less likely to perceive MBA student athletes as reliant on others. WBB. There were five independent variables significantly related to genera l cheating for t he WBB subgroup ( Tables 5 19 5 20 5 21 and 5 22 ). Faculty statu s attributes were four of the five significant variables, which included age, service involving athletics, attendance at MBA events, and contact with student athletes ( Table 5 19 ). Older faculty were significantly less likely to perceive WBB student athlet es as general cheaters (r = 0.20, p < .05). Faculty that were involved in service to athletics were significantly less likely to perceive WBB athletes as general cheaters compare d to faculty that were not involved in service to athletics (r = 0.17, p < 05). Additionally, faculty that attended more MBA events (r = 0.20, p < .05) and had more contact with student athletes on their campuses (r = 0.26, p < .001) were significantly less likely to perceive WBB as general cheaters compared to faculty that att ended less MBA events and had less contact with student athletes on their campus Only o ne of the university status attribute variables was significantly related to general cheating for the WBB subgroup ( Table 5 20 ). Faculty at UGA were significantly

PAGE 168

168 less likely to perceive WBB athletes as general cheaters compared to faculty at the three other universities (r = 0.20, p < .05). There were only two independent variables significantly related to the relying on others academic deviance scale for t he WBB sub group ( Tables 5 19 5 20 5 21 and 5 22 ). These included academic rank (assistant professor) and perception of percentage of male athletes on their campus. Assistant professors were significantly less likely to perceive WBB athletes as reliant on others c ompared to faculty of other ranks (b = 0.16, p < .05) ( Table 5 19 ). Faculty that estimated a larger number of male athletes on their campus were significantly less likely to perceive WBB student athletes as reliant on others (r = 0.17, p < .05) ( Table 5 20 ). OLS Regression by Sporting Group and Clogg Coefficient Comparison Test Predicting general cheating. Variables that were significantly associated with at least one of the sporting groups were entered into OLS regression models to predict general cheati ng. More specifically, OLS regression models were run for each sporting group using age, academic rank (associate professor), academic discipline (engineering), service involving athletics, sports fandom, attendance MBA events, contact with student athlete s, UI, UGA and UF 31 Table 5 23 shows OLS regression results predicting general cheating by sporting group. For MFB, the overal l model is significant (F = 2.31 p < .05, R = 0.16 ) ( Table 5 23 ). Two variables remained significant after controlling for othe r variables in the MFB model, academic rank (associate professor) and UF Associate professors were 31 The significant university athletic status attributes were not included in the models due to multicollin earity. These variables were highly correlated with the university status attributes, especially OSU. Additionally, the significant perceptions of student athlete status attributes were not included in the analysis for power issues.

PAGE 169

169 significantly more likely to perceive MFB athletes as general cheaters compared to faculty of other ranks (b = 0.38, p < .01 ). Additionally, faculty at UF w ere significantly more likely to perceive MFB athletes as general cheaters compared to faculty at OSU (b = 0.37, p < .05 ). For MBA, the overall model is significant (F = 3.16 p < .01, R = 0.23 ) ( Table 5 23 ). Three variables remained significant after co ntrolling for other variables in the MBA model, ac ademic discipline (engineering), sports fandom and UI Faculty in engineering were significantly less likely to perceive MBA student athletes as general cheaters compared to faculty in other disciplines (b = 0.68, p < .01 ). F aculty that were bigger sports fans of their university sports program were significantly less likely to perceive MBA athletes as general cheat er s (b = 0.11, p < .01). Additionally, faculty at UI were significantly more likely to perc eive MBA athletes as general cheaters compared to faculty at OSU (b = 0.37, p < .05). For WBB, the overall model is significant (F = 2.98 p < .01 R = 0.20 ) ( Table 5 23 ). Two variables remained significant after controlling for other variables in the WBB model, attendance at MBA events and contact with student athletes. Faculty that attended more MBA events (b = 0.13, p < .05) and had more contact with student athletes (b = 0.11, p < .05) were less likely to perceive WBB athletes as general cheaters. Th e regression coefficient comparison test developed by Clogg, Petkova, and Haritou (1995) determined there are significant differences concerning some independent variables b y sporting group. Table 5 24 report s the z values associated with each independent variable First, the effect of academic rank (associate professor

PAGE 170

170 status) on perceptions of student athletes general cheating was significantly different MFB compared to WBB (z = 3.15 p < .05). The coefficient for MFB was significant and positive, meaning associate professors were more likely to perceive MFB student athletes as involved in general cheating. However, the coefficient for WBB was non significant. Second, th e effect of academic discipline (engineering) on perceptions of student athletes genera l cheating was significantly different for MBA compared to MFB (z = 2.48 p < .05) and WBB (z = 2.09, p < .05) The coefficient for MBA was negative and significant, meaning faculty in engineering were less likely to perceive general cheating among MBA st udent athletes. However, for MFB and WBB the coefficient was non significant. Third, the effect of sports fandom on perceptions of student athletes general cheating was significantly different for MBA compared to WBB (z = 2.80, p < .05) The coefficient f or MBA was negative and significant, meaning faculty that were bigger fans of their university sports program were less likely to perceive MBA student athletes as general cheaters. However, for WBB, the coefficient was non significant. Last, the effect of contact with student athletes on perceptions of student athlete cheating was significantly different for WBB than MBA (z = 2.34 p < .05). The coefficient for WBB is negative and significant, meaning faculty with more contact with student athletes are less likely to perceive WBB athletes as general cheaters. However, the coefficient for MBA was non significant. Predicting relying on others. Variables that were significantly associated with relying on others for at least one of the sporting groups were ente red into OLS regression models to predict relying on others. More specifically, OLS regression models were run for each sporting group using age, academic rank (associate professor

PAGE 171

171 and full professor), academic discipline (business and other category 32 ), ti me at current institution, sports fandom, attendance MFB events, UI, UGA and UF 33 Table 5 25 shows OLS regression results predicting relying on others by sporting group. For MFB, the overall model is significant (F = 2.08 p < .05, R = 0.14 ) ( Table 5 25 ). There is only one variable that remain s significant after controlling for other variables, which is sport fandom. F aculty that were bigger fans of their university sports program were significant ly less like to perceive MFB student athletes as reliant on others (b = 0.10, p < .05). For MBA, the overal l model is significant (F = 2.96, p < 0 1, R = 0.23 ) (Table 5 25 ). There are two variables that remain significant after controlling for other variables, which are academi c discipline (business) and sport fa ndom. Faculty in business were significantly more likely to perceive MBA student athletes as reliant on others compared to facult y in other disciplines (b = 0.43, p < .05 ). Additionally, faculty that were bigger fans of their university sports programs wer e significant ly less likely to perceive MBA student athletes as reliant on others (b = 0.19 p < .001). For WBB, t he overall model is signific ant (F = 2.25 p < .05, R = 0.16 ) ( Table 5 25 ). There are two variables that remain significant after controllin g for other variables, attendance at MFB events and UF. Faculty that attended more MFB events were significantly less likely to perceive WBB student athletes as reliant on others (b = 0.16, 32 Although the law aca demic discipline was significant at the bivariate level for at least one of the sporting groups, I chose not to include the variable in the regression models because there were too few cases in the sample (n = 4). 33 Again, student athlete status attribut e variables were not included in the regression models by sport because of power issues. The athletic status attribute variables that were significant at the bivariate level were also not included in the regression analysis because of multicollinearity. Th ey are highly correlated with the university status attributes, especially OSU.

PAGE 172

172 p < .05). Additionally, faculty at UF were significantly more lik ely to perceive WBB student athletes as reliant on others compared to faculty at OSU (b = 0.49, p < .01). The regression coefficient comparison test developed by Clogg, Petkova, and Haritou (1995) determined there are significant differences concerning so me variables b y sporting group. Table 5 26 report s the z values associated with each independent variable in the models First, the effect of academic rank (associate professor status) on faculty perceptions of student athletes relying on others was signif icantly different for WBB than MBA (z = 1.96 p < .05). The coefficient for WBB was negative, meaning associate professors were less likely to perceive WBB student athletes as reliant on others compared to faculty of other ranks. Alternatively, the coeffic ient for MBA was positive, meaning associate professors are more likely to perceive MBA athletes as reliant on others. Second the effect of academic discipline (business) on perceptions of student athletes relying on others was significantly different for MBA than WBB (z = 2.76 p < .05). The coefficient for MBA was positive and significant, meaning faculty in business are significant more likely to perceive MBA athletes as reliant on others. However, for WBB the coefficient is non significant. Third the effect of sports fandom on faculty perceptions of relying on others was significantly different for MFB and MBA compared to WBB ( MFB: z = 2.12 p < .05 ; MBA: z = 3.28, p < .05 ). The coefficients for MFB and MBA were negative and significant, meaning f acu lty that were bigger fans of their university sports programs were significantly less likely to perceive MFB and MBA student athletes as reliant on others. However, the coefficient for WBB was non significant. Fourth, the effect of attendance at MFB events on perceptions of student athletes relying on others was significantly different for WBB than MBA (z = 2.50, p <

PAGE 173

173 .05). For WBB, the coefficient is positive and significant, meaning faculty that attended more MFB events were more likely to perceive WBB ath letes as reliant on others. However, the coefficient for M B A was non significant. Finally, the effect of UI on perceptions of student athletes relying on others was significantly different for WBB than MFB (z = 2.06, p < .05). Although neither coefficient was significant, they were in different directions. For WBB, the effect was positive, meaning faculty at UI were more likely to perceive WBB student athletes as reliant on others. However, for MBA, the effect was negative, meaning faculty at UI were less likely to perceive MBA student athletes as reliant on others.

PAGE 174

174 Table 5 1. Factorial ANOVA of university and sporting group for general cheating Source Subjects df MS F Model 17.81 11 1.62 3.84*** 0.10 University 5.65 3 1.89 4.47** 0.03 Sport 10.74 2 5.37 12.74*** 0.06 University*Sport 3.01 6 0.50 1.19 0.02 Residual 168.97 401 0.42 Total 186.78 412 Note. *p < .05, **p < .01, ***p < .001 Table 5 2. Mean differences and confidence intervals of perceptions of general cheating by university Mean Difference SE p value 95% Confidence Interval Lower Bound Upper Bound Ohio State University University of Florida 0.16 0.09 0.242 0.39 0.06 University of Georgia 0.03 0.09 0.979 0.19 0.26 University of Illinois 0.22 0.09 0.076 0.46 0. 01 University of Florida Ohio State University 0.16 0.09 0.242 0.06 0.39 University of Georgia 0.20 0.09 0.138 0.04 0.43 University of Illinois 0.06 0.10 0.929 0.31 0.19 University of Georgia Ohio State University 0.03 0.09 0.979 0.26 0.19 Un iversity of Florida 0.20 0.09 0.138 0.43 0.04 University of Illinois .25 0.10 0.040 0.50 0.01 University of Illinois Ohio State University 0.22 0.09 0.076 0.01 0.46 University of Florida 0.06 0.10 0.929 0.19 0.31 University of Georgia .25 0 .10 0.040 0.01 0.50 Note. *p < .05

PAGE 175

175 Table 5 3. Mean differences and confidence intervals of perceptions of general cheating by sport Mean Difference SE p value 95% Confidence Interval Lower Bound Upper Bound Football Baseball 0.06 0.08 0.694 0.12 0.25 Women's Basketball 0.34 0.08 0.000 0.16 0.52 Baseball Football 0.06 0.08 0.694 0.25 0.12 Women's Basketball 0.28 0.08 0.002 0.09 0.46 Women's Basketball Football 0.34 0.08 0.000 0.52 0.16 Baseball 0.28 0.08 0.002 0.46 0.09 N ote. *p < .05 Table 5 4 Factorial ANOVA of university and sporting group for relying on others Source Subjects df MS F Model 27.83 11 2.53 4.27*** 0.10 University 7.66 3 2.55 4.31** 0.03 Sport 16.57 2 8.29 13.99*** 0.06 University*Sport 3.99 6 0.67 1.12 0.02 Residual 245.83 415 0.59 Total 273.67 426 Note. *p < .05, **p < .01, ***p < .001 Table 5 5. Mean differences and confidence intervals of perceptions of relying on others by university Mean Difference SE p value 95% Confidence I nterval Lower Bound Upper Bound Ohio State University University of Florida 0.29 0.10 0.023 0.55 0.03 University of Georgia 0.04 0.10 0.973 0.30 0.22 University of Illinois 0.12 0.11 0.666 0.40 0.15

PAGE 176

176 Table 5 5. Continued Mean Difference SE p value 95% Confidence Interval Lower Bound Upper Bound University of Florida Ohio State University 0.29 0.10 0.023 0.03 0.55 University of Georgia 0.24 0.11 0.099 0.03 0.52 University of Illinois 0.17 0.11 0.445 0.12 0.46 University of G eorgia Ohio State University 0.04 0.10 0.973 0.22 0.30 University of Florida 0.24 0.11 0.099 0.52 0.03 University of Illinois 0.08 0.11 0.899 0.37 0.21 University of Illinois Ohio State University 0.12 0.11 0.666 0.15 0.40 University of Florid a 0.17 0.11 0.445 0.46 0.12 University of Georgia 0.08 0.11 0.899 0.21 0.37 Note. *p < .05 Table 5 6. Mean differences and confidence intervals of perceptions of relying on others by sport Mean Difference SE p value 95% Confidence Interval Lower Bound Upper Bound Football Baseball 0.22 0.09 0.045 0.00 0.44 Women's Basketball 0.47 0.09 0.000 0.26 0.68 Baseball Football 0.22 0.09 0.045 0.44 0.00 Women's Basketball 0.25 0.09 0.021 0.03 0.47 Women's Basketball Football 0.47 0.09 0 .000 0.68 0.26 Baseball 0.25 0.09 0.021 0.47 0.03 Note. *p < .05

PAGE 177

177 Table 5 7 Correlations of academic deviance and faculty status attributes General cheating Relying on others Faculty status attributes Age 0.11* 0.13** Gender (Ma le = 1; Female = 0) 0.01 0.01 Race (White = 1; Non white = 0) 0.06 0.09* Academic Rank Lecturer = 1 0.11* 0.01 Assistant Professor = 1 0.05 0.09* Associate Professor = 1 0.08 0.02 Full Professor = 1 0.05 0.12* Other = 1 0.04 0.03 Tenure status (Tenure = 1; Non tenure = 0) 0.02 0.10* Administrative position (Yes = 1; No = 0) 0.05 0.07 Academic Discipline Architecture = 1 0.00 0.06 Arts and Humanities = 1 0.03 0.03 Business = 1 0.03 0.02 Education = 1 0.09 0. 07 Engineering = 1 0.06 0.00 Law = 1 0.01 0.08 Life Sciences = 1 0.01 0.01 Medicine and Health Sciences = 1 0.05 0.02 Physical sciences and mathematics = 1 0.06 0.02 Social and Behavior sciences = 1 0.01 0.04 Other = 1 0.03 0.04 Time at current institution 0.05 0.11* Service involving athletics 0.11* 0.02 Sports fandom 0.22*** 0.20*** Attendance at MFB events 0.13** 0.14** Attendance at MBA events 0.12* 0.08 Attendance at WBB events 0.10 0.08 Contact wi th student athletes 0.12* 0.06 Note. *p < .05, **p < .01, ***p < .001

PAGE 178

178 Table 5 8 Correlations of academic deviance and university status attributes General cheating Relying on others University status attributes University OSU 0.08 0.09 UF 0.07 0.13** UGA 0.09 0.04 UI 0.11* 0.01 Region (Midwest = 1; South = 0) 0.02 0.07 Undergraduate Student Population 0.02 0.05 Faculty population 0.00 0.06 Table 5 9 Correlations of academic deviance variables and stude nt athlete status attributes General cheating Relying on others Sporting group assigned MFB 0.16*** 0.22*** MBA 0.07 0.01 WBB 0.23*** 0.22*** Faculty perception of student athlete attributes Estimate % of student athlete gender Male 0.00 0.07 Female 0.05 0.11* Estimate % of student athlete race Black 0.08 0.18*** White 0.03 0.06 Hispanic 0.04 0.02 Asian 0.07 0.04 Other 0.09 0.06 Estimate % of student athlete sport MFB 0.00 0.08 MBA 0.00 0.03 WBB 0.02 0.02

PAGE 179

179 Table 5 10 Correlations of academic deviance and university athletic status attributes General cheating Relying on others Athletic status attributes Student athlete population 0.09 0.09 Varsity athletic teams 0.0 8 0.09 NCAA infractions 0.08 0.09 Athletic revenue 0.09 0.05 Directors cup standing 0.07 0.02 Table 5 11 OLS regression predicting general cheating academic deviance for the entire sample b SE B Faculty status attributes Age 0.00 0.00 0.08 Academic Rank (Lecturer = 1) 0.06 0.09 0.03 Service involving athletics 0.03 0.07 0.02 Sports fandom 0.06* 0.02 0.14 Attendance at MFB events 0.03 0.03 0.07 Attendance at MBA events 0.10* 0.04 0.13 Contact with student athlet es 0.04 0.03 0.07 University status attributes UI 0.24** 0.09 0.15 UGA 0.00 0.08 0.00 UF 0.29** 0.09 0.20 Student athlete status attributes MFB 0.34*** 0.07 0.25 MBA 0.29*** 0.08 0.21 Constant 2.26*** 0.18 R square 0.14 df 12 F 5.25*** N 392 Note. *p < .05, **p < .01, ***p < .001

PAGE 180

180 Table 5 12 OLS regression predicting relying on others academic deviance for the entire sample Variable b SE B Faculty status attributes Age 0.01 0.01 0.10 Race (White) 0. 20* 0.13 0.12 Academic Rank (Assistant Professor) 0.12 0.13 0.07 Academic Rank (Full Professor) 0.03 0.11 0.01 Tenure status (Tenure) 0.08 0.13 0.05 Time at current institution 0.00 0.01 0.05 Sports fandom 0.10** 0.03 0.22 Attendance at MFB events 0.03 0.04 0.05 University status attributes UI 0.09 0.11 0.05 U GA 0.12 0.11 0.07 UF 0.30** 0.11 0.18 Perceptions of student athlete attributes Gender (Female) 0.01* 0.00 0.12 Race (Black) 0.01*** 0.00 0.21 Student athlete st atus attributes MFB 0.38*** 0.09 0.24 MBA 0.25* 0.10 0.15 Constant 2.88*** 0.33 R square 0.19 df 15 F 4.39*** N 306 Note. *p < .05, **p < .01, ***p < .001

PAGE 181

181 Tab le 5 13 Correlations of academic deviance and faculty status attrib utes UI UF UGA OSU General cheating Relying on others General cheating Relying on others General cheating Relying on others General cheating Relying on others Faculty status attributes Age 0.07 0.04 0.10 0.12 0 .11 0.21* 0.16 0.11 Gender (Male = 1; Female = 0) 0.07 0.10 0.10 0.11 0.05 0.05 0.00 0.04 Race (White = 1; Non white = 0) 0.02 0.22* 0.22* 0.19* 0.03 0.15 0.07 0.13 Academic Rank Lecturer = 1 0.22* 0.02 0.06 0.08 0 .09 0.02 0.00 0.11 Assistant Professor = 1 0.00 0.09 0.05 0.08 0.02 0.07 0.01 0.02 Associate Professor = 1 0.27* 0.13 0.02 0.05 0.15 0.10 0.00 0.13 Full Professor = 1 0.07 0.17 0.04 0.21* 0.04 0.11 0.02 0.01 Other = 1 0.05 0 .00 0.09 0.07 0.07 0.12 0.02 0.02 Tenure status (Tenure = 1; Non tenure = 0) 0.10 0.09 0.06 0.19 0.08 0.04 0.00 0.09 Administrative position (Yes = 1; No = 0) 0.00 0.06 0.10 0.16 0.04 0.04 0.07 0.10 Academic Discipline Architecture = 1 0.01 0.07 Arts and Humanities = 1 0.14 0.05 0.07 0.00 0.07 0.10 0.06 0.05 Business = 1 0.04 0.03 0.15 0.14 0.02 0.01 0.07 0.03 Education = 1 0.09 0.09 0.14 0.07 0.21* 0.16 0.06 0.09 Enginee ring = 1 0.12 0.10 0.02 0.03 0.00 0.03 0.05 0.02 Law = 1 0.13 0.28** Life Sciences = 1 0.07 0.06 0.04 0.02 0.14 0.07 0.02 0.01 Medicine and Health Sciences = 1 0.04 0.08 0.06 0.11 0.12 0.02 0.09 0.04 Physical sciences and mathematics = 1 0.01 0.06 0.01 0.07 0.28** 0.15 0.03 0.09 Social and Behavior sciences = 1 0.01 0.04 0.09 0.15 0.09 0.07 0.00 0.04 Other = 1 0.01 0.14 0.14 0.11 0.13 0.00 0.09 0.09 Note. *p < .05, **p < .01, *** p < .001

PAGE 182

182 Table 5 13 Continued UI UF UGA OSU General cheating Relying on others General cheating Relying on others General cheating Relying on others General cheating Relying on others Time at current institution 0.07 0.01 0. 02 0.16 0.13 0.23* 0.02 0.01 Service involving athletics 0.00 0.15 0.15 0.03 0.33** 0.21* 0.00 0.02 Sports fandom 0.32** 0.23* 0.11 0.14 0.13 0.09 0.26** 0.32*** Attendance at MFB events 0.19 0.01 0.10 0.00 0.00 0.18 0.16 0.20* Attendance at MBA events 0.26* 0.25* 0.04 0.11 0.30** 0.34*** 0.14 0.10 Attendance at WBB events 0.06 0.10 0.17 0.18 0.03 0.06 0.08 0.08 Contact with student athletes 0.23* 0.00 0.04 0.05 0.10 0.08 0.11 0.1 4 Note. *p < .05, **p < .01, ***p < .001

PAGE 183

183 Table 5 14 Correlations of academic deviance and perceptions of student athlete status attributes by university UI UF UGA OSU General cheating Relying on others General cheating Relying on others General cheating Relying on others General cheating Relying on others Sporting group assigned MFB 0.08 0.04 0.23* 0.18 0.20** 0.34*** 0.03 0.26** MBA 0.21 0.09 0.08 0.12 0.02 0.04 0.05 0.06 WBB 0.28** 0.13 0.28** 0.28** 0.33*** 0.32** 0.09 0.21* Faculty perceptions of student athlete status attributes Estimate % of student athlete gender Male 0.23 0.06 0.02 0.10 0.10 0.09 0.00 0.13 Female 0.17 0.03 0.02 0. 08 0.10 0.19 0.14 0.18 Estimate % of student athlete race Black 0.24* 0.15 0.02 0.02 0.11 0.29* 0.02 0.24* White 0.17 0.02 0.17 0.11 0.14 0.04 0.06 0.12 Hispanic 0.23 0.15 0.09 0.11 0.09 0.05 0.03 0.04 Asian 0 .16 0.19 0.07 0.04 0.17 0.06 0.04 0.01 Other 0.02 0.21 0.12 0.08 0.29 0.11 0.21 0.05 Estimate % of student athlete sport MFB 0.00 0.23 0.22* 0.08 0.13 0.08 0.10 0.08 MBA 0.02 0.20 0.20 0.14 0.17 0.14 0.06 0.03 WBB 0.07 0.23 0.23* 0.13 0.05 0.14 0.06 0.05 Note. *p < .05, **p < .01, ***p < .001

PAGE 184

184 Table 5 15 OLS regression predicting general cheating academic deviance by university Variable UI UF UGA OSU b SE B b SE B b SE B b SE B Faculty status attributes Race (White = 1) 0.20 0.18 0.11 0.46 0.32 0.16 0.09 0.20 0.05 0.07 0.18 0.04 Academic Rank (Lecturer = 1) 0.33 0.24 0.15 0.00 0.25 0.00 0.17 0.16 0.11 0.17 0.16 0.11 Academic Rank (Associate Professor = 1) 0.31 0 .17 0.20 0.02 0.17 0.01 0.10 0.16 0.06 0.00 0.14 0.00 Academic Discipline (Education = 1) 0.21 0.27 0.08 0.16 0.26 0.07 0.31 0.21 0.15 0.02 0.22 0.01 Academic Discipline (Physical sciences and mathematics = 1) 0.11 0.21 0.06 0.09 0.26 0.04 0.16 0.25 0.06 0.03 0.21 0.01 Service involving athletics 0.14 0.15 0.11 0.18 0.15 0.14 0.29* 0.14 0.23 0.20 0.13 0.15 Sports fandom 0.09 0.05 0.20 0.02 0.04 0.06 0.02 0.05 0.05 0.07 0.04 0.19 Attendance at MBA events 0.13 0.09 0.17 0.01 0.07 0.02 0.14 0.08 0.17 0.04 0.11 0.04 Contact with student athletes 0.10 0.07 0.18 0.04 0.07 0.07 0.01 0.05 0.02 0.09 0.06 0.15 MFB 0.46** 0.16 0.35 0.35* 0.16 0.25 0.36* 0.14 0.28 0.03 0.15 0.02 MBA 0.43* 0.17 0.31 0.28 0.17 0.19 0.20 0.15 0 .15 0.11 0.15 0.09 Constant 2.28*** 0.31 1.54*** 0.41 1.96*** 0.30 2.20*** 0.32 R square 0.34 0.14 0.28 0.09 df 11 11 11 11 F 3.13** 0.32 3.11** 0.92 N 79 94 99 118 Note. *p < .05, **p < .01, ***p < .001

PAGE 185

185 Table 5 16 Z values comparing beta coefficients predicting general cheating by university UI v. UF UI v. UGA UI v. OSU UF v. UGA UF v. OSU UGA v. OSU z z z z z z Race 0.71 1.08 0.51 1.46 1.06 0.59 AR Lecturer 0.95 0.55 1.73 0.57 0.57 1.50 AR Associate Professor 0.82 0.90 1.41 0.51 0.09 0.47 AD Education 0.99 1.52 0.55 0.45 0.53 1.09 AD Physical sciences and mathematics 0.60 0.15 0.27 0.69 0.36 0.40 Service involving athletics 1.51 2.10* 0.30 0.54 1.91 2.56* Sports fandom 1.09 1.56 0.31 0.62 0.88 1.41 Attendance at MBA events 1.23 0.08 0.63 1.41 0.38 0.74 Contact with student athletes 1.41 1.05 0.11 0.58 1.41 1.02 MFB 0.49 0.47 1.96* 0.05 1.46 1.61 MBA 0.62 1.01 1.41 0.35 0.75 0.42 Note. *p < .05

PAGE 186

186 Table 5 17 OLS regression predicting relying on others academic deviance by university Variable UI UF UGA OSU b SE B b SE B b SE B b SE B Faculty status attributes Age 0.00 0.01 0.01 0.01 0.01 0.13 0.01 0.01 0.09 0.01 0.01 0.11 Race (White = 1) 0.44* 0.20 0.23 0.29 0.30 0.10 0.22 0.26 0.08 0.14 0.20 0.06 Academic Rank (Full professor = 1) 0.14 0.20 0.10 0.25 0.18 0.16 0.10 0.20 0.06 0.03 0.19 0.02 Time at current institution 0.01 0.01 0.11 0.01 0.01 0.15 0.02 0.01 0.21 0.00 0.01 0.00 Service involving athletics 0.14 0.15 0.10 0.04 0.15 0.03 0.07 0.17 0.04 0.11 0.15 0.06 Sports Fandom 0.15* 0.07 0.29 0.10 0.05 0.21 0.01 0.07 0.02 0.11* 0.05 0.24 Attendance at MFB events 0.16 0.08 0.27 0.06 0.07 0.10 0.02 0.06 0.04 0.04 0.07 0.06 Attendance at MBA events 0.22* 0.11 0.26 0.04 0.08 0.05 0.25* 0.12 0.22 0.04 0.13 0.03 Student athlete status attributes MFB 0.23 0.18 0.16 0.28 0.18 0.18 0.64** 0.19 0.38 0.52** 0.17 0.31 MBA 0.33 0. 19 0.22 0.26 0.19 0.16 0.37 0.20 0.20 0.22 0.17 0.13 Constant 2.66*** 0.52 2.63*** 0.54 3.44*** 0.53 3.08*** 0.45 R square 0.24 0.13 0.27 0.18 df 10 10 10 10 F 2.24* 1.32 3.32** 2.40* N 84 100 101 124 Note. *p < .05, **p < .01, ***p < .001

PAGE 187

187 Table 5 18 Z values comparing beta coefficients predicting relying on others by university UI v. UF UI v. UGA UI v. OSU UF v. UGA UF v. OSU UGA v. OSU z z z z z z Age 0.71 0.71 0.71 1.41 1.41 0.00 Race (White = 1) 0.42 2.01* 1.06 1.28 0.42 1.10 Academic Rank (Full professor = 1) 0.41 0.85 0.62 1.30 1.07 0.25 Time at current institution 1.41 2.12* 0.71 0.71 0.71 1.41 Service involving athletics 0.47 0.93 0.14 0.49 0.33 1.02 Sports Fandom 0.58 1.62 0. 46 1.28 0.14 1.39 Attendance at MFB events 0.94 1.80 1.88 0.87 1.01 0.22 Attendance at MBA events 1.91 0.18 1.53 2.01* 0.00 1.64 MFB 0.20 1.57 1.17 1.38 0.97 0.47 MBA 0.26 0.15 0.43 0.40 0.16 0.57 Note. *p < .05

PAGE 188

188 Table 5 19 Correlations of academic deviance and faculty status attributes by sporting group MFB MBA WBB General cheating Relying on others General cheating Relying on others General cheating Relying on others Faculty status attributes Age 0.07 0.18* 0.08 0.16 0.20* 0.07 Gender (Male = 1) 0.02 0.05 0.05 0.08 0.02 0.07 Race (White= 1) 0.07 0.13 0.05 0.01 0.14 0.10 Academic Rank Lecturer = 1 0.14 0.01 0.15 0.07 0.05 0.07 Assistant Profess or = 1 0.03 0.12 0.05 0.14 0.13 0.04 Associate Professor = 1 0.25** 0.10 0.10 0.12 0.13 0.16* Full Professor = 1 0.12 0.19* 0.00 0.18* 0.02 0.03 Other = 1 0.11 0.04 0.08 0.01 0.06 0.04 Tenure status (Tenure = 1; Non tenure = 0) 0 .08 0.12 0.04 0.09 0.06 0.09 Administrative position (Yes = 1; No = 0) 0.12 0.08 0.01 0.09 0.03 0.01 Academic Discipline Architecture = 1 0.09 0.02 0.10 0.04 0.02 Arts and Humanities = 1 0.00 0.10 0.07 0.09 0.06 0.08 Business = 1 0.03 0.03 0.07 0.17* 0.03 0.11 Education = 1 0.05 0.00 0.01 0.05 0.09 0.10 Engineering = 1 0.00 0.00 0.29*** 0.15 0.05 0.10 Law = 1 0.06 0.15* 0.00 0.08 . Life Sciences = 1 0.01 0.01 0.07 0.04 0.11 0 .07 Medicine and Health Sciences = 1 0.04 0.09 0.06 0.01 0.06 0.02 Physical sciences and mathematics = 1 0.08 0.05 0.02 0.05 0.07 0.04 Social and Behavior sciences = 1 0.01 0.10 0.09 0.13 0.13 0.15 Other = 1 0.02 0.17* 0.08 0.04 0.03 0.09 Note. *p < .05, **p < .01, ***p < .001

PAGE 189

189 Table 5 19 Continued MFB MBA WBB General cheating Relying on others General cheating Relying on others General cheating Relying on others Time at current institution 0.0 6 0.18* 0.05 0.15 0.04 0.04 Service involving athletics 0.04 0.04 0.11 0.02 0.17* 0.04 Sports fandom 0.24** 0.25** 0.34*** 0.31*** 0.08 0.06 Attendance at MFB events 0.16 0.22** 0.12 0.09 0.09 0.07 Attendance at MBA even ts 0.10 0.14 0.04 0.11 0.20* 0.16 Attendance at WBB events 0.05 0.12 0.02 0.03 0.11 0.03 Contact with student athletes 0.03 0.03 0.08 0.12 0.26*** 0.14 Note. *p < .05, **p < .01, ***p < .001

PAGE 190

190 Table 5 20 C orrelations of academic deviance and student athlete status attributes by sporting group MFB MBA WBB General cheating Relying on others General cheating Relying on others General cheating Relying on others Faculty perceptions of stud ent athlete status attributes Estimate % of student athlete gender Male 0.08 0.00 0.07 0.02 0.12 0.17* Female 0.02 0.13 0.01 0.08 0.11 0.13 Estimate % of student athlete race Black 0.08 0.17 0.12 0.26** 0 .06 0.14 White 0.03 0.07 0.05 0.13 0.11 0.02 Hispanic 0.13 0.09 0.09 0.13 0.09 0.14 Asian 0.05 0.01 0.11 0.16 0.10 0.03 Other 0.06 0.09 0.21 0.11 0.11 0.09 Estimate % of student athlete sport MFB 0.08 0.02 0.10 0.21* 0.06 0.16 MBA 0.02 0.04 0.19* 0.18* 0.11 0.06 WBB 0.03 0.01 0.04 0.14 0.00 0.01

PAGE 191

191 Table 5 21 Correlations of academic deviance and university status attributes by sporting group MFB MBA WBB General ch eating Relying on others General cheating Relying on others General cheating Relying on others University status attributes University OSU 0.18* 0.04 0.11 0.18* 0.03 0.09 UF 0.13 0.11 0.08 0.23** 0.09 0.15 UGA 0.02 0.05 0.15 0.09 0.20* 0.13 UI 0.05 0.12 0.22* 0.07 0.07 0.08 Region (Midwest = 1; South = 0) 0.12 0.13 0.07 0.11 0.09 0.02 Undergraduate Student Population 0.14 0.05 0.03 0.11 0.10 0.02 Faculty population 0.01 0.09 0.04 0.07 0.10 0.07 Table 5 22 Correlations of academic deviance and university athletic status attributes by sporting group MFB MBA WBB General cheating Relying on others General cheating Relying on others General cheating Relying o n others Athletic status attributes Student athlete population 0.18* 0.03 0.14 0.19* 0.01 0.11 Varsity athletic teams 0.18* 0.04 0.11 0.18* 0.03 0.09 NCAA infractions 0.18* 0.04 0.11 0.18* 0.03 0.09 Athletic revenue 0 .13 0.05 0.17 0.13 0.00 0.08 Directors cup standing 0.04 0.12 0.16 0.02 0.00 0.01

PAGE 192

192 Table 5 23 OLS regression predicting general cheating academic deviance by sporting group Variable MFB MBA WBB b SE B b SE B b SE B Faculty sta tus attributes Age 0.00 0.00 0.03 0.00 0.01 0.04 0.01 0.00 0.16 Academic Rank (Associate Professor = 1) 0.38** 0.14 0.23 0.07 0.13 0.05 0.20 0.12 0.14 Academic Discipline (Engineering = 1) 0.12 0.19 0.05 0.68** 0.26 0.24 0.03 0.17 0 .02 Service involving athletics 0.13 0.13 0.10 0.16 0.12 0.12 0.06 0.10 0.05 Sports fandom 0.06 0.04 0.14 0.12** 0.04 0.30 0.02 0.03 0.07 Attendance at MBA events 0.08 0.07 0.10 0.02 0.08 0.03 0.13* 0.06 0.19 Contact with student athlet es 0.01 0.05 0.02 0.04 0.05 0.07 0.11* 0.04 0.22 University status attributes UI 0.27 0.16 0.16 0.37* 0.16 0.23 0.08 0.14 0.06 UGA 0.15 0.15 0.10 0.09 0.15 0.06 0.23 0.14 0.17 UF 0.37* 0.16 0.23 0.22 0.17 0.14 0.07 0.13 0.05 Constant 2.22*** 0.35 2.28*** 0.31 2.59*** 0.30 R square 0.16 0.23 0.20 df 10 10 10 F 2.31* 3.16** 2.98** N 137 116 134 Note. *p < .05, **p < .01, ***p < .001

PAGE 193

193 Table 5 24 Z values comparing beta coefficients by sporting group predicting general cheating MFB vs. MBA MFB vs. WBB MBA vs. WBB z z z Age 0.00 0.00 1.00 Academic Rank (Associate Professor = 1) 1.62 3.15* 1.53 Academic Discipline (Engineering = 1) 2.48* 0.59 2.09* Service involving athletics 1.64 1.16 0.64 Sports fandom 1.06 1.60 2.80* Attendance at MBA events 0.94 0.54 1.50 Contact with student athletes 0.71 1.56 2.34* UI 0.44 0.89 1.36 UGA 1.31 1.85 0.68 UF 0.64 1.46 0.70 Note. *p < .05

PAGE 194

194 Table 5 25 OLS regression predi cting relying on others academic deviance by sporting group Variable MFB MBA WBB b SE B b SE B b SE B Faculty status attributes Age 0.00 0.01 0.02 0.00 0.01 0.05 0.00 0.01 0.05 Academic rank (associate professor = 1) 0.08 0.17 0.04 0.16 0 .18 0.09 0.34 0.18 0.17 Academic rank (full professor = 1) 0.16 0.16 0.10 0.11 0.23 0.07 0.02 0.19 0.01 Academic discipline (business = 1) 0.16 0.23 0.06 0.43* 0.22 0.18 0.49 0.25 0.17 Academic discipline (other = 1) 0.24 0.25 0.08 0.18 0.2 4 0.06 0.09 0.26 0.03 Time at current institution 0.00 0.01 0.06 0.00 0.01 0.05 0.01 0.01 0.15 Sports fandom 0.10* 0.04 0.21 0.19*** 0.05 0.38 0.02 0.04 0.05 Attendance at MFB events 0.07 0.08 0.07 0.16 0.10 0.16 0.16* 0.08 0.17 Universi ty UI 0.20 0.18 0.10 0.09 0.18 0.05 0.34 0.19 0.17 UGA 0.06 0.16 0.04 0.09 0.18 0.05 0.02 0.18 0.01 UF 0.23 0.17 0.12 0.25 0.20 0.19 0.49** 0.17 0.29 Constant 3.61*** 0.38 3.04*** 0.42 2.57*** 0.40 R square 0 .14 0.23 0.16 df 11 11 11 F 2.08* 2.96** 2.25* N 149 121 138 Note. *p < .05, **p < .01, ***p < .001

PAGE 195

195 Table 5 26 Z values comparing beta coefficients by sporting group predicting relying on others MFB vs. MBA MFB vs. WBB MBA vs. WBB z z z Age 0.00 0.00 0.00 Academic rank (associate professor = 1) 0.32 1.70 1.96* Academic rank (full professor = 1) 0.18 0.72 0.44 Academic discipline (business = 1) 0.84 1.91 2.76* Academic discipline (other = 1) 1.21 0.91 0.25 Time at current inst itution 0.00 0.71 0.71 Sports fandom 1.41 2.12* 3.28* Attendance at MFB events 1.80 0.80 2.50* UI 1.14 2.06* 0.96 UGA 0.12 0.33 0.43 UF 0.08 1.08 0.91 Note. *p < .05

PAGE 196

196 Figure 5 1. Main effects of university for general ch eating academic deviance Figure 5 2. Main effects of sporting group for general cheating academic deviance

PAGE 197

197 Figure 5 3. Main effect of university for relying on others academic deviance Figure 5 4. Main effect of sport for relying on others academi c deviance

PAGE 198

198 CHAPTER 6 RESULTS PREDICTING NORMATIVE DEVIANCE Normative Deviance by University and Sporting Group Factorial ANOVA for Criminal Deviance A factorial ANOVA was conducted to compare the effect of university and sporting group and the interaction effect between university and sporting group on perceptions of criminal deviance ( Table 6 1 6 2, 6 3, and Figure 6 1). U niversity included four levels: U niversity of Illinois (U I ) University of Florida ( UF ) University of Georgia ( UGA ), and Ohio State Un iversity ( OSU) and sporting g roup consisted of three MFB ) ( MBA ) and ( WBB). The interaction effect of university and sport was not significant for criminal deviance. Sporting group was the only eff ect that was significant (F(2, 388) = 30.92, p < .001 ). Therefore, this indicates that there is a significant difference in perceptions of criminal deviance bet ween MFB (M = 1.88, SD = 0.48), MBA (M = 1.69 SD = 0.46), and WBB (M = 1.39, SD = 0. 51 ). Significant differences using Tukey HSD found faculty believe MFB student athletes were more likely to commit criminal deviance compared to MBA and WBB student athletes (Table 6 3) Additionally, faculty believed MBA student athletes are significantly more likely to commit criminal deviance compared to WBB student athletes. Factorial ANOVA for Drinking Related D eviance A factorial ANOVA was conducted to compare the effect of university and sporting group and the interaction effect between university an d sporting group on perceptions of drinking related deviance ( Table 6 4, 6 5, and 6 6 ). University included

PAGE 199

199 four levels (UI, UF, UGA and OSU) and sporting group consisted of three levels (MFB, MBA, and WBB). All effects were s ignificant, including the inte raction effect. The main effect for university yielded an F ratio of F(3, 380) = 3.46, p < .05, 0.03 indicating a significant difference between UI (M = 2.86, SD = 0.61), UF (M = 2.64, SD = 0.75), UGA (M = 2.63, SD = 0.86), and OSU (M = 2.62, SD = 0.70 ). However, the post hoc Tukey HSD test did not find any significant differences between the g roups (Table 6 5) The main effect for sporting group yielded an F ratio of F(2, 380) = 35.74, p < .001 0.16 indicating a significant difference between MFB (M =2.95 SD = 0.71), MBA (M = 2.75, SD = 0.72), and WBB (M = 2.22, SD = 0.70 ). Significant differences using Tukey HSD found faculty believe MFB student athletes were more likely to commit drinking related deviance compared to MBA and WBB student athletes (Table 6 6) Additionally, faculty believed MBA student athletes specifically are signific antly more li kely to commit drinking related deviance compared to WBB student athletes. There was also a significant interaction between the two factors, F(6, 380) = 2.35, p < .05 0.04 The interaction suggests that the effect of sporting group on perceptions of drinking related deviance is not the same at each university. The interaction plot in Figure 6 2 shows that MBA student athletes have different perceptions of drinkin g related deviance depending on the university. At OSU, UI and UGA, MBA athletes are the team with the second highest perceptions of drinking

PAGE 200

200 related deviance. However, at UF, MBA athletes have the highest perceptions of drinking related deviance. 34 Bivari ate Relationships for the Entire Sample Bivariate correlations were run for the two normative deviance scales (criminal deviance and drinking related deviance) and the independent variables reflecting individual status attributes of faculty, student athlet e attributes, university attributes, and university athletic attributes ( Table 6 7 6 8 6 9 and 6 10 ). Following the correlational analysis, OLS regres sions were run for any variable correlated at the bivariate level. Criminal Deviance There were four fa culty status a ttributes that were significantly related to criminal deviance at the bivariate level, which include academic discipline (education), sports fandom, and attendance at MFB and WBB events ( Table 6 7 ) Faculty in education were less likely to pe rceive student athletes as participants in criminal deviance ( r = 0.11, p < .05) compared to faculty in other disciplines. Faculty that were bigger fans of their university sports programs were less likely to perceive student athletes as participants in c riminal deviance (r = 0.15, p < .01) Additionally, faculty that attended more MFB and WBB events were less likely to perceive student athletes as participants in criminal deviance ( MFB: r = 0.10, p < .05; WBB: r = 0.11, p < .05). There were no universi ty status attributes significantly related to criminal devia nce at the bivariate level ( Table 6 8 ). However, there were significant relationships between the criminal deviance scale and sporting groups ass igned through 34 A MANOVA revealed t here was not a statistically significant multivariate main effect for university, F(6, 730 ) = 1.75 p < .10 7 2 = 67 There was a statistically significant main effect for sport F(4 730 ) = 21.98 80 2 = 1.00

PAGE 201

201 manipulation ( Table 6 9 ). Faculty tha t were randomly assigned to the MFB sporting group were more likely to predict criminal deviance (r = 0.32, p < .001). Alternatively, faculty that were randomly assigned the WBB sporting group were less likely to predict criminal deviance (r = 0.36, p < 001). There were no university athletic status attributes significantly related to normative devia nce at the bivariate level ( Table 6 10 ). Drinking R elated Deviance There were eight faculty status attributes significantly relate to drinking related devianc e at the bivariate level, which include age, race, academic rank (assistant professor), academic discipline (education), time at current institution, sports fandom, and attendance at M BA and WBB events ( Table 6 7 ) Older faculty were less likely to perceiv e drinking related deviance compared to younger faculty (r = 0.16, p < .001). White faculty were more likely to perceive student athlete drinking related deviance compared to non white faculty (r = 0.14, p < .01). Assistant professors were more likely to perceive student athlete drinking related deviance compared to faculty of other ranks (r = 0.10, p < .05). Faculty in education were less likely to perceive student athletes as participants in drinking related deviance (r = 0.11, p < .05) compared to facu lty in other disciplines. Faculty that were bigger fans of their university sports programs were less likely to perceive student athletes as participants in drinking related deviance (r = 0.16, p < .01). Additionally, faculty that attended more MBA and WB B events were less likely to perceive student athletes as participants in drinking related deviance ( MBA: r = 0.12, p < .05; WBB: r = 0.10, p < .05). Only one university status attribute variable was significantly related to the normative deviance scales ( Table 6 8 ). Faculty at UI were more likely to perceive

PAGE 202

202 student athletes as engag ed in drinking related deviance compared to faculty at the three other universities (r = 0.13, p < .05). There were significant relationships between the drinking related d eviance scale and sporting groups ass igned through manipulation ( Table 6 9 ). Faculty that were randomly assigned to the MFB sporting group were more likely to predict drinking related deviance compared to those assigned to the other sporting groups (r = 0. 29, p < .001). Alternatively, faculty that were randomly assigned the WBB sporting group were less likely to predict drinking related deviance compared to those assigned to the other sporting groups (r = 0.39, p < .001). Two university athletic status att ribute variables were significantly associated with the drinking related deviance s cale of normative deviance ( Table 6 10 ). Faculty at institutions with smaller athletic revenues were significantly more likely to believe student athletes engage in drinking related deviance (r = 0.11, p < .05). Also, faculty at universities with higher were more likely to believe student athletes engage in drinking related deviance (r = 0.12, p < .05). However, these variables are not added to the regression model because of concerns with multicollinearity. When added to the regression model, both of these variables have tolerance levels are below .20, meaning more than 80% of the variance of these variables is shared with another variable in the mo del (Allison, 1999). Additionally, both variables have Variance Inflation Factors (VIF) above 5, which is strong evidence for multicollinearity (Allison, 1999).

PAGE 203

203 OLS Regression M odels for the Entire S ample Predicting Criminal D eviance An OLS r egression mo del was run predicting the criminal deviance scale with academic discipline (education), sports fandom, attendance at M FB and WBB events, UI, UGA, UF, MFB and MBA sporting group s ( Table 6 11 ). The overall model predicting criminal deviance was signific ant (F = 8.85 p < .0 01, R = 0.17 ). S ports fandom remained significant after controlling for other variables Fa culty that were bigger fans of their university sports program were significantly less likely to perceive student athletes as criminally deviant ( b = 0.05 p < .05). Additionally, MFB and MBA sporting groups had a positive significant relationship with criminal deviance controlling for other variables, meaning faculty randomly assigned MFB and MBA sporting groups were significant more likely to per ceive criminal deviance compared to those randomly assigned to the WBB sporting group (MFB: r = 0.46, p < .001; MBA: r = 0.27, p < .001). Predicting D rinking Related D eviance An OLS regression model was run predicting the drinking related deviance scale wi th age, race, academic rank ( assistant professor ) academic discipline (education), time at current institution, sports fandom, attendance at M BA and WBB events, UI, UF, UGA, MFB and MBA sporting groups ( Table 6 12 ). The overall model predicting drinking r elated deviance was significant (F = 8.58 p < .001, R = 0. 23 ). Five independent variables remained significant after controlling for other faculty status attributes, which include: faculty race sports fandom UI, MFB and MBA White faculty were signific ant ly more likely to perceiv e drinking related deviance by student athletes compare d to non white faculty (b = 0.33 p < .01). F aculty that are

PAGE 204

204 bigger fans of their university sports programs were significantly less likely to perceive student athletes as d rin king related deviants (b = 0.05 p < .05 ). Faculty from UI were significantly more likely to perceive student athletes as drinking related deviants compared to faculty at OSU (b = 0.25, p < .05). Finally, faculty randomly assigned MFB or MBA sporting g roups were significant more likely to perceive drinking related deviance compared to faculty assigned WBB (MFB: b = 0.65, p < .001; MBA: b = 0.50, p < .001). Normative Deviance by Individual University Bivariate R elationsh ips for Individual U niversities Un iversity of Illinois There was only faculty status attribute significantly related to the criminal deviance scale for UI, which was attendance at MBA events ( Table 6 13 ). Faculty at UI that attended more MBA events were less likely to perceive student at hletes as criminal deviants (r = 0.22, p < .05). Additionally, there were two student athlete attribute variable s significantly related to criminal deviance WBB sport group and perceptions of Hispanic student athletes ( Table 6 14 ). UI faculty that were r andomly assigned the WBB sporting group were less likely to perceive criminal deviance compared to faculty assigned the other sporting groups (r = 0.22, p < .05). Additionally, UI faculty that estimated a higher percentage of Hispanic student athletes on their campus were more likely to believe student athletes are involved in criminal deviance (r = 0.31, p < .01). There were three faculty status attributes si gnificantly related to drinking related deviance for UI academic rank (lecturer), sports fandom, and attenda nce at MBA events ( Table 6 13 ). L ecturers at UI were less likely to perceive student a thletes as involved in drinking related deviance compared to faculty of other ranks (r = 0.22, p <

PAGE 205

205 .05). Additionally, faculty that were larger fans of UI a thletics (r = 0.25, p < .05) and attended more MBA events (r = 0.26, p < .05) were significantly less likely to perceive drinking related deviance by student athletes. Only one student athlete status attribute was significantly related to drinking relate d deviance at the bivariate le vel for UI ( Table 6 14 ). Similar to the criminal deviance scale, UI faculty that were randomly assigned to the WBB sporting group were less likely to perceive drinking related deviance compared to faculty assigned the other sp orting groups (r = 0.25, p < .05). University of Florida. There were two faculty status attributes significantly related to criminal deviance at the bivariate level for UF, facult y gender and race ( Table 6 13 ). Male (r = 0.25, p < .05) and white (r = 0.24 p < .05) faculty at UF were significantly more likely to perceive that student athletes engage in normative deviance compared to female and non white faculty. There were two student athlete status attributes significantly related to criminal deviance at the bivariate level for UF, which were MFB and WBB sport groups ( Table 6 14 ) UF faculty that were randomly assigned to the MFB sporting group were more likely to perceive criminal deviance compared to faculty assigned the other sporting groups (r = 0.45, p < .001). Alternatively, UF faculty that were randomly assigned the WBB sporting group were less likely to perceive criminal deviance compared to faculty assigned the other sporting groups (r = 0. 53, p < .001 ). There was only one faculty status attribu te si gnificantly related to drinking related deviance at the bivariate level for UF which was faculty race ( Table 6 13 ). White faculty at UF were significantly more likely to perce ive drinking related deviance by student athletes compared to non white fac ulty (r = 0.35, p < .001). Additionally, three student

PAGE 206

206 athlete attribute s were si gnificantly related to drinking relat ed d eviance at UF ( Table 6 14 ). UF faculty that were randomly assigned to the MBA sporting group were more likely to perceive drinking rel ated deviance compared to faculty assigned the other sporting groups (r = 0.28, p < .01). Alternatively, UF faculty that were assigned the WBB sporting group were less likely to perceive drinking related deviance compared to faculty assigned the other spor ting groups (r = 0.42, p < .001). Finally faculty at UF that estimated a larger percentage of black athletes on their campus were less likely to perceive student athletes as drinking related deviants (r = 0.22, p < .05 ). University of Georgia. There were three faculty status attributes significantly related to criminal deviance at UGA, which include d : academic discipline (education), academic discipline (physical sciences and mathematics), and attenda nce at MBA events ( Table 6 13 ). Faculty in the acad emic discipline of education at UGA are significantly less likely to perceive student athletes as criminal deviants compared to faculty in other disciplines (r = 0.21, p < .05). Alternatively, faculty in physical sciences and mathematics were significantl y more likely to perceive student athletes at criminal deviants compared to faculty in other disciplines (r = 0.24, p < .05). Last, faculty that attended more MBA events at UGA were significantly less like ly to perceive student athletes as criminal deviant s (r = 0.25, p < .05). There were also five student athlete status attributes significantly related to perceptions of criminal deviance for fa culty at UGA ( Table 6 14 ). First, UGA faculty that were randomly assigned to the MFB sporting group were more li kely to perceive criminal deviance compared to faculty assigned the other sporting groups (r = 0.37, p < .01). Alternatively, UGA faculty that were randomly assigned to the WBB sporting group

PAGE 207

207 were less likely to perceive criminal deviance compared to facul ty assigned the other sporting groups (r = 0.48, p < .001). Also, higher estimates of both male (r = 0.25, p < .05) and female (r = 0.22, p < .05) student athletes on UGA campus was negatively related to criminal deviance. Also, higher estimates of Asi an student athletes on the UGA campus was significantly related higher perceptions of criminal deviance (r = 0.31, p < .05). There were five faculty status attributes significantly related to drinking related deviance at the bivari ate level for UGA ( Table 6 13 ). These include d : age, academic discipline (education), time at current institution, service involving athletics, and attendance at MBA events. Older faculty (r = 0.24, p < .05) and faculty that had been at UGA longer (r = 0.20, p < .05) were less likely to perceive student athletes involved in drinking related deviance. Similar to the criminal deviance scale, faculty in the academic discipline of education at UGA were significantly less likely to perceive student athletes as involved in drinking re lated deviance compared to faculty in other disciplines (r = 0.22, p < .05). Faculty that participated in service involving athletics were significantly less likely to perceive student athletes being involved in drinking related deviance compared to facul ty that did not participate (r = 0.25, p < .05). Finally, faculty at UGA that attended more MBA events were significantly less likely to perceive student athletes as drinking related deviants (r = 0.27, p < .01). There were also four student athlete sta tus attributes significantly related to drinking relate d deviance at UGA ( Table 6 14 ). Similar to the criminal deviance scale, UGA faculty that were randomly assigned to the MFB sporting group were more likely to perceive drinking related deviance compared to faculty assigned the other sporting

PAGE 208

208 groups (r = 0.44, p < .001). Alternatively, UGA faculty that were randomly assigned to the WBB sporting group were less likely to perceive criminal deviance compared to faculty assigned the other sporting groups (r = 0.40, p < .001). Additionally, h igher estimates of both black (r = 0.22, p < .05) and Asian (r = 0.28, p < .05) student athletes on UGA campus was positively related to drinking related deviance. Ohio State University. There was one faculty status attrib ute significantly relate d to both criminal and drinking related deviance at OSU, which was sports fandom ( Table 6 13 ). Faculty that were bigger fans of OSU athletics were less likely to perceive student athletes as involved in criminal (r = 0.28, p < .01) and drinking related deviance (r = 0.29, p < .01). Additionally, faculty in the discipline of education were significantly less likely to perceive student athletes as involved in drinking related deviance compared to faculty in other disciplines (r = 0. 23, p < .05). There were also two student athlete status attributes significant at the bivariate level for both normat ive deviance scales at OSU ( Table 6 14 ). OSU faculty that were randomly assigned to the MFB sporting group were more likely to perceive c riminal and drinking related deviance compared to faculty assigned the other sporting groups (criminal: r = 0.19, p < .05; drinking related: r = 0.33, p < .001 ). Alternatively, OSU faculty that were randomly assigned to the WBB sporting group were less lik ely to perceive criminal deviance compared to faculty assigned the other sporting groups ( criminal: r = 0. 21, p < .05; drinking related: b = 0.45, p < .001). OLS Regression by U niversity and Clogg Coefficient Comparison Test Predicting criminal deviance Faculty and student athlete status attributes that were significantly associated with at least one of the university groups were entered into

PAGE 209

209 OLS regression models to predict criminal deviance 35 More specifically, OLS regression models were run for each university using gender, race, academic discipline (education and physical sciences and mathematics), sports fandom, attenda nce at MBA events and MFB and MBA ( Table 6 15 ). The overall models for UF, UGA and OSU were significant ( UF: F = 6.07, p < .001, R = 0.36; UGA: F = 6.32, p < .001 R = 0. 35 ; OSU: F = 2. 10 p < .05, R = 0.14 ) ( Table 6 15 ). For OSU, the only variable that remained significant while controlling for other variables was sports fandom (OSU: b = 0.09, p < .01), where faculty that were l arger fans of OSU athletics were less likely to perceive student athletes as criminal deviants. For UF, the only significant variable s after controlling for other faculty status attributes was the manipulation sporting groups of MFB and MBA, where faculty at UF that were assigned MFB and MBA sporting groups were significantly more likely to perceive criminal deviance compared to faculty assigned WBB (MFB: b = 0.60, p < .001; MBA: b = 0.35, p < .01). For UGA, academic discipline (physical sciences and mathem atics), MFB and MBA were significant predictors of criminal deviance controlling for other variables. Faculty in the discipline of physical sciences and mathematics at UGA were significantly more likely to perceive student athletes as criminal deviants tha n faculty in other disciplines (b = 0.50, p < .05) Additionally, faculty at UGA that were assigned MFB and MBA sporting groups were significantly more likely to perceive criminal deviance compared to faculty assigned WBB (MFB: b = 0.64, p < .001; MBA: b = 0.30, p < .05). 35 The models did not include the student athlete status attributes because there was a high f requency of missingness, which results in power issues.

PAGE 210

210 The overall model for UI was not significant (F = 1.49, p > .05, R = 0.14). One variable was still significant after controlling for other variables, which was the MFB sport group Faculty at UI that were randomly assigned the MFB sportin g group were significantly more likely to perceive criminal deviance compared to faculty assigned the WBB sporting group (b = 0.26, p < .05). The regression coefficient comparison test developed by Clogg, Petkova, and Haritou (1995) determined there are s ome significant differences concerning some variables by university group. Table 6 16 report s the z values associated with each independent variable and university First, the effect of academic discipline (physical sciences and mathematics) on perceptions of student athlete criminal deviance is significantly different for UGA compared to UI (z = 2.20, p < .05) For UGA, the coefficient is significant and positive, meaning faculty in the discipline s of physical sciences and mathematics are more likely to p erceive student athletes as criminally deviant compared to faculty in other disciplines. However, at UI, the coefficient is negative and non significant. Second the effect of sports fandom on perceptions of criminal deviance is stronger for OSU compared to UF (z = 2.12, p < .05) Last, the effect of the MFB manipulation was stronger for UGA and UF compared to UI and OSU. Predicting drinking related deviance Faculty and student athlete status attributes that were significantly associated with at least on e of the university groups were entered into OLS regression models to predict drinking related deviance. More specifically, OLS regression models were run for each university using age, race, academic rank (lecturer), academic discipline (education), time at current institution,

PAGE 211

211 service involving athletics, sports fandom, attendance at MBA events, MFB and MBA 36 ( Table 6 17 ). The overall model for UI was significant (F = 3.31 p < .01, R = 0.32) ( Table 6 17 ). For UI, race, academic rank (lecturer), and MFB remained significant after controlling for other variables. White faculty at UI are significantly more likely to perceive student athletes as engaging in drinking related deviance compared to non white faculty (b = 0. 47 p < .05). Lecturers at UI were sign ificantly less likely to pe rceive student athlete drinking related deviance compared to non lecturer faculty (b = 0.51 p < .05). Additionally, UI faculty that were randomly assigned to the MFB sporting group were significantly more likely to perceive dri nking related deviance compared to those assigned to the WBB sporting group (b = 0.45, p < .01). For UF, the overall model for was significant (F = 3.25, p < .01, R = 0.29 ) ( Table 6 17 ). Three variables remained significant after controlling for others, w hich were race, MFB and MBA. White faculty from UF were significantly more likely to perceive drinking related deviance compared to non white faculty (b = 0.91, p < .01). UF faculty that were randomly assigned to the MFB or MBA sporting groups were signifi cantly more likely to perceive drinking related deviance compared to those assigned to the WBB sporting group (MFB: b = 0.45, p < .01; MBA: b = 0.69, p < .001). For UGA, the overall model for was significant (F = 5.25, p < .001, R = 0.37) ( Table 6 17 ). On ly one variable remained significant after controlling for others, which was the MFB sporting group UGA faculty that were randomly assigned to the MFB 36 Again, I chose not to include the perception of student athlete variables because there were many cases of missing data, which decreased the power of the regression analysis.

PAGE 212

212 sporting group were significantly more likely to perceive drinking related deviance compared to those as signed to the WBB sporting group (b = 0.89, p < .001). For OSU, the overall mo del is also significant (F = 4.12, p < .001 R = 0.14 ) ( Table 6 17 ). T hree variables remained significant after controlling for other variables, which include sports fandom, MFB and MBA. F aculty at OSU that were bigger fans of the univ ersity sports program were significantly less likely to perceive student athletes as involve d in drinking related deviance (b = 0.08, p < .05). OSU faculty that were randomly assigned to the MFB o r MBA sporting groups were significantly more likely to perceive drinking related deviance compared to those assigned the WBB sporting group (MFB: b = 0.69, p < .001; MBA: b = 0.52, p < .01). The regression coefficient comparison test developed by Clogg, P etkova, and Haritou (1995) determined there are some significant differences concerning some variables by university group. Table 6 18 reports the z values associated with each variable and university The effect of academic rank ( lecturer ) on perceptions of drinking related deviance is significantly different for UI compared to UF ( z = 2.29, p < .05). For UI, the coefficient is negative and significant, meaning lecturers were less likely to perceive student athletes as drinking related deviants. However, at UF, the coefficient is positive and non significant Additionally, t he effect of attendance at MBA events on perceptions of drinking related deviance is significantly different for UI compared to UF (z = 2.16, p < .05). For UI the coefficient is nega tive, meaning faculty that attended more MBA events were less likely to perceive student athletes as drinking r elated deviants. However, at UF the coefficient for attendance at MBA events was positive.

PAGE 213

213 Normative Deviance by Student Athlete Sporting Group B ivariate Relationships by Sporting G roup MFB. There were two independent variables that were significantly related to the criminal deviance scale at the bivariate level for the MFB subgroup ( Tables 6 19 6 20 6 21 and 6 22 ). Both variables were faculty status attributes, which include d : academic discipline (physical sciences and mathematics) and sports fandom ( Table 6 19 ) Faculty in the physical sciences and mathematics disciplines were significantly more likely to perceive MFB student athletes as crimi nal deviants compared to faculty in other disciplines (r = 0.22, p < .01). Faculty that are bigger fans of their university sports program were significantly less likely to perceive MFB athletes as criminal deviants (r = 0.17, p < .05). There were four in dependent variables that were significantly related to the drinking related deviance scale at the bivariate level for the MFB subgroup ( Tables 6 19 6 20 6 21 and 6 22 ) All fou r were faculty status attributes which include: age, race, time at current i nstitution, a nd sports fandom ( Table 6 19 ). Older faculty were significantly less likely to perceive MFB student athletes as involved in drinking related deviance (r= 0.19, p < .05). White faculty were significantly more likely to perceive MFB athletes as involved in drinking related deviance compared to non white faculty (r = 0.17, p < .05). Faculty that had been at their institution longer were less likely to perceive MFB student athletes as involved in drinking related deviance (r = 0.19, p < .05). Fi nally, faculty that were bigger fans of their university sports program were significantly less likely to perceive MFB athletes as involved in drinking related deviance (r = 0.22, p < .01).

PAGE 214

214 MBA. There was only one independent variable significantly relate d to criminal deviance at the bivariate level for the MBA subgroup, which was sports fandom ( Tables 6 19 6 20 6 21 and 6 22 ) Again, this relationship is negative, where faculty that are bigger fans of their university sports program were significantly less likely to perceive MBA student athletes involved in criminal deviance (r = 0.23, p < .05) ( Table 6 19 ) There were five indep en dent variables related to drinking related deviance at the bivariate level for the MBA subgroup ( Tables 6 19 6 20 6 21 and 6 22 ). These variables include three faculty status attributes (age, time at current institution, and sports fandom), one student athlete status attribute (percentage of other race student athletes), and one university status attribute (UGA). Older fac ulty were significantly less likely to perceive MBA student athletes as involved in drinking related deviance (r= 0.20, p < .05). Faculty that had been at their institution longer were less likely to perceive MBA student athletes as involved in drinking r elated deviance (r = 0.23, p < .01). Faculty that were bigger fans of their university sports program were significantly less likely to perceive MBA athletes as involved in drinking related deviance (r = 0.28, p < .01). Faculty that estimated a higher pe rcentage of student athletes in the other race category were significantly less likely to believe MBA students were involved in drinking related deviance (r = 0.27, p < .05) Finally, f aculty at UGA were significantly less likely to perceive MBA students were involved in drinking related deviance compared to faculty at the three other institutions (r = 0.19, p < .05) WBB. There were five independent variables significantly related to criminal deviance at the bivari ate level for the WBB subgroup ( Tables 6 19 6 20 6 21 and 6 22 ). There were two faculty status attributes (academic discipline social and behavioral

PAGE 215

215 sciences and service involving athletics), one student athlete status attribute (percentage of Asian student athletes), and two university status attributes (UI and region). Faculty in the social and behavioral sciences are significantly less likely to perceive WBB athletes involved in criminal deviance compared to faculty in other disciplines (r = 0.18, p < .05). Faculty involved in service to a thletics were also significantly less likely to perceive WBB athletes involved in criminal deviance compared to faculty that were not involved in any service to athletics (r = 0.17, p < .05). Faculty that estimated a higher percentage of Asian student ath letes on their campus were significantly more likely to believe WBB students were involved in criminal deviance (r = 0.25, p < .05). Faculty from UI are significantly more likely to perceive WBB athletes in volved in criminal deviance (r = 0.18, p < .05). I n fact, faculty from the Midwest region or Big 10 conference are significantly more likely to perceive WBB athletes as involved in criminal deviance (r = 0.20, p < .05). There were also four variables that were significantly related to alcohol related dev i ance for the WBB subgroup ( Tables 6 19 6 20 6 21 and 6 22 ). There was one faculty status attribute (academic rank associate professor), one university status attribute (UI), and two university athletic status attribute s (athletic revenue and directors cup standing). Associate professors were significantly less likely to perceive WBB athletes are involved in drinking related deviance compared to faculty in other ranks (r = 0.23, p < .01). Faculty from UI were significantly more likely to perceive WBB at hletes are involved in drinking related deviance compared to the other three institutions (r = 0.27, p < .01). Faculty from universities with less athletic revenue were significantly more likely to perceive WBB student athletes are involved in alcohol rela ted deviance (r

PAGE 216

216 = ranking (meaning less overall athletic success) were significantly more likely to perceive WBB are involved in alcohol related deviance (r = 0.23, p < 05) OLS Regression by U niversity and Clogg Coefficient Comparison Test Predicting criminal deviance Independent variables that were significantly associated with at least one of the sporting groups were entered into OLS regression models to predict crimi nal deviance More specifically, OLS regression models were run for each sporting group using academic discipline (physical sciences and mathematics and social and behavioral sciences), service invo lving athletics, sports fandom, UI, UGA, and UF 37 ( Table 6 23 ). For MFB, the overall model is significant (F = 3.61 p < .05, R = 0.16 ) ( Table 6 19 ). There are three variables that remain significant after controlling for other variables, which are academic discipline (physical sciences and mathematics), sports f andom, and UF. F aculty in the physical sciences and mathematics disciplines are significantly more likely to perceive MFB athletes as criminal de viants compared to faculty in other disciplines (b = 0.51, p < .01 ). Faculty that were bigger fans of their uni versity sports programs were significantly less likely to perceive MFB student athletes as criminally deviant (b = 0.09, p < .01). Additionally, faculty at UF are significantly more likely to perceive MFB student athletes as criminal deviants compared to faculty at OSU (b = 0.23 p < .05). 37 I chose not to inclu de the one significant student athlete status attribute (perceptions of Asian student athletes) in the regression models because of missing data resulting in power issues. I also did not include region in the regression analyses even though it was signific ant at the bivariate level because I am included the university dummy variables, which would be redundant.

PAGE 217

217 For MBA, the overall model is not significant (F = 1. 68, p > .10, R = 0.10 ) ( Table 6 23 ) Although the overall model is not significant, th ere is one independent variable that remains significant controlling for other variables, which is sports fandom. Faculty that are bigger fans of their university sports programs are significantly less likely to perceive MBA students as involved in criminal d eviance (b = 0.06 p < .05). The overall model is not significant for WBB a t the .05 p value cutoff, but is does have a p value below .10 (F = 2.02, p < .10, R = 0.10 ) ( Table 6 23 ). Additionally, none of the individual variables remain significant after controlling for other variables. The regression coefficient comparison test developed by Clogg, Petkova, and Haritou (1995) determined there are some significant differences concerning some v ariables by sporting group. Table 6 24 reports the z values associated with each independent variable First, the effect of academic discipli ne (physical sciences and mathematics) was significantly different for MFB compared to WBB (z = 2.83, p < .05). For MFB, the coefficient was significant and positive, meaning faculty in the disciplines of physical sciences and mathematics were significantl y more likely to perceive MFB student athletes as criminally deviant. However, for WBB, the coefficient was non significant and negative. Second, the effect of service involving athletics on perceptions of criminal deviance was stronger for WBB than MFB (z = 2. 51 p < .05). Predicting drinking related deviance Independent variables that were significantly associated with at least one of the sporting groups were entered into OLS regression models to predict drinking related deviance 38 More specifically, OL S 38 The models did not include percentage of other race athletes, even though it was significantly related to substance related deviance at the bivaria te level because of a high frequency of missingness and lack of power Additionally, I did not include the significant university athletic status attributes because they are

PAGE 218

218 regression models were run for each sporting group using age, race, academic rank (associate professor), time at current institution, sports fandom, UI, UGA and UF. Table 6 25 shows OLS regression results predicting drinking related deviance by sporting group. For MFB, the overall model is significant (F = 3.36, p < 01 R = 0. 17 ) ( Table 6 25 ) There were two variables that remained significant after controlling for other variables, race and sports fandom. White faculty were significantly more likely to perceive MFB athletes involved in drinking related deviance compared to non white faculty (b = 0.79, p < .01). Additionally, faculty that are bigger fans of their university sports programs were significantly less likely to perceive that MFB athletes were involved in drinking related deviance (b = 0.08, p < .05). For MBA, the overall model is significant (F = 2. 74 p < 01, R = 0.16) ( Table 6 25 ). There were two variable s that remained significant after controlling for other variables, which were sports f andom and UF Again, faculty that are bigger fans of their university sports programs were significantly less likely to perceive that MBA athletes are involved in drin king related deviance (b = 0.09 p < .05). Additionally, faculty at UF were significantl y more likely to perceive drinking related deviance compared to faculty at OSU (b = 0.35, p < .05). The overall model was also significant for the WBB subgro up regression analysis (F = 2.96, p < .01, R = 0.17) ( Table 6 25 ). Two variables remained signific ant after controlling for other variables, academic rank (associate professor) and UI. Associate highly correlated with the university status attributes, which would bring redund ancy and multicollinearity issues.

PAGE 219

219 professors were significantly less likely to perceive WBB student athletes as drinking related deviants compared to f aculty in other ranks (b = 0.34, p < .05 ) Additionally, faculty UI are significantly more likely to perceive WBB student athletes as drinking related deviants compared to faculty at OSU (b = 0.53 p < .001). The regression coefficient comparison test developed by Clogg, Petkova, and Haritou (19 95) determined there are some significant differences concerning some variables b y sporting group. Table 6 26 report s the z values associated with each independent variable by sporting group First, the effect of race on perceptions of substance related de viance was stronger for MFB compared to MBA (z = 2.04, p < .05) and WBB (z = 2.34 p < .05). Secon d, the effect of academic rank associate professor on perceptions of drinking related deviance was stronger for WBB than MBA (z = 2. 03, p < .05).

PAGE 220

220 Table 6 1 Factorial ANOVA of university and sporting group for criminal deviance Source Subjects df MS F Model 18.93 11 1.72 7.16*** 0.17 University 0.50 3 0.17 0.69 0.01 Sport 14.88 2 7.44 30.92*** 0.14 University*Sport 2.67 6 0.44 1.85 0.03 Residual 93.34 388 0.24 Total 112.27 399 Note. *p < .05, **p < .01, ***p < .001 Table 6 2 Mean diffe rences and confidence intervals of perceptions of criminal deviance by university Mean Difference SE p value 95% Confidence Interval Lower Bound Upper Bound Ohio State University University of Florida 0.01 0.0 7 0.999 0.16 0.19 University of Geo rgia 0.01 0.0 7 0.997 0.19 0.16 University of Illinois 0.08 0.07 0.698 0.26 0.11 University of Florida Ohio State University 0.01 0.0 7 0.999 0.19 0.16 University of Georgia 0.02 0.0 7 0.986 0.20 0.15 University of Illinois 0.09 0.07 0.628 0. 27 0.10 University of Georgia Ohio State University 0.01 0.0 7 0.997 0.16 0.19 University of Florida 0.02 0.0 7 0.986 0.15 0.20 University of Illinois 0.06 0.07 0.816 0.25 0.12 University of Illinois Ohio State University 0.08 0.07 0.698 0.11 0.26 University of Florida 0.09 0.07 0.628 0.10 0.27 University of Georgia 0.06 0.07 0.816 0.12 0.25

PAGE 221

221 Table 6 3. Mean differences and confidence intervals of perceptions of criminal deviance by sport Mean Difference SE p value 95% Confidence Inte rval Lower Bound Upper Bound Football Baseball .19 0.06 0.004 0.05 0.34 Women's Basketball .48 0.06 0.000 0.34 0.61 Baseball Football .19 0.06 0.004 0.34 0.05 Women's Basketball .28 0.06 0.000 0.14 0.43 Women's Basketball Football .48 0.06 0.000 0.61 0.34 Baseball .28 0.06 0.000 0.43 0.14 Note. *p < .05 Table 6 4 Factorial ANOVA of university and sporting group for alcohol related deviance Source Subjects df MS F Model 44.61 11 4.06 8.99*** 0.21 University 4.68 3 1.56 3.46* 0.03 Sport 32.25 2 16.12 35.74*** 0.16 University*Sport 6.37 6 1.06 2.35* 0.04 Residual 171.41 380 0.45 Total 216.01 391 Note. *p < .05, **p < .01, ***p < .001 Table 6 5. Mean differences and confidence intervals of perceptions of drinking related deviance by university Mean Difference SE p value 95% Confidence Interval Lower Bound Upper Bound Ohio State University University of Florida 0.02 0.09 0.996 0.26 0.22 Uni versity of Georgia 0.01 0.09 0.999 0.25 0.22 University of Illinois 0.24 0.10 0.064 0.49 0.01

PAGE 222

222 Table 6 5. Continued Mean Difference SE p value 95% Confidence Interval Lower Bound Upper Bound University of Florida Ohio State University 0.02 0. 09 0.996 0.22 0.26 University of Georgia 0.01 0.10 1.000 0.24 0.25 University of Illinois 0.22 0.10 0.122 0.48 0.04 University of Georgia Ohio State University 0.01 0.09 0.999 0.22 0.25 University of Florida 0.01 0.10 1.000 0.25 0.24 Univer sity of Illinois 0.23 0.10 0.099 0.48 0.03 University of Illinois Ohio State University 0.24 0.10 0.064 0.01 0.49 University of Florida 0.22 0.10 0.122 0.04 0.48 University of Georgia 0.23 0.10 0.099 0.03 0.48 Table 6 6. Mean differences an d confidence intervals of perceptions of drinking related deviance by sport Mean Difference SE p value 95% Confidence Interval Lower Bound Upper Bound Football Baseball .20 0.08 0.043 0.01 0.39 Women's Basketball .70 0.08 0.000 0.51 0.89 Base ball Football .20 0.08 0.043 0.39 0.01 Women's Basketball .50 0.09 0.000 0.30 0.70 Women's Basketball Football .70 0.08 0.000 0.89 0.51 Baseball .50 0.09 0.000 0.70 0.30 Note. *p < .05

PAGE 223

223 Table 6 7 Correlations of normative deviance and faculty status attributes Criminal deviance Drinking related deviance Faculty status attributes Age 0.02 0.16*** Gender (Male = 1; Female = 0) 0.05 0.00 Race (White = 1; Non white = 0) 0.07 0.14** Academic Rank Lecturer = 1 0.06 0.07 Assistant Professor = 1 0.01 0.10* Associate Professor = 1 0.00 0.02 Full Professor = 1 0.03 0.05 Other = 1 0.01 0.00 Tenure status (Tenure = 1; Non tenure = 0) 0.02 0.04 Administrative position (Yes = 1; No = 0) 0.06 0.02 Acad emic Discipline Architecture = 1 0.04 Arts and Humanities = 1 0.06 0.00 Business = 1 0.01 0.06 Education = 1 0.11* 0.11* Engineering = 1 0.06 0.04 Law = 1 0.00 0.04 Life Sciences = 1 0.04 0.02 Medicine and Health Sciences = 1 0.02 0.00 Physical sciences and mathematics = 1 0.08 0.01 Social and Behavior sciences = 1 0.06 0.05 Other = 1 0.04 0.02 Time at current institution 0.05 0.12* Service involving athletics 0.08 0.04 Sports fandom 0.15** 0.16** A ttendance at MFB events 0.10* 0.10 Attendance at MBA events 0.09 0.12* Attendance at WBB events 0.11* 0.10* Contact with student athletes 0.05 0.01

PAGE 224

224 Table 6 8 Correlations of normative deviance and university status attributes Criminal deviance Drinking related deviance University status attributes University OSU 0.02 0.05 UF 0.03 0.03 UGA 0.00 0.04 UI 0.06 0.13* Region (Midwest = 1; South = 0) 0.03 0.06 Undergraduate Student Population 0.01 0.02 Faculty population 0.05 0.08 Table 6 9 Correlations of normative deviance variables and student athlete status attributes Criminal deviance Drinking related deviance Sporting group assigned MFB 0.32*** 0.29*** MBA 0.04 0.09 WBB 0.36*** 0.39*** Faculty perception of student athlete attributes Estimate % of student athlete gender Male 0.06 0.01 Female 0.05 0.01 Estimate % of student athlete race Black 0.02 0.07 White 0.00 0.02 Hispanic 0.06 0.03 As ian 0.11 0.09 Other 0.11 0.13 Estimate % of student athlete sport MFB 0.01 0.04 MBA 0.00 0.03 WBB 0.01 0.03

PAGE 225

225 Table 6 10 Correlations of normative deviance and athletic status attributes Criminal deviance Drinking relat ed deviance Athletic status attributes Student athlete population 0.03 0.06 Varsity athletic teams 0.02 0.05 NCAA infractions 0.02 0.05 Athletic revenue 0.05 0.11* Directors cup standing 0.06 0.12* Table 6 11 OLS regression pre dicting criminal deviance for the entire sample Variable b SE B Faculty status attributes Academic discipline (education) 0.14 0.09 0.07 Sports fandom 0.05* 0.02 0.14 Attendance at MFB events 0.00 0.02 0.01 Attendance at WBB events 0.03 0.04 0.04 University status attributes UI 0.02 0.07 0.02 UGA 0.02 0.07 0.02 UF 0.02 0.07 0.02 Student athlete status attributes MFB 0.46*** 0.06 0.41 MBA 0.27*** 0.06 0.23 Constant 1.67*** 0.10 R square 0.17 df 9 F 8.8 5*** N 394

PAGE 226

226 Table 6 12 OLS regression models predicting drinking related deviance for the entire sample Variable b SE B Faculty status attributes Age 0.01 0.00 0.11 Race (White = 1) 0.33** 0.11 0.14 Academic rank (Assistan t professor = 1) 0.01 0.10 0.00 Academic discipline (Education = 1) 0.20 0.13 0.07 Sports fandom 0.05* 0.02 0.10 Time at current institution 0.00 0.01 0.03 Attendance at MBA events 0.09 0.05 0.09 Attendance at WBB events 0.00 0.05 0.00 Un iversity status attributes UI 0.25* 0.10 0.14 UGA 0.02 0.09 0.01 UF 0.18 0.10 0.11 Student athlete status attributes MFB 0.65*** 0.08 0.42 MBA 0.50*** 0.09 0.31 Constant 2.60*** 0.25 R square 0.23 df 13 F 8.58 *** N 379

PAGE 227

227 Table 6 13 Correlations of normative deviance and faculty status attributes by university UI UF UGA OSU Criminal devianc e Drinking related devianc e Criminal devianc e Drinking related devianc e Criminal devianc e Drink ing related devianc e Criminal devianc e Drinking related devianc e Faculty status attributes Age 0.02 0.01 0.06 0.14 0.05 0.24* 0.03 0.08 Gender (Male = 1; Female = 0) 0.13 0.13 0.25* 0.15 0.00 0.01 0.10 0.00 Race (White = 1 ; Non white = 0) 0.00 0.20 0.24* 0.35*** 0.01 0.02 0.07 0.10 Academic Rank Lecturer = 1 0.14 0.22* 0.13 0.09 0.00 0.08 0.03 0.03 Assistant Professor = 1 0.14 0.12 0.04 0.12 0.08 0.02 0.09 0.09 Associate Professor = 1 0.08 0.14 0.00 0.05 0.01 0.02 0.04 0.15 Full Professor = 1 0.13 0.11 0.09 0.10 0.10 0.02 0.11 0.06 Other = 1 0.03 0.02 0.07 0.05 0.12 0.19 0.09 0.04 Tenure status 0.06 0.01 0.03 0.14 0.09 0.04 0.09 0.05 Administrative position 0.16 0.07 0.03 0.07 0.16 0.02 0.04 0.03 Academic Discipline Architecture = 1 0.07 0.04 Arts and Humanities = 1 0.02 0.19 0.11 0.01 0.05 0.06 0.14 0.01 Business = 1 0.11 0.00 0.05 0.16 0.04 0.12 0.0 2 0.03 Education = 1 0.03 0.07 0.08 0.08 0.21* 0.22* 0.17 0.23* Engineering = 1 0.03 0.07 0.07 0.09 0.03 0.05 0.11 0.03 Law = 1 0.08 0.20 Life Sciences = 1 0.04 0.05 0.11 0.00 0.12 0.04 0.05 0.02 Medicine a nd Health Sciences = 1 0.13 0.06 0.05 0.03 0.16 0.01 0.01 0.03 Physical sciences and mathematics = 1 0.02 0.20 0.11 0.00 0.24* 0.07 0.07 0.06 Social and Behavior sciences = 1 0.12 0.05 0.16 0.01 0.12 0.05 0.01 0.09 Other = 1 0.1 1 0.04 0.07 0.03 0.11 0.09 0.06 0.12 Note. *p < .05, **p < .01, ***p < .001

PAGE 228

228 Table 6 13 Continued UI UF UGA OSU Criminal devianc e Drinking related devianc e Criminal devianc e Drinking related devianc e Criminal devianc e Drinking related devianc e Criminal devianc e Drinking related devianc e Time at current institution 0.09 0.02 0.05 0.15 0.15 0.20* 0.06 0.02 Service involving athletics 0.12 0.15 0.15 0.13 0.18 0.25* 0 .03 0.06 Sports fandom 0.15 0.25* 0.02 0.13 0.11 0.06 0.28** 0.29** Attendance at MFB events 0.12 0.12 0.09 0.06 0.04 0.02 0.15 0.16 Attendance at MBA events 0.22* 0.26* 0.14 0.09 0.25* 0.27** 0.13 0.17 Attendance at WB B events 0.08 0.12 0.13 0.14 0.07 0.09 0.11 0.12 Contact with student athletes 0.15 0.01 0.03 0.06 0.01 0.03 0.02 0.10 Note. *p < .05, **p < .01, ***p < .001

PAGE 229

229 Table 6 14 Correlations of normative deviance and s tudent athlete status attributes by university UI UF UGA OSU Criminal deviance Drinking related deviance Criminal deviance Drinking related deviance Criminal deviance Drinking related deviance Criminal deviance Drinking related de viance Sporting group assigned MFB 0.13 0.16 0.45*** 0.18 0.37** 0.44*** 0.19* 0.33*** MBA 0.09 0.08 0.11 0.28** 0.12 0.07 0.02 0.09 WBB 0.22* 0.25* 0.53*** 0.42*** 0.48*** 0.40*** 0.21* 0.45*** Faculty perceptions of s tudent athlete status attributes Estimate % of student athlete gender Male 0.03 0.15 0.04 0.13 0.25* 0.02 0.01 0.01 Female 0.06 0.10 0.02 0.05 0.22* 0.09 0.02 0.04 Estimate % of student athlete race Black 0.07 0.18 0.12 0.22* 0.19 0.22* 0.04 0.15 White 0.03 0.02 0.09 0.13 0.12 0.01 0.03 0.00 Hispanic 0.31** 0.21 0.10 0.19 0.09 0.11 0.03 0.01 Asian 0.21 0.12 0.16 0.06 0.31* 0.28* 0.14 0.12 Other 0.01 0.18 0.04 0. 02 0.03 0.03 0.23 0.24 Estimate % of student athlete sport MFB 0.06 0.02 0.07 0.11 0.03 0.03 0.05 0.02 MBA 0.07 0.02 0.02 0.19 0.06 0.08 0.09 0.03 WBB 0.19 0.04 0.03 0.14 0.05 0.05 0.10 0.03 Note. p < .05, **p < .01, ***p < .001

PAGE 230

230 Table 6 15 OLS regression predicting criminal deviance by university Variable UI UF UGA OSU b SE B b SE B b SE B b SE B Faculty status attributes Gender (Male = 1; Female = 0) 0.09 0.10 0.10 0.18 0.10 0.17 0.06 0.08 0.07 0.08 0.09 0.09 Race (White = 1; Non white = 0) 0.02 0.14 0.01 0.31 0.20 0.14 0.02 0.16 0.01 0.01 0.17 0.01 Academic Discipline (Education = 1) 0.09 0.21 0.05 0.02 0.17 0.01 0.28 0.19 0.13 0.17 0.20 0.08 Academic Disci pline (Physical sciences and mathematics = 1) 0.06 0.17 0.04 0.16 0.21 0.07 0.50* 0.19 0.23 0.19 0.19 0.10 Sports fandom 0.03 0.04 0.09 0.00 0.03 0.01 0.04 0.04 0.10 0.09** 0.03 0.28 Attendance at MBA events 0.12 0.07 0.21 0.03 0.05 0.06 0. 06 0.07 0.07 0.02 0.10 0.02 Student athlete status attributes MFB 0.26* 0.12 0.29 0.60*** 0.10 0.56 0.64*** 0.12 0.56 0.23 0.14 0.19 MBA 0.21 0.13 0.21 0.35** 0.12 0.30 0.30* 0.13 0.24 0.11 0.14 0.09 Constant 1.87*** 0.2 1 0.92*** 0.24 1.49*** 0.24 1.90*** 0.25 R square 0.14 0.36 0.35 0.14 df 8 8 8 8 F 1.49 6.07*** 6.32*** 2.10* N 84 94 102 109 Note. *p < .05, **p < .01, ***p < .001

PAGE 231

231 Table 6 16 Z values comparing beta coefficients predicti ng criminal deviance by university UI v. UF UI v. UGA UI v. OSU UF v. UGA UF v. OSU UGA v. OSU z z z z z z Gender (Male = 1; Female = 0) 1.91 1.17 1.26 0.94 0.74 0.17 Race (White = 1; Non white = 0) 1.35 0.19 0.05 1.13 1.22 0.13 Academic Di scipline (Education = 1) 0.26 1.31 0.90 1.18 0.72 0.40 Academic Discipline (Physical sciences and mathematics = 1) 0.81 2.20* 0.98 1.20 0.11 1.15 Sports fandom 0.35 0.18 1.20 0.80 2.12* 1.00 Attendance at MBA events 1.74 0.61 0.82 1.05 0.45 0 .33 MFB 2.18* 2.24* 0.16 0.26 2.15* 2.22* MBA 0.79 0.49 0.52 0.28 1.30 0.99 Note. *p < .05

PAGE 232

232 Table 6 17 OLS regression predicting drinking related deviance by university Variable UI UF UGA OSU b SE B b SE B b SE B b SE B Facul ty status attributes Age 0.01 0.01 0.09 0.00 0.01 0.05 0.02 0.01 0.24 0.00 0.01 0.06 Race (White = 1; Non white = 0) 0.47* 0.18 0.25 0.91** 0.33 0.28 0.15 0.25 0.05 0.17 0.19 0.08 Academic Rank (Lecturer = 1) 0.51* 0.23 0.24 0.25 0.24 0.11 0.13 0.20 0.06 0.06 0.18 0.03 Academic Discipline (Education = 1) 0.15 0.25 0.06 0.26 0.26 0.10 0.17 0.31 0.05 0.21 0.23 0.08 Time at current institution 0.01 0.01 0.15 0.01 0.01 0.13 0.01 0.01 0.09 0.00 0.01 0.01 Service involv ing athletics 0.23 0.12 0.19 0.11 0.15 0.07 0.17 0.17 0.09 0.17 0.14 0.11 Sports fandom 0.09 0.05 0.22 0.09 0.05 0.19 0.06 0.06 0.09 0.08* 0.04 0.21 Attendance at MBA events 0.15 0.08 0.20 0.08 0.07 0.11 0.17 0.12 0.14 0.10 0.11 0.09 S tudent athlete status attributes MFB 0.45** 0.15 0.36 0.45** 0.16 0.29 0.89*** 0.19 0.51 0.69*** 0.18 0.46 MBA 0.28 0.15 0.22 0.69*** 0.18 0.41 0.38 0.20 0.20 0.52** 0.16 0.37 Constant 2.81*** 0.43 2.10*** 0.54 3.07*** 0.5 1 2.63*** 0.39 R square 0.32 0.29 0.37 0.30 df 10 10 10 10 F 3.31** 3.25** 5.25*** 4.12*** N 82 91 99 105 Note. *p < .05, **p < .01, ***p < .001

PAGE 233

233 Table 6 18 Z values comparing beta coefficients predicting drinking related dev iance by university UI v. UF UI v. UGA UI v. OSU UF v. UGA UF v. OSU UGA v. OSU z z z z z z Age 0.71 0.71 0.71 1.41 0.00 1.41 Race (White = 1; Non white = 0) 1.17 1.04 1.15 1.84 1.94 0.06 Academic Rank (Lecturer = 1) 2.29* 1.25 1.95 1.22 0 .63 0.71 Academic Discipline (Education = 1) 1.14 0.80 1.06 0.22 0.14 0.10 Time at current institution 1.41 1.41 0.71 0.00 0.71 0.71 Service involving athletics 1.77 1.92 0.33 0.26 1.36 1.54 Sports fandom 0.00 1.92 0.16 1.92 0.16 1.94 Atten dance at MBA events 2.16* 0.14 0.37 1.80 1.38 0.43 MFB 0.00 1.82 1.02 1.77 1.00 0.76 MBA 1.75 0.40 1.09 1.25 0.71 0.55 Note. *p < .05

PAGE 234

234 Table 6 19 Correlations of normative deviance and faculty status attributes by sporting g roup MFB MBA WBB Criminal deviance Drinking related deviance Criminal deviance Drinking related deviance Criminal deviance Drinking related deviance Faculty status attributes Age 0.15 0.19* 0.11 0.20* 0.12 0.13 Gen der (Male = 1; Female = 0) 0.03 0.11 0.01 0.02 0.14 0.14 Race (White= 1) 0.13 0.17* 0.07 0.13 0.09 0.12 Academic Rank Lecturer = 1 0.08 0.10 0.13 0.14 0.01 0.01 Assistant Professor = 1 0.07 0.14 0.13 0.16 0.01 0.05 A ssociate Professor = 1 0.02 0.12 0.08 0.11 0.12 0.23** Full Professor = 1 0.06 0.13 0.07 0.12 0.12 0.12 Other = 1 0.08 0.01 0.13 0.08 0.04 0.06 Tenure status (Tenure = 1; Non tenure = 0) 0.07 0.06 0.03 0.04 0.02 0.07 Administr ative position (Yes = 1; No = 0) 0.03 0.04 0.15 0.02 0.03 0.12 Academic Discipline Architecture = 1 0.00 . 0.10 Arts and Humanities = 1 0.07 0.07 0.07 0.03 0.09 0.03 Business = 1 0.06 0.11 0.05 0.08 0.06 0.00 Educ ation = 1 0.02 0.03 0.03 0.04 0.11 0.11 Engineering = 1 0.07 0.07 0.12 0.16 0.02 0.06 Law = 1 0.10 0.09 0.06 0.09 . Life Sciences = 1 0.01 0.03 0.08 0.07 0.14 0.09 Medicine and Health Sciences = 1 0.05 0.06 0.03 0.02 0.02 0.04 Physical sciences and mathematics = 1 0.22** 0.01 0.08 0.00 0.06 0.03 Social and Behavior sciences = 1 0.11 0.05 0.03 0.11 0.18* 0.04 Other = 1 0.01 0.06 0.03 0.08 0.06 0.05 Note. *p < .05, **p < .01, ***p < .001

PAGE 235

235 Tab le 6 19 Continued MFB MBA WBB Criminal deviance Drinking related deviance Criminal deviance Drinking related deviance Criminal deviance Drinking related deviance Time at current institution 0.14 0.19* 0.11 0.23** 0.05 0.03 Service involving athletics 0.05 0.09 0.03 0.04 0.17* 0.04 Sports fandom 0.17* 0.22** 0.23* 0.28** 0.10 0.02 Attendance at MFB events 0.13 0.13 0.13 0.12 0.05 0.02 Attendance at MBA events 0.13 0.16 0.02 0.06 0.10 0.10 At tendance at WBB events 0.00 0.04 0.08 0.10 0.14 0.06 Contact with student athletes 0.01 0.13 0.05 0.09 0.13 0.09 Note. *p < .05, **p < .01, ***p < .001

PAGE 236

236 Table 6 20. Correlations of normative deviance and student athlet e status attributes by sporting group MFB MBA WBB Criminal deviance Drinking related deviance Criminal deviance Drinking related deviance Criminal deviance Drinking related deviance Student athlete status attributes Gender Male 0.10 0.04 0.03 0.07 0.09 0.07 Female 0.12 0.06 0.02 0.06 0.05 0.01 Race Black 0.13 0.16 0.01 0.07 0.04 0.01 White 0.02 0.04 0.04 0.18 0.06 0.03 Hispanic 0.05 0.02 0.10 0.00 0.01 0.13 Asian 0.03 0.11 0.12 0.04 0.25* 0.16 Other 0.00 0.15 0.23 0.27* 0.01 0.20 Sport MFB 0.10 0.04 0.02 0.00 0.01 0.16 MBA 0.03 0.01 0.13 0.06 0.05 0.07 WBB 0.09 0.08 0.06 0.04 0.02 0.04

PAGE 237

237 Table 6 21 Correlations of normative deviance and university status attributes by sporting group MFB MBA WBB Criminal deviance Drinking related deviance Criminal deviance Drinking related deviance Criminal deviance Drinking related deviance University status attributes University OSU 0.09 0.03 0.04 0.08 0.07 0.13 UF 0.09 0.11 0.04 0.16 0.12 0.00 UGA 0.09 0.12 0.09 0.19* 0.11 0.13 UI 0.09 0.02 0.10 0.12 0.18* 0.27** Region (Midwest = 1; South = 0) 0.16 0.02 0.04 0.03 0.20* 0.11 Undergraduate Student Population 0.011 0.07 0.00 0.01 0.10 0.05 Faculty population 0.04 0.12 0.01 0.07 0.10 0.12 Table 6 22 Correlations of normative deviance and university athletic status attribute s by sporting group MFB MBA WBB Criminal deviance Alcohol related deviance Criminal deviance Alcohol related deviance Criminal deviance Alcohol related deviance Athletic status attributes Student athlete population 0.08 0.02 0.06 0.10 0.04 0.17 Varsity athletic teams 0.09 0.03 0.04 0.08 0.07 0.13 NCAA infractions 0.09 0.03 0.04 0.08 0.07 0.13 Athletic revenue 0.00 0.05 0.07 0.09 0.07 0.24** Directors cup standing 0.07 0.07 0.07 0.05 0.16 0.23*

PAGE 238

238 Table 6 23 OLS regression predicting criminal deviance by sporting group Variable MFB MBA WBB b SE B b SE B b SE B Faculty status attributes Academic discipline (physical sciences and mathematics) 0.51** 0.16 0.25 0.21 0.14 0.14 0.15 0.17 0.08 Academic discipline (social and behavior al sciences) 0.17 0.10 0.14 0.10 0.11 0.08 0.21 0.13 0.14 Service involving athletics 0.14 0.09 0.14 0.07 0.09 0.07 0.18 0.09 0.17 Sports fandom 0.09** 0.03 0.26 0.06* 0. 03 0.21 0.02 0.03 0.07 University status attributes UI 0.01 0.11 0.01 0.11 0.12 0.10 0.13 0.14 0.10 UGA 0.21 0.11 0.20 0.04 0.11 0.03 0.11 0.13 0.09 UF 0.23* 0.12 0.20 0.10 0.12 0.09 0.10 0.12 0.08 Constant 2.07*** 0 .14 1.94*** 0.15 1.66*** 0.16 R square 0.16 0.10 0.10 df 7 7 7 F 3.61** 1.68 2.02 N 140 120 132 Note. *p < .05, **p < .01, ***p < .001

PAGE 239

239 Table 6 24 Z values comparing beta coefficients predicting criminal deviance by sportin g group MFB vs. MBA MFB vs. WBB MBA vs. WBB z z z Academic discipline (physical sciences and mathematics) 1.41 2.83* 1.63 Academic discipline (social and behavior sciences) 0.47 0.24 0.65 Service involving athletics 1.65 2.51* 0.86 Sports fando m 0.71 1.65 0.94 UI 0.74 0.79 0.11 UGA 1.61 1.88 0.41 UF 0.77 1.94 1.18 Note. *p < .05

PAGE 240

240 Table 6 25 OLS regression predicting drinking related deviance by sporting group Variable MFB MBA WBB b SE B b SE B b SE B Faculty stat us attributes Age 0.01 0.01 0.15 0.00 0.01 0.02 0.01 0.01 0.15 Race (white = 1) 0.79** 0.22 0.30 0.21 0.18 0.10 0.11 0.19 0.05 Academic rank (associate professor = 1) 0.06 0.14 0.04 0.09 0.15 0.06 0.34* 0.15 0.23 Time at current inst itution 0.01 0.01 0.09 0.01 0.01 0.19 0.00 0.01 0.02 Sports fandom 0.08* 0.04 0.18 0.09* 0.04 0.20 0.02 0.04 0.04 University status attributes UI 0.06 0.16 0.04 0.18 0.16 0.11 0.53*** 0.19 0.31 UGA 0.18 0.14 0.13 0.09 0.16 0.06 0 .07 0.18 0.04 UF 0.07 0.15 0.05 0.35* 0.18 0.20 0.18 0.17 0.12 Constant 2.99*** 0.35 2.97*** 0.42 2.48*** 0.41 R square 0.17 0.16 0.17 df 8 8 8 F 3.36** 2.74** 2.96** N 139 121 122 Note. *p < .05, **p < .01, ***p < .0 01

PAGE 241

241 Table 6 26 Z values comparing beta coefficients predicting drinking related deviance by sporting group MFB vs. MBA MFB vs. WBB MBA vs. WBB z z z Age 0.71 0.00 0.71 Race (white = 1) 2.04* 2.34* 0.38 Academic rank (associate professo r = 1) 0.15 1.95 2.03* Time at current institution 0.00 0.71 0.71 Sports fandom 0.18 1.77 1.94 UI 0.53 1.89 1.41 UGA 1.27 1.10 0.08 UF 1.79 1.10 0.69 Note. *p < .05

PAGE 242

242 Figure 6 1. Main effect of sport for criminal deviance Figure 6 2. Interaction effect of university and sport for alcohol related deviance

PAGE 243

243 CHAPTER 7 DISCUSSION This research exa mined how individual and university status attributes as w ell as characteristics of perceived athletes (e.g., gender, race, and sport) affect fac ulty perceptions of both academic and normative deviance Tables 7 1, 7 2, and 7 3 summarize the overall findings of this study Using data from an online survey from faculty at four institutions, results suggest that faculty perceptions of student athlete academic and normative deviance is low overall However, some group differences show important and statistically significant relationships. The following highlights the key findings and discusses implications for theory and practice. Revisiting the Hypot heses Intergroup Contact When looking at results for the entire sample, the variable that was consistent ly a predictor for all four deviance scales was sports fandom. F aculty who were bigger fans of their university sports programs were significantly less likely to perceive both academic and normative deviance by student athletes. T his shows support for the intergroup contact hypothesis where the more contact different groups of people have with each other the less prejudice and better social relations the re will be between the groups (Allport, 1954). Faculty that are fans have opportunity to see student athletes in a positive light through their sport, which lessen s the perception of student athlete deviance Although faculty generally reported low levels of fandom, another possible explanation for fandom being a consistent predictor of lower perceptions of academic and normative de viance may be the idea of sport hero worship This is where fans are

PAGE 244

244 loyal followers of individual athletes based on the positi ve marketing of the athlete (Schwarz, 1973; Crepeau, 1981) The sports industry (i.e., NCAA or the individual universities) m arkets athletes to identify with fans and increase their brand 2006). According to Schwarz (1973) through marketing and public relations of the sports industry, the athlete becomes a symbol of success for the group in which they belong to not in other areas of life, like the classroom or the communi ty Therefore, sport fans tend to place athletes on a pedestal which may lead them to think athletes are less deviant, compared to someone who is not a consumer or fan of athletics. There were also some universi ty and sport differences for the effect of fandom on the deviance variables. Sports fandom had a stronger positive effect in explaining perceptions of academic deviance for the two basketball. This difference between the sporting groups is not surprising because wome female and non revenue producing sport at each of the universities and has the least fan support and media attention generally of the three groups (Adams & Tuggle, 2004). Therefore, faculty could be fans of some of the university sports revenue producing sports. Additionally, f different from fans of basketball, in terms of their commitment and identity salience. Lavarie and A rnett (2000) argue that sport fan salient identities are an important factor in explaining fan related behavior, like attendance and involvement at games. For faculty, this may also extend to fan behavior in the classroom. Therefore, a faculty member that is a committed fan of the football team may not perceive academic

PAGE 245

245 deviance among football players in their courses because they feel connected to the may not feel any attach team. athletes from the other two males sports. For example, one faculty member emailed me after taking the survey sa I typically do not know in what sports students know what sport a student plays if I recognize their name from my knowledge of the different teams. Additi onally, sports fandom had a stronger effect on explaining perceptions of criminal deviance for facu lty at Ohio State compared to University of Florida Again, it could be that faculty at OSU are more connected and identify with their university sports team s compared to faculty at UF. OSU consistently rates high on the college football fan index (Harris 2016). With these high levels of fandom, perhaps faculty are more likely to worship ath letes, which may lead them to think their student athletes are less deviant, compared to faculty at UF. Alternatively, faculty at UF may not be as connected to the ir university sports program due to some high profile incidents of crime. One example is Aar on Hernandez, the former Gator Football All American, who was convicted of first degree murder and sentenced to life in prison without parole (Thompson & Romero, 2015). Once the news of Hernandez being suspect ed for murder broke there were also several re ports of Hernandez being investigated for crimes that occurred during his time

PAGE 246

246 at UF (Thompson & Romero, 2015). Additionally, a n ESPN report claimed from 2009 to 2014 that Florida had the most athletes as crime suspects and repeat offenders compared to 10 (Lavigne, 2015 a ). Faculty at UF could read these reports and it impact their perceptions of criminal deviance, regardless of their fandom. There were also other variables significantly related to deviance that supports the intergroup contact hypothesis (Allport, 1954). events was negatively related to perceptions of general cheating for the entire sample Therefore, faculty who the last year were less likely to perceive student athletes as general cheaters Again, the intergroup contact hypothesis could be one explanation for this relationships, where faculty who watched student at hletes at their sporting events, were able to see student athletes in a positive light, which decreases misconceptions about them. Adding to that explanation for th is relationship is that baseball has a different eligibility system than other college sports That is, college baseball players, have the op portunity to play professionally right out of high school, where athletes in other sports (football and basketball ) do not have that option. If faculty are aware of this difference, it may lead faculty to believe that baseball players are more serious stud ents that want to complete their degree because they chose to come to college rather than play professionally right out of high school (Billings, 2012) Therefore, faculty who attend their baseball events, might be remi nded of the fact that these students chose college over the professional route when they think of these student athletes.

PAGE 247

247 It is important to note that there were some university differences with the relationship between attendance at baseball events and deviance. A ttendance at baseball events was an important predictor of perceptions that student athletes rely on others in terms of academic deviance at UGA and drinking related deviance at UI where the more faculty attended men s baseball events, the less likely they were to believe the student athletes were deviant Both of these relationships at UGA and UI were significantly different from UF, which had a positive but non significant relationship with deviance and attendance. One reason for this difference may be that in the last year before the survey was administered UF had two baseball players arrested for trespassing and climbing a crane on a construction site late at night, which was widely publicized (Thompson, 2015). Faculty at UF may have remembered this event, which may have l ed them to perceive more deviance by baseball student athletes compared to faculty at UI and UGA. Another variable that is related to the academic deviance scale of general cheating that could be explained by the intergroup hypothesis is participation in s ervice involving athletics. Service had a bigger effect explaining perceptions of general cheating at UGA compared to UI and OSU. Therefore, my hypothesis that faculty involved in service to athletics will have lower perceptions of student athlete deviance than faculty not involved in service to athletics was true for faculty at UGA talking about general cheating only Therefore, UGA may have some faculty service programming that is making a difference in perceptions of academic deviance among their student athletes When looking at the text responses that faculty specified for service, no faculty at UGA listed filling out academic progress reports as their service. However, it was the

PAGE 248

248 second highest response from the other three universities. Perhaps fillin g out reports on student athletes draws faculty attention to the negative performance in the class, which may lead them to think they are more likely to commit academic deviance. Although none of the academic rank variables remained significant in overall models for deviance, it was an important predictor by university and sporting group. I hypothesized that faculty with higher academic ranks ( associate professors and full professors ) would have lower perceptions of deviance than faculty of lower ranks ( le cturers and assistant professors ). This hypothesis was not supported for one university, UI. Faculty at UI who were lecturers were significantly less likely to perceive drinking related deviance among athletes compared to faculty who were not. The hypothes is was somewhat supported by sporting group. A ssociate professors were less likely to perceive drinking related deviance for WBB athletes compared to MBA student athletes. Alternatively, associate professors were more likely to perceive general cheating fo Therefore rank did not matter as much as the sporting group, which in this case was a female group versus a male group. That is, for associate professors by sporting group women were perceived to be less d eviant than men. Academic discipline is another variable that was important by university that supports the intergroup contact hypothesis. I hypothesized that faculty affiliated with STEM disciplines would have increased perceptions of student athlete dev iance compared to faculty in other disciplines. Student athletes are more likely to be overrepresented in non STEM majors (Fountain & Finley, 2009). Therefore, faculty in the STEM disciplines would have less intergroup contact increasing their perceptions of

PAGE 249

249 deviance. This hypothesis was supported only at UGA, where faculty in the disciplines of physical sciences and math were more likely to perceive student athlete criminal deviance compared to faculty in other disciplines. Sporting Group Faculty were als o significantly less likely to perceive student athletes as deviant compared to athletes, which supports one of my hypotheses This was found for the each of the overall sample models predicting both a cademic and normative deviance Again, w a non revenue producing sport that receives the least fan support and media attention of the three groups (Adams & Tuggle, 2004). These results support other academic research, where faculty hold more prejudicial attitudes towards male athletes compared to female athletes (Engstrom et al., 1995). Additionally, a n undisputed fact in criminology is that males commit more crime than females female sport (Lauritsen et al. 2009). Added to that is the public pe rception that most crime is committed by black males, not females (Drummond, 1990; Barlow, 1998; Russell, 2002). Therefore, these results are similar to the reality of deviance in general as well as perceptions more b roadly. Specifically, males are seen as more deviant than than baseball, which could explain the greater perceptions of deviance among this group (Billings, 2012) T depending on the type of deviance. First, the effect of the MFB manipulation on perceptions of general cheating was stronger for UI compared to OSU. Second, this effect was stronger on criminal deviance for UGA and UF compared to UI and OSU.

PAGE 250

250 There may be some contextual factors at the universities that I am unaware of affecting this relationship. SEC football is routinely called out in the media for arrest data, which may exp lain why for UGA and UF faculty perceive as more criminal ly deviant compared to faculty at the Big Ten schools of OSU and UI ( Cooper, 2015; Lavigne, 2015 b ). Faculty may see or hear about th ese articles impacting their perceptions of studen t athlete deviance For example, one faculty member at UI Our football team is pretty bad on the field; but the student athletes have a very high graduation rate and almost never get into trouble that finds its wa y to the news or social media. As I was thinking about your questions, my main conclusion is that our football players are basically regular students our team which still looks like a group of amateurs. We embarrassed ourselves last [do not] embarrass us as a school. Additionally, contact with student athletes was also found to matter more in predicting perceptions o f baseball. That is, faculty that reported they had more interaction with student athletes generally, were less likely to perceive WBB athletes as general cheaters. However, for the MBA subgroup, th is relationship was positive and non significant Therefore, faculty that indicated having more interaction with student athletes seem to have fewer perceptions of deviance for the female sporting group compared to a male sporting group. In other words ev en for those with more contact with student athletes faculty still perceive male student athletes are more deviant than female ones

PAGE 251

251 There is also a n effect of gender in the sport seen in how the academic discipline of faculty explains student athlete d eviance. First, faculty in business were significantly more likely to perceive MBA student athletes as reliant on others in academic settings compared to faculty in other disciplines. However, for the WBB subgroup, this relationship was negative and non si gnificant. Second, the faculty in the physical sciences and mathematics disciplines were significantly more likely to perceive MFB student athletes as criminally deviant compared to faculty in other disciplines. Again, the WBB subgroup the coefficient for this variable was negative and non significant. Both of these could be considered STEM disciplines, which fits with the intergroup hypothesis that faculty in these majors may have negative perceptions of student athletes because they have less contact. How ever, again, there are differences by sporting group, where the male group has more negative perceptions than the female group. 39 Dumb Jock Another interesting finding for the entire sample was that white faculty and those who estimate d a higher percentage of black student athletes on their campus es were significantly more likely to perceive student athletes as reliant on others to commit academic deviance Therefore, faculty may have created a coding scheme schema for the dumb jock stereotype, which generally focuses on black student athlete s (Skolnick, 1993). This finding is supportive of the possibility of an implicit racial bias, which is when a person stereotypes or judges another race outside 39 There is an additional sporting effect seen in how the academic discipline of faculty impacts perceptions of student athlete deviance that does not support this hypothesis. Faculty in the engineering discipline were sig nificantly less likely to perceive MBA general cheating compared to faculty in other disciplines. The effect of engineering discipline mattered more in explaining perceptions of general cheating for MBA compared to both MFB and WBB.

PAGE 252

252 of conscious awareness or control (B anaji & Greenwald, 2016) Theref ore, even if faculty say white and non white students are equally good students, it is possible that they unintentionally associate needing to rely on others academically with non white students In this case, it was white f aculty and those that have perceptions of student athletes generally as black, that were more likely to perceive student athletes as reliant on others. T he education literature shows similar findings to this, where minority students are more at risk for ne gative expectations and implicit prejudiced attitudes by teachers compared to white students (van den Bergh et al. 2010). Additionally, this research claims these negative teacher expectancy effects c ontribute to poor student performance (van den Bergh et al. 2010). That is, c reating a self fulfilling prophecy or stereotype threat Implicit race biases are concerning because several studies have found them to contribute to the perpetuation of discrimination (McConnell & Leibold, 2001; Bertrand & Malainath an, 2004; Green et al., 2007). Although this research did not examine the effect of a stereotype threat for student athletes, it does imply that faculty may have detrimental stereotypes about specific student a thlete groups. The dumb jock stereotype of st udent athletes was also more substantial at certain universities compared to others. More specifically, f aculty at UF were significantly more likely to perceive student athletes as both general cheaters and reliant on others than faculty at OSU. Faculty at UI were also significantly more likely to perceive student athlete general cheating compared to faculty at OSU. A possible athlete population is much larger than both UF and UI, so OSU faculty have the opportunity to meet more student athletes there. Another possible explanation is that OSU hosts faculty appreciation

PAGE 253

253 events through their Student Athlete Advisory Board (SAAB), also referred to as SAAC at other universities, where UI and UF do not hav e similar events ( SAAB and SASSO Host 2014 Staff Appreciation Dinner, 2014). There may also be other unmeasured variables that could account for this difference. Race of Faculty Another interesting finding was that the race of faculty m attered in predicti ng perceptions of deviance. This was already discussed above in terms of implicit bias for the entire sample. However, there were also some differences by university and sporting group. First, the effect of being a white faculty member mattered more in exp laining perceptions that student athletes rely on others at UI compared to UGA. Additionally, the effect of being a white faculty member mattered more in predicting perceptions of drinki ng related deviance for MFB compared to MBA and WBB. O ther studies pre dicting prejudicial attitudes toward student athletes have found that white faculty members are more negative about student athletes compared to non white faculty members (Comeaux, 2011) A possible explanation presented for this relationship by Comeaux (2 011) is that non white faculty are more understanding of student athletes and the athletic subculture, where as white faculty are not. The next section discusses the patterns of non response on the outcome measure, which may have impacted the results by un iversity. Non R esponse P atterns Finally there were interesting findings of non response patterns for the academic and normative deviance items. There was strong evidence of a social desirability bias or did not respond to items at all. In general, faculty that had more fandom, involvement and experience with athletics were

PAGE 254

254 more likely to respond to the deviance items. Alternatively, faculty who had less fandom, involvement and experience were less like ly to respond to the deviance items. Additionally, several faculty reached out to me after completing the survey explaining deviance items. Although these findings show th ere may be a flaw in the design of the study, I believe social desirability bias would be an issue regardless of the design because of the sensitivity of questions, asking about deviant behavior, and the group of students I was asking about, student athlet es. Tourangeau and Yan (2007) talk about how error that arises from sensitive topics in surveys is most often a deliberate process by respondents rather than design flaws. The authors also claim that social desirability is contextual, meaning having to do with the situation of the respondent and the privacy offered by revealing sensitive information. One response I received from a respondent Student athletes have a very poor record of performance in my courses, and I am not willing to speak freely on th for this person, the context of the survey was not perceived as safe. This could be because I am employed by the athletic association at UF and a subject could identify me as such. For example, one respon dent looked me up on LinkedIn and messaged me that he participated in my survey. My Linkedin profile shows that I currently work for the athletic association, which could have made respondents feel unsafe taking part in my survey or answering items truthfu lly. In addition, the survey respondents were faculty, meaning that they were likely cognizant of scientific methods as well as the importan ce of treating students fairly (Smith, 1990) and possibly therefore more careful about responses than someone in

PAGE 255

255 t he general public might be. Still I believe more research needs to be done on this topic to increase more candid responses. Limitations Although the current study is a step forward in understanding the how deviant labels form for student athletes on coll ege campuses, t he study has several limitations. F irst, data were gathered using an online survey While internet surveys provide flexibility, timeliness, convenience, and a low cost of administration, there are also weaknesses to their use ( Evans & Mathur 2005 ). First, internet surveys could easily be perceived as junk mail to respondents, which may be an explanation for why the response rate for this survey was low. Second, respondents may have had concerns about their privacy, even though responses were anonymous. For example, one if you are going to ask questions about department, that information coupled with information about the university (which I presume you know based on the sampling fra me) and basic demographics (age, sex) would enable fairly easy identification of participants. Have you considered what level of aggregation of the findings you report would be necessary to prevent this? Faculty may have been concerned about their respons e anonymity and either not responded truthfully or participate in the survey at all. Finally the survey had many items with forced response options Respondents may not feel like their attitudes and perceptions of student athletes were accurately reflecte d by these answer options. Future research on this topic should consider inter viewing faculty to get their responses to the se issues and questions. Another limitation is the low number of grouping variables sampled, which only included four universities. I was not able to use aggregate data like university status

PAGE 256

256 attribute variables, in the regression models because of multi collinearity issues For example, th e athletic revenue variable was highly correlated with the individual university variables. This is because all faculty at each university have the same level of athletic revenue, so there is very strong linear relationship between the variables. This is common issue in aggregate data (Allison, 1999). In addition because the number of groups were so small I could not use hierarchical methods to account for aggregate data, which most require at least 35 groups (Bryk & Raudenbush, 1992 ) Additionally, the NCAA presented research at the most recent Faculty Athletics Representative Association (FARA) Con ference on perceptions of college sports amongst faculty and staff, college affinity, students, and general population and found that faculty and staf f had the lowest perceptions regarding the opportunities provided for student athletes, prioritization of student athlete well being, commitment to academics, commitment to fairness, and overall opinion of the NCAA (Dunham & Williams, 2016). Although this research did not look at deviance specifically, it does show that faculty perceptions of student athletes and the NCAA are a concern across the nation and need to be improved. The number of sporting groups asked about is another limitation of the survey. Faculty were randomly assigned only one of three sporting groups about which to answer questions This app roach was used to prevent an y ordering effect s of asking about more than one and to keep the survey length reasonable However, some faculty indicated to me in email after taking the survey that they wanted to comment on other student athlete groups and we re concerned about the one they were randomly assigned

PAGE 257

257 B y only including three groups in the study, there may be different results for other sporting groups that I am missing. Some other limitation s of the study have to do with the outcome variable First, the outcome variables had low means and variances, which resulted in a floor effect, where it was hard to find differences between groups. In addition, for the outcome measure of deviance I did not i I answered a whole slew of questions near the survey's end as "prefer not to answer" -but only because there was no "don't know" option. I do not know enough student athletes to have the vaguest idea about their morals or their behavior. This may explain some of the reason for the high percentage of missingness for the depend ent variables items. Participants may have selected a have made the data more valid if they d id not actually know, but also may have made it harder to analyze the con struct of interest by essentially increasing missing data Finally a possible limitation of the study was my affiliation with University of here were several indications from facult y and other staff that they had looked up my status with the athletic department For example, my supervisor at the athletic department contacted me about the purpose of my study the other institutions reac hed out to him with concerns show that there was some stress felt by faculty and administrator in athletics about the topic of my study, which I believe show the importance of studying the topic. However,

PAGE 258

258 other faculty may h ave felt concern with my affiliation in athletics and chose not to participate due to their concerns Implications for Theory This study is couched in the labeling perspective. However, I only focused on one aspect of the theory which was the differenti a l enforcement /status characteristics hypothesis. According to labeling perspective theorists it is crucial to understand the creation of the label to understand deviance ( Becker, 1963). The result s of this study show that there are status attributes of pe ople (in this case faculty) that matter in the labeling process of student athletes as deviant s Tittle (1980) argues that social characteristics should be the most important factor in determining the outcomes of deviant labels, more important than the act ual rule breaking. The status attributes that were most consistently related to perceptions of deviance can be explained by the idea of intergroup contact in social psychology, which is the more contact different groups of people have with each other the l ess prejudice and better social relations there will be between the groups (Allport, 1954). As applied to this study, opportunities for faculty and student athletes to have more contact (i.e., fandom, attendance at sporting events, and participating in ser vice involving athletics) lowered perceptions of student athlete deviance. However, the labeling perceptions (Gibbs, 1966, p. 50). The core of the labeling perspective is that negati ve aspects of societal perceptions lead to more negative behavior. Although this study did not examine the effect of the se perceptions on student behavior it did find for this sample there are aspects that decrease per ceptions of deviance by fac ulty. Therefore, labeling may

PAGE 259

259 not be the best perspective to explain the relationship between perceptions and behavior. Instead other social psychological frameworks may better explain faculty perceptions of student athlete deviance, like the intergroup contact hypothesis or attitude accessibility. Intergroup contact has to do with increasing contact between groups to foster positive attitudes (Allport, 1954). However, faculty may not always have direct contact with student athletes, but still have attitu des about them based on other experiences. A ttitude accessibility is the ease with which an attitude is activated from memory when seeing or hearing about an object (Fazio, 1995). In this case, the object would be student athletes. It may be that attitudes about certain student athlete groups come to mind faster than other student athlete groups which influences faculty attitudes That is, stereotypes about groups may come to mind first. Therefore, more research should be done using social cognitive attitu de frameworks to explain perceptions of student athletes. Although status attributes are influential and may have some effect on labeling outcomes, they are not the only factors that attribute to a deviant label (Bernstein et al., 1977). There may be ot her factors that were not accounted for in this study. One factor that may have had an effect on perceptions of deviance is actual student athlete deviance. Perhaps faculty have these perceptions of student athlete deviance because that is a reflection of their actual deviance at their university. The media is another important source of images of deviance that is not accounted for in this study and could influence faculty labels of student athlete deviance either positively or negatively

PAGE 260

260 ( Hawkins & Tiedema n, 1975). Therefore, this research could be expanded by examining how these factors impact perceptions of deviance. With a better understanding of how the label is formed, this research could be expanded to explore the deviance amplification hypothesis or stereotype threat for student athletes A gain, a athletes can feel a negative stereotype and conform to it (Steele, 1997) 20). This study shows for this sample football and the s basketball. Fu ture research should examine whether student athletes generally feel these labels from fac ulty and if that impacts their academic performance. Additionally, this research should examine differences in stereotype threat between male and female student athlete groups. Implications for Practice The findings from the current study emphasize the imp ortance of increasing contact between faculty and student athletes as well as finding ways to expand the knowledge base and general contextual knowledge of student athletes among faculty Again, v ariables that were associated with the intergroup contact h ypotheses seemed to have the biggest effect in perceptions of deviance. Specifically, fandom was the most consistent predictor of perceptions of deviance where fans had fewer negative perceptions Therefore, athletic departments should work to promote fan dom among faculty to improve the relationship with their student athletes. Activities that promote fandom in college football include tailgating but could also include free tickets to (Koch & Wann, 2013)

PAGE 261

261 Research in soc ial psychology also shows support for decategorization, recategorization, and subcategorization as ways to reduce intergroup conflict (Brewer, 1996) Decategorization is the idea of creating interactions that are personal or individualized rather than inte ractions where people are category based members of a group as a way to decrease outgroup biases (Brewer & Miller, 1984). This means creating opportunities for student athletes to connect with faculty members on an individual basis, not as a member of thei r sport team. There are programming efforts at certain universities that may be making a difference in perceptions of student athlete deviance because the student athletes are only being presented to faculty as part of their athletic team, instead of as in dividuals. For example, faculty at UF, OSU, and UI indicated filling out progress reports as a form of service. However, no faculty at UGA listed that as a form of service. Faculty at UGA also had lower perceptions of general cheating compared to the other groups. It may be that activities where athletes are singled out from other students and presented as a member of the outgroup make them stick out to faculty thereby creating negative perceptions. Perhaps athletic departments should find ways to minimize singling out athletes from the normal student population. For example, one faculty indicat ed in an open response question in the survey about requests form their counsel ors about their progress, but none of these students have A decategorization technique could be for student athletes to meet with faculty as their own person rather than as just an athlete A second technique to improve intergroup rela tions is r ecategorization which involves enhancing the salience of a common team identity between the two groups to

PAGE 262

262 improve intergroup contact (Brewer, 1996). For the present study, that could include both faculty and student athletes being seen as member s of the overall university community. For example, this might include talking about faculty and student athletes at university a better place. A major goal at UF is to beco me one of the top 10 public research universities in the country (Ordway, 2014). Graduation rates of student athletes is one variable that is used to rank universities for the US News Best Colleges (Morse, Brooks, & Mason, 2016). Therefore, athletic depart ments and faculty can work together to achieve this goal as success for the Gator Nation as a whole. Finally, s ubcategorization is the idea of creating positive and cooperative experiences through distinct social identities to improve attitudes towards th e outgroup as a whole (Hewstone & Brown, 1986). This is achieved by highlighting the distinct roles each group has to work towards a common goal. For example, the common goal faculty and student athletes share is education, but the roles each group has are different to obtain that goal. The role of faculty is to teach and the role of student athletes is to learn. Therefore, it may be beneficial to create opportunities for student athletes to show faculty their role as learners. Activities that may promote t he student role of the student athlete to faculty could be studying abroad, service learning activities, and getting involved with faculty research. Conclusion This study finds that faculty have the most negative perceptions of deviance ll student athletes and the least negative perceptions about athletes. In addition, the more familiarity or closeness

PAGE 263

263 faculty have with student athletes the less likely they are to have negative perceptions. Below is a summary of major findings: Sport and university matter for perceptions of academic deviance Sport matters for perceptions of criminal deviance. There is an interaction effect between sport and university in predicting perceptions of drinking related deviance. Multi variate analyses show similar results for the sample as whole. Sports fandom is a consistent predictor for all deviance scales for the entire sample, where the bigger fans faculty were of their university sports programs the less deviance they perceived st udent athletes engage in Attendance at MBA events is a predictor of general cheating for the entire sample, where faculty that attended more MBA events were less likely to perceive student athlete general cheating Faculty were less likely to perceive WBB as d eviants compared to the male sports. Faculty at UF were significa ntly more likely to perceive student athletes as academic deviants compared to OSU Faculty at UI were signific antly more likely to perceive student athlete s as general cheaters and drin king related deviants compared to faculty at OSU Race matters, too. White faculty were more likely to perceive student athlete relying on others and drinking related deviance compared to non white faculty White faculty and those who estimated a higher pe rcentage of black and male student athletes on their campuses were more likely to perceive student athletes as reliant on others. Sometimes university location matters in the strength of relationships. At UI, being white mattered more in predicting percept ions of relying on others compared to faculty at UGA. Being a lecturer mattered more in predicting drinking related deviance compared to UF. Finally MFB student athletes matter more in predicting perceptions of general cheating compared to faculty at OSU At UF, MFB student athletes matter more in predicting perceptions of criminal deviance compared to UI and OSU. At OSU, sports fandom was important predictor of criminal deviance compared to UF. Some intergroup contact variables stuck out more at UGA compa red to the other universities Service involving athletes matter ed more in predicting general cheating compared to both OSU and UI. Going to more MBA events also matter more in predicting perceptions of relying on others compared to faculty at UF. Being in the discipline of physical sciences and math mattered more at UGA compared to UI. Finally, MFB student athletes mattered more to UGA faculty in perceptions compared to those from the big ten schools. Sometimes sport matters in strength of relationships. For faculty randomly assigned to the MFB group, s ports fandom, being an associate professor, in the discipline of physical sciences and math, and being a white faculty mattered

PAGE 264

264 more compared to those randomly assigned the WBB group. Being a white faculty w as also more important in predicting perceptions of drinking for MFB compared to MBA For faculty assigned to the MBA group, s ports fandom, being in the discipline of engineering and business mattered more compared to those assigned to WBB. Engineering was also a more important predictor compared to faculty assigned to MFB Contact with student athletes and being an associate professor mattered more compared to MBA

PAGE 265

265 Table 7 1. Summary of results for the entire sample Variable General cheating Relying o n others Criminal deviance Drinking related deviance Faculty status attributes Age NS NS NS Race (White) + + Academic Rank (Lecturer) NS Academic Rank (Assistant Professor) NS NS Academic Rank (Full Professor) NS Tenure status ( Tenure) NS Time at current institution NS NS Academic discipline (Education) NS NS Service involving athletics NS Sports fandom Attendance at MFB events NS NS NS Attendance at MBA events NS Attendance at WBB events NS NS Contact with student athletes NS University status attributes UI + NS NS + UGA NS NS NS NS UF + + NS NS Perceptions of student athlete attributes Gender (Female) Race (Black) + Student athlete status attributes M FB + + + + MBA + + + + Note. NS = not significant

PAGE 266

266 Table 7 2. Summary of results by university Variable General cheating Relying on others Criminal deviance Drinking related deviance UI UF UGA OSU UI UF UGA OSU UI UF UGA OSU UI UF UGA OSU Fac ulty status attributes Age NS NS NS NS NS NS NS NS Gender (Male) NS NS NS NS Race (White) NS NS NS NS + NS NS NS NS NS NS NS + + NS NS Academic Rank (Lecturer) NS NS NS NS NS NS NS Academic Rank (Asso ciate professor) NS NS NS NS Academic Rank (Full professor) NS NS NS NS Academic Discipline (Education) NS NS NS NS NS NS NS NS NS NS NS NS Academic Discipline (Physical sciences and math) NS NS NS NS NS NS + NS Time at current institution NS NS NS NS NS NS NS NS Service involving athletics NS NS NS NS NS NS NS NS NS NS NS Sports fandom NS NS NS NS NS NS NS NS NS NS NS NS Attendance at MFB events NS NS NS NS Attendance at MBA events NS NS NS NS NS NS NS NS NS NS NS NS NS NS Contact with student athletes NS NS NS NS Student athlete status attributes MFB + + + NS NS NS + + + + + NS + + + + MBA + NS NS NS NS NS NS NS NS + + NS NS + NS + Note. NS = not significant

PAGE 267

267 Table 7 3. Summary of results by sport Variable General cheating Relying on others Criminal deviance Drinking related deviance MFB MBA WBB MFB MBA WBB MFB MBA WBB MFB MBA WBB Faculty status attributes Age N S NS NS NS NS NS NS NS NS Race (White) + NS NS Academic Rank (Associate professor) + NS NS NS NS NS NS NS Academic Rank (Full professor) NS NS NS Academic Discipline (Business) NS + NS Academic Discipline (Physi cal sciences and math) + NS NS Academic Discipline (Social and behavioral sciences NS NS NS Academic Discipline (Engineering) NS NS Academic Discipline (Other) NS NS NS Time at current institution NS NS NS NS NS NS Service involving athletics NS NS NS NS NS NS Sports fandom NS NS NS NS NS Attendance at MFB events NS NS Attendance at MBA events NS NS Contact with student athletes NS NS Universi ty status attributes UI NS + NS NS NS NS NS NS NS NS NS + UGA NS NS NS NS NS NS NS NS NS NS NS NS UF + NS NS NS NS + + NS NS NS + NS Note. NS = not significant

PAGE 268

268 APPENDIX A EMAIL TEMPLATE S Introduction Email Template Subject: Disserta tion Survey about Student Athletes September 14, 2016 Dear Faculty Member, My name is Ashley Kuhn and I am a doctoral student of Criminology & Law at the University of Florida. To complete my dissertation, I am undertaking a study of faculty perceptio ns of student athletes at NCAA Division I institutions. As part of that study, you have been selected to participate in a survey of faculty to discern those attitudes; it is my hope that you will agree to be part of this study. In the next 5 7 days, you will be sent an email with a website link to that survey. The survey is composed of 35 opinion based questions. It should take 10 15 minutes to complete the survey. Participation in this study is voluntary and all responses will be kept anonymous. Results of the survey should be available in the spring of 2017 It is my hope that institutions and perhaps the NCAA itself will be able to utilize these results to better understand the voice of faculty in regards to student athletes. Thank you again for your time. Again, I will be sending another email within a week with the link to the survey. Sincerely, Ashley Kuhn Ph.D. Candidate Department of Sociology and Criminology & Law University of Florida

PAGE 269

269 Survey Link Email Template Subject: Dissertation Survey about Student Athletes September 21, 2016 Dear Faculty Member: My name is Ashley Kuhn and I am a doctoral student at the University of Florida. I recently sent you an introductory letter inviting you to be a part of my dissertation study on the athletes at Division I institutions. As mentioned before, your participation in the study is voluntary and all information you submit will be kept confidential. Your participation will involve completing a 35 questi on opinion survey. Completion of the survey should take no more than 10 15 minutes. Please click the link below to go to the survey: [Link] Your opinions in this survey are quite valuable. It is my hope that the results will be utilized by institutions as well as the NCAA to better understand the voice of faculty in the role and governance of intercollegiate athletics. I greatly appreciate your time. If you would like to see the results of the study, you may contact me at the email address below. Sinc erely, Ashley Kuhn Ph.D. Candidate Department of Sociology and Criminology & Law University of Florida

PAGE 270

270 Follow Up Email Template Subject: Dissertation Survey about Student Athletes October 5, 2016 Dear Faculty Member: Hello again. I recently sen t you an introductory email inviting you to be a part of my athletes at Division I institutions. If you have already completed the survey, please ignore this message; I have no way of tracki ng the identity of those who have already participated. Your opinions in this survey are quite valuable and I hope you take a few minutes to complete the survey. Completion of the survey should take no more than 10 15 minutes. If you have not done so ye t, you may go to the following link to complete the survey: [Link] Many thanks for your time and opinions! Sincerely, Ashley Kuhn Ph.D. Candidate Department of Sociology and Criminology & Law University of Florida

PAGE 271

271 APPENDIX B IRB PROTOCOL UFIRB 02 Social & Behavioral Research Protocol Submission Form T HIS FORM MUST BE TYP ED DO NOT STAPLE Send this form and the supporting documents to IRB02, PO Box 112250, Gainesville, FL 32611. Should you have questions about completing this form, call 3 52 392 0433. Title of Protocol: Faculty Perceptions of Student Athletes Principal Investigator: Kuhn Ashley UFID #: (Last Name) (First Name) Degree / Title: MA/Doctoral Student Mailing Address: ( If on campus provide PO Box address ): Email : Department: Sociology & Criminology and Law Telephone #: Co Investigator(s): Coordinator: Research Asst.: UFID#: (Last Name) (First Name) Degree/Title Mailing Address: ( If on campus provide PO Bo x address ): Email: Department: Telephone #: Supervisor (If PI is student) : Lane Jodi UFID#: (Last Name) (First Name) Degree / Title: PhD/Professor Mailing Address: ( If on campus provide PO Box address ): Email: D epartment: Sociology & Criminology and Law Telephone #: Date s of Proposed Research: April 2016 August 201 7 Source of Funding (A copy of the grant proposal must be subm itted with this protocol if funding is involved): NOTE: If your study has curr ent or pending funding, AND your research involves comparison of different kinds None

PAGE 272

272 of treatment or interventions for behavior, cognition or mental health, you must submit the Clinical Trial Assessment Form Describe the Scientific Purpose of the Study : athletes at NCAA Division I institutions. Specifically, we are exploring whether faculty and university level attributes impact perceptions of student athlete normative and academic de viance. Describe the Research Methodology in Non Technical Language: ( Explain what will be done with or to the research participant. ) The sampling frame includes a list of all faculty listed on department websites at the four institutions selected ( Univ ersity of Florida, University of Georgia, Ohio State University, and University of Illinois ). Only faculty listed on Faculty participants will be sent the series of emails discussed in the recruitment section (see Appendix A) Participants will be provided a link to a confidential on line Qualtrics survey. Upon opening the email, participants will be brought to an informed consent screen that the y read and sig n electronically (see Appendix B ). Following that, participants will be taken to the questionna ire (see Appendix C), which takes no more than 20 minutes to complete. Describe the Data You Will Collect: ( what are you collecting, where will it be stored, how will it be stored ) The data will be collected using the Qualtrics website. It will be exported from Qualtrics and stored on the researchers locked personal computer for analysis. There will be no identifying information gathered from part icipants. Please List all Locations Where the Research Will Take Place: ( if doing an on line survey then just state line survey ) The on line survey will be sent to faculty at four institutions of higher education These include University of Florida, University of Georgia, Ohio State University, and University of Illinois. Describe Potential Benefits: There are no benefits to participating. Describe Potential Risks: ( If risk of physical, psychological or economic harm may be involved, describe the s teps taken to protect participant.) There is no more than minimal risk in participating. Describe How Participant(s) Will Be Recruited: (flyers, email solicitation, social media websites, etc ) Recruitment of participants will follow the guidelines of Dil lman, Smyth, and Christian (2009) for Internet surveys First, the researcher will send an introduction recruitment email to the sample of faculty at the four institutions. This email will introduce the researcher, inform faculty members about the study, a nd participation is voluntary and confidential. This email will also let them know that the link to the survey will be provided in an a dditional email one week later ( See Appendix A 1 for the Introduction Email Template ) One week after the Introduction Email, the researcher will send a Survey Link Email to the sample of faculty at the four institutions. This email will remind the faculty about the study and provide the link to a confidential online survey through www.qualitrics.com ( See Appendix A 2 for the Survey Link Email Template ) Approximately two weeks after the Survey Link Email is sent, a Follow Up Email will be sent to all faculty reminding them to take the survey if they have not. If participants have taken the survey, they are asked to ignore e survey for their convenience ( See Appendix A 3 for the Follow Up Email Template ) Maximum Number of Partici pants (to 9,000 Age Range of Participants: 18 and over Amount of Compensation/ course credit: None

PAGE 273

273 be approached with consent) Describe the Informed Consent Process (How will informed consent be obtained? Attach a c opy of the Informed Consent Document) After the participant has signed up for the study, the informed consent document will be shown on screen penalty for choosing not to partici pate. (SIGNATURE SECTION) Principal Investigator(s) Signature: Date: Co Investigator(s) Signature(s): Date: Date: Department Chair Signature: Date:

PAGE 274

274 APPENDIX C INFORMED CONSENT Informed Consent Title: Faculty Perceptions of Student Athletes Please read this consent document carefully before you decided to participate in the study. Purpose of the research study: feelings and opinions abou t student athletes at NCAA Di vision I institutions. What you will be asked to do in the study: If you decide to participate in the research study, you will be asked to answer a series of questions about your background characteristics, university climate contact and perceptions of student athletes on your campus. Time required: It should take no more than 20 minutes to complete the survey but it could be longer depending on your pace. Benefits: There are no known benefits to you. Compensation: The re is no payment or compensation for participating in this study. Risks and confidentiality: Taking part in this study create s little risk for you. Researchers will not receive any identifying information; so your responses will be anonymous. Only the res earchers will have access to the information we collect online. There is a minimal risk that security of any online data may be breached, but since no identifying information will be collected, and the online host (Qualtrics) uses several forms of encrypti on other protections, it is unlikely that a security breach of the online data will result in any adverse consequence for you. It is not anticipated that talking about issues related to student athletes at your institution will cause any psychological or emotional discomfort. However, you do not have to answer any questions you do n ot feel comfortable answering (i.e., you can skip any question if you do not want to answer it) or you can stop the survey at any time. Voluntary Participation: You do not ha ve to answer the questions. Your participation is completely voluntary. Nothing negative will happen if you do not want to join this study. Right to withdraw from the study: You have the right to leave the study and stop the survey at any time. You also can choose to answer some questions and not answer other questions.

PAGE 275

275 Whom to contact if you have questions about the study: Ashley Price Kuhn, Ph.D. Student, Department of Sociology and Criminology & Law Jodi Lane, Ph.D., Professor, Department of Sociolo gy and Criminology & Law Whom to contact about your rights as a research participant in this study: UFIRB Office, Box 112250, University of Florida, Gainesville, FL 32611 2250, telephone (352) 392 0433 Agreement: I have read this consent form. I volunt ary AGREE to answer the questions for this study (knowing that I can cease participation at any time without penalty) ___ Agree ___ Disagree

PAGE 276

276 APPENDIX D ONLINE SURVEY INSTRUMENT First, I would like to know some background information about you. Please answer the following demographic questions. 1. What is your age? Enter age (in years) below ____ 2. What is your sex? Please select one of the answer options. 1 Male 2 Female 3 Other (Specify if you wish) ____ _______________________ 4 Prefer not to answer 3. Which of the following comes closest to describing your racial/ethnic identity? Please select one of the answer options. 1 Black/African American 2 White/Caucasian 3 Latino/Hispanic 4 Asian/Pacific Islander 5 Mixed race/Biracial (Please speci fy below) _____________________________ 6 Other (Please specify below) _____________________________ 4. What university are you affiliated with? Please select one from the drop down list. 1 Ohio State University 2 University of Florida 3 University of Georgi a 4 University of Illinois 5. What is your academic rank? Please select one of the answer options. 1 Lecturer 2 Assistant Professor 3 Associate Professor 4 Full Professor 5 Other (Please specify below) _____________________________ 9 NA 6. What is your tenure status? Please select one of the answer options. 1 Tenured 2 Not yet tenured 3 Not in tenure track 9 NA 7. Do you hold an administrative position? Please select one of the answer options. 1 No 2 Yes, Department/Program Head 3 Yes, Assistant Dean 4 Yes, Associate Dean 5 Other (Please specif y below)

PAGE 277

277 _____________________________ 8. What is your academic discipline? Please select one of the answer options. 1 Architecture 2 Arts and Humanities 3 Business 4 Education 5 Engineering 6 Law 7 Life Sciences 8 Medicine and Health Sciences 9 Physical Sciences and Mathemat ics 10 Social and Behavioral Sciences 11 Other (Please specify below) _____________________________ 99 NA 9. What undergraduate majors does your discipline serve? _____________________________________________________________________________ _____________________ ________________________________________________________ _____________________________________________________________________________ _________ 10. How many years have you been at your current institution? Please enter years in box below. ____ 11. Have you ever p articipated in service that involves athletics? 1 Yes 2 No 12. [If yes selected for Q11] Please explain the service you participated in that involves athletics. _____________________________________________________________________________ ________________________ _____________________________________________________ _____________________________________________________________________________ _________ 13. Have you ever served in an institutional governance role with responsibilities for intercollegiate athletics? 1 Yes 2 No 14. [If yes selected for Q13 ] Which of the following institutional governance roles did you have with responsibilities for intercollegiate athletics? (Please check all that apply) 1 Faculty Athletic Representative 2 Campus Advisory Board 3 tification Team 4 Other (Please specify below) _____________________________ 15. Please indicate how strongly you agree with the following statements. 1 2 3 4 5 6 7 Strongl y disagre e Disagre e Somewha t disagree Neither agree or disagre e Somewha t agree Agre e Strongl y agree

PAGE 278

278 I consider myself to be a (Gator/Buckeye/Bulldog/Illi ni) fan 1 2 3 4 5 6 7 My friends see me as a (Gator/Buckeye/Bulldog/Illi ni) fan 1 2 3 4 5 6 7 I believe that following (Gator/Buckeye/Bulldog/Illi ni) sports is the most enjoyable form of entertainment 1 2 3 4 5 6 7 16 H ow often have you attended the following (Gator/Buckeye/Bulldog/Illini) sporting events in the 2015 2016 academic year ? Never Rarely Sometimes Often Very Often 1 2 3 4 5 baseball 1 2 3 4 5 Wome 1 2 3 4 5 17. Are there any other (Gator/Buckeye/Bulldog/Illini) sporting events that you have gone to in the 2015 2016 academic year ? ____________________________________________________________________________________ ________________________ ____________________________________________________________ ____________________________________________________________________________________ ______ 18. How often did you interact with student athletes on your campus during the 2015 2016 academic year ? N ever Rarely Sometimes Often Very Often NA Student athletes are in your courses 1 2 3 4 5 9 Student athletes communicate with you by email or in person 1 2 3 4 5 9 Student athletes interact with you during class sessions 1 2 3 4 5 9 19. Have you had an y other interaction with student athletes on your campus during the 2015 2016 academic year? Please explain. ____________________________________________________________________________________ _______________________________________________________________ _____________________ ____________________________________________________________________________________ ______ The next few questions ask you to estimate based on your experience. Please answer even if you are unsure. 20. In the current academic year and t o the best of your knowledge, what percentage of students in your undergraduate courses are student athletes? ____ In the current academic year, please give your best guess based on your experience. Please answer even if you are unsure. 21. Please estimate the percentage of student athletes on your campus who are male and female: Male _____ Female _____ Prefer not to answer _____

PAGE 279

279 22. Please estimate the percentage of student athletes on your campus who are in each of the following racial categories: Black/Af rican American ______ White/Caucasian ______ Latino/Hispanic ______ Asian/Pacific Islander ______ Other _____ Prefer not to answer _____ 23. Please estimate the percentage of student athletes on your campus who participate in (Qualtrics will randomly assign sport group) Basketball ______ Baseball _____ Prefer not to answer _____ 24. What is your best guess of the percentage of (Qualtrics will randomly assign sport groups ) student athletes that graduate on your campus? ________ 25. Do you know of any NCAA violations for your institution? a. Yes b. No 26. [If yes to Question 25 ] What NCAA violations has your institution received? Please describe what happened below. _______________ ______________________________________________________________ _____________________________________________________________________________ _____________________________ _______________________________________ _________ 27. Were you a varsity student athlete in college? a. Yes b. No 28. [If yes to Question 2 7 ] Please specify the varsity sport you participated in? ___________________________________ _____________ _____________________________ 29. Please indicate how strongly you agree with the following statements about studen t athletes on your campus. Not at all Slightly Moderately Very much Know Student athletes are motivated to earn their degrees 1 2 3 4 9 Missed class time due to athletic obligations detracts from the quality of student asses 1 2 3 4 9 Student athletes are more burdened than other students on my campus by demands of their out of class time 1 2 3 4 9 Student athletes are not prepared academically to keep pace with other students in my classes 1 2 3 4 9 30. Do you have any other comments about student athletes on your campus? ________________________________________________________________________________ ________________________________________________________________________________ _________________________________________ _______________________________________ _________

PAGE 280

280 31 Using your best estimate I would like you to indicate how often you think in the last year ( FOOTBALL student athletes on your campus have engaged in following b Never Rarely Sometimes Often All of the time Prefer not to answer Copied from other students 1 2 3 4 5 9 Passed answers to other students during a test 1 2 3 4 5 9 Used prohibited notes 1 2 3 4 5 9 Obtained the test questions illegally 1 2 3 4 5 9 Used unauthorized electronic equipment on a test or assignment 1 2 3 4 5 9 1 2 3 4 5 9 Worked on assignment with others when asked for individual work 1 2 3 4 5 9 Got extra help on an assignment from a tutor 1 2 3 4 5 9 Provided a paper or assignment for another student 1 2 3 4 5 9 Gave forbidden help to others on their assignments 1 2 3 4 5 9 Did less of their share of work in group project 1 2 3 4 5 9 Copied materials without citing them 1 2 3 4 5 9 Falsified ath letic travel letters to postpone exams or assignments 1 2 3 4 5 9 32 Using your best estimate I would like you to indicate how often you think in the last year ( FOOTBALL student athletes on your campus have enga Never Rarely Sometimes Often All of the time Prefer not to answer Purposely damaged or destroyed property belonging to others? 1 2 3 4 5 9 Stolen (or tried to steal) something worth more than $50? 1 2 3 4 5 9 Thrown objects (such as rocks, bottles, etc.) at cars or people? 1 2 3 4 5 9 Lied about their age to gain entrance or to purchase something: for example, lying about their age to buy liquor? 1 2 3 4 5 9 Drank alcohol? 1 2 3 4 5 9

PAGE 281

281 Drank more than 5 alcoholic drinks at once? 1 2 3 4 5 9 Stolen (or tried to steal) things worth $50 or less? 1 2 3 4 5 9 Had sexual relations with a person other than their significant other? 1 2 3 4 5 9 Been involved in a group fight? 1 2 3 4 5 9 Sold marijuana or hashish ("pot", "grass" "hash")? 1 2 3 4 5 9 Used marijuana or hashish ("pot", "grass", "hash")? 1 2 3 4 5 9 Stolen money or other things from their friends, neighbors, or roommates? 1 2 3 4 5 9 Taken money or gifts from alumni 1 2 3 4 5 9 Hit (or threatened to hit) other p eople? 1 2 3 4 5 9 Been loud, rowdy, or unruly in a public place (disorderly conduct)? 1 2 3 4 5 9 Sold harsh drugs such as heroin, cocaine, and LSD? 1 2 3 4 5 9 Used hard drugs such as heroin, cocaine, and LSD? 1 2 3 4 5 9 Bought or provided liquor fo r a minor? 1 2 3 4 5 9 Had (or tried to have) sexual relations with someone against their will? 1 2 3 4 5 9 Avoided paying for such things as movies, clothing, and food? 1 2 3 4 5 9 Been drunk in a public place? 1 2 3 4 5 9 Broken into a building or ve hicle (or tried to break in) to steal something or just look around? 1 2 3 4 5 9

PAGE 282

282 LIST OF REFERENCES AAPOR. (2016). Standard definitions: final dispositions of case codes and outcome rates for surveys. Retrieved from: http://www.aapor.org/AAPOR_Main/media/publications/Standard Definitions20169theditionfinal.pdf 248. Agnew, R. (1991). The interactive effects of peer variables on delinquency. Criminology 29 (1), 47 72. Allison, P. D. (1999). Multiple regression: A primer Pine Fo rge Press. Allison, P. D. (2002). Missing data series: Quantitative applications in the social sciences. Thousand Oaks. CA Allport, G. W. (1954). The nature of prejudice New York: Addison. Arbaugh, J. B. (2001). How instructor immediacy behaviors affe ct student satisfaction and learning in web based courses. Business Communication Quarterly 64 (4), 42 54. Barlow, M. H. (1998). Race and the problem of crime in Time and Newsweek cover stories, 1946 1995. Social Justice, 25 149 183. Baucom, C., & Lant z, C. D. (2001). Faculty attitudes toward male Division II student athletes. Journal of Sport Behavior 24 (3), 265 276. Becker, H. S. (1963). Outsiders: Studies in the sociology of deviance New York: The Free Press. Banaji, M. R., & Greenwald, A. G. (20 16). Blindspot: Hidden biases of good people Bantam. Beard, A. (2014). Review: UNC athlete literacy findings flawed. CNS News. Retrieved from: http://www.cnsnews.com/news/article/review unc athlete literacy findings flawed Bernstein, I. N., Kelly, W. R ., & Doyle, P. A. (1977). Societal reaction to deviants: The case of criminal defendants. American Sociological Review 743 755. Bertrand, M., & Mullainathan, S. (2004). Are Emily and Greg more employable than Lakisha and Jamal? A field experiment on labo r market discrimination. American Economic Review. 94, 991 1013.

PAGE 283

283 Billings, A. C. (2012). Talking around race: Stereotypes, media, and the twenty first century collegiate athlete. Wake Forest JL & Pol'y 2 199. Brewer, M. B. (1996). When contact is not en ough: social identity and intergroup cooperation. International Journal of Intercultural Relations, 20, 291 303. Brewer, M. B., & Miller, N. (1984). Beyond the contact hypothesis: Theoretical perspectives on desegregation. In N. Miller & M. Brewer (Eds.), Groups in contact: The psychology of desegregation. New York: Academic Press. Buckman, J. F., Yusko, D. A., Farris, S. G., White, H. R., & Pandina, R. J. (2011). Risk of marijuana use in male and female college student athletes and nonathletes. Journal o f studies on alcohol and drugs 72 (4), 586. Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models: Applications and data analysis methods. Newbury Park, CA: Sage Publications. Chung, D. J. (2013). The dynamic advertising effect of collegiat e athletics. Marketing Science 32 (5), 679 698. Clinard, M., & Meier, R. (2010). Sociology of deviant behavior. Cengage Learning. Clogg, C. C., Petkova, E., & Haritou, A. (1995). Statistical methods for comparing regression coefficients between models. American Journal of Sociology 100 (5), 1261 1293. Coakley, J. J. (1990). Sport in society: Issues and controversies Boston: Times Mirror/Mosby. Cockley, T., & Roswal, G. knowledge and satisfact ion regarding NCAA. Journal of Sport Behavior, 17(4), 217. Cohan, M. (2016). Remembering and misremembering the Duke Lacrosse case. Vanity Fair. Retrieved from: http://www.vanityfair.com/news/2016/03/duke lacrosse case fantastic lies documentary Cohen, A. K. (1959). The study of social disorganization and deviant behavior. Sociology Today, 461 484. Comeaux, E. (2011). A study of attitudes toward college s tudent athlete: Implications for faculty athletics engagement. The Journal of Negro Education, 80(4), 521 532. Duke Chronicle. Retrieved from:

PAGE 284

284 http://www.dukechronicle.com/articles/2006/11/07/group 88 faculty hears criticism wake lax scandal Crepeau, R. C. (1981). Sport, heroes and myth. Journal of Sport & Social Issues 5 (1), 23 31. Crosset, T. W., Benedict, J. R., & McDonald, M. A. (1995). Male student athletes reported for sexual assault: A survey of campus police departments and judicial affairs offices. Journal of Sport & Social Issues 19 (2), 126 140 Saturday Down South Retrieved from: http://www.saturdaydownsouth.com/sec football/coll ege football arrests 2015/ Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment Research & Evaluation. 10(7). 1 9. Retrieved from: http://pareonline.net/pdf/v10n7.pdf Cullen, F. T., & Agnew, R. (2003). Labeling, interaction, and crime: Societal reaction to the creation of criminals. In F. Cullen & R. Agnew (Eds.), Criminology theory: P ast to present (2 nd ed.) (pp. 295 303). Los Angeles, CA: Roxbury Publishing Company Digital Commons Three Tiered List of Academic Disciples. (2015, December 11). Retrieved from http://scholarw orks.uark.edu/assets/taxonomy.pdf Dillman, D. A., Smyth, J. D., & Christian, L. M. (2009). Internet, Mail, and Mixed Mode Surveys: The Tailored Design Method. Hoboken, NJ: Wiley. Drummond, W. J. (1990). About face: Blacks and the news media. The Ameri can Enterprise, 1, 23 29. Dunham, A., & Williams, B. (2016, November). College sports: The pathway to opportunity. Presented at the annual FARA Conference, Indianapolis, IN. Abstract retrieved from: http://farawebsite.org/wp content/uploads/2012/10/Preliminary Agenda.pdf people as delinquent. Legal and Criminological Psychology, 5(2) 153 164. Eitzen, D. S. (2012). Fair and foul: Beyond the myths and paradoxes of sport Rowman & Littlefield Publishers. Enders, C. K. (2010). Applied missing data analysis Guilford Press: New York, NY.

PAGE 285

285 Engstrom, C. M., & Sedlacek, W. E. (1991). A st udy of prejudice toward university student athletes. Journal of Counseling & Development 70 (1), 189 193. Engstrom, C. M., Sedlacek, W. E., & McEwen, M. K. (1995). Faculty attitudes toward male revenue and nonrevenue student athletes. Journal of College S tudent Development 36 (3), 217 227. ESPN. (2016). 2015 college football rankings postseason. Retrieved on October 21, 2016 from: http://www.espn.com/colle ge football/rankings/_/poll/1/seasontype/3/year/2015/week/1 Evans, J. R., & Mathur, A. (2005). The value of online surveys. Internet research 15 (2), 195 219. Fazio, R. H. (1995). Attitudes as object evaluation associations: Determinants, consequences, and correlates of attitude accessibility. Attitude strength: Antecedents and consequences 4 247 282. Feltz, D. L., Schneider, R., Hwang, S., & Skogsberg, N. J. (2013). Predictors of Collegiate Student Athletes' Susceptibility to Stereotype Threat. Jour nal of College Student Development 54 (2), 184 201. Fountain, J. J., & Finley, P. S. (2009). Academic majors of upperclassmen football players in the Atlantic Coast Conference: An analysis of academic clustering comparing white and minority players. Journ al of Issues in Intercollegiate Athletics 2 (1), 13. Gage, E. A. (2008). Gender Attitudes and Sexual Behaviors Comparing Center and Marginal Athletes and Nonathletes in a Collegiate Setting. Violence Against Women 14 (9), 1014 1032. Gibbs, J. P. (1966). Conceptions of deviant behavior: The old and the new. Pacific Sociological Review 9 (1), 9 14. Gove, W. R. (1980). The labeling of deviance Beverly Hills, CA: Sage. Graham, J. W., & Hoffer, S. W. (2000). Multiple imputation in multivariate research. In T. D. Little & K. U. Schnabel (Eds.), Modeling longitudinal and multilevel data: Practical issues, applied approaches, and specific examples (pp. 201 218, 269 281). Mahwah, NJ: Lawrence Erlbaum. Green, A. R., Carney, D. R., Pallin, D. J., Ngo, L. H., Ra ymond, K. L., Iezzoni, L. I., & Banaji, M. R. (2007). Implicit bias among physicians and its prediction of thrombolysis decisions for black and white patients. Journal of general internal medicine 22 (9), 1231 1238.

PAGE 286

286 Harris, J. (2016). Who has the best fan s in college football?. NCAA. Retrieved from: http://www.ncaa.com/news/football/article/2016 08 23/who has best fans college football Hawkins, R., & T iedeman, G. (1975). The creation of deviance: Interpersonal and organizational determinants Merrill Publishing Company. Hewstone, M., & Brown, R. J. (1986). Contact is not enough: An intergroup perspective In M. Hewston & R. Brown (eds.), Contact and conflict in intergroup encounters. Oxford: Basil Blackwell. Hirko, S. (2009). Intercollegiate athletics and modeling multiculturalism. New Directions for Higher Education 2009 (148), 91 100. Humphrey, S. E., & Kahn, A. S. (2000 ). Fraternities, athletic teams, and rape. Journal of interpersonal violence 15 (12), 1313 1320. Johnson, B. D. (2010). Multilevel analysis in the study of crime and justice. In Handbook of quantitative criminology (pp. 615 648). Springer New York Johnso n, K. C. (2008). The Duke Lacrosse Case and the Blogosphere. Law and Contemporary Problems 155 170. Kaiser, H. F. (1960). The application of electronic computers to factor analysis. Educational and psychological measurement 20 (1), 141 151. Kluger, J. ( 2014). Why on field violence continues off field. Time. Retrieved from: http://time.com/3398946/nfl violence psychology/ Koch, K., & Wann, D. L. (2013). Fans' identification and commitment to a sport team: the impact of self selection versus socialization processes. Athletic Insight 5 (2), 129. LaBrie, J. W., Grossbard, J. R., & Hummer, J. F. (2009). Normative misperceptions and marijuana use among male and female college athletes. Journal of Applied Sport Psychology 21 (S1), S77 S85. Larimer, M. E., Turner, A. P., Mallett, K. A., & Geisner, I. M. (2004). Predicting drinking behavior and alcohol related problems among fraternity and sorority members: examining the role of descriptive and in junctive norms. Psychology of Addictive Behaviors 18 (3), 203. Lauritsen, J. L., Heimer, K., & Lynch, J. P. (2009). Trends in the gender gap in violent offending: New evidence from the National Crime Victimization Survey. Criminology, 47(2), 361 399.

PAGE 287

287 La verie, D. A., & Arnett, D. B. (2000). Factors affecting fan attendance: The influence of identity salience and satisfaction. Journal of leisure Research 32 (2), 225. Lavigne, P. ( 2015 b ). OTL: Florida Gators had most athletes as crime suspects from 2009 14 ESPN. Retrieved from: http://www.espn.com/blog/ncfnation/post/_/id/111513/otl study gators had most athletes as crime sus pects from 2000 14 Lavigne, P. (2015 a ). Lawyers, status, public backlash aid college athletes accused of crimes. ESPN Retrieved from: http://www.espn.com/espn/otl/story/_/id/13065247/college athletes major programs benefit confluence factors somes avoid criminal charges Lawrence, J. H., Hendricks, L. A., & Ott, M. C. (2007). Faculty Perceptions of Int ercollegiate Athletics: A National Study of Faculty at NCAA Division I Football Bowl Subdivision Institutions. Online Submission Learfield Sports. (2015). 2014 Standings. Retrieved from: http://grfx.cstv.com/photos/schools/nacda/sports/directorscup/auto_pdf/2014 15/misc_non_event/DIJune25.pdf Lemert, E. M. (1972). Human deviance, s ocial problems, and social control Englewood Cliffs, NJ: Prentice Hall Public Relations Review, 32, 386 394. Letawsky, N. R., Schneider, R. G., Pedersen, P. M., & Palmer, C. J. (200 3). Factors influencing the college selection process of student athletes: Are their factors similar to non athletes. College Student Journal 37 (4), 604. Liptak, A. (2007). Prosecutor becomes prosecuted. New York Times. Retrieved from: http://www.nytimes.com/2007/06/24/weekinreview/24liptak.html Lin, C. S., & Wen, L. M. (2007). Academic dishonesty in higher education: a nationwide study in Taiwan. Higher Education, 54, 85 97. Little, R. J. A. (1988). A test of missing completely at random for multivariate data with missing values. Journal of the American Statistical Association, 83, 1198 1202. McConnell, A. R., & Leibold, J. M. (2001). Relations among the implicit association test, discriminatory behavior, and explicit measure of racial attitudes. Journal of Experimental Social Psychology, 37, 435 442.

PAGE 288

288 McCray, K. L. (2014). Intercollegiate Athletes and Sexual Violence A Review of Literature and Recommendations for Future Stud y. Trauma, Violence, & Abuse 1524838014537907. Miller L. E., & Smith, K. (1983). Handling non response issues. Journal of Extension On line, 21(5). Available at: http://www.joe.o rg/joe/1983september/83 5 a7.pdf Mitchell, O. (2005). A meta analysis of race and sentencing research: Explaining the inconsistencies. Journal of Quantitative Criminology 21 (4), 439 466. Morse, R., Brooks, E., & Mason, M. (2016). How U.S. News calculat ed the 2017 best colleges rankings. US News. Retrieved from: http://www.usnews.com/education/best colleges/articles/how us news calculated the rankings National Center for Educational Statistics. (2015). Integrated Postsecondary Education Data System (IP EDS). Retrieved from: https://nces.ed.gov/ipeds/datacenter/InstitutionByName.aspx NCAA. (2005). News release: University of Illinois at Urbana Champaign public infractions report. Retrieved from: https://web1.ncaa.org/LSDBi/exec/miSearch?miSearchSubmit=publicReport&key =607&publicTerms=T HIS%20PHRASE%20WILL%20NOT%20BE%20REPEATE D NCAA. (2006). Ohio State University public infractions report. Retrieved from: https://web1.ncaa.org/LSDBi/exec/miSearch?miSearchSubmit=publicReport&key =620&publicTerms=THIS%20PHRASE%20WILL%20NOT%20BE%20REPEATE D NCAA. (2011). The Ohio State University public infractions report. Retrieved from: https://web1.ncaa.org/LSDBi/exec/miSearch?miSearchSubmit=publicReport&key =726&publicTerms=THIS%20PHRASE%20WILL%20NOT%20BE%20REPEAT E D NCAA. (2014a). NCAA, White House partner to combat sexual violence on campus. Retrieved from: http://www.ncaa.org/about/resources /media center/news/ncaawhitehouse partner combat sexual violence campus support of healthy and safe campuses. Retrieved from: http://www.ncaa.org/sites/default/files/Sexual Violence Prevention.pdf NCAA. (2014c). Drug testing program. Retrieved from: htt p://www.ncaa.org/sites/default/files/DT%20Book%202014 15.pdf

PAGE 289

289 NCAA. (2014d). University of Georgia public infractions decision. Retrieved from: https://web1.ncaa.org/LSDBi/exec/miSearch?miSearchSubmit=publicReport&key =854&publicTerms=THIS%20PHRASE%20WILL%20NOT%20BE%20REPEATE D NCAA. (2015a). Legislative Service Database LSDBi. Retrieved from: https://web1.ncaa.org/LSDBi/exec/miSearch NCAA (2015b). University of Florida public infractions decision. Retrieved from: https://web1.ncaa.org/LSDBi/exec/miSearch?miSearchSubmit=publicReport&key =875&publicTerms=THIS%20PHRASE%20WILL%20NOT%20BE%20REPEATE D Nishimoto, P. A. (1997). Touchdowns and term papers: Telescopin g the college student athlete culture. College Student Affairs Journal, 16(2), 97 103. Ordway, D. (2014). UF aims to crack top 10 of U.S. public universities. Orlando Sentinel Retrieved from: http://articles.orlandosentinel.com/2014 01 07/news/os university florida top 10 20140107_1_uf president bernie machen state university system ranking s Osborne, J. W. (2000). Advantages of hierarchical linear modeling. Practical Assessment, Research, & Evaluation, 7, Retrieved January 18, 2014 from http://PAREonline.net/getvn.asp?v=7&n=1 Osgood D. W., McMorris, B. J., & Potenza, M. T. (2002). Analyzing multiple item measures of crime and deviance I: Item response theory scaling. Journal of Quantitative Criminology 18 (3), 267 296. Page, R. M., & Roland, M. (2004). Misperceptions of the prevale nce of marijuana use among college students: athletes and non athletes. Journal of Child & Adolescent Substance Abuse 14 (1), 61 75. Paternoster, R., & Iovanni, L. (1989). The labeling perspective and delinquency: An elaboration of the theory and an asses sment of the evidence. Justice Quarterly 6 (3), 359 394. Pettigrew, T. F., & Tropp, L. R. (2006). A meta analytic tes t of intergroup contact theory. Journal of Personality and Social Psychology, 90(5), 751 783. Peugh, J. L., & Enders, C. K. (2004). Missi ng data in educational research: A review of reporting practices and suggestions for improvement. Review of Educational Research, 74(4), 525 556. Putler, D. S., & Wolfe, R. A. (1999). Perceptions of intercollegiate athletic programs: priorities and tradeo ffs. Sociology of Sport Journal 16 (4), 301 325.

PAGE 290

290 Radhakrishna, R., & Doamekpor, P. (2008). Strategies for generalizing findings in survey research. Journal of Extension, 46 (2), Available at: https: //www.joe.org/joe/2008april/tt1.php Rubie Davies, C. M (2006). Teacher expectations and student self perceptions: Exploring relationships. Psychology in the Schools 43 (5), 537 552. Rubin, D. B. (1976). Inference and missing data. Biometrika 581 592. Russell, K. K. (2002). The racial hoax as crime: The law as affirmation. In Shaun L. Gabbidon, Helen Taylor Greene, and Vernetta D. Young, (eds.), African American classics in criminology and criminal justice. Thousand Oaks, CA: Sage Publications. SAAB and SASSO Host 2014 Staff Appreciation Dinne r. (2014). Retrieved February 05, 2017, from http://www.ohiostatebuckeyes.com/sports/sasso/spec rel/120214aab.html Sailes, G. A. (1998). The African American athlete: Social myths and stereotypes. African Americans in sport: Contemporary themes 183 198. Tittle, C. R. (1980). Sanctions and social deviance: The question of deterrence New York: Praeger. Schwartz, J. M. (1973). Causes and effects of spectator sports. International Review of Sport Sociology 8 (3), 25 45. Schwartz, J. A., & Beaver, K. M. (2014). A biosocial analysis of the sources of missing data in criminological research. Journal of Criminal Justice 42 (6), 452 461. Skolnick, J. H. (1993). Justice without trial: Law enforcement in Democratic Society 3 rd ed. New York. Macmillian. Skol nick, J. H. (2007). Racial profiling Then and now. Criminology & Public Policy 6 (1), 65 70. Smith, P. (1990). Killing the spirit: Higher education in America Viking Penguin Publishing, 375 Hudson St., New York, NY 10014. Sparent, M. E. (1988). The stud ent athlete in the classroom: Developmental issues affecting college athletes and their impact on academic motivation and performance. Steele, C. M. (1997). A threat in the air: How stereotypes shape intellectual identity and performance. American Psycho logist, 52, 613 629

PAGE 291

291 Steffensmeier, D., Ulmer, J., & Kramer, J. (1998). Interaction of Race, Gender, and Age in Criminal Sentencing: The Punishment Cost of Being Young, Black, and Male, The. Criminology 36 763. Sternberg, R. J. (2016). College athletics : Necessary, not just nice to have. Retrieved from: http://www.nacubo.org/Business_Officer_Magazine/B usiness_Officer_Plus/Bonu s_Material/College_Athletics_Necessary_Not_Just_Nice_to_Have.html effect of identity priming on stereotype threat for academically engaged A frican American college athletes. Basic and applied social psychology 34 (2), 99 106. Tennenbaum, F. (1938). Crime and the Community New York: Columbia University. Thompson, E. ( 2015). UF pitchers A.J. Puk and Kirby Snead arrested, charged with third de gree criminal trespass. Orlando Sentinel. Retrieved from: http://www.orlandosentinel.com/sports/florida gators/swamp things blog/os gators pitchers aj puk kirby snead arrested 20150413 post.html Thompson, E., & Romero, I. L. (2015). Former Gator Aaron Hernandez guilty of murder, gets life in prison. Associated Press. Retrieved from: http://www.orlandosentinel.com/sports/florida gators/os gator aaron hernandez guilty murder life in prison 20150415 story.html Tittle, C. R. (1980). Labelling and crime: An empirical evaluation. The labelling of deviance 241 63. Tourangeau R., & Yan, T. (2007). Sensitive questions in surveys. Psychological bulletin 133 (5), 859. Toma, J. D. (1999), The Collegiate Ideal and the Tools of External Relations: The Uses of High Profile Intercollegiate Athletics. New Directions for Higher Education 81 90. Toma, J. D., & Cross, M. E. (1998). Intercollegiate athletics and student college choice: Exploring the impact of championship seasons o n undergraduate applications. Research in Higher Education, 39, 633 661. Tracy, M. (2014). N.C.A.A. is reopening inquiry into academic violations at North Carolina. New York Times. Retrieved from: http://www.nytimes.com/2014/07/01/sports/ncaa is reopening university of north carolina inquiry.html

PAGE 292

292 Triplett, R. (1993). The conflict perspective, symbolic interactionism, and the status charact eristics hypothesis. Justice Quarterly 10 (4), 541 558. Umbach, P. D., & Wawrzynski, M. R. (2005). Faculty do matter: The role of college faculty in student learning and engagement. Research in Higher Education 46 (2), 153 184. Urdan, T., & Schoenfelder, E. (2006). Classroom effects on student motivation: Goal structures, social relationships, and competence beliefs. Journal of school psychology 44 (5), 331 349. U.S. Department of Education (2015). The Equity in Athletics Data Analysis Cutting Tool. Retr ieved from: http://ope.ed.gov/athletics/GetOneInstitutionData.aspx U.S. News (2016). Top public schools: National universities. Retrieved October 21, 2016, from http://colleges.usnews.rankingsandreviews.com/best colleges/rankings/national universities/top public?_mode=table Van den Bergh, L., Denessen, E., Hornstra, L., Voeten, M., & Holland, R. W. (2010). The implicit prejudiced attitudes of teachers: Relations to teacher expectations and the ethnic achievement gap. American Educational Research Journal 47 (2), 497 527. Wadhwani, A. (2016a). Sweeping sex assault suit filed against University of Tennessee. The Tennessean Retrieved from: http://www.tennessean.com/story/news/2016/02/09/sweep ing sexual assault suit filed against ut/79966450/ Wadhwani, A. (2016 b ). Butch Jones, Dave Hart among potential witnesses in UT lawsuit. The Tennessean Retrieved from: http://www.tennessean.com/story/news/2016/03/09/butch jones dave hart among potential witnesses university tennessee suit/81527966/ Wann, D. L. (2002). Preliminary validation of a measure for assessing identification as a sport fan: The Sport Fandom Questionnaire. International Journal of Sport Management 3 (2), 103 115. Watt, S. K., & Moore, J. L. (2001). Who are student athletes?. New Directions for Student S ervices 2001 (93), 7 18. Wechsl er, H., Davenport, A. E., Dowdall, G. W., Grossman, S. J., & Zanakos, S. I. (1997). Binge drinking, tobacco, and illicit drug use and involvement in college athletics: A survey of students at 140 American colleges. Journal of American College Health 45 (5) 195 200.

PAGE 293

293 Wellford, C. (1975). Labeling theory and criminology: An assessment. Social Problems, 22, 332 345. Wellford, C. F., & Triplett, R. A. (1993). The future of labeling theory: Foundations and promises. In F. Alder & W. S. Laufer (Eds.), Advances in criminological theory (Vol 4.) (pp. 1 19). New Brunswick, NJ: Transaction Press. What does a social disaster sound like? (2006, April 6). Retrieved March 26, 2016, from htt p://today.duke.edu/showcase/mmedia/pdf/socialdisasterad.pdf Wilson, D. (2007). Former Duke players cleared of all charges. The New York Times. Retrieved from: http://www.nytimes.com/2007 /04/11/us/12dukecnd.html backlash. The Chronicle of Higher Education. Retrieved from: http:// chronicle.com/article/Chapel Hill Researcher s/144169/ Wininger, S., & White, T. (2008). The Dumb Jock Stereotype: To What Extent Do Student Athletes Feel the Stereotype?. Journal for the Study of Sports and Athletes in Education 2 (2), 227 237. Yopyk, D. J., & Prentice, D. A. (2005). Am I an athlete or a student? Identity salience and stereotype threat in student athletes. Basic and Applied Social Psychology 27 (4), 329 336. Yusko, D. A., Buckman, J. F., White, H. R., & Pandina, R. J. (2008). Alcohol, tobacco, illicit drugs, and performance enhancers: a comparison of use by college student athletes and nonathletes. Journal of American College Health 57 (3), 281 290.

PAGE 294

294 BIOGRAPHICAL SKETCH Ashley Price Kuhn is originally from Naples, Florida. She gradua ted with a Bachelor of Science degree in psychology and Bachelor of Arts degree in criminology from the University of Florida in 2010. She also earned a Master of Arts (2012) and Ph.D. (2017) in criminology from the University of Florida.