fE‘K‘K‘JF'Ol

Sf

i

a. r a.
5.2;”;
, a

:
s

.23

3% 5%

 

C i) (7} LIBRARY
Michigan State
University

This is to certify that the
thesis entitled
EVALUATING THE EFFECTIVENESS OF A SEXUAL ASSAULT
PREVENTION PROGRAM TARGETING MALE COLLEGIATE
ATHLETES
presented by

Wendi L. Siebold

has been accepted towards fulfillment
of the requirements for

M.A. degreein Psychology

£4,

Major professor

Date M

0-7639 MSU is an Affirmative Action/Equal Opportunity Institution

 

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

DUE DATE DUE DATE DUE

 

6/01 c:/ClRC/DaleDue.p65-p.15

EVALUATING THE EFFECTIVENESS OF A SEXUAL ASSAULT PREVENTION
PROGRAM TARGETING MALE COLLEGIATE ATHLETES

By

Wendi Lyn Siebold

A THESIS

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

MASTER OF ARTS
Department of Psychology

2000

ABSTRACT

EVALUATING THE EFFECTIVENESS OF A SEXUAL ASSAULT PREVENTION
PROGRAM TARGETING MALE COLLEGIATE ATHLETES

By

Wendi Lyn Siebold

Male collegiate athletes are one of the groups on college campuses most
commonly assumed to be at a high risk for perpetration of sexual assault, because of their
direct involvement in a culture that has been known to foster an aggressive and
competitive environment (Curry, 1991; Messner, 1992; Nelson, 1994). Yet, there exist
only a handful of prevention and education programs that are specifically designed for
male student athletes, none of which have been evaluated and published in the literature.
The present study evaluated a sexual assault prevention program targeting male collegiate
athletes. Initial results from this evaluation indicated that the program did not
significantly reduce men’s rape myth acceptance or likelihood of sexually aggressing, or
increase their likelihood of confronting. Additionally, none of the sport-specific
characteristics of being an athlete significantly predicted participants’ change in attitudes
post-intervention. Open-ended program utilization questions gave insight into these
findings, as well as the possibility of group-level change within a male peer supportive
environment (Schwartz & DeKeseredy, 1997). Implications for fiiture programming
targeting male collegiate athletes are discussed, along with a general discussion about the
efficacy of such efforts that target individuals who operate within a larger system of

accountability.

 

ACKNOWLEDGEMENTS

This study would not have been possible without the enduring support of
numerous people who worked with me through both the process and the product of this
project. I would particularly like to thank Jayne Schuiteman, the Women’s Resource
Center, and all of the members of the program planning committee for allowing me to
come on board and be a part of the this project from the start. I appreciate all of the
support they gave to both me and the evaluation of this program — it was truly a joy to get
to know and work with all of you. I am sincerely grateful to Cris Sullivan, the chair of
my thesis committee, for advising me through the changing tides of evaluating such a
politically tempestuous program. Her wealth of experience in program evaluation and
working with campus groups helped guide me through the process. And if it weren’t for
Cris, this paper would have probably been twice as long with multiple punctuation
rnistakesl My other two committee members also deserve accolades, namely, Deborah
Bybee for the amount of time she worked with me on my methodology and statistical
analyses, and Pennie Foster-Fishman for guiding me through the “larger worth”
conceptual portions of this project. Last, but certainly not least, I thank Carolyn for all
of her emotional and practical support. She was my solid ground for not only the
emotional aspects of this project, but also by being my own personal editor through
reading numerous drafts of this paper. Thanks for putting up with all of my mood swings

and supporting me throughout!

iii

TABLE OF CONTENTS

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

INTRODUCTION ...........................................................................

CHAPTER 1
OVERVIEW OF THE LITERATURE ....................................................

Defining Sexual Assault .............................................................
The Etiology of Sexual Assault ....................................................
Male Collegiate Athletics as a Rape-Prone Sub-Culture ........................
Rape Prevention Education ..........................................................
F emale-Focused Programs .................................................
Mixed-Sex Programs ......................................................
Male-Focused Programs ...................................................
Programs Targeting Collegiate Athletes ................................
Group-Level Change via Individual Change ....................................
Conclusion to the Review ...........................................................
Hypotheses ...........................................................................

CHAPTER 2
METHOD .......................................................................................

Sample .................................................................................
The Program: Out of Bounds: An Anti- Violence Workshop for

S tudent-A thletes at MS U ............................................................
Evaluation ............................................................................
Measurement ..........................................................................
Statistical Analyses ..................................................................

CHAPTER 3
RESULTS .....................................................................................

Pre-test Descriptives ................................................................
Multivariate Analyses ...............................................................
Post-test Descriptives ...............................................................
Predictors of Post-Intervention Change ..........................................
Program Utilization ..................................................................

CHAPTER 4
DISCUSSION .................................................................................

iv

iv

1

6

6

10
22
28
29
31
34
37
40
44
46

47
47

47
S4
S6
73

75
78
84
88
93
100

104

LIST OF REFERENCES .................................................................... 131

APPENDICES ................................................................................ 140
Appendix A: Pre-Test Questionnaire ............................................. 141
Appendix B: Post-Test Questionnaire ........................................... 156
Appendix C: Sexual Assault Prevention Programs Targeting College

Students .............................................................. 165

LIST OF TABLES

Table 1 - Sport Orientation Subscale Item-to-Total Correlations ...................... 58
Table 2 - Individual Utilization Coding Categories ...................................... 69
Table 3 - Team Utilization Coding Categories ........................................... 70

Table 4 — Coding Categories for Factors Influencing Participants’
Confrontation of a Teammate ................................................... 72

Table 5 - Independent Sample T-Tests for Cases Excluded From
Analyses Based on Selection Criteria .......................................... 76

Table 6 - Independent Sample T-Tests for Estimated Cases Based

on Missing Partial Post-Test ...................................................... 77
Table 7 - Predictor Variable Scale Percentages at Pre-Test (measured on a 7-point
scale ................................................................................... 79
Table 8 - Sport Orientation Scale Percentages at Pre-Test ........... . .................. 80
Table 9 - Percent Sexual Aggression Severity at Pre-Test .............................. 81
Table 10 — Percent Rape Myth Acceptance at Pre-Test ................................. 82
Table 11 - Percent Likelihood of Sexually Aggressing at Pre-Test .................... 82
Table 12 — Percent of Responses By Item for Likelihood of
Confronting at Pre-Test ......................................................... 84
Table 13 - Correlation Matrix for Variables Used in Multivariate Analyses 85

Table 14 - Mean Scale Scores for Outcome Variables at Pre-Test
and Post-Test ..................................................................... 88

Table 15 - Percent Rape Myth Acceptance at Pre and Post-Test ....................... 89

Table 16 - Percent Likelihood of Sexually Aggressing at Pre and
Post-Test ......................................................................... 90

vi

Table 17 - Percent Likelihood of Confronting at Pre and Post-Test .... ............... 93

Table 18 - Summary of Sequential Regression Analysis for
Variables Predicting Participants’ Likelihood of Sexually

Aggressing (Post-Test) ......................................................... 96
Table 19 - Summary of Sequential Regression Analysis for Variables

Predicting Participants’ Likelihood to Confront (Post-Test) .............. 97
Table 20 - Summary of Sequential Regression Analysis for Variables

Predicting Participants’ Rape Myth Acceptance (Post-Test) .............. 99
Table 21 - Factors Determining Individual Utilization .................................. 101
Table 22 - Factors Determining Team Utilization ....................................... 102

Table 23 - Factors Influencing Respondents’ Decision to Confront
a Teammate ...................................................................... 103

Table 24 - Sexual Assault Prevention Programs Targeting College
Students ........................................................................... 166

vii

INTRODUCTION

It is now known that sexual assault on college campuses is a major social
problem. Recent studies suggest that undergraduate women are sexually victimized at
three times the rate of women in the general United States population, and at least one in
three undergraduate women experience an attempted or completed rape before leaving
college (Boeringer, 1999; Koss, Gidycz, & Wisniewski, 1987; Meuhlenhard & Linton,
1987). Based on these findings, a number of colleges and universities across the country
have implemented sexual assault prevention and education programs targeting the general
student population. Although college students have been the recipients of a variety of
prevention programs, specific sub-groups of the college population (e. g. fraternities and
sports teams) have traditionally either missed general programming or have not received
programming that is specifically tailored to their group’s reality. For example, sexual
assault awareness programs are often targeted to first-year students on a voluntary basis,
so students who elect to not attend these programs (who may ofien be men), or who are
hard to access because of a busy schedule (e. g. athletes) do not receive any programming.

However, recent theories about the etiology of sexual assault have increased the
focus of speculation about the perpetration of sexual assault to certain “high—risk” groups
of students (Sanday, 1990). Male student athletes are one of the sub-groups most
commonly assumed to be at a high risk for perpetration of sexual assault, because of their
direct involvement in a culture that has been known to foster an aggressive and

competitive environment (Curry, 1991; Messner, 1992; Nelson, 1994). Factors that are

both coached and expected within the world of sport, such as aggression and conquest
over others, as well as the enforced belief that male power over women is natural and
sexualized (Messner, 1992; Messner & Sabo, 1994), have led the general public, media,
and some researchers to believe there is a connection between participation in all-male
sports and the perpetration of sexual assault.

Although there is a growing body of literature that makes connections between
the sports culture and sexual assault, there is only a handful of prevention and education
programs that work directly and specifically with male student athletes. Although these
programs have reported having good outcomes and reputations with university
administrators and students, none of the programs published in the literature has been
evaluated for its effectiveness.

The present study undertakes the evaluation of a college-based rape prevention
program targeting male student athletes. The program is one of the few that targets males
as a single-sex group in team-based workshops incorporating group facilitation and
interactive presentation materials. It was developed in an effort to specifically address
the unique culture that male collegiate athletes are a part of, and deliver the information
in an effective way. The current evaluation adds not only to the general knowledge base
of research on program content and presentation of information, but allows for a more
critical look at what extraneous factors may predict the effectiveness of education
programs that target male collegiate athletes. Considering the tailored nature of this
program, the effectiveness of such an approach is important to understand to direct the
efforts of fiiture programs targeting male collegiate athletes. It is necessary to not only

determine if specific factors related to being a male collegiate athlete are related to

program effectiveness, but also whether or not targeting male collegiate athletes for rape
prevention education is even a worthwhile endeavor.

The first section of this paper examines the definitional process that sexual assault
has undergone over the past 30 years. There have been a variety of opinions about how
sexual assault should be defined for use in research, and the specific definition that is
used for this study, in addition to the development of presentation materials, is explained.

In addition to there being a debate about how sexual assault should be defined, an
even larger debate has surrounded the explanation for the perpetration of it. The second
section gives an overview of the main theories of the etiology of sexual assault, and the
underlying rationale for targeting male collegiate athletes with specifically designed rape
prevention programs that differ from programs targeting the general male student body.
The main theory reviewed is one put forth by Peggy Sanday (1990) in which America as
a whole is deemed as a “rape culture,” or a culture that condones and facilitates the
existence of rape. She and other researchers have hypothesized that certain groups within
this rape culture (e. g. fraternities and male student athletes) are more prone to condone or
commit sexual assault. A review of the research supporting the male collegiate athlete
sub-culture as a rape culture is provided.

Although there is evidence that the male collegiate athletic culture is conducive to
increased sexual aggression against women, the factors connecting sports participation
with sexual assault have been equivocal in the research. The third section of this paper
reviews the recent research, theory, and debate on the specific connection between male
collegiate athletes and sexual assault. The quantitative evidence of a variety of specific

factors related to sexual aggression is presented, and suggestions of other possible factors

are presented for firrther research. It is these sport-specific factors that may serve as
predictors of a prevention program’s effectiveness. Therefore, these sport-specific
factors were measured and tested as predictors of the current programs’ effectiveness at
attitude change.

Because the general link between student athletes and increased sexual aggression
has provided evidence that athletes are indeed a higher risk population, there have been a
few prevention and education programs developed that target male student athletes.
However, in order to draw a distinction between these programs and programs that target
the general student population, it was necessary to first gain an understanding of the rape
prevention programs targeting the general college student population. There have been
numerous programs developed targeting both male and female college students, and
certain program formats and content have been found to be more effective with specific
groups than with the general student body. For example, researchers have shown that,
under certain conditions, men have less attitude change after a workshop than do women.
As a result, there are programs targeting only men that incorporate specific presentation
styles and formats to which men are more responsive. These findings within the general
rape prevention literature help in the development of programs specifically targeting male
collegiate athletes. Although specifically designed for male athletes, programs are based
largely on techniques and materials used by programs targeting the general student body.
It is therefore essential that the effectiveness of using general programmatic planning
techniques for the implementation of a program specifically targeting male collegiate

athletes is determined.

The last section of the literature review describes the two main programs that
have been developed and implemented specifically for male collegiate athletes. Although
both of these programs report having great rapport with the school administration and
athletes, neither one has been formally evaluated for effectiveness at attitude change or
knowledge increase. Additionally, although one of the programs emphasizes the peer
culture in which male student athletes exist, no evaluation of the effectiveness of this type
of program focus has yet been evaluated. A discussion of the possibilities that exist for
group change when individuals participate in programs like those that target collegiate

athletes is then presented, as well as a description of the current program evaluation.

Chapter 1

OVERVIEW OF THE LITERATURE

Defining Sexual ASM

Over the past 30 years, there have been a variety of opinions about how sexual
assault should be defined for use in research. Much of the debate about the definition of
sexual assault started and has centered on the largest and most widely cited survey of the
prevalence of sexual assault on college campuses was administered by Koss and
colleagues in 1987. In this landmark study, a self-report questionnaire was filled out by
6,159 students (3,187 women and 2,972 men) in 32 institutions of higher learning located
across the United States. About 25% of the men in Koss’s original sample (Koss et al.,
1987) reported engaging in some form of coercive sexual activity since the age of 14.
Part of what has been most controversial about Koss et al.’s (1987) study is how sexual
assault was measured. A range of coercive sexual behaviors were included in the scale,
in order to reflect the dimensional view that sexual assault is a behavior that lies at one
extreme of human sexual behavior. The two original goals for the creation of the scale
were to 1) develop an instrument to collect data supporting the dimensional viewpoint
and 2) develop an instrument capable of reflecting the large number of rapes that go
unreported in the United States (Koss & Oros, 1982). Therefore, the scale captures a
wide range of sexual behaviors that some people have considered unrelated to rape
(Gilbert, 1991b; Sommers, 1994), while others consider them to be rape, whether or not

the woman perceives it as so (e. g. women talked into having sex against their wishes).

For example, Sommers (1994) claims that if a woman does not label an incident as rape,
the researcher has no right to call her a rape victim, based on certain behaviors that took
place in the situation. Despite this conceptual turf war, multiple people in the field of
sexual assault research have maintained a dimensional view, which is still the most
widely cited and used measure of rape prevalence in the US. (Bohmer & Parrot, 1993;
Schwartz & DeKeseredy, 1997).

More recent studies on the prevalence of college sexual assault have measured
women’s and men’s experiences of a broad range of behaviors, rather than the classic
legally-bound definitions of rape or criminal sexual conduct. Studies investigating the
prevalence of not just rape, but all forms of sexual assault, including sexual coercion,
have helped enlighten our understanding of the prevalence of sexual assault in America’s
universities. With the use of a range of sexual behaviors to determine prevalence rates,
many other researchers have found rates to range between 15% and 25% (Kanin, 1957;
Koss & Gidycz, 1985; Koss et al., 1987; Makepeace, 1986; Muehlenhard & Linton,
1987; Rivera & Regoli, 1987).

Expanding the concept of sexual assault has taken years. For years American
society has perceived rape through stereotypic images (Lonsway, 1996). Rapists have
been believed to be the outcasts of society — those who are often linked to crimes because
of their income level, ethnicity, or mental health. We have been overwhelmed with news
reports of serial rapists and deranged attackers — images that reinforce the belief that if
women just avoid dark alleys and stay safe in their own homes and private spaces, rape

will not affect them.

This “stranger rape” ideology shaped American society’s understanding of sexual
assault for many years, and only recently is beginning to change with more evidence that
rape and sexual assault are much more commonly acted out among acquaintances. We
now know that the majority of rapes (80-90%) are perpetrated by known acquaintances
(Koss,1987; Lonsway, 1996; National Victim Center,1992; Warshaw,1988), and
gradually, the general public as well as the media, medical, and legal professions are
becoming more aware of the pervasiveness of acquaintance rape.

However, broadening peOple’s understanding and definition of sexual assault to
include behaviors such as sexual coercion and sex that is not penile-vaginal has been
problematic. Many researchers and theorists still vary on their definition of sexual
assault, with the main commonality being an emphasis on nonconsensual sexual conduct
(Bohmer & Parrot, 1993; Schwartz & DeKeseredy, 1997). Feminist scholars and
activists have argued for years that a broader definition of sexual assault that includes a
range of unwanted sexual experiences is less judgmental and more encompassing of the
violations that women endure throughout their lives, both emotionally and physically.
For example, the use of alcohol and other drugs to facilitate a rape is increasingly
common on college campuses (Koss & Gaines, 1993). Many women who have been
drugged are not aware that what they experienced would be considered rape, largely
because our society is slow to name unwanted sexual contact as “rape.” Often women
expect sex to be somewhat “unpleasant” or “aggressive” with men, since American
mainstream culture stresses heterosexual sex as being a man’s ultimate goal and source of
pleasure, rather than something a woman should actively want or pursue (Brownmiller,

1975). This presents barriers to naming unwanted sex as rape, since these beliefs are so

embedded in American culture. By capturing the totality of unwanted experiences that

women endure, we gain a more representative picture of what college sexual life can be
like, rather than excluding those equally devastating experiences that may not fit strictly
legal definitions of rape or criminal sexual conduct (e.g. men’s use of alcohol or threats
to gain sex).

For the purposes of the current project, the definition of sexual assault that was
developed by Koss et al. in 1987 and used by many other researchers since then (e. g.
Schwartz & DeKeseredy, 1997), was used. This understanding breaks sexual assault into
four categories:

1) Unwanted sexual contact — includes unwanted sex play (fondling, kissing, or
petting) arising from menacing verbal pressure, misuse of authority, threats of
harm, or actual physical force.

2) Sexual coercion — includes unwanted sexual intercourse arising from the use
of menacing verbal pressure or the misuse of authority.

3) Attempted rape — includes attempted unwanted sexual intercourse arising from
the use of or threats of force, or the use of drugs or alcohol.

4) Rape — includes unwanted sexual intercourse arising from the use of or threats
of force and other unwanted sex acts (anal or oral intercourse or penetration
by objects other than the penis) arising from the use of or threat of force, or

the use of drugs or alcohol.

The Etiology of Sexual Assa_ult_

In addition to the debates surrounding the definition of sexual assault, there have
been a variety of theories about its etiology. Most theories of perpetration have focused
on one or more of the following factors: 1) the individual, psychological make-up of
perpetrators, 2) group dynamics, and 3) sociocultural phenomena. At this time, most
researchers and theorists have come to an agreement that sexual assault is
multidimensional in nature (Schwartz & DeKeseredy, 1997; Thorne-F inch, 1992;
Malamuth, Stockloskie, Koss, & Tanaka, 1991; Craig, 1990; Baron & Strauss, 1989;
Stanko, 1990). Individual, group, and sociocultural factors play a role in sexual and
physical aggression by men against women.

One example of the interrelatedness of these three factors is how group and
sociocultural factors support the individual factors associated with sexual assault, by
giving a person who is more prone to perpetration the environment and cultural support
to actually carry out the assault. The individual component of a rape culture that is
studied the most in relation to sexual assault is a person’s adherence to rape myths. Rape
myths are “prejudicial, stereotyped, or false beliefs about rape, rape victims, and rapists”
that are present in the general American culture, yet widely held by certain individuals
(Burt, 1980, p. 217). Burt (1980) developed the first scale to measure an individual’s
acceptance of rape myths, which included statements such as “A woman who goes to the
home or apartment of a man on their first date implies that she is willing to have sex,”
and “Many women have an unconscious wish to be raped, and may then unconsciously
set up the situation in which they are likely to be attacked.” Lonsway & Fitzgerald

(1994) have refined the definition of rape myths as “attitudes and beliefs that are

10

generally false but are widely and persistently held, and that serve to deny and justify
male sexual aggression against women” (p.134). Rape myths are the most commonly
measured variable in the evaluation of rape prevention programs, as an individual’s level
of adherence to rape myths has been found to be associated with self-reported likelihood
to sexually aggress (Koss & Gaines, 1993).

Another aspect of an individual’s belief system associated with increased sexual
aggression is an individual’s acceptance of the use of violence in both general society and
interpersonal relationships. Burt (1980) also was the first to develop a scale that measured
a person’s acceptance of interpersonal violence, and found that this construct was the
strongest attitude predictor of rape myth acceptance. These findings have been replicated
numerous times within the literature on rape prevention (Lonsway, 1996). Individual
factors involved with rape perpetration, such as those described above, have been the
main focus of empirical sexual assault research, as well as rape prevention interventions.
However, there is a wide range of literature that focuses on how the group and cultural
components of a society play a role in an individual’s choice of behaviors. It is these
group/cultural components that have added to the speculation that male athletes are
especially prone to sexually assault women, since they exist in a culture that is so
dependent on peer-group support and teamwork. The understanding of the contribution
of group or cultural factors to the occurrence of sexual assault was developed in the past
fifteen years, mainly through qualitative research.

One of the most prominent and widely cited cultural theories of rape perpetration
was developed by anthropologist Peggy Sanday in 1990 in her landmark qualitative study

of fraternities, Fraternity Gang Rape (Sanday, 1990). Sanday studied the existence of

11

rape in different cultural communities across the world, and based on those experiences
abroad and her interviews and observations of fraternity members on American college
campuses, she dubbed certain cultures “rape cultures.” Sanday’s claim that American
society is a “rape culture” that supports the facilitation of rape is supported by previous
writings about sexual assault. In her landmark book, Against Our Will, Susan
Brownmiller (1975) asserted that rape is a “conscious process of intimidation by which
all men keep all women in a state of fear” (p. 15). Martin and Hummer (1989) published
a qualitative look at fraternity culture, and suggested that fraternities “foster and
strengthen beliefs and attitudes supportive of sexual violence” (Boeringer, 1999, p.83).
The theory of “rape culture” supports the contention that sexual assault is not just a result
of one person’s biological make-up or psychological disposition; rather it is a behavior
that is socially encouraged (Crosset, et al, 1995; Brownmiller, 1975; Russell, 1975;
Sanday, 1981, 1990). Most definitions of a rape culture stem from Sanday’s (1981) study
of ninety-five band and tribal societies, in which she classified 47% as rape-free and 18%
as rape-prone (Sanday, 1996, p.193). Cultures that displayed a high level of tolerance for
violence, male dominance, and sex segregation had the highest frequency of reported
rape (both individual and gang) (Crosset, 1995). Sanday (1996) describes societies that
are rape prone as those in which “the incidence of rape is reported by observers to be
high, or rape is excused as a ceremonial expression of masculinity, or rape is an act by
which men are allowed to punish or threaten women” (Sanday, 1996, p. 193).

Schwartz and DeKeseredy (1997) have expanded Sanday’s (1996) concept by
specifying the cultural components that make up a rape-prone culture. Among the

multiple factors involved, there are a few that are more salient to the world of male sports

12

than others. First, the existence of a “courtship patriarchy” is an important component of
the rape culture (Schwartz & DeKeseredy, 1997). Courtship patriarchies exist in
societies that hold customs, beliefs, and opinions about the man’s role of dominance in a
dating relationship. Expecting men to pay for dinner and a movie (the date), and
expecting women to wait for men to open car doors are both examples of how the
courtship process consists of traditional rules of behavior. Many people think these rules
make life easy for women, because the man is “taking care” of her. However, these
behaviors set up a dangerous power dynamic, as societies often are accustomed to
engaging in behavioral “exchanges,” rather than a simple favor without anything
expected in return. Schwartz & DeKeseredy (1997) comment that “the problem here as it
relates to [sexual assault] is that people often feel that under the rules of what we will call
‘courtship patriarchy,’ the male under many circumstances is entitled to sex provided by
the female” (p.62). Both men and women may at times feel pressured into ending a date
with sexual intercourse or contact, although one or both parties may not want to engage
in such behavior.

Although the concept of a courtship patriarchy was originally developed in
reference to general societal practices, there are some components of the male collegiate
athlete experience that, when combined with a general culture supportive of courtship
patriarchies, may exacerbate women’s vulnerability. Particularly, the loss of
accountability and personal responsibility that often accompanies being a male student
athlete (particularly on scholarship), may be a prominent factor involved with an athlete’s
perception that he has the right to sex from a woman he dates or with whom he comes

into intimate contact. Jeffrey Benedict, in his book Athletes and Acquaintance Rape

l3

(1998) describes that in addition to multiple monetary and academic-related benefits,
certain athletes also are given a type of privilege that may play the largest role in a
player’s misuse of his power in an intimate situation:
In addition to the isolation from typical student responsibilities, athletes
are also relieved from virtually all personal accountability as well...As
citizens, student athletes see their behavior condoned or overlooked when
it violates either campus codes of conduct or local laws. Coaches and

athletic administrators are often content to tolerate otherwise socially
unacceptable behavior in exchange for superb athletic performances.

(p.13)

Other components defining a rape culture in our general American society may
also be exacerbated through the world of male sports. Two of them stem from slightly
different roots, yet result in a complementary oppression of women. The extent to which
parents raise their children with traditional sex role beliefs and a mass media that
sexualizes women as objects, combine to characterize the perceptions and expectations
that men. and women have for their own role and the role of the opposite sex in intimate
relationships and situations. A rape culture is often easily defined by its media images of
women and their bodies in sexually demeaning and/or objectified roles. Examples of
women in objectified positions often arise in all-male social gatherings, where fratemal
bonding is at its highest point. Ice cube trays in the shape of nude women, glasses in the
shape of a breast or with pictures of sexy women whose clothes disappear when the glass
gets wet are all examples of objects turned into a representation of women’s bodies
(Schwartz & DeKeseredy, 1997). Women are often reduced to just a part of their body —
a breast, a nipple, armless, or headless. The pornographic industry and some
advertisements for commercial products often depict women as sexual objects. In fact,

pornography has been a primary target of feminists and anti-rape advocates for years.

14

They have argued that in heterosexual male-produced pornography, women exist to serve
and service men, and that to some men “that means that women are fair game to rape or
sexually abuse” (Schwartz & DeKeseredy, 1997, p.85). MacKinnon (1993) has made
some of the most widely cited and publicized arguments against pornography, contending
that pornography silences women and creates a hostile environment for women within
general societal living (e. g. attaining jobs, housing, etc). Baker (1992) clearly describes
the danger imposed to women specifically relating to sexual abuse:

Pornography expresses more than the view that women are mere sex

objects. It tells men that women enjoy sex and are always available for it,
even when they deny it. It tells men that women secretly enjoy rape...

(p. 140)

While this view of women’s sexuality is often portrayed to the general American
public as part fantasy, the availability of women for the sexual taking is often a reality for
certain male student athletes. Highly recruited players are often “exposed to women who
are used to entice players to their school [and] this initial experience with women as a
benefit is expanded over the course of a college career” (Benedict, 1998, p. 14). Benedict
(1998) notes that this practice is not carried out by college athletic departments, but rather
by other players on the team. An NFL player interviewed by Benedict (1998), recalled
the number of women (usually referred to as “groupies”) who were essentially at the
sexual disposal of the team:

There were groupies in college — college groupies. Athletic department

college groupies are usually the ones that you meet on a recruiting trip.

And they’re there and they sleep with all the recruits, or they sleep with

some freshmen. It’s either someone that they [coaches and recruiters] are

trying to entice to come to the school or someone that’s at the school and
they want to keep happy. (p. 15)

15

Not all male student athletes have the “luxury” of available women and additional
privileges, as “groupies” associate more with higher profile athletes and teams (e. g.
basketball, football). However, other teams and athletes may assume that women who
were known to be groupies are also sexual objects for the taking. When societal all-male
group objectification of women is combined with the sexual privileges of certain male
student athletes, it is possible that the extent to which men are trained to objectify women
on the basis of sex creates a danger for those women who say “no.” When women are
seen as objects and not as individuals, the boundaries get blurred between a woman who
wants sex and a woman who wants companionship. This belief is reflected in our
society’s practice of doubting the motives of women who claim higher profile athletes
sexually abused them, particularly if the women are known to be “groupies” or have a
promiscuous sexual history. Unfortunately, athletes’ experiences with groupies have not
yet been empirically studied in relation to their likelihood of sexually aggressing, nor to
the severity of their sexual aggression.

A complementary cultural practice (that impacts individual beliefs) to the sexual
objectification of women is the raising of children with extreme sex-role specific beliefs.
For example, boys and girls who are raised to believe that maternal duties are for girls
and being the family breadwinner is for boys often grow up with certain expectations for
the other sex’s behavior in an intimate relationship (Thorne, 1993). Socialization
theorists have written for years about the harm of growing up with such polarized
expectations of gender, and some have tied this genderization to violence in intimate
heterosexual relationships. Berkowitz et a1. (1994, p.7) argue that “men’s sexuality and

their relationships with women provide a sphere for enactment and confirmation of these

16

traditional gender-role expectations, which assign men the role of aggressor and women
the role of gatekeeper in sexual intimacy.” When the societal sexual objectification of
women is concomitant with the engendered socialization of its girls and boys, individuals
grow up into adults with belief systems supportive of a rape culture.

It has been suggested that male collegiate athletes hold more stereotypic sex role
beliefs for both themselves and the women in their lives (Parrot et al., 1994). Research
has found that men who subscribe to traditional sex roles and male sexual dominance are
more likely than other men to engage in verbal sexual coercion, sexual assault, and rape
(Muehlenhard et al., 1990 as cited in Parrot et al., 1994). It is hard to argue with the
connection of traditional sex role stereotypes and male athletes, since part of America’s
definition of masculinity and “what it means to be a man” includes participation in sports.
The implications for sexual assault are encountered in the fact that “masculinity in sport
has been defined in contrast with, if not wholly separate from, women and womanhood”
(Crosset et al., 1996, p.175). If a man hones his athletic talent enough to make a college
team, he has “made it” into a world that defines its masculinity not only by stereotypical
masculine behaviors (e.g. yelling, grabbing his crotch, being violent and unemotional
unless the emotion is anger), but mainly to the extent it is not feminine. For example, it
is common for coaches and teammates to call a player derogatory names when he fails to
do something correctly in practice or a game. Some of the more common choices for
names are “pussy,” “bitch,” and “girl” (Messner, 1992; Nelson, 1994). One of the
accusations that boys in sports commonly grow up hearing is that they “play like a girl”
or “throw like a girl” if their athleticism is not at par with the expectations of the activity.

This polarization of gender is so extreme in male sports that even homosexuality between

17

men is engendered into being “feminine” and derogatory words about homosexual acts
are often targeted at athletes. Nelson (1994), who has written extensively about the male
sports culture, gives an example of how influential coaches can be in the process of
defining athletes as non-feminine:

Coaches who would be fired for calling athletes “nigger” employ with

impunity such terms as “cunt.” “Faggot” is another popular derogation. It

refers not so much to sex between men as to weakness, timidity,
cowardice — and femininity. . .In response to “wimpy” performances, male
coaches have been known to deposit tampons, sanitary napkins, and bras

in young men’s lockers. “You have debased yourself to the level of a

woman” is the message. (p.87)

Cultural practices such as those described above, have led researchers to label
certain sub-groups within the American rape culture as more “rape prone” because of
their group characteristics often associated with sexual assault (e. g. male bonding,
distinct hierarchies of power and dominance) (Sanday, 1996). The two most studied sub-
groups are fraternities and male sports teams (Boeringer, 1999). Although the current
study focuses on male collegiate athletes as a sub-culture, it is important to review some
of the research regarding the fraternity sub-culture. Fraternities were the first sub-group
to be studied, and likewise, most of the research on the dynamics of rape-prone sub-
cultures comes from the fraternity literature. There are also some important connections
and distinctions to be drawn between fraternity sub-groups and athlete sub-groups.

Initial research with fraternities was more qualitative in nature, which helped
describe and explain the type of culture that existed within fraternity houses. Sanday
(1990) and Martin & Hummer (1989) have been the two main proponents (via

ethnography) of there being a rape sub-culture within fraternities. Both provided

convincing evidence that “fraternities may provide an environment in which rape-

18

supportive ideologies can thrive and be transmitted to others” (Boeringer, 1999, p.83).
Social get-togethers in which fraternity men use alcohol, drugs, and physical and verbal
coercion to “work a yes out,” as well as male bonding around the objectification of
women and “scoring” sexually, all lend to the suggestion that fraternities are conducive to
sexual assault (Boeringer, 1999).

However, quantitative research that has tested the hypothesized connection
between rape myth acceptance, sexual aggression, and affiliation with fraternities has not
provided the convincing support that was expected. While many researchers have not
been able to find a connection between fraternity affiliation and sexual aggression
(Lenihan & Rawlins, 1994; Schwartz & Nogrady, 1996; Boeringer, Shehan, & Akers,
1991; Koss & Gaines, 1993), others have (Boeringer, 1999). Some have even found that
fraternity members were less accepting of rape myths than a non-Greek comparison
group (Schwartz & Nogrady, 1996). Both Schwartz & Nogrady (1996) and Koss &
Gaines (1993) found that alcohol and substance use were stronger predictors of sexual
aggression than fraternity affiliation. However, both had methodological problems,
which may have had an impact on their findings. For example, Koss & Gaines (1993)
warn against inferring too much from their study (1993), because of its low sample size
of fraternity members. Schwartz & Nogrady (1996) caution that “it may well be that the
large number of fraternity men who do not engage in sexually coercive behavior is
masking the effect of those who do” (p. 159).

Other studies have found that fraternity members were more likely to use alcohol,
drugs, and other substances, even though they were less accepting of rape myths. Despite

Koss & Gaines’ (1993) and Schwartz & Nogrady’s (1996) unsupportive findings,

19

Boeringer (1999) did find a relationship between fraternity and athletic affiliation and
rape-supportive attitudes. Boeringer (1999) sampled general attitudes in a sample of 477
college men, and found that men who were members of fraternities scored significantly
higher than the control group (of non-Greek and non-student-athlete men) on 18% of
statements endorsing rape-supportive beliefs. The control group agreed with 8% of rape-
supportive belief statements. Male collegiate athletes endorsed the most rape-supportive
beliefs, agreeing with 56% of the statements (16 out of 28 statements). Boeringer’s
(1999) study joins other quantitative and qualitative studies that have found similar
evidence that fraternity groups hold more rape-supportive ideologies (Bohmer & Parrot,
1993; Schaeffer & Nelson, 1993; Sanday, 1990, Martin & Hummer, 1989). Because the
research findings on fraternities and sexual aggression have been so inconsistent, some
researchers and theorists have turned to a more detailed analysis of what is involved in
the experience of a being fraternity member. Currently, peer group influence is one of
the most prominent theories put forth, which suggests that an individual’s acceptance of
rape myths or increased sexual aggression is largely influenced by a male peer group
culture, in addition to other factors. Schwartz & Nogrady (1996) state quite simply that it
may not be just fraternities, but rather the components that define a fraternity that make
certain groups more rape-prone:

It seems to be possible that other groups on campus may be just as likely

as fraternities to provide extensive male peer support for the sexual

objectification of women, and access to alcohol, that encourages some

men to engage in victimizing behaviors. (p. 159)

Schwartz & DeKeseredy (1997) are two of the main proponents of this idea and

have written extensively about it in their book Sexual Assault On The College Campus:

The Role of Male Peer Support (1997). They note that there are a variety of factors

20

associated with male peer group membership that have a relationship with being rape
prone. For example, a group’s narrow conception of masculinity (e. g. quite accepting of
the idea that males are dominant and females are submissive), group secrecy (e. g. hiding
their peers’ deviant activities), and sexual objectification of women all are part of some
male peer support groups. The concept of peer group influence particularly plays into the
reasoning behind sports teams’ reputation as being the second most common group
implicated in gang rapes (behind fraternities) (N eimark, 1991).

Certain male collegiate athletes exist in a peer group culture that is quite similar to
that of a fraternity group, including the male bonding that leads to the objectification of
women and tight vows of secrecy that prevent exposure (Schwartz & DeKeseredy, 1997).
As discussed earlier, certain male athletes are in an environment that continuously
debases women and polarizes the sexes. However, the most salient part of the collegiate
team athlete peer group culture is the daily training to sacrifice everything to a group goal
and to exist “for the team.” Not only is each athlete a part of a team, he is part of a team
that is trained to react without question to the orders and authoritarian directions of
coaches and team captains. The hierarchy of team membership is set right from the start,
as first year team members frequently go through “initiation rights” at the hands of older
players (much like in fraternities), and the simple fact that most team captains are older,
starting players. The male bonding that is essentially required on collegiate sports teams
can be very powerful, since the athletes live, eat, and work with each other. Schwartz &
DeKeseredy (1997) note that “this group bonding can be so strong that such men are

willing to take part in rape, or to observe rape, or at least to take part in a cover-up,

21

because the alternative is to go against the group. It becomes more important to be part
of the group than it is to do the right thing” (p. 126).

This group influence also may account for the lack of research findings linking
group members’ individual factors to sexual aggression. Boeringer (1999) alluded to the
possibility that individual level factors (e.g. rape myth acceptance, acceptance of
interpersonal violence) may not always have a direct relationship to sexual aggression,
because such strong peer influence may essentially drown out individual decision-making
and behavior in the guise of “being a team player.” Despite this, the relation that group
factors have to sexual assault by members of the group has not yet been empirically
explored. Because athletes are part of such a group-dominated culture (i.e. teams), it is
necessary to explore the group influences involved with their level of sexual aggression
or intervention in a sexually aggressive situation. More discussion about the effects of

group influence will be discussed later in this chapter.

Male Collegiate Athletics as_a Rape-Prone Sub-Culture

There is little doubt among researchers and theorists that the male sports culture is
indicative of a rape culture (Crosset et al., 1996). Male sports are highly sex segregated
and often promote hostile attitudes toward rivals and physical domination over others
(Kidd, 1990; Messner, 1992; Messner & Sabo, 1994). Organized male sports have been
described as supporting male dominance and sexist practices (Crosset et al., 1996;
Bryson, 1987; Kidd, 1990; MacKinnon, 1987; Messner, 1992; Whitson, 1990). Curry
(1991) qualitatively assessed the language and culture within male locker rooms, and

found statements consistent with what would exist within a rape culture. There are

22

perhaps two additional factors that are specific to the collegiate male athlete’s experience
that are not accounted for in the peer group model presented above: lack of accountability
and a struggle for autonomy. Issues of power, dominance, and privilege intertvvine in the
daily life of a high profile collegiate athlete. On one hand, the athlete is often submissive
to the orders of the coaching staff and older players on the team. Sport sociologists have
noted that athletes exist in a life of “structured inequality” in which they are dependent on
their coaches for playing time and exposure. This struggle for social placement has been
suggested to be a contributing factor to male athletes’ sexual aggression against women
off the field (Crosset, 1996). On the other hand, much of the general student body, some
school administrators and staff, and numerous people affiliated with the university often
treat athletes with more favor than non-athlete students, which serves as a constant
reminder that the athlete is in a position of power and acquisition. It is possible that a
athletes’ struggle for autonomy coupled with a heightened status on the college campus
may play a role in some athletes’ violence against women.

Research with college student athletes seems to be taking a similar path as
research on fraternities. There is rich qualitative evidence that the male sports culture is a
rape supportive environment (Messner, 1992, Curry, 1991; Nelson, 1994). However,
there have been equivocal quantitative findings related to collegiate athletic involvement
and sexual aggression. Despite some research supporting a connection between
participation in collegiate sport and rape perpetration, the specific factors that contribute
to sexual aggression remain undetermined (Koss & Cleveland, 1996). In fact, this lack of
specific evidence has created quite a debate among researchers and the general public as

to how or why student athletes may have a higher propensity to rape women. Some

23

claim that athletes are scrutinized more because of their notoriety, which creates a
distorted perception that athletes are assaulting women at a higher rate than the general
population (Dershowitz, 1994). However, members of the press have pointed out that
athletic teams foster a sense of elitism, disrespect for women, and physical domination
over others — which all can lead to a supportive environment for sexual aggression
(Eskenazi, 1990). It turns out that both perspectives are probably accurate. As Crosset et
al. (1996) stated, “The press has overstated the extent of the problem of athletes and
violence against women, but evidence nonetheless suggests an association between
athletic involvement and violence against women” (p. 164).

Studies have found evidence to support both assertions (Crosset, Benedict, &
McDonald, 1995; Koss & Gaines, 1993). In their study on the prediction of sexual
aggression by alcohol use, athletic participation, and fraternity affiliation, Koss & Gaines
(1993) found that the strongest predictor of self-reported sexual aggression in a sample of
530 male undergraduates, was drinking intensity, followed by nicotine use, hostility
toward women, and athletic involvement. No effect was found for fraternity affiliation,
which might be accounted for by the low sample size of fraternity members. For
example, nicotine use contributed almost twice as much Beta weight power to the
prediction of sexual aggression as athletic involvement. However, it must be noted that
athletic involvement included a variety of levels of sports participation. Spectatorship,
informal participation (e. g. pick-up games), club sports participation, and formal sports
participation (i.e. varsity intercollegiate athletics) were all included in the variable

athletic involvement. When this was broken down into specific components, formal

24

sports participation was more strongly associated with sexual aggression than other forms
of athletic involvement.

Crosset et al. (1995) followed up the work of Koss & Gaines by reviewing
campus judicial records of sexual assaults on 30 campuses. Varsity athletes were found
to be over-represented as reported perpetrators of simple sexual assault (one perpetrator
as opposed to a group perpetration). For the combined three years that the judicial files
encompassed, male student athletes accounted for 3.3% of the total male student
population, but were 19% of the reported perpetrators. The authors note that caution
must be taken when interpreting these results, as sample sizes were low (perpetrators n =
69), and the action of women’s reporting a sexual assault to campus police is not yet firlly
understood. The notoriety of a student athlete may either encourage or discourage a
woman’s decision to report the incident to campus police. The authors note that these
findings “do not support the contention that athletes’ violence against women only
appears to be a problem because athletes are being targeted by the media” (Crosset et al.,
1996, p. 175). However, the media may still play a large role in the perceived notoriety of
an athlete, which may or may not affect a woman’s reporting of the incident to campus
pohce.

The emphasis in current research has been to determine whether it is simply
participation in varsity college sports that is related to violence against women, or
whether violence is the result of behavior indirectly related to sport (e. g. hostile attitudes
toward women, peer group support for violence) (Crosset et al., 1996). Part of the debate
about whether or not athletes indeed have a higher rate of sexual assault is possibly due to

the lack of specificity about what components of athletes’ experiences lead to sexual

25

assault. The direct relationship of sport participation to sexual aggression is blurred with
additional factors that need investigation (Crosset et al., 1996). A variety of strategies
have been suggested and used to distinguish between groups on campus that may have a
higher or lower propensity to sexual assault. One suggestion has been to label athletic
teams based on college student perceptions of them as “high-risk” or “low-risk”, with the
recent finding that teams perceived as “high-risk” actually do seem to have a higher rate
of sexual assault (Humphrey & Kahn, 2000).

One of the strategies used to specify a particular component of the athletic peer
group experience has been to group sports according to whether they are contact or non-
contact. Contact sports, which involve players’ physical contact with one another, are
sometimes assumed to be more fostering of aggressive behavior, and therefore, sexual
assault (Caron et al., 1997). One study has shown that collegiate male athletes on
competitive teams that also happen to be contact sports (e. g. football, hockey) are less
egalitarian in their views toward women’s roles than their peers who are part of non-
contact sports (e. g. track and field, swimming, golf) (Caron et al., 1985). However, a
specific connection of participation in contact sports with sexual aggression or
responsiveness to sexual assault prevention education has not yet been supported in the
research (Koss & Cleveland, 1996).

It has been suggested that it is not just whether a man participates in sports (or
even contact sports), but that something else within the athlete’s experience might
contribute to their increased likelihood to sexually aggress (Benedict, 1998; Koss &
Cleveland, 1996). Researchers have reduced the variable of sports participation into

various types of sport participation (e. g. contact/non-contact, club sports, varsity sports),

26

but very few have gone the step further to investigate what exactly about the athlete’s
experience - not just the level of sport played — has a role in possible sexual aggression.
Perhaps it is not just simply the type of sports participation, but rather a combination of
specific athlete characteristics or behaviors associated with sport participation and the
team “type” that are associated with increased sexual aggression toward women.

The only study to date that has looked at additional sport-specific individual
factors is Caron et a1 (1997). Using the Sport Orientation Questionnaire (Gill & Deeter,
1988), they measured not just type of sport participation, but also level of
competitiveness, win orientation, and goal orientation. Their study is the most recent to
distinguish between level of sports involvement (high school v. college), and compare
level of involvement to specific athlete characteristics (e. g. competitiveness). A
significant positive correlation was found between athletic participation in high school
and competitiveness, but no significant correlations were found between athletic
participation and win orientation or goal orientation. However, the relationships between
sport participation in college and scores on all three of the Sport Orientation
Questionnaire’s subscales were positively significant, indicating that intercollegiate
athletes have more extreme levels of sport-specific personality characteristics (6. g.
competitiveness).

It is important to view these results in consideration of how such athlete-specific
traits may play into abusive behaviors toward women (e. g. sexual assault). Many athletes
are taught to use force to settle a conflict or problem (Caron et al., 1997), and
competitiveness is necessary to win over an opponent. Winning is known to be the most

important part of competition for most athletes (Kang et al., 1990; Toufexis, 1990) and

27

 

athletes learn to use their bodies as a “tool, machine, or even a weapon” in order to
“defeat the objectified opponent” (Messner, 1987, p.59). When men see women as an
opponent, or see sex as an accomplishment or competition, such views at most may
seriously increase a man’s ability to behave in sexually assaultive ways toward women.
At least, such views may impede a man’s likelihood of understanding issues related to
sexual assault against women, especially if such aggressive behaviors are considered
normative in the athletic culture in which he exists.

Although the above findings are helpfirl in determining the intra-individual factors
involved in an athlete’s sexual aggression, there has been no empirical research on the
group factors that may play a role in an athlete’s decision to sexually aggress or prevent a
teammate from sexually aggressing. Although most of the qualitative accounts of the
male athlete culture describe the strength with which the team holds together and
enforces uniform behavior among its members, empirical studies have only measured
intra-individual characteristics of each player (Crosset et al., 1996). It would be helpfiil
to measure to what extent the team component of an athlete’s collegiate sport experience
has on his willingness to take part in sexually aggressive activities or to intervene with
teammates who are doing something sexual to a woman who is not willingly

participating.

Rape Prevention Education
Although researchers have not identified the specific causal link between collegiate male
sport culture and sexual aggression, enough evidence of a prevalence of sexual assault by

athletes has been found to support the need for prevention education among student

28

athletes (Crosset et al., 1996; Crosset et al., 1995; Koss & Cleveland, 1996; Koss &
Gaines, 1993). There are a few rape and dating violence prevention programs in the
country that have targeted college student athletes (Parrot, 1994; Katz, 1995). These
programs will be discussed in more detail in later sections, as first a discussion of the
rationale and findings related to sexual assault prevention education is necessary.
Although most prevention education programs have not focused on collegiate athletes,
there are findings within the field that can be used in the design of effective programs
with collegiate athletes.

Over the past twenty-five years there has been an increasing number of studies
and theoretical papers focusing on rape and its prevention. A variety of disciplines,
including social work, psychology, criminology, and sociology have looked at the issue
of rape prevention through different lenses of analysis. Traditionally, researchers and
theorists have focused the prevention of sexual assault on either individual “risk-factors”
(a man’s poor impulse control, or a woman’s ‘provocative’ attire), or specific situational
factors (drinking and dark alleys). A sample of prevention program strategies has
included: increased campus lighting, self-defense training, educational presentations for
high school and college students, media campaigns, and role-playing. In their 1987
review of rape prevention strategies, Fischoff, Furby, and Morgan found a total of 1,140
possible strategies.

Femle-focused prgggms. In the last ten years, in response to the findings that
college women are more at risk for being sexually assaulted than any other age group
(Koss, 1987), there has been an increase in the number of programs that target female

college students. Many colleges have attempted to educate women about the risks

29

associated with sexual assault and “protective measures” they can take to increase their
safety. Unfortunately, most colleges lack programs aimed at men and their own role in
campus sexual assault. Despite some philosophical support of programs targeting
females (Breitenbecher & Scarce, 1999; Hanson & Gidycz, 1993; Yeater & O’Donohue,
1999) there has been a sea of debate surrounding the ethical underpinnings of these
programs as well as doubt as to their ultimate effectiveness in preventing rape. A wide
number of researchers and theorists have been involved in the discourse around the
efficacy of female-focused interventions (Katz, 1995; Funk, 1993; Berkowitz,1994;
Lee,l987; Schmidt & Peter, 1996; Matabuti,1993; Lonsway et al.,1998; Foubert &
Marriot,1996; Ring & Kilmartin, 1992; Schewe & O’Donohue,1993 and 1996;
Dworkin,1993). These writers have all contended that when we limit intervention
programs to women and girls, and teach them to behave in “rape appropriate” ways, we
are sending the message that they can prevent a rape, and that they themselves are in
control of whether they are raped or not. As F eltey et al. (1991) put it, “When only
females are educated and informed, they are ultimately being held responsible for their
own victimization.” Lonsway (1996) has called such “prevention” efforts more of a
deterrent approach, rather than actual prevention:

Because men who rape select potential victims on the basis of

vulnerability (Brownmiller, 1975), it makes sense that a deterred

attempt will only result in the victimization of another, more

vulnerable individual. Rape deterrence strategies can therefore

only protect individual women (albeit with no guarantees), but can

never reduce the vulnerability of women as a group (p.232)

Schewe & O’Donohue (1996) reason that:

Not only are these forms of “prevention” unacceptable on the basis
of a severe restriction of basic human rights, but also these tactics

30

will never be completely successful. Despite the fact that women

take these and other precautions, the fact remains that no one can

be constantly and perfectly vigilant. Thus, no matter how well

trained potential victims become in avoidance, escape, and

physical self-defense, they will be vulnerable to rape to the extent

that there are men who will attempt to commit acts of sexual

violence (p.456)

While such programs give women the responsibility for stopping rape, they also
ignore the fact that violence against women concerns men as well as women. Katz
(1995), who has deve10ped an extensive violence prevention program for male collegiate
athletes, considers this lack of male inclusion one of the main factors involved with
violence against women. He states, “Calling this violence a “women’s issue”. . .is in fact
part of the problem. Why? It sends a signal to guys that it is not our concern: Why would
a man concern himself with women’s issues?” (p.163)

Mixed-sex programs. Katz, as well as a good number of other researchers and
activists (Berkowitz,1994; Dworkin,1984; Foubert & Marriot,1996; Funk, 1993;

Lee, 1987; Lonsway,1998; Matabuti,1993; Ring & Kilmartin,1992; Schewe &
O’Donohue, 1993 and 1996; Schmidt & Peter,l996) has started approaching the issue of
sexual violence prevention from a male inclusionary perspective. These researchers have
acknowledged the fact that when a woman is raped it affects men as individuals and as a
whole, and there is sufficient reason for men to become involved in the movement to end
rape. As an effort to include males in the prevention of rape, mixed-sex programs as well
as male—focused programs have been developed and some evaluated over the past fifteen
years.

Operating on the assumption that both males and females will benefit from a

change in rape-supportive attitudes, many programs have been set up that target a mixed-

3l

sex audience, usually in a classroom setting. This notion of mutual benefit has been a
topic of theoretical discussion within the literature for years (Berkowitz, 1994), and has
more recently been applied to the study of rape prevention education.

A common format for mixed-sex interventions has been a lecture-based
presentation with or without the use of brochures and efforts to reduce rape myth
adherence. One of the most widely used techniques in rape prevention programs is the
challenging of adherence to rape myths, as well as a measure of a person’s acceptance of
interpersonal violence and hostility toward women. Most mixed-sex programs that entail
some form of dispelling rape myths report positive short-term outcomes among women
and occasionally some men. However, men in general are more resistant to rape
prevention programs, adhere more strongly to rape myths than do women, and support
sexually coercive behavior (Buckley & Masters, 1992; Feltey, Ainslie, & Geib, 1991;
Heppner, Humphrey, Hillenbrand-Gunn & DeBord, 1995; Lenihan, Rawlins, Eberly,
1992; Proto-Campise et al., 1998). So while women leave the program with a significant
amount of positive attitude change, men do not always have the same magnitude of
attitude change, and may even become more resistant to attitude change (Berg, 1993;
Borden, Karr, & Caldwell-Colbert,1988; Krulewitz & Kahn,1983; Lenihan et al., 1992;
Proto-Campise, 1998) . In addition, it has been shown through follow-up surveys that
men have a higher “rebounding” rate, or return to original ideals than women after an
intervention (Heppner et al., 1995). This has consistently been found in the literature for
over thirty years (Borden et al., 1988; Lonsway, 1996).

One way of teasing apart the gender difi’erences has been to differentiate between

rape-specific attitudes and more general, contextual level variables, such as sex role

32

stereotyping (F onow et al., 1992). Rape specific attitudes have mainly been measured by
the rape myth acceptance scale developed by Burt in 1980, or a variation of the scale into
some related form. Contextual-level variables have included measurement of attitudes on
a variety of social and interpersonal beliefs (e. g. sex role stereotyping, hostility toward
women, and acceptance of interpersonal violence). It has been widely supported in the
literature that adherence to contextual-level variables is strongly related to adherence to
rape myths and more rape supportive attitudes (Fonow et al., 1992). However, the
majority of evaluated mixed-sex programs that report some reduction of rape-supportive
attitudes also have found little or no significant change in contextual-level attitudes for
both males as well as females (Dallager & Rosen, 1993; F onow et al., 1992; Lonsway,
1996). Because contextual-level variables make up the day-to-day worldview of an
individual, it has been suggested that only programs that include an in-depth discussion
of societal norms and stereotypes can accomplish this level of attitude change (F onow et
al., 1992; Lonsway, 1996).

The gender gap that exists in males’ and females’ understanding and acceptance
of rape-supportive attitudes has been well documented (Proto-Campise,1998), and
researchers are now beginning to acknowledge that rape prevention education needs to
take into account the specific needs of males and females (F onow et al., 1992). However,
this shift has been slow in coming. The majority of programs still have outcomes that
reflect this gender difference in attitudes, and researchers have continuously
acknowledged that it is a problem, yet have rarely gone beyond a mention to create a

conducive solution.

33

 

_M_ale-focused programs. A number of strategies have been implemented to tap
into the complex attitude changes required of males. The recent emergence of all-male
(or male-focused) programs has brought some insight into more effective strategies for
attitude change with males. As a whole, male-focused programs are more intense and
interactive than mixed-sex or all-female programs, with the intent of helping men
recognize and understand the implications rape has on themselves and the loved ones in
their lives. Researchers are now starting to recommend the use of all-male programs
mainly because their interactive and intense atmosphere affords the focus and attention
on male-specific culture and constructs (sex roles, masculinization) that may promote
significant behavioral change (Yeater & O’Donohue, 1999).

One factor that seems to be stressed more than others in male—focused programs
is generation of empathy for a victim of sexual assault. Lonsway (1996) notes that
“because empathy leads to more situational attributions for the behavior of others, it is
reasonable to suggest that individuals with more empathy for victims will be less likely to
attribute rape to personal characteristics such as victim appearance or behavior, and more
likely to blame the man or society for the attack” (p.249). Empathizing with the victim
and women’s constant threat of rape also helps men understand that any woman in their
lives may have been or may be in the fiiture a victim of rape.

In her review of acquaintance rape education programs, Lonsway (1996) found
that three programs for men have reported desirable attitude change with a strategy of
empathy induction (Gilbert et al, 1991; Gray et al, 1990; Lee, 1987), whereas another two
have not (Berg, 1993; Ellis et a1, 1992). The reason for this difference is not very clear,

since both sets of programs used similar techniques for inducing empathy. All of these

34

programs used empathy inducing techniques that included having the group of male
participants either listen to, read about, or watch a video of another man or boy tell his
story and experience of being a victim of sexual assault. This strategy of “seeing oneself
as a potential victim” was used in these programs to induce empathy fiom the
participants, and conclusively lead to higher sensitivity to issues around rape and less
rape myth acceptance. However, only the first three programs (Gilbert et al, 1991; Gray
et al, 1990; and Lee, 1987) reported a positive outcome and increase in empathy for rape
victims, whereas the other two programs (Berg, 1993; Ellis et a1, 1992) reported no
increase in empathy. In fact, Berg et al (1993) had divided the intervention into two
groups of men: one that heard a woman’s story of being assaulted, and another that heard
a man’s story of being assaulted. Both groups reported no significant change in victim
empathy, while the group that heard the woman’s story actually reported a greater
likelihood of sexual aggression than before hearing the story.

Berg (1993) and Lonsway (1996) have both hypothesized as to why men may
have trouble at first with victim empathy. One idea is that since acquaintance rape is
largely seen as a crime in which women are the victims, men automatically assume the
role of observer or aggressor (Berg, 1993). Programs that are more intense and involve
personal interaction may allow men the freedom to work out feelings of anxiety and
resistance to prevention programs in general.

This need for increased intensity in education programs involving men is
something that has been voiced by many researchers. Borden et al. (1988), after
unsuccessful efforts at changing attitudes through lecture presentations, argued for “new,

more dynamic, vivid interactive program formats to enhance the desired efl’ects of

35

consciousness raising, attitude change, and empathy toward rape” (p. 135). This
suggestion, coupled with the finding that men do not easily empathize with rape, may
account for the fact that there is so much resistance and lack of attitude change among
men.

A common strategy for enhancing intensity is interactive participation by the
participants. Programs that include some form of participant interaction generally report
outcomes with desirable attitude change (Lonsway, 1996). Interactive program
techniques may entail small discussion groups, role-playing (Gilbert et al., 1991; Gray et
al., 1990; Ring & Kilmartin, 1992), and even interactive dramatic performances (Mann et
al, 1988). The importance of such interactive participation in programs that involve
changing men’s attitudes, is just now being fully understood. For example, a program
evaluated by Gilbert, Heesacker, & Gannon (1991) presented a combination of lecture-
based information about rape as well as empathy induction techniques to 75
undergraduate men. Male participants reported feeling more comfortable with the
presented lecture material only after partaking in the discussion group. Lonsway (1996)
notes that “such interaction apparently played an important role in addressing men’s
concerns, thereby easing their discomfort” (p.248). This “discomfort” may be part of the
resistance seen from men to empathy-inducing strategies, as well as prevention programs
in general. Since most interventions are short-term and are held in a mixed-sex
classroom setting, men may be dealing with a wide range of emotions that are brought up
by the setting of the program itself. Theorists like Funk (1993) have noted how intense
feelings of guilt often encompass men when the subject of rape is addressed. This guilt is

uncomfortable, and to avoid this discomfort, men often choose to resist what messages

36

are causing their feelings of guilt. Discussion groups and more interaction among
audience members may help to override men’s initial guilt that provokes resistance to
sexual assault prevention programs.

Another benefit of all-male groups is that in some ways they replicate the all-male
groups in which rape-prone male bonding occurs. Especially in relation to student
athletes, who exist as a unified group (i.e. team) through most of their years in college,
this all-male group structure is a place to more appropriately address many of the
activities than male peer groups engage in and feelings that men may have that are
conducive to a rape prone sub-culture. Yet, only one program targeting male collegiate
athletes has discussed sexual violence against women in the context of peer group
influence, and this program has not yet evaluated its effectiveness at attitude or behavior
change (Katz, 1995).

groggms targeting collegiate athletes. There are a few rape and dating violence
prevention programs in the country that work with male collegiate athletes. A few have
been described in the literature, yet none of those published have been formally
evaluated. One prevention and education program was designed for and used with
student athletes at Cornell University in 1990. Both the men’s rowing team and football
team participated in the workshop, the goals of which were to “educate students about the
risks and consequences of committing sexual assault, the behaviors that can lead to rape,
and the consequences of these acts” (Parrot et al., 1994, p. 180). The program also
attempted to “challenge attitudes that demean or objectify women and the messages and
norms that condone and perpetuate a rape-prone environment” (Parrot et al., 1994,

p.180).

37

According to the authors, this program was well received by both the players and
the athletic department, and was given firnding to continue in firture years with certain
teams. Although the authors were able to suggest certain steps to take in the development
and implementation of a program with collegiate athletes, actual attitude or knowledge
change among the athletes was not formally evaluated. In addition, the narrow scope of
the program (only rowing and football took part in the program) disallowed any
exploration of sport-specific factors that may play a role in the effectiveness of such a
program (e. g. contact or non-contact sport, team influence).

Katz and his colleagues (1995) have designed and implemented perhaps the most
extensive sexual assault education program with male student athletes. The Mentors in
Violence Prevention Project (MVP) trains student athletes (both male and female) to
work on campus to prevent rape, battering, sexual harassment, and other forms of
violence against women (Northeastern University Student Handbook, 1999).

The MVP program is focused on the underlying assumption that the
“sociocultural construction of manhood is central to the problem of men’s violence
against women, as well as the basis of potential sources of prevention” (Katz, 1995,
p.163). For the men’s program, two male facilitators go into the locker rooms of men’s
college teams and do a series of workshops with each team. They work in an all-male
environment and incorporate exercises that induce empathy, increase interaction among
the team members, and directly address issues of masculinity, male peer groups, and
athletes’ celebrity role and power on the college campus. One of the more innovative
characteristics of the MVP program is its focus on the influence that the team atmosphere

has on collegiate athletes’ decision-making. Specifically, MVP directly discusses the

38

“bystander effect” that occurs when one or more teammates socially misbehave in group
settings (e. g. a party) and other teammates, although not directly participating in the
behavior (e. g. raping the woman), stand to the side and watch, without intervening to stop
the other teammates’ actions. Although many of the men who are bystanders may be
uncomfortable with the behaviors of their teammates, the social pressure to conform to
the group (i.e. team), and not deviate from the group norm is especially strong in a team
environment that’s purpose is to achieve the goal of athletic success (Hackman, 1976).

Berkowitz (1994) has found evidence that many men feel uncomfortable when
other men brag about their sexual exploits or comment on women’s bodies in sexually
suggestive ways. These men have been labeled as belonging to a “silent majority” of
men who “keep their discomfort to themselves rather than express disagreement or
intervene in an environment which they perceive as unsympathetic” or as threatening
their status in the group itself. Although not directly participating in the behavior, lack of
intervention is a serious problem that allows the behavior to continue without question.
The MVP Program talks with men about their bystander role in such situations, and helps
them work on ways in which they can intervene, as well as the consequences of such
intervention.

Although Katz et al. (1995) report “success” through the acceptance of their
program and the use of the Playbook, a manual of activities used in the program, they
have not yet published a formal evaluation of attitude or knowledge change. Even more
striking is the fact that the MVP program is the only published program that directly
targets the peer-group component of the male collegiate athlete culture, yet this

component, to date, has gone without evaluation.

39

 

   

Group—Level Qhange viflndividual Chflge

Considering how much peer support is involved in the male athletic culture, it is
necessary to be familiar with and use the various theories pertaining to group-level
change via the individual person. The following section reviews some of the work within
the social psychological field of minority dissent and how this may affect group norms of
behavior. According to this literature, there are two ways that a training such as the one
evaluated in the current study can help create group-level change: 1) programs targeting
collegiate athletes can help individuals to withstand group pressure and be minority
dissenters, and 2) the training can help minority dissenters feel confident enough to stand
up to group pressure and gradually change group norms of behavior. Both components
will be discussed in the following section, starting with a discussion about the group
dynamics through which norms and social pressure occur on a team. Understanding
these dynamics will help clarify how a group can be influenced by one or more
individuals within that group, which will be discussed later in this section.

Group pressure and norms of behavior. Athletic teams are especially prone to
intense levels of group (i.e. peer) pressure, mainly because of their structural make-up
and purpose. The main purpose of a team is to attain the goal of being successfirl against
opponents in the game of sport. As Hackman (1976) notes, one of the conditions for
which a group often exerts more social pressure on its members is to produce uniformity
for the purpose of attaining a common goal. Hackman (1976) explains, “it may be usefirl
for group members to hold similar beliefs about the external environment, especially if

the group must respond as a unit to that environment” (pg. 1473). Collegiate sports

40

Group—Level Change via Individual Change

 

Considering how much peer support is involved in the male athletic culture, it is
necessary to be familiar with and use the various theories pertaining to group-level
change Via the individual person. The following section reviews some of the work within
the social psychological field of minority dissent and how this may affect group norms of
behavior. According to this literature, there are two ways that a training such as the one
evaluated in the current study can help create group-level change: 1) programs targeting
collegiate athletes can help individuals to withstand group pressure and be minority
dissenters, and 2) the training can help minority dissenters feel confident enough to stand
up to group pressure and gradually change group norms of behavior. Both components
will be discussed in the following section, starting with a discussion about the group
dynamics through which norms and social pressure occur on a team. Understanding
these dynamics will help clarify how a group can be influenced by one or more
individuals within that group, which will be discussed later in this section.

Group pressure and norms of behavior. Athletic teams are especially prone to
intense levels of group (i.e. peer) pressure, mainly because of their structural make-up
and purpose. The main purpose of a team is to attain the goal of being successfiil against
opponents in the game of sport. As Hackman (1976) notes, one of the conditions for
which a group often exerts more social pressure on its members is to produce uniformity
for the purpose of attaining a common goal. Hackman (1976) explains, “it may be usefirl
for group members to hold similar beliefs about the external environment, especially if

the group must respond as a unit to that environment” (pg. 1473). Collegiate sports

40

teams eat meals, travel, room, and socialize with each other throughout the calendar year,
essentially sharing both the sports world, as well as their external social environment.
Outside of their task of playing a sport, teams often have a vested interest in keeping one
another uniform and acting as a whole to preserve the integrity of the team. Hackman
(1976, pg. 1474) again explains that “groups may ... seek uniformity for purely
“maintenance” reasons; that is, keeping the group intact and functioning as a unit,
independent of task-related activities. Too much individualistic or idiosyncratic behavior
on the part of a few members, for example, can threaten the very survival of the group.”
One of the ways that groups attain and maintain uniformity is by establishing group
norms of behavior and attitudes, with which members of the group are expected to
follow, or risk exclusion from the group (Hackman, 1976; Moskovici, 1980).

Changing group norms: the role of the minority dissenting opinion. Despite the
lack of independence among group members for fear of group exclusion (Hackman,
1976; Moskovici, 1980), it is possible for a member of a group to dissent from the
majority opinion and change group norms. This person takes the role of a rrrinority
dissenter, or one who holds a minority dissenting opinion. One of the main factors that
determines whether or not a member of a group will comply with group norms is whether
or not the behavior expected by the norm is congruent with the personal attitudes and
beliefs of the individual group members (Hackman, 1976). This is perhaps the
component of group norms relating to behavior change that is most relevant to
intervention with male student athletes. It is possible that by changing or enforcing a
team member’s attitudes or beliefs about sexual assault issues, that he would have the

tools needed to deviate from the group norm. In doing so, he would be one step closer to

41

providing the rrrinority dissenting opinion that is the first step to group-level change
(Nemeth et al., 1988).

It has been shown that exposing a group that holds a majority viewpoint to a
person who is willing to offer a dissenting opinion is a very powerful mechanism by
which to increase the chances that other group members will develop independent
judgments, rather than relying on group norms (Nemeth et al., 1988). Although people
sometimes assume minority opinions are incorrect, and therefore dismissed, perseverance
and consistency by the minority person increases the chances that the majority group
takes the minority viewpoint seriously and considers alternate opinions (e. g. “How can
that person be wrong, and yet so sure of themselves?”). The majority group is
“stimulated to reappraise the entire situation, which involves a consideration of numerous
alternatives, one of which is the position proposed by the majority” (N emeth, 1986, p.
26). Essentially, it is not whether or not the minority opinion is correct, but rather simply
that he or she spoke up against the majority view, resulting in the majority opinion being
opened to questioning from all sides. As Nemeth (1986, p. 26), explains, “as such, the
thought processes [of the majority group] are marked by divergence and hence, the
potential for detecting novel solutions or decisions.”

This divergence within the group is what contributes to individual group
members’ level of independence and ability to withdraw from group normative behavior
(N emeth et al., 1988). Individuals are faced with confrontation and a divergent
viewpoint, which “makes each individual a better problem solver or decision maker by
stimulating him or her to examine and reexamine premises” (Nemeth, 1986 p. 30).

Nemeth (1986, p. 30) again explains that “the presence of dissenting minority views and

42

the expression of those views thus aid the consideration of alternatives at a group level.”
Hence, although educational programs like the one evaluated for this study function at
changing individuals’ attitudes and behaviors, individuals are perfectly capable of
creating group-level change within their social group (i.e. team).

As a last point, it has been found that “ingroup” minorities (e. g. teammates) are
more effective than “outgroup” persons (e. g. non-teammates or prevention program
facilitators) at altering group normative behavior or attitudes (Clark & Maas, 1988;
Nemeth, 1995; Volpato et al., 1990). This underscores the importance that trainings
involving athletes work to increase individual team member’s propensity to act with a
minority dissenting opinion when involved in a situation that contradicts their own
personal judgments. To increase the chances that an individual can actually change
group norms of behavior, educational programs such as the one evaluated for this study
should be designed to help individuals gain the knowledge and tools necessary to hold a
consistent minority dissenting opinion when faced with group normative pressure.

Only one sexual assault prevention program to date emphasizes the importance of
minority dissent in the prevention of male sexual violence, and even so, the focus is on
fraternity groups (Mahlstedt & Jacobson, unpublished manuscript 2000). In the
Fraternity Violence Education Project, Mahlstedt & Jacobson (unpublished manuscript,
2000) explain that “the goal [of educating men to become minority dissenters] is that they
become a rrrinority group that creates tension in the larger fraternity culture, provide an
alternative model, will be present at the parties to influence the climate so that it will be a

safe one for women, and intervene when necessary” (p. 1). The pressure that minority

43

groups and individuals exert on the group norms of a rape-supportive culture is exactly

what changes the norms of that culture over time.

9mm

Considering that there is now some evidence that college sport participation may
have some connection to increased sexual aggression (Crosset, 1996), it is LISBfili to see if
programs aimed at changing collegiate athletes’ attitudes and knowledge about sexual
assault are making any kind of impact. The programs that have been implemented with
athletes seem to be of good quality and philosophy and it is important to start
investigating the effectiveness of such potentially groundbreaking programs. As with the
numerous prevention and education programs targeting the general college student body,
programs targeting student athletes need to be evaluated for their usefirlness so that
efforts at improvement can be made and successfiil formats passed on to others. Perhaps
it is even more important to explore programs targeting athletes, as student athletes’
schedules are extremely pressed for time, and more effective workshops will undoubtedly
have more support from athletic department administrators and coaches.

There are certain factors associated with an individual’s change in attitudes or
knowledge due to an intervention that have been evaluated within the general rape
prevention literature. For example, acceptance of interpersonal violence and hostility
toward women are two of the factors most often associated with a person’s level of
attitude or knowledge change. However, because student athletes have additional factors
that may play a role in their level of sexual aggression (e. g. discouragement from

identifying with anything considered “feminine” — extreme sex role socialization, high

44

need for achieving a goal or winning a competition) it is necessary to explore whether
these additional factors also play a role in their attitude and knowledge change after
participation in an educational intervention. Although they have not been as thoroughly
investigated, many of these variables have been outlined above, in addition to a few that
have been suggested by researchers as having an impact (e. g. contact versus non-contact
sport participation).

One of the most salient issues in working with a group culture such as collegiate
athletes is how important it is to understand and acknowledge if intervention targeting a
group-level change in behavior is possible. Considering the literature documenting the
influence peer support has on individual behavior (Hackman, 1976), it is especially
important to attempt change at the group, as well as the individual level. The current
program that specifically targeted collegiate athletes was uniquely equipped with the
exercises and environment that would make group change possible. One of the
program’s goals was to give men the tools they need to confront peer pressure and act as

minority dissenters, which is an essential component of any group-level effects.

45

Hypotheses

The current study evaluated attitude and knowledge change in male student

athletes after they had taken part in a sexual assault prevention and education program at

a large mid-western university. The evaluation addressed the following research

hypotheses:

l.

2.

Involvement in this program will decrease men’s reports of their rape myth
acceptance and likelihood of sexually aggressing.
Involvement in this program will increase men’s reports of their likelihood of

confronting a situation in which there is sexual aggression.

Effectiveness of the workshop (defined as significant positive change on rape myth

acceptance, likelihood of sexually aggressing, and likelihood of confronting) is expected

to be predicted by a number of men’s prior attitudes and experiences. Specifically, the

workshop is expected to be less effective for those men who report pre-workshop:

3.

4.

8.

9.

Higher goal orientation, win orientation, and competitiveness
More extreme levels of prior sexual aggression

Higher hostility toward women

Higher sex role stereotyping

Higher acceptance of interpersonal violence

Higher acceptance of general violence

Involvement in contact sports

10. More years participated in intercollegiate sports

46

Chapter 2

METHOD

Sam 1e

The workshop that was evaluated targeted all male student athletes at a large mid-
westem university who at the time were not members of the freshman academic class.
Ten program sessions were conducted, each of which was an hour and a half in duration.
Approximately 180 peOple attended the program, and 150 evaluation packets were
collected. Power analyses indicated that a minimum sample size of 125 would give
power of .80 at p<.05 for all analyses used. Power analyses were conducted using PASS:
Power Analysis and Sample Size (Version 6.0) Kaysville, UT: NC SS. More information
about the power required for each analysis is available in the statistical analysis section at

the end of this chapter.

The Program

 

Out of Bounds: An Anti—— Violence Workshop for Student Athletes at MS U

The workshop was mandated by the Provost of the University to provide student
athletes with education about sexual assault, sexual harassment, and relationship
violence. In addition to sexual assault, other topics of abuse were added to the training
content, to capitalize on the opportunity of having a training of this nature. A planning
committee of campus stakeholders was formed to develop and implement the trainings

with the student athletes. All of the people on the planning committee had multiple years

47

of experience in and/or knowledge about relationship violence education and prevention.
Members included the following: representatives from the campus sexual assault
counseling and crisis center (each with over fifteen years of experience), the director of
the campus domestic violence shelter (over fifteen years of experience), two male
outreach coordinators from the campus shelter and counseling program (over ten and four
years of experience, respectively), three campus police officers (about four years of
experience each), a representative from the campus women’s resource center (about
seven years of experience), two members of the student athlete support services (non-
athletic department), one program evaluator (author of this paper), as well as additional
people involved with the facilitation of the workshops.

Each workshop session was approximately one and a half hours in length, single-
sex (all men). Having an all-male audience followed on the suggestion of previous
researchers that a male majority environment lends for more open, safe, and honest
discussion about the sensitive tOpic of sexual assault (e. g. Katz, 1995). Often male
participants in mixed-sex workshops feel targeted by just the presence of women in the
audience and the presentation of the issue of sexual assault. By keeping the participants
in a single-sex environment, the chances for men being defensive and unwilling to
participate due to other women athletes’ presence was hopefirlly lowered. Facilitators
were men and women of varying ethnicities, and care was taken to ensure that all
sessions included male facilitators.

The workshop activities were developed by using a combination of reports from
other college programs working with athletes, and the various programming experience

of the members of the planning committee. A pilot workshop was held with all freshman

48

 

 

 

student athletes, and feedback from the evaluation of that session was also used to revise
certain activities and program format. The current program’s workshops were held in an
auditorium with tables and movable chairs, which made mobility of participants for
activities easier than in most auditorium settings. The auditorium was also used as a
study hall and team meeting room for the athletes on a regular basis, and therefore was
known to the athletes as “their territory.” It was the desire of the planning comnrittee to
hold the program in a setting that was well known to the athletes, to increase the chances
of the program discussion being seen as something that mattered in their lives as an
athlete, rather than as a separate and therefore forgettable event.

The main goal of this workshop was to increase education about relationship
violence, and help men learn ways to intervene with teammates or fiiends if they thought
something inappropriate was occurring. The planning committee decided unanimously
that a lack of “freely given consent” was the underlying common thread between all
forms of intimate violence. The concept of “freely given consent” was understood by the

facilitators as incorporating the following main points:

1. Consent is based on choice

2. Consent is active, not passive

3. Consent is possible only when there is equal power

4. Giving in because of fear is not consent

5. Going along with something because of wanting to fit in with the group, being
deceived or feeling bad, is not consent

6. If you can’t say “no” comfortably, then “yes” has no meaning

7. If you are unwilling to accept a “no,” then “yes” has no meaning

(Vanburen-Hay, 1999)
As such, the whole framework of the presentation was centered on the concept of
consent, and each of the points above were explained with examples to the audience at

various times during the program.

49

At the beginning of the workshop, a brief introduction video was shown to the
participants. The video was professionally produced for this particular group of
presentations, as an effort to not only capture participants’ attention from the start, but
also as an explanation of the importance of their attendance at the workshop that day.
The video featured a former football star from the university who later played
professional football before becoming a campus police officer at the university these
sessions were held. With a backdrop of contemporary music, video clips of sporting
events, and newspaper clippings from past stories of athletes who were arrested or
accused of committing intimate partner violence, the officer explained the special role
that athletes have in their community and the importance of them taking their prestige
seriously. It was explained that not only did their prestige on the college campus and in
the community buy them privileges; their fame also focused the spotlight of media
attention to personal wrongdoings more intensely than on other students at the university.
He thanked them for coming, and urged their attention throughout the workshop.

After the video, the facilitators introduced themselves and reviewed an overhead

of the five main points of the workshop:

1. Definitions
2. Consent
3. Laws

4. Choices
5. Resources

It was explained to the participants that each person in the room had the ability to make
his own decisions, and every action he took as an individual was his own choice, despite
how much group pressure he felt. In addition, it was emphasized that every choice has

good and/or bad consequences, and that more discussion about consequences and ways to

50

intervene would be explained later in the program. A brief explanation of how each main
point would be addressed throughout the program activities was explained by one of the
facilitators.

After this initial introduction, participants raised their hands as a group to signify
on what athletic team they participated. Certain workshops had more than one team
present, so the planning committee felt it was important to have the players be familiar
with each other enough to know that they were accounted for and not invisible to the
group or the facilitators. A facilitator, at this point, explained what the purpose of this
workshop was, and the importance of talking about sexual assault. They also
acknowledged that some people in the room may have already been abused, either as
children or adults, and that sexual abuse is not only a crime against women.

Because attendance at workshops ranged between 30-75 participants per
workshop, the decision was made to incorporate interactive discussion-based activities.
The first activity was the “Ball Toss,” in which the participants were seated as a larger
group. One of the facilitators stood on the front stage and threw a miniature soft ball to
the crowd. Whoever caught the ball responded to the question “How do you know if
someone wants to kiss you?” The first person gave his response and then threw it to
another person in the audience who answered the same question. This process repeated
for about six responses, and then the ball was thrown back to the facilitator for
discussion. About three questions were asked and answered in this manner as an
icebreaker and warm up for the group. This activity also encouraged participation right
fiom the start of the workshop and sensitized the participants to the topics of sexual

assault and relationship violence.

51

The next activity was a discussion about the myths and facts of intimate partner
violence. There were a wide variety of myths and facts displayed on individual
overheads, with the sexual assault myths paralleling the myths used on the pre and post-
test questionnaires. For each myth, participants were asked to raise their hand signifying
if they thought it was true or false. The complimentary “fact” slide was shown after the
“myth” slide. The facilitator discussed each myth with individuals who brought up
concerns.

The next part of the program consisted of a short video clip and discussion about
consent. The video clip was a scene from the recent popular series “Party of Five,” in
which one of the college age female characters is involved in an attempted rape by her
current boyfiiend at a party. After the video, facilitators engaged the audience in a
discussion about what happened in the video and whether or not she freely gave consent
to have sex. This activity usually had the most interaction, as many people had opinions
about the woman’s behavior in the situation and whether or not drinking (which the
characters were doing at the party) affected someone’s ability to give or receive consent.

After the consent piece was completed, participants were presented with a short
computer overhead presentation about the laws regarding criminal sexual conduct,
domestic violence, and sexual harassment. Questions were taken from the group.

The last exercise was another discussion-based interactive activity that was aimed
at helping men decide what their options for action were when faced with being a
bystander in a situation where something sexually abusive was (or would soon be)
occurring. Three scenarios were developed specifically about situations where male

athletes may be involved. This activity was based on a similar, yet different exercise

52

created by Katz (1995), which has been used by the Mentors in Violence Prevention
Project. The scenarios were real-life situations that most men have experienced, or at
least known about in their families or communities. All scenarios were read from the
bystander’s perspective, not as a perpetrator or victim. As mentioned previously, it is
essential in a workshop with student athletes to discuss the issue of being a bystander to
situations that might bring up ethical and moral concerns for a person. Perhaps the most
salient reason for targeting athletes for prevention/education trainings is the research
documenting that persons who would not typically victimize a person by themselves are
often pressured into doing so in a group situation heavily guided by male bonding and
team membership (Boeringer, 1999). Asking participants to consider their opinions
about what they would and possibly should do when faced with group pressure is
essential to tap into this phenomenon.

For example, one of the scenarios was “You see a teammate of yours with a
woman who is dancing very provocatively and appears to be really drunk. After they
finish dancing, he grabs her by the hand to lead her upstairs. She doesn’t seem like she is
into it and appears to be having trouble standing and walking.” After reading the
scenario out loud, the facilitator walked participants through their thought processes if
they were actually in that situation. Steps included a discussion about the following:

1. Facts of the scenario (e. g. he is your teammate, she is drunk)
2. His options (for this scenario, options usually included walking away, intervening
between the two of them, taking your teammate to the side and telling him what

he’s doing is wrong, etc.)

53

3. The consequences of his decision for each of the options presented (e. g. she is not
assaulted, your teammate may get upset, she may be assaulted and you witnessed
it enough to be prosecuted, you may feel stupid in front of your teammates).
This exercise always opened up a lively discussion about a person’s options for action
when he is faced with tough, often peer influenced decisions in real life.

After the scenarios were completed, facilitators explained what campus resources
were available to participants and/or anyone in their lives who may need help with any of

the issues discussed in the program.

Evaluation

The evaluation consisted of a pre-test and post-test that were given matching
numbers to ensure the ability to match them in the event that they became separated. One
pre-test and one post-test were placed into a manila envelope, which was labeled with the
same identifying number. A cover sheet on both the pre-test and post-test explained that
participation in the survey was voluntary, and that all information would be kept strictly
confidential. A similar cover sheet was used for both test administrations, and explained
that the project was interested in learning more about students’ perceptions about
relationships and themselves. Prior to the start of the program, the pre—test and the post-
test were placed in an envelope and set on the tables at which the participants sat for the
duration of the presentation. Care was taken to ensure that there was ample space
between participants to ensure confidentiality. All questionnaires were filled out in the
room in which the presentation was held. As participants entered the room for the

training, a facilitator at the entrance greeted them and instructed them to take a seat where

54

there was already an envelope and start filling out the green packet (pre-test). The
remaining facilitators stood at the front of the room to ensure that no one spoke or looked
at the post-test, and that each person kept his answers to himself. As soon as people were
finished filling out the green packet, they placed them back into the envelope and pushed
the envelope to the side until the presentation was over. The entire pre-test took each
person between 15 and 20 minutes. To minimize any coercion a participant would feel to
complete the test packet, no coaches or persons with athletic department authority were
present in the room.

After the presentation content was completed, one of the facilitators introduced
the principal investigator of the evaluation, who then explained to participants the process
of completing the post-test. The evaluator attended and explained the post-test procedure
at every presentation to ensure that participants were given identical information at each
presentation. Participants filled out the post-test in the same fashion in which they
completed the pre-test, placed it into the envelope, and dropped the envelope in a box by
the door on their way out. As participants left the room, they were given a sheet with a
list of resources and information about obtaining counseling, which was meant to
minimize individuals’ embarrassment or uneasiness at having to request such resources.
The decision to hand out these sheets at the door was made to ensure that those people
who opted to not complete the survey would still receive the list of resources and
information. A file was created for each participant in a locked file cabinet at the
university that was only accessible to the project’s investigators and research assistants.
No names or any other identifying information were required or collected. Only the

numbers printed on the pre-test and post-test in each packet with no identifying

55

information were entered into the data set. Post-test completion time ranged from 10-15
rrrinutes, and procedures for ensuring confidentiality were the same as with the pre-test
conditions.

In addition to administering the pre and post-tests, the principal investigator
attended each of the sessions to observe the interactions of the audience and facilitators,
as well as to observe the overall tone of each presentation. As noted later in the
discussion section of this paper, the researcher was able to observe the apparent
receptiveness of participants to their attendance at the workshop, in addition to their
overall response to the pre and post-test. The researcher was only visible to the audience
at the end of the presentation, when she explained the procedures related to the
completion of the post-test. Otherwise, the researcher stayed at the back of the room, far
away from any of the participants to eliminate the possibility that participants felt they

were being “watched.”

Measurement

 

Ten measures were used in this study. Demographics, life experiences, and
attitudinal scales related to predictor variables were included only in the pre-test packet
because they were not a target for change via the workshop. Pre-test attitudinal variables
included intra-individual factors that have typically been measured in rape prevention
program evaluations (e. g. sex role stereotyping, hostility toward women), in addition to
sport-specific factors that have not yet been measured in the context of an evaluation (e. g.
competitiveness). Post-test attitudinal variables included all outcome variables that

were a target for change via the workshop. Also included in the post-test were group-

56

utilization variables that tested whether or not group-level change was possibly achieved
as a result of the workshop.
Demoggrphics

Data was collected on the type of sport played (e. g. contact/non—contact), and
number of years the student participated as an intercollegiate athlete. These measures
addressed the environmental factors that may have played a role in not only a man’s
sexual aggression, but also his level of attitude change due to the intervention.
Contact vs. Non-Contact Sport Participation

All sports offered as intercollegiate varsity sports at the participating university
that were classified as non-contact were listed in a box, and participants were asked to
check the box if it contained the sport they participated in as an intercollegiate athlete.
Teams classified as non-contact included: tennis, volleyball, cross country, track & field,
golf, baseball, swimming, gymnastics and crew. Contact sports included: football,
basketball, wrestling, soccer, and ice hockey. To ensure confidentiality, participants were
not asked their specific team membership, as it would have been difficult to ensure
anonymity when reporting results on a team-by-team basis.
Competitiveness, Goal Orientation, Win Orientation

Competitiveness, Goal Orientation, and Win Orientation were measured using the
Sport Orientation Questionnaire (SOQ) developed by Gill & Deeter (1988). Original
reliability analyses on each of the three subscales (win orientation, goal orientation, and
competitiveness) indicated alpha coefficients of .72, .78, .91, respectively (Gill &
Deeter, 1988). The alpha coefficient for the combined three subscales was .92. The

questionnaire was developed as a “sport-specific multi-dimensional measure of

57

achievement orientation” (Gill & Deeter, 1988, p. 192). It captures not only an
individual’s motivation for achievement within the world of sports, but also a person’s
level of interpersonal achievement. The authors suggest that such personal qualities as
competitiveness and achievement orientation may easily cross over from sports into an
individual’s personal life. Respondents were asked to indicate how strongly they agreed
with 25 statements about winning (6 items), goal attainment (6 items), and
competitiveness (13 items) on a scale ranging from 1 (strongly agree) to 5 (strongly
disagree). Examples of statements included on the questionnaire were: “I am a
competitive person,” “Losing upsets me,” “I look forward to the opportunity to test my
skills in competition,” and “The best way to determine my ability is to set a goal and try
to reach it.” For the current study, reliability analyses on each of the three subscales (win
orientation, goal orientation, and competitiveness) indicated alpha coefficients of .79,
.72, .93, respectively. Range of corrected item-to-total correlations are listed below in
Table 1. The alpha coefficient for the combined three subscales was .93. The mean
scores for each of the three subscales (or the mean score of the entire scale with all three
dimensions included) were utilized for analyses in this study.

Table 1

Sport Orientation Subscale Item-to-Total Correlations

 

 

 

 

 

 

 

 

 

 

# Items Lowest Item-to- Highest Item-to- Coefficient
in total Correlation total Correlation Alpha
Subscale
Win orientation 6 .36 .71 .79
Goal Orientation 6 .42 .57 .72
Competitiveness 13 .56 .84 .93

 

58

Attitudes Toward General Violence and Interpersonal Violence

Respondents’ general attitudes toward violence, as well as their attitudes toward
interpersonal violence were both measured by the Attitudes Toward Violence Scale
developed by Lonsway & Fitzgerald (1995). This scale had recently been used in rape
prevention evaluations, as a result of Lonsway & Fitzgerald’s (1995) finding that other
attitudinal scales commonly used (e.g. hostility toward women) confounded a
respondent’s attitudes toward violence. This scale more accurately differentiates a
respondent’s acceptance of general violence from his or her acceptance of interpersonal
violence in a way that is not as confounded by other attitudinal variables (e. g. hostility
toward women). Sample items included “Violent crimes should be punished violently,”
“The government should send armed soldiers to control violent university riots,” “Giving
mischievous children a quick slap is the best way to quickly end trouble,” and “It is
alright for a partner to hit the other if they are unfaithfirl.” Respondents indicated using a
scale of 1 (not at all agree) to 7 (very much agree) how much they agreed with each
statement. Lonsway & Fitzgerald reported an or of .87 for the entire scale, which consists
of 20 items (10 general violence, 10 interpersonal violence). Corrected item-to-total
correlations for the entire 20-item scale ranged from .23 to .65. Although the questions
were split equally between conceptual constructs, the original developers of the scale did
not analyze or report subscale psychometrics. To explore the possibility of analyzing this
scale as two separate subscales, factor analyses and reliability tests were run.

Exploratory factor analysis with varimax rotation yielded three factors. The first
factor included the items measuring acceptance of violence against children, the second

factor included items measuring acceptance of violence against intimate partners, and the

59

third factor included items measuring the acceptance of general violence. Since this scale
was developed to measure two factors, including acceptance of general violence and
acceptance of interpersonal violence, reliability tests were run to determine if the first two
factors could be analyzed as one subscale, despite their possible separation into two
factors. The alpha coefficient for the first two factors together was .88 (corrected item-
to-total correlations ranged from .40 to .78). The alpha coefficient for the third factor
was .77 (corrected item-to-total correlations ranged from .21 to .55). Since it was
conceptually desirable to use the first two factors together as a measure of acceptance of
interpersonal violence, the decision was made to use the mean score of the first two
factors to create one subscale that measured acceptance of interpersonal violence. The
mean score of the last factor was used to create the second subscale that measured
acceptance of general violence.
Sex Role Stereotyping

Sex role stereotyping was measured using the Sex Role Stereotyping Scale
developed by Burt (1980). This 9-item scale was created for the purpose of measuring
the degree to which a person assigns men and women to traditional sex-specific roles. In
Burt’s (1980) study, Cronbach’s alpha was reported as .80. Example statements
included: “It is acceptable for the woman to pay for the date” and “A wife should never
contradict her husband in public.” Respondents indicated using a scale of 1 (not at all
agree) to 7 (very much agree) how much they agreed with each statement. Three items
were reverse coded, and reliability analyses indicated an alpha coefficient of .60.
Corrected item-to-total correlations ranged from .16 to .43. The mean score of all nine

items was used for analyses in this study.

60

Hostility Toward Women

Hostility toward women was measured by the Hostility Toward Women Scale,
which was originally developed by Check, Malamuth, Elias, and Barton (1985), and later
modified by Lonsway & Fitzgerald (1995). Coefficient alpha for this 10-item scale was
reported by Lonsway and Fitzgerald (1995) as .83. A list often statements was used to
assess a person’s level of hostility toward women. Respondents indicated using a scale
of 1 (not at all agree) to 7 (very much agree) how much they agreed with each statement.
Examples of statements included “I believe that most women tell the truth,” “I am easily
angered by women,” and “Sometimes women bother me just by being around.” Two
items were reverse coded, and the alpha coefficient for all items was .83. Corrected item-
to-total correlations in this study ranged from .37 to .68. The mean score of all ten items
was used for analyses in this study.
Sexual Aggxession Severity

A respondent’s severity of prior sexual aggression was measured by a revised 14-
item version of the original ll-item version of the Sexual Experiences Survey that was
developed by Koss et al. in 1985 and revised by Koss & Gaines in 1993. Internal
consistency ratings for the original SES (Koss et al., 1985) were reported at .89, with test-
retest agreement between l-week apart test administrations reported to be 93%. The
1993 version includes more concise wording of items and incorporates a wider variety of
sexually coercive and abusive behaviors (including not just individually coercive
behaviors, but also incorporating gang rapes). For example, questions were added asking
if the respondent had ever made “catcalls” (i.e. sexual harassment) to a woman or had

ever stood in line to have sex with a “party girl” (i.e. gang rape). Two additional

61

questions were added for the current study that specifically asked about a respondent’s
involvement with “groupies,” since these women are often prominent figures in an
athlete’s social life. These two questions asked how often a respondent has “had sex with
a groupie” and “stood in line to have sex with a groupie.” Using a 5-point scale ranging
from “Never” to “Often”, respondents indicated how often since they were 14 years old

they had engaged in each of the listed behaviors.

Dependent Vaiables

Three instruments were used to measure the three dependent variables including
1) Rape Myth Acceptance, 2) Likelihood of Sexually Aggressing (i.e. Likelihood to Use
Force, Likelihood to Rape, Likelihood to Use Coercion, Likelihood of Using Alcohol),
and 3) Likelihood of Confronting someone in or after a sexually aggressive situation.
Group (Team) Utilization was also measured at post-test, as an attempt to capture the
possibility that group change (i.e. across a team) would occur after the presentation.
Since the three dependent variables measured the outcome of the intervention, all of
them, with the exception of the group utilization items, were measured at both pre and
post-test. Group utilization, described later in this section, was measured only at post-
test. For the remaining five outcome variables, scale construction was conducted

independently with pre and post-test scores. These variables were measured as follows:

62

QflMXth Acceptance

 

Rape myth acceptance was measured using the Illinois Rape Myth Acceptance
Scale — Short Form (IRMA-SF), developed by Payne, Lonsway, and Fitzgerald in 1999.
This scale was developed as a more accurate measure of rape myth acceptance than
Burt’s (1980) original Rape Myth Acceptance Scale. Coefficient alpha for the scale
constructed by Lonsway & Fitzgerald (1994) was reported as .87 (corrected item-to-total
correlations ranged from .34 to .65). The definition of rape myths that are measured
using the IRMA-SF are an expanded version of Burt’s (1980) original cultural definition
of rape myths, and include multiple dimensions of rape myth acceptance. The IRMA-SF
measures the extent to which a respondent adheres to rape myths, defined as “attitudes
and beliefs that are generally false but are widely and persistently held, and that serve to
deny and justify male sexual aggression against women,” (Lonsway & Fitzgerald, 1994,
p. 134). Sample statements include “It is usually only women who dress suggestively
who are raped” and “Rape happens when a man’s sex drive gets out of control.”
Respondents indicated how much they agreed with each statement using a scale of 1 (not
at all agree) to 7 (very much agree). For both pre and post-test scores, three items were
reverse coded, and the mean was taken of all items on each test occasion to create a scale
score. Alpha coefficient for the pre-test was .82 and .86 for the post-test. Corrected
item-to-total correlations ranged from -. 18 to .76 for the pre-test and from -.11 to .74 for

the post-test.

63

Behavioral Intentions (Likelihood to Use F orce, Likelihood to Rape, Likelihood to Use

 

Coercion, Likelihood of UsingAlcohol. Likelihood to Confi'ont)

All behavioral intentions were measured as part of a questionnaire developed by
Berg, Lonsway, and Fitzgerald (1999). The first two behavioral intentions, Likelihood to
Use Force (LF) and Likelihood to Rape (LR), were originally developed by Briere and
Malamuth (1983), and reworded to their current form by Berg et al. (1999). Two of the
other behavioral intentions, Likelihood to Use Coercion (LC) and Likelihood to Confront
(LCon) were originally developed by Berg et al (1999). All items measuring the first
three behavioral intentions (LF, LR, LC) were included as statements within a list of
various sexual behaviors and actions. The fourth variable, Likelihood of Using Alcohol
(LA), was created for the current study in response to the growing number of studies that
have found alcohol and/or other substance use as significant predictors of sexual
aggression or sexual force among both athletes as well as non-athletes (Boeringer, 1996;
K038 and Gaines, 1993; Schwartz & Nogrady, 1996).

There were a total of eleven statements, seven of which were distracter items.
Respondents rated on a scale of 1 (never) to 5 (always) how often they would engage in
each behavior if they could be assured that no one would know. Higher scores indicated
an individual’s higher self-reported likelihood of engaging in sexually coercive behaviors
(Berg et al., 1999). Although previous researchers have reported each of the behavioral
intentions as a separate variable, it was a possible reliability concern that each intention
was measured by using only a one-item statement. For example, the construct
“Likelihood to Rape” has typically been measured only by the following statement: “If

you could be assured that no one would know, how often would you force a woman to

64

have sex when she doesn’t want to?” Other researchers have just explicitly asked a man
if he would rape if he was assured no one would know and he would not be punished
(Boeringer, 1999). The other behavioral intention measures consisted of similar
statements that capture the variety of ways that men may attempt to engage in sexual
behavior with an unwilling woman (e. g. coercion, force).

An exploratory factor analysis using varimax rotation was used to determine if
these items would be more valid as several indicators of a central construct of
“Likelihood of Sexually Aggressing.” This analysis indicated that at pre-test, all four
items loaded highly on one component (.81, .76, .82, .86) and accounted for 66% of the
total variance. At post-test, the same four items loaded even higher on the first
component (.93, .96, .98, .90) and accounted for 89% of the total variance. Additionally,
reliability tests with pre-test data indicated that the four items taken together as one scale
resulted in an alpha coefficient of .82 (corrected item-to—total correlations ranging from
.58 to .72), while the post-test alpha coefficient was .96 (corrected item-to-total
correlations ranging from .87 to .95). Therefore, for conceptual ease of purpose, as well
as psychometric clarity, the mean of the scores for each of the four behavioral intention
measures (likelihood to use force for sexual acts, force for intercourse, coercion, and
alcohol), was used to create a scale score measuring the construct of “likelihood of
sexually aggressing.”

The fifth behavioral intention measure, Likelihood to Confront, was designed by
Berg et al., (1999) “to assess the level of confrontation that participants would engage in
when they have reason to believe that a woman is being made to do something that she

does not want to do” (Berg et al., 1999, p. 224). Respondents were asked to respond to

65

six scenarios by choosing one of four different options. Relationships of either the
person instigating the behavior or receiving the behavior were varied. For example, the
person instigating the sexual behavior was either the respondent’s “teammate” or “ a
man” (i.e. stranger), and the woman in the scenario was either “a good fiiend” of his or
someone he “does not know.” The use of the relationship of a “teammate” in the
scenario was reworded for the current study from “roommate” to “teammate” in order to
make the question more relevant to student athletes’ experiences of being involved in a
team culture. In addition, different intensities of sexual acts were also varied for each
scenario (e. g. kissing or making her do “something sexual” she doesn’t want to do).
Response options ranged from “do nothing” to “interrupting the situation to see if
everything is alright.” Although each scenario was created to be conceptually unique
from the other scenarios, reliability analyses indicated that all six items held together as
one scale measuring a person’s “likelihood to confront” (scale coefficient alpha = .74;
corrected item-to-total correlations ranged from .40 to .68). For ease of multivariate
analyses, all six items were combined to make one scale measuring the outcome variable
“likelihood to confront.” This scale was used in all multivariate analyses for this study.

However, items were also analyzed separately (i.e. not as a scale) for descriptive

purposes. Descriptive analyses are discussed later in the results section.

Program Utilization
A selection of questions gauged at measuring projected group-level change
(within each team), as well as individual utilization of program information was also

developed for this study. The questions developed for this study followed the concept of

66

a type of evaluative technique termed “utilization-focused evaluation” (Patton, 1990).
This final collection of “group utilization” outcome variables is unique fi'om the other
outcome variables for this study because it was only measured at post-test, rather than at
both pre-and post-test. One of the purposes of the utilization-focused evaluation
technique is to ensure that the event that is evaluated (e. g. an informational workshop) is
actually relevant to the population that is expected to use the information from the event
(Patton, 1990). Although this type of evaluation is often considered a process that
requires participant participation throughout the entire evaluation planning and
implementation (Patton, 1990), for this study, only the practicality of this concept was
used, so program participants were not directly involved in the development of evaluation
questions. For example, the practicality of asking participants how much they will use
the information that was provided at the workshop was what was deemed important for
the current study, whereas in other studies the participants might have been involved in
the practicality of the evaluation itself, rather than just the program inforrnation’s
practicality to them.

The group utilization items for this study focused on exploring whether or not the
information provided by the program was meaningful and usefirl to the program
participants. In this sense, this technique offers a very practical way of asking how
useful the workshop was for participants, as well as a check on the accuracy of findings
from pre-post test measures. For example, since the current study evaluated change by
administering pre and post-tests only an hour apart, it was a good additional measure to
find out if participants directly believed that the information presented was something

they would put to use or not, regardless of attitudinal or abstract pre-post test attitudinal

67

measures. The practicality of just asking participants directly whether or not they will
use the program information is regarded as one of the strengths of the utilization-focused
evaluation approach. Patton (1990, p. 405) states, “A utilization-focused approach to
analysis can help keep the findings from becoming too abstract, esoteric, or theoretical.”
The level of program usefulness that the program participants report is an important and
direct way of knowing whether or not the workshop information will actually be used and
make a diflerence in their lives.

Aside from exploring individuals’ use of the information, utilization questions for
this study were used to help determine the likelihood that other team members would use
the information. Asking an individual team member how much his team would use the
information from the workshop goes a step farther into a participant’s estimation of
group-level change. As noted earlier, no other program targeting athletes has attempted
to measure whether change at the group level of interaction may occur as a result of an
intervention (Katz, 1995; Parrot et al., 1994). Other ways of determining group-level
change (e. g. participant observation of group behaviors, focus groups held months after
the intervention) were impractical for this type of study. Therefore, the decision to use
utilization questions was made to ensure that at least some estimation of group behavior
was attained.

General program usefulness. Items developed for this study measured varying
components of behavioral impact. For example, to find out the general usefulness of
such a presentation (and therefore likelihood of making an actual behavioral impact) the
question was asked “How much do you think you will use some of the information that

was presented in this workshop?” To capture not only individual utilization, but also the

68

 

 

possibility of team utilization, the question was asked “How much do you think members
of your team will use some of the information that was presented in this workshop?”
Respondents answered both of these questions using a scale ranging from 1 “very much”
to 4 “not much at all.” Blank space for clarification on both of these questions was also
available to respondents (asked in the form of the question “what makes you say that?”).
All open-ended responses for individual utilization items were reviewed and then coded
into six emergent categories of response by the researcher (see Table 2). Two coders
were trained in the use of the coding categories, then independently read and coded the
responses according to the predetermined categories. Initially, out of sixty-three coded

responses, coders agreed on fifty-nine. Thus, initial inter-coder agreement was 94%.

Table 2

Individual Utilization Coding Categories (n = 65)

 

Coding Category

Example of response included in category

 

. No new information/
common sense (n = 29)

“I know most of it”
“Because I didn’t learn anything new”
“It’s common sense!”

 

. These situations do not
happen often (11 = 7)

“Not confronted with it a lot”
“Hasn’t happened yet”
“Don’t go to places it would happen”

 

. Would behave in the ways
suggested by the program

anyways (n = 4)

“I already would do that”
“Cuz I’m not gonna sexually harass
anyone” (sic)

 

. Would not behave in the
ways suggested by the
program anyways (n = 3)

“I wouldn’t do it anyway”
“Because sexual assault will always be
present”

 

. New information (11 = 15)

“It’s good to know the definitions”
“There is some information I did not
know”

 

 

. These situations happen
often (n = 4)

 

“I’m in these situations often”
“A lot of these instances occur”
“It happens a lot (these situations)”

 

69

 

As with the individual utilization items, all open—ended responses for team
utilization items were reviewed and then coded into six emergent categories of response
by the researcher (see Table 3). Two coders were trained in the use of the coding
categories, then independently read and coded the responses according to the
predetermined categories. Twelve responses were thrown out because they did not
answer the question, were incomplete, or did not make sense to either of the coders.
Initially, out of fifty-five coded responses, coders agreed on fifty-one. Thus, initial inter-

coder agreement was 93%.

Table 3

Team Utilization Coding Categories (n = 55)

 

 

Coding Category Example of response included in category
1. Lack of attentiveness or “I don’t think a lot of my teammates would
caring for program care”
material (n = 16) “They all thought it was stupid we had to

attend this meeting”
“It all depends on how well they paid

 

 

attention”
2. Information was “common “We learned this is third grade” (sic)
sense” (n = 9) “Common sense”
“We all know”
3. Team is inherently bad “There are some dirty pigs on my team”
(n = 6) “We’re gross!”

“They do not care. NO respect”
“They are pretty bad guys who think they
deserve it all”

 

 

4. Team lacks intelligence “Because they are idiots!”
(n = 5) “Some of them aren’t very smart”
“Their jocks” (sic)
5. New information (n = 10) “It was informative”

“Because they learned”
“It seemed like everyone got something
from what was said”
6. Team is inherently good “We don’t do this crap”
(n = 9) “We are a good team”

 

 

 

 

 

70

Confrontation with a teammate. Two items were created to more directly
determine if the intervention helped participants receive the behavioral or attitudinal tools
they would need to assume the role of a minority dissenter. As a follow-up to the
scenario questions involving confrontation (described above), the first item asked
respondents “What factors would influence your decision to confront a teammate who
was making a woman do something sexual that she does not want to do?” All open-
ended responses were reviewed and then coded into seven emergent categories of
response by the researcher (see Table 4). Two coders were trained in the use of the
coding categories, then independently read and coded the responses according to the
predetermined categories. Eleven responses were thrown out because they did not answer
the question, were incomplete, or did not make sense to any of the coders. Initially, out
of seventy-five coded responses, coders agreed on sixty-six. Thus, initial inter-coder
agreement was 88%. Coding categories were fiirther discussed between the researcher
and the two coders and revised so that all three were in agreement with the defining
features of each category. Responses were again independently reviewed by each coder,
and as a result, coders agreed on seventy-three out of the seventy-five responses,
indicating a 97% inter-coder agreement.

An additional item was also added that asked respondents “How much do you
think the information that was presented to you in this workshop would help you stand up
to a teammate if he was doing something with a woman that you thought was wrong?”
For this item, respondents indicated on a scale ranging from “won’t help me at all in any

situation” to “will help me in all situations,” in addition to one answer choice “I already

71

 

 

 

would have stood up to a teammate in any situation.” Respondents who answered that it
would not help them in any situation were asked to briefly explain why they didn’t think
the information would help them. Very few respondents (n= 7) answered this last open-

ended question, so it was not used in any of the analyses for this study.

Table 4

Coding Categories for Factors Influencing Participants’ Confrontation of a Teammate

 

 

 

 

 

(n = 75*)
Coding Category Example of response included in category
1. Cues from the woman “The way the female is acting”
(n = 19) “If she was upset or mad”
“If she was screaming help or something else”
“Amcunt of resistance by the woman”
2. Situational factors “Environment and time”
(n = 15) “Atmosphere and setting and extent of the
situation”
“Where it is at”
3. Protecting teammate “If he was doing something criminal”
(n = 13) “If he was going to get himself in trouble”
“Don’t want him to get in trouble”
4. Personal morals “I don’t care who it is, I don’t think it’s right”
(n = 12) “It is not right”

“Rights of everyone — it’s common sense”
5. Relationship with teammate “Who the teammate was”

 

 

(n = 12) “How well you know him, if they’re violent”
“How close of a friend he is”
6. Relationship with woman “How well I know her”
(11 = 9) “If she was my sister or relative, someone
close”

“Who the woman was, if she could handle
herself or not”
7. Cues from teammate “The way he is touching her”

(n = 8) “If he was forceful”

“If they’re violent”
*n = 73 responses, but some responses listed multiple factors. The total number of factors given was 11 = 93.

 

 

 

 

 

72

 

 

Statistical Analyses

 

Multivagiate Analyses

To test the hypotheses that involvement in this program would decrease men’s
reports of their rape myth acceptance and likelihood of sexually aggressing and increase
men’s reports of their likelihood of confronting a situation, repeated measures
multivariate analysis of variance (MANOVA) was used to initially compare pre and post-
test scores. Power analyses indicated that for MAN OVA, N =125 (answering both pre
and post-tests) gave power = .80 to find pre-post change significant at p < .05 if the effect
size is at least as big as .25 standard deviations. Power analyses were conducted using
PASS: Power Analysis and Sample Size (Version 6.0) Kaysville, UT: NCSS.

When preparing to run the repeated measures MANOVA, it became apparent that
there were a number of respondents who were listwise deleted from the analysis. Case
summaries were run to determine what cases were missing from the dataset, which
revealed that the majority of cases that were listwise deleted consisted of seventeen
football players and ten members of others teams who did not complete their entire post-
test. Although these players had completed most scales on their post-test, because they
were missing one or more scales, listwise deletion procedures (used in general
multivariate analyses) deleted them from all analyses.

To more accurately estimate the values of this missing data, an expectation
maximization procedure was run. Expectation maximization (EM) is an estimating
procedure that is known for producing more realistic estimates of variance than other
methods of missing data estimation (e. g. mean imputation) (Tabachnick & Fidell, 2000).

Tabachnick and Fidell (2000) explain: “EM forms a missing data correlation (or

73

covariance) matrix by assuming the shape of a distribution (such as normal) for the
partially missing data and basing inferences about missing values on the likelihood under
that distribution” (p.63). By using estimates based on the distribution of existing data,
this procedure of estimating missing data not only more accurately reflects the actual
sample population, but also avoids impossible matrices and overfitting (making the
estimation look better than it actually is). The results of t-tests that were run to determine
if cases in which data were estimated were different than cases without estimated data are
provided at the beginning of the results section. A total of 60 out of 720 values (or 8.33%
of the data matrix) were estimated using EM procedures.
Predictors of Post-Intervention Changa

To examine the relationships between predictor variables and participants’ post-
scores, a sequential regression (also called hierarchical regression) of predictors on post
scores was conducted for each outcome variable. Pre-scores were co-varied out of the
equation by entering them into the first block of each regression equation. Sample size
for all sequential regressions was N = 120, which yielded only slightly lower power than
had been anticipated. For sequential regression analyses, power analyses indicated that N
=125 should give power = .80 at (p < .05) to find significant an R square change = .05
when pre-scores account for 25% of the variance (implying pre-post stability, or the
correlation between pre and post-scores = .5). Power analyses were conducted using
PASS: Power Analysis and Sample Size (Version 6.0) Kaysville, UT: NCSS. All data
were entered into and analyzed with the Statistical Package for Social Scientists (SPSS),
version 10.0. A total of 1 16 out of 1920 values (or 6.04% of the data matrix) were

estimated using EM procedures.

74

Chapter 3

RESULTS

Of the 150 survey packets collected, 120 were used for analyses in the current
study. Six surveys were deleted from the dataset due to a lack of information on both the
pre and post-test. Twenty-four surveys could not be used for one or more of the
following reasons:

1. Incomplete post-test (less than one page completed). n = 4

2. Incomplete pre-test (less than one page completed). n = 3

3. Answers to one or more of the measurements were clearly fabricated or

answered with sarcasm. n = 17
Cases that fell under the last criterion above were determined through inspection of
numerical answers to items on the scales included in either the pre or post-test. Items that
were numerically answered with an obvious pattern were included under this third
criterion for exclusion. For example, if all twenty-five items on a scale were marked as
“undecided” and/or answers increased numerically in order from lowest option to highest,
and then repeated, such as 1,2,3,4,5,6,7,1,2,3,4,5,6,7. . .), or if items were answered
without regard to the reverse coded items (e.g. answering the reverse coded items within
the same range as the non-reverse coded items were answered), the case was excluded
from analyses. Surveys that fell under the third criterion for exclusion above at either the
pre-test or the post-test were excluded from analyses. This decision was made with the

assumption that scores that were fabricated or answered sarcastically by a person on

75

either the pre or post-test could not be assuredly deemed as reliable on the opposite test
administration.

Independent sample t-tests indicated that people who were excluded from
analyses using the above selection criteria were significantly difi’erent than people who
were included in analyses. As Table 5 indicates, participants who were excluded from
analyses had slightly higher sex role stereotyping (excluded participant mean score: 3.68;
included mean scale score: 3.10) and severity of prior sexual aggression (excluded mean

scale score: 1.87; included mean scale score: 1.27) than participants who were included

in the analyses.

Table 5

Independent Sample T-Tests for Cases Excluded From Analyses Based

on Selection Criteria

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variable T df Sig. (two-tailed)
Rape Myth Accemnce -.307 142 .759
Likelihood of Sexually -l .849 142 .067
Aggressing
Acceptance of -1.466 142 .145
Interpersonal Violence
Likelihood of Confronting .763 142 .446
Acceptance of General -1.636 142 .104
Violence
Contact vs. Non-Contact .298 142 .766
Sport Participation
Hostility Toward Women -.471 142 .639
Sex Role Stereotyping —2.392 142 .018
Severity of Sexual -2.386 142 .018
Aggression
Competitiveness -.011 142 .992
Goal Orientation .451 142 .653
Win Orientation .418 142 .677
Years played -.553 109 .581
intercollegiate sports

 

 

76

 

 

 

 

Independent sample t-tests were also run to determine if cases for which post-test
data were estimated (through EM procedures) were significantly different from non-
estimated cases. T-tests indicated that people who did not completely fill out their post-
tests were significantly different from the group that did complete the post-test on only
one of the thirteen variables. As Table 6 indicates, participants who did not complete
their entire post-test scored significantly higher on acceptance of interpersonal violence
than participants who completed the entire post-test (uncompleted post-test mean scale

score: 2.97; completed post—test mean scale score: 2.47).

Table 6

Independent Sample T-Tests for Estimated Cases Based on Missing Partial Post-Test

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variable T df Sig. (two-tailed)
Rape Myth Acceptance .056 142 .955
Likelihood of Sexually .883 142 .379
Aggressing
Acceptance of -2.008 142 .047
Interpersonal Violence
Likelihood of Confronting .710 142 .479
Acceptance of General -.704 142 .483
Violence
Contact vs. Non-Contact 1.287 142 .200
Sport Participation
Hostility Toward Women -1.240 142 .217
Sex Role Stereotyping -1 .7 87 142 .076
Severity of Sexual -.7 96 142 .427
Aggression
Competitiveness -. 170 142 .865
Goal Orientation -.430 142 .668
Win Orientation -.142 142 .887
Years played -.812 109 .419
intercollegiate sports

 

 

 

77

 

 

 

 

Pre-Test Descriptives

F ifty-seven percent of the sample were members of a contact-sport, and forty-
three percent were members of a non-contact sport. Respondents ranged between one
and four years of collegiate sports participation, with most having played for one year
(33%). Twenty-three percent had participated for two years, 12% had participated for
three years, and 9% had participated for four years or more. Twenty-three percent of the
respondents did not answer how many years they played sports. This is most likely due
to the location of the question on the questionnaire page (very top of second page), which
probably made it easier to skip than subsequent questions.

Descriptive findings for variables measured on the same 7-point scale are listed in
Table 7. Overall, respondents were somewhat less supportive of the use of interpersonal
violence than they were of general violence. However, the majority of responses
indicated that respondents were generally unsupportive of all types of violence.
Respondents generally reported low hostility toward women, although only 3% indicated
that they did not at all agree with statements endorsing hostility toward women. The
majority of respondents (66%) reported that they mostly or somewhat did not agree with
statements indicating hostility toward women. Respondents were also generally
unsupportive of sex role stereotypes. Only seven percent of respondents indicated that
they at least somewhat agreed with sex role stereotypes, and not one person indicated that
they very much agreed with the statements. Twenty-five percent of respondents were

undecided about the statements.

78

 

 

 

Table 7

Predictor Variable Scale Percentages at Pre-Test (measured on a 7—point scale]

 

 

 

 

 

 

l 2 3 4 5 6 7 . Scale
Variable Not Mostly Some- Un- Some- Mostly Very i .. SD
at all do not what decided what agree much
agree agree do not agree agree """
agree
Acceptance
ofgeneral 1 10 26 39 19 1 l 3.74 1.00
violence
Acceptance
of 22 37 24 14 2 0 l 2.50 1.11
interpersonal
violence
Hostility 03 30 35 22 8 1 0 3.05 1.03
toward
women
Sex role 06 23 38 25 4 2 0 3.05 1.07
stereotypes ’

 

 

 

 

 

 

 

 

 

 

Not surprising with collegiate-level athletes, there was not a large amount of

variance among responses on all three of the sport-orientation variables (see Table 8).

The most variability was seen with respondents’ win orientation, while the least

variability was seen with respondents’ goal orientation. No one indicated that they

strongly disagreed with any of the statements on any of the three subscales, although a

few people indicated some disagreement with some of the items.

79

 

Table 8

Sport Orientation Scale Percentages at Pre-Test

 

 

 

 

 

l

1 r 2 3 4 5
Variable Strongly Slightly Neither Slightly Strongly
agree agree agree nor disagree disagree

disagree

Competitiveness 88 l l l l 0

Win orientation 5 8 i 36 5 1 0

Goal orientation 79 l 20 0 1 0

 

 

 

 

 

 

 

 

To measure sexual aggression severity, items were sealed in accordance with the
classification procedures outlined by Koss & Gaines (1993) in order to score a man
according to the most severe level of sexual aggression he had perpetrated (4-point scale,
0 = sexual nonaggression, 4 = attempted or completed rape). A score of 0, representing
sexual nonaggression, was assigned to the men who responded “never” to all of the SES
items. This score was obtained from 18% of the current study’s sample (see Table 9). A
score of 1 represented uninvited sexual advances (e. g. cat calls and wolf whistles to
women) and was obtained by 56% of the sample. A score of 2, representing unwanted
sexual contact (e. g. unwanted touching of a woman’s buttocks, breasts, or genital area),
was obtained by 13% of the sample. A score of 3, representing sexual coercion, was
obtained by men who answered yes to items indicating that they had attempted or
completed unwanted sexual intercourse by using continual arguments and menacing
verbal pressure (2%). Finally, men who answered yes to items indicating that they had
attempted or achieved unwanted penetration by giving a woman more alcohol or drugs
than she could handle, threatening bodily harm, using physical force, or overcoming her

by a group of men (Koss & Gaines, 1993) received a score of 4, representing attempted

80

or completed rape. The score of 4 was obtained by 11% of the sample. The sample

mean for this scale was 1.31, SD = 1.13.

Table 9

Percent Sexual Aggression Severity at Pre-Test

 

 

 

 

l 1
0 1 2 ‘ 3 4 Scale Scale
Sexual , Uninvited Unwanted Sexual Attempted Mean SD
non- ‘ sexual sexual coercion or
aggression advances contact completed
l rape
%atPre-test 18 l 56 13 l 2 11 1.31 1.13

 

 

 

 

 

 

The scale mean for pre-test scores of rape myth acceptance was 2.55, SD = .76 (7-
point scale; 1 = not at all agree, 7 = very much agree). Respondents were generally
unsupportive of rape myths, although only 4% said that they did not at_all agree with rape
myth supportive statements. Eighty-two percent of respondents said that they either
mostly did not agree or somewhat did not agree with statements endorsing rape myths.
Seven-percent stated that they were undecided, and only one percent somewhat agreed
with the statements. No one stated that he mostly agreed or very much agreed with the
statements.

At pre-test, most participants reported a relatively low likelihood of sexually
aggressing (scale mean = 1.15, SD = .47) (5-point scale; 1 = never, 5 = always). Eighty—
eight percent of participants stated that they would never do any of the sexually
aggressive behaviors listed, even if they thought no one would know. Over 9% of
participants stated that although it would be rarely, they would be willing to force a

woman into doing something sexual by using verbal coercion, alcohol, or physical force.

81

One percent stated that they would sometimes do any of the behaviors (score of 4). The

remaining two percent stated that they were unsure whether they would do any of the

sexually aggressive behaviors listed, and no one stated that he would always do any of

the behaviors (see Table 11).

Table 10

Percent Rape My1h Acceptance at Pre—Test

 

 

 

 

 

 

 

 

 

 

 

 

1 2 3 4 5 6 7 Scale Scale
Not Mostly Some- Un- Some- Mostly Very 1 Mean SD
at all do not what do decided what agree much
agree agree not agree agree agree
% at
Pre- 4 5 1 3 l 7 1 O 0 2-55 .76
test "
Table 11

Percent Likelihood of Sexually Aggressing at Pre-Test

 

 

1 2 3 4
Never Rarely Unsure Sometimes
% at 88 9 2 l
Pre-test

 

 

 

 

 

 

At pre-test, overall, participants reported a relatively high proclivity of

 

confronting a person if they thought a woman was being forced to do something sexual

that she did not want to do (see Table 12). Percentage of respondents clairrring they

would interrupt the situation while it was happening ranged from 56% to 85% across a_ll

scenarios. Percentage of respondents who stated that they would wait and talk with the

82

man ranged from 6% to 29% across all scenarios. The percentage of respondents who
stated that they would wait to talk with the woman was slightly lower, ranging from 4%
to 9% across all situations. A range of 3% to 29% of respondents claimed that they
would do nothing. Participants were most likely to intervene when both the man and the
woman in the scenario were known to the participant. For example, in the scenario in
which a teammate is kissing a woman who is a good friend of the participant, 85% of
respondents claimed that they would interrupt the situation. Likewise, in the scenario in
which a teammate is making a woman do something sexual and she is a good fiiend of
the participant, 82% of respondents stated that they would interrupt the situation. The
above two scenarios were also the two scenarios in which participants were least likely to
do nothing. Only 3% of respondents stated that they would do nothing if they knew both

people in the situation.

83

 

 

 

Table 12

Percent of Responses by Item for Likelihood of Confronting at Pre-Test

 

 

 

 

 

 

 

Do Wait & Wait & Interrupt
nothing talk with talk with the
woman man situation
Sit. 1
See a teammate kissing a woman. . .do not 20% 4% 20% 56%
know woman
Sit. 2
Know a man is making a woman do 14% 9% 10% 68%
something sexual. . . do not know man
Sit. 3
Know a teammate is making a woman do 3% 5% 10% 82%
something sexual. . . woman is good friend
Sit. 4
See a teammate kissing a 3% 6% 6% 85%
woman. . .woman is good friend
Sit. 5
Know a teammate is making a woman do 18% 4% 19% 59%
something sexual. . . do not know woman
Sit. 6
See a man kissing a woman. . . do not know 29% 9% 6% 57%

man

 

 

 

 

 

Multivariate Analyses

To test the hypotheses that involvement in this program would decrease men’s

reports of their rape myth acceptance and likelihood of sexually aggressing and increase

men’s reports of their likelihood of confronting a situation, repeated measures

multivariate analysis of variance (MANOVA) was used to initially compare pre and post-

test scores. Dependent variables for the MAN OVA were rape myth acceptance,

likelihood of confronting, and likelihood of sexually aggressing. The independent

variable was time (two time-points of pre-test and post-test). Total N = 120.

84

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mm 7 2b.- #2. woo. onor mmo. 23v socficoto 53
E _. coo. ovo. moor _mo.- ovo. 3x409 couficoto .80
who. vwo. moo. So. So: woo. 3208 305253800
oo _ .- mm _ .- _m _. wvo. we _ .- MN o .- nZOUo cocaomfitam room 83:00-52 co 68:00
So: 5.. Ear who. o2. omor K L .- ago 38% BEwBBEQE 5 oofiamomtmm as»; u
......mm m 533. ... mmmr wwor ...o o N. 1. mvm. 95$ :o_mm8ww< 338m SFCO bigom
......owm 1&3. moor moor .....Nmm. tQmm. 305 mamboocoom 20m xom
Iowv .....onm. mm _ .- mm? :53. zoom. 89$ 5an “US$81 bzcmom
......omm .....onm. Rot voor .....owm. 1.3 m. 90$ 85.05 36:00 mo 85383..
.583: .....omm. nmor mo_. .....oom. z. 5v. A33 8:205 1385985 we 853686.
ooo; 55%. 1.30. cm _ .- 1. wmv. 2.94m. Ago “mom 3 8.5383. 532 comm
......ono ooo; .....oomr $03.. $an. 1.3%. 2325 8m “a 83383. :52 and
2.30. 2.80. ooo._ 1.3». 153.. we? Amud “mom 8 wEEomEoU mo 80:28:;
om? .....ommr Imam. ooog wmo... mgr God 05 «a mad—Hoodoo «0 8228—5
......wmv. 1.an ERR: wmor ooo._ Iowv. Am<md “mom 3 wEmmoEw< .23:me mo “59:3on
1.9%. 5.9%. we o .- NS .- Iowa. ooo._ 2.4.on 2m 3 mammocwa. bfizxom mo BEES—5
mg 232 NZOU :AOU N<mq Fem;

 

 

month—«5.. 33.35:: 5 633 83.3.2; .8 55a: zeta—2.80 2 ~35.

85

 

 

J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

nmo. mmo. .1. mom: moor .....ommr o5: 9:26 00380200 0:5
......oom. moo. Eor o2. who. omor @209 00550020 .80
5o. mi. moor mo _. oofi- no? 3208 3000353800
E m- ...oo fl .- 1.0mm.- .. mm _ .- om o .- .....wnmr 208 00:36:53 ton—m 003000-002 00 008000
ooo._ S _. ...vmmr mmor :o.- for $56 30% 80%26285 E Boaamomtam £00.» a
S o. ooo._ *3 o. .....Nwm. moor 30 o. AMT/mo :0_mm8ww< Ream SE m0 bt0>0m
*var 3% o. ooo._ .2..me 2.3m. Immm. 30% wfinboocoom 20m xom
mmor .....Nwm. .3..me ooo._ EL. 3.. 5w. 605 00603 @330..- 5:503
_ 5.- moor wamvm. 2:. ooo; 1. omv. 90$ 0000—05 18050 o0 0003303..
o _ o.- 3:. .....mmm. 5.. 5w. .....omv. ooo._ A>H<o 000205 13020825 .0 0053000<
N o o.- .....mmm. .....owm. .....owv. .....omm. 13m. ago 50m “0 00053006V :32 comm
...... SN- zoom. 1&3. 1.03. .....onm. .....omm. Ego 0.5 3 0008000050. fibz 000m
who. .38.. moor mm _ .- For nmor AND-C “mom 3 wEEE-Eou o0 82:60::
om o. moor ooor MS .- woo; mo _. 206 05 H0 wEEEEoU ,0 “50:28:;
omor ...SN. .....Nmm. .....omm. .....owm. ......oom. Redd 30m :0 wEmmewaa bigom .0 “0005—045
Km- :mvm. *ava. 1.23. 1.3 m. 5.. 3w. C<md 0E 00 wEmm00w< Quaxom m0 “0005—35
my; m>m 40m mom >O< >2

llllllllllllllllllllllllllllllllllllllllllllll
ween—«.3. 30:02:32 E comb 833:3 m0 N232 00:29:00 2 035—.

86

 

 

 

 

 

325$ 35 8.0 2: a 285%: m: 8:20.80 ..
235$ .92 3.0 2: a 389%: 2 8:288 ......

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0002 .....m 3. 20$. 1.31m. 23v 5:00:35 53
3.0;. 000.: 1.03. m: _. 3<OOV :0_§:2:O .000
......onc. swoon. coo; if. 3208 3083280800
3.3%. E _. *5 _. 0002 AZOUV :289093 toqm 82:00:02 :0 88:00

go. zoom. So. 2 _.- AmMCC mtoqm 328—3285 E :2022EE 28> n

mmo. N3. mi. .80 _ .- 95$ :0_m8:ww< 330m :0th m0 £85m
......womr 05.- moor :93... 305 wEaonEm 20m xom

30.. of. wofi ...mw _ .- $05 :0803 0330.2. @238:
1.03.- who. oofi- om _ .- 904$ 8:203 18:00 .20 85380.0.

on 2 .- 03.. 32.- .....wbmr A23 8:203 3:028:85 no 853083:

mm _ .- :2. mg. 00 2 .- @325 :8: :0 8:8:80< :32 03%

mgr coo. 30. m m 2 .- 2325 05 :0 8:8080< :52 000%

K 2. 0:0. moo. _m _. ANUAV 50m :0 w50:0¢:00 m0 2000:2025

wwo. 30.. .30. wvo. Cod 8m :0 wE::0¢:0U .«0 0005—025

08.- 28.- :o... wit Am<md 50m :0 mammbw/x 3?:me mo 002:025

mmo. mg. wmo. mm? A 2<mnc 2m :0 w:_mm8w< Ewsxom m0 0005—025

7:3 $00 $200 ZOO

 

 

89:02 82332:: E 0003 832:5» no its: 5:29:00 9 030,—.

87

 

 

After the training, participants were marginally significantly more likely to
sexually aggress (F = 3.87, p = .051), and less likely to confront a situation in which a
person was making a woman do something sexual that she did not want to do (F = 3.66, p
= .058). Participants did not experience a significant change in their level of rape myth
acceptance (F = .685, p = .409). These results suggest the opposite of the first two
hypotheses purported at the beginning of this study. Namely, the hypotheses that 1)
involvement in this program would decrease men’s reports of their rape myth acceptance
and likelihood of sexually aggressing and 2) increase men’s reports of their likelihood of
confronting a situation. Table 14 displays the means and standard deviations of each of

the three outcome measures at pre and post-test.

Table 14

Mean Scale Scores for Outcome Variables at Pre-Test and Post-Test

 

 

 

 

 

Outcome variable Pre-test Pre-test Post-test Post-test
mean standard mean standard Univariate p-

deviation deviation F value
Likelihood of
sexually aggressing 1.21 .49 1.31 .63 3.87 .051
Likelihood of
confronting 3.33 .68 3.22 .94 3.66 .058
Rape myth 2.55 .74 2.50 .85 .68 .409
acceptance

 

 

 

 

 

 

 

Post—Test Descriptives

 

Due to the complex nature of the constructs being measured as outcome variables,

it is important and interesting to explore individual scale differences between pre-test and

88

 

post-test. Scale means for all post-test outcome variables can be found in Table 14.
Broken down by responses given, there was an increase of three percentage points from
pre-test of participants who indicated that they did not alt—all agree with rape myth
supportive statements (7% post-test) (see Table 15). Fifty-two percent of respondents
indicated that they mostly did not agree with rape myth supportive statements (a 1%
increase from 51% at pre-test), and 26% stated that they somewhat did not agree (a 5%
decrease from 31% at pre-test). Fourteen-percent stated that they were undecided,
doubling the amount of people who stated they were undecided at pre-test (7%).
Similarly to the pre-test, one percent somewhat agreed with the statements, and no one

stated that he mostly agreed or very much agreed with the statements.

Table 15

Percent Rape Myth Acceptance at Pre and Post-Test

 

 

 

l 2 3 4 5 6 7
Not at Mostly Some- Un- Some- Mostly Very Scale Scale
all do not what do decided what agree much Mean SD
agree agree not agree agree agree
% at
Pre— 4 51 31 7 l 0 0 2.55 .76
test
% at
Post- 7 52 26 14 l 0 0 2.50 .85
test

 

 

 

 

 

 

 

 

 

 

At post-test, the scale mean for likelihood of sexually aggressing actually slightly

increased, rather than decreased as had been anticipated. Similar to the pre-test, 88%

percent of participants stated that they would never do any of the sexually aggressive

behaviors listed, even if they thought no one would know. Two percent of participants

 

 

stated that although it would be rarely, they would be willing to force a woman into doing

something sexual by using verbal coercion, alcohol, or physical force. This reflects a 5%

decrease in response as compared to pre-test (7%). The percentage of participants who

stated that they were unsure whether they would do any of the sexually aggressive

behaviors increased from 2% (pre-test) to 8% at post-test. Two-percent stated that they

would sometimes do any of the behaviors (score of 4), and no one stated that he would

always do any of the behaviors.

Although the scale mean was used for all analyses in this study, it is interesting to

look at each of the listed behaviors individually, since they represent different although

psychometrically and logically similar types of sexual aggression. Table 16 is a percent

breakdown of each type of behavior endorsed at pre and post-test.

Table 16

Percent Likelihood of Sexually Aggressing at Pre and Post-Test

 

 

 

 

 

 

Never Rarely Unsure Sometimes Always
% % % % % % % % % %
Pre Post Pre Post Pre Post Pre Post Pre Post
Force a woman to do 93 88 2 2 4 8 1 2 0 0
something sexual she
doesn’t want to do
Talk a woman into doing 81 81 12 7 3 9 3 3 1 0
something sexual she
doesn’t want to do
Force a woman to have 95 90 3 0 l 8 0 2 1 0
sex when she doesn’t
want to

 

 

 

 

 

 

 

 

 

 

 

90

 

 

 

 

 

Use alcohol in hopes of 82 82 12 8 3 8 2 2 l 0
having a woman give into
something sexual she
wouldn’t give in to if not
drinking

 

 

 

 

 

 

 

 

 

 

 

The post-test results for likelihood of confronting were somewhat disappointing,
as well as surprising. Although the scale mean did not experience a large change from
pre to post-test, it is interesting to note which situations did experience change, and in
what direction. Even though participants were again most likely to interrupt the
situations in which they knew both of the people involved, they were less likely to
interrupt these situations after participating in the intervention than before the
intervention (see Table 17). For example, participants’ likelihood of interrupting the
situation in which a teammate is making a woman who is the participant’s good friend do
something sexual reduced from 82% at pre-test to 73% at post-test.

For a similar scenario, in which a teammate is kissing a woman who is the
participants’ good friend, participants’ likelihood of interrupting the situation reduced
from 85% at pre-test to 74% at post-test. Even more disturbing, it appears as though for
this latter scenario, most of the variance from the pre-test option of interrupting the
situation shifted to the option of doing nothing (3% doing nothing at pre-test up to 15%
doing nothing at post-test).

It is also possible that these shifts in variance are partially due to the fact that the
above-mentioned scenarios were the items most highly endorsed at pre-test, which
implies that they may have experienced some regression to the mean. However, the

lowest response both for these items (“do nothing”) did not shift toward the mean to the

91

same extent from pre-test to post-test, suggesting that regression to the mean may not
have been very prevalent.

As opposed to the above scenarios in which both peOple were known to the
participant, at post-test respondents were more likely to interrupt two of the situations in
which one or both of the people was unknown to the participant. For one situation, in
which a teammate is making an unknown woman do something sexual, participants
indicated more that they would interrupt the situation after the workshop (68% at post-
test) than before the workshop (59% at pre-test). This stands in contrast to the similar
situation described above in which the woman is known to the participant and he was less
likely to interrupt after the workshop. Additionally, for another situation in which both
the people are unknown to the participant (man is kissing a woman and man is unknown),
respondents were also more likely to interrupt the situation at post-test (65%) than at pre-
test (5 7%).

It is worth noting that at post-test respondents indicated they were more likely to
do nothing for every scenario described, except the last one in which the man was
unknown and the woman was unspecified. Percentage change scores ranged from 1% to
12%. The highest change (12%) was indicated for the scenario in which a teammate is
kissing a woman who is a good friend of the participant, and the least change was seen
for the scenario in which a teammate is making an unknown woman do something sexual

(1%).

92

Table 17

Percent Likelihood of Confronting at Pre and Post-Test

 

 

 

 

 

 

 

 

 

Do Wait & Wait & talk Interrupt
nothing talk with with man the
woman situation
% % % % % % % %
Pre Post Pre Post Pre Post Pre Post
Sit. 1
See a teammate kissing a 20 23 4 4 20 15 56 58
woman. . .do not know woman
Sit. 2
Know a man is making a woman 14 22 9 5 10 10 68 63
do something sexual. . .do not
know man
Sit. 3
Know a teammate is making a
woman do something 3 8 5 7 10 12 82 73
sexual. . .woman is good friend
Sit. 4
See a teammate kissing a 3 15 6 6 6 5 85 74
woman. . .woman is good friend
Sit. 5
Know a teammate is making a
woman do something sexual. . .do 18 19 4 1 19 12 59 68
not know woman
Sit. 6
See a man kissing a woman. . .do 29 24 9 5 6 6 57 65
not know man

 

 

 

 

 

 

 

 

 

Predictors of Post-Intervention Change

 

To more fully examine the relationships between participants’ characteristics and

pre-post change, a sequential regression (also called hierarchical regression) of predictors

on post scores was conducted for each outcome variable. Pre-scores were co-varied out

of the equation by entering them into the first block of each regression equation. Sample

93

size for all sequential regressions was N = 120, which yielded only slightly lower power
than had been anticipated. These analyses tested the remaining hypotheses # 3 — 10
stating that men who had higher levels of goal orientation, win orientation,
competitiveness, prior sexual aggression severity, hostility toward women, sex role
stereotyping, acceptance of interpersonal violence, acceptance of general violence,
participated in more years in intercollegiate sports, and involvement in contact sports
would experience less attitude change due to the workshop (post-test scores).

A total of 10 predictors were entered into each of the three regression equations.
The order of variables that were entered into all three of the equations was the following
(the first block changed in each equation to reflect the pre-score of the outcome variable
post score being used as the dependent variable): first block: rape myth acceptance (pre-
score); second block: acceptance of interpersonal violence, acceptance of general
violence, hostility toward women, sex role stereotyping, sexual aggression severity; third
block: win orientation, number of years participated in intercollegiate sports, goal
orientation, contact vs. non-contact sports participation, competitiveness. The sport-
specific variables were entered into the last block to more closely determine whether the
sport-specific characteristics of athletes make a significant contribution to the variance of
the equation, over and above the non-sport-specific variables typically associated with the
effectiveness of rape prevention education workshops.

The first sequential regression predicted how much variance the ten predictor
variables accounted for in a participants’ post-score on likelihood of sexual aggression
(with pre-scores for likelihood of sexual aggression co-varied out). None of the predictor

variables significantly contributed to participants’ outcome scores for likelihood of sexual

94

 

 

 

aggression. Only the first block, participants’ pre-intervention likelihood of sexually
aggressing, significantly increased R-square (R-square change = .23, F (1,118), p = .000).
As Table 18 shows, the second block that included all of the predictor variables did not
significantly increase R-square (R-square change = .047, F (5,113), p = .201). The third
block, which included only sport-specific variables (e. g. competitiveness), did not
significantly increase R-square (R-square change = .012, F (5,108), p = .863), nor did it
include any coefficients that were significantly different from zero.

The second sequential regression predicted how much variance the ten predictor
variables accounted for a participants’ post-score on likelihood to confront (with pre-
scores for likelihood to confront co-varied out). As Table 19 shows, the first block,
participants’ pre-intervention likelihood of confronting, significantly increased R—square
(R-square change = .525, F (1,118), p = .000). Although the second block did not
significantly increase R-square (R-square change = .036, F (5,113), p = .1 13), the
coefficient for one of the variables in the block, severity of prior sexual aggression, was
significantly different from zero ([3 = -.15, p = .021). None of the other variable
coefficients were significantly different from zero. The third block, which included only
sport-specific variables (e. g. competitiveness), did not significantly increase R-square (R-
square change = .015, F (5,108), p = .574), nor did it include any coefficients that were

significantly different from zero.

95

Table 18

Summary of Sequential Regression Analysis for Variables Predicting Participants’

Likelihood of Sexuallj Aggressing (Post-Test) (n = 120)

 

 

 

 

B B
Model 1
Likelihood of sexually aggressing (pre-test) .62** .48**
Model 2
Likelihood of sexually aggressing (pre-test) .47** .36**
Acceptance of interpersonal violence .03 .05
Acceptance of general violence .06 .09
Hostility toward women -.01 -.01
Sex role stereotyping .08 .13
Severity of sexual aggression .05 .09
Model 3
Likelihood of sexually aggressing (pre-test) .48** .37**
Acceptance of interpersonal violence .02 .03
Acceptance of general violence .07 .11
Hostility toward women .00 .00
Sex role stereotyping .10 .16
Severity of sexual aggression .05 .09
Win orientation .02 .02
Year participated in collegiate sports .06 .10
Goal orientation -.21 -. 13
Contact or non-contact sport participation .02 .01
Competitiveness .05 .03
** p<.00 Model 1: R2 = .230
*p<.05 Model 2: R2 =
Model 3: R2

96

.278 A R2: .047
= .290 AR2= .012

Table 19

Summary of Sequential Regression Analysis for Variables Predicting Participants’

Likelihood to Confront (Post-Test) (n = 120)

 

 

 

 

B [3
Model 1
Likelihood to confront (pre-test) .99* * .72* *
Model 2
Likelihood to confront (pre-test) .99* * .72**
Acceptance of interpersonal violence -.09 -. 10
Acceptance of general violence .00 .00
Hostility toward women .00 -.00
Sex role stereotyping .04 .05
Severity of sexual aggression -.13* -. 15*
Model 3
Likelihood to confront (pre-test) .97* * .71 * *
Acceptance of interpersonal violence -.07 -.08
Acceptance of general violence .00 .00
Hostility toward women -.01 -.02
Sex role stereotyping .09 .10
Severity of sexual aggression -.14* -.16*
Win orientation .18 .1 1
Year participated in collegiate sports .01 .01
Goal orientation .06 .02
Contact or non-contact sport participation .07 .04
Competitiveness - . 03 -.01

** p<.00 Model 1: R2 = .525
*p<.05 Model 2: R2 = .561 A R2: .036
Model 3: R2 = .576 AR2= .015

97

The third sequential regression predicted how much variance the ten predictor
variables accounted for in a participants’ post-score on rape myth acceptance (with pre-
scores for rape myth acceptance co-varied out). As Table 20 shows, the first block,
participants’ pre-intervention rape myth acceptance, significantly increased R-square (R-
square change = .448, F (1,118), p = .000). The second block also significantly increased
R-square (R-square change = .080, F (5,113), p = .003). The coefficients for two of the
variables in the block, severity of prior sexual aggression ([3 = .19, p = .007) and
acceptance of general violence (B = .21, p = .005), were significantly different from zero.
This finding suggests that acceptance of general violence and severity of prior sexual
aggression predict pre-post change in rape myth acceptance. In other words, participants
who had a higher acceptance of general violence and more severe prior sexual aggression
before the workshop were less likely to have decreased acceptance of rape myths after
participating in the workshop. None of the other variable coefficients were significantly
different from zero.

With the addition of the third block, which included only sport-specific variables
(e. g. competitiveness), the model did not significantly increase R-square (R-square
change = .023, F (5,108), p = .357), nor did it include any coefficients that were

significantly different from zero.

98

Table 20

Summary of Sequential Regression Analysis for Variables Predicting Participants’ Rape

 

Myth Acceptance (Post-Test) (n = 120)

 

 

 

 

 

B 13
Model 1
Rape myth acceptance (pre-test) .77** .67**
Model 2
Rape myth acceptance (pre-test) .58* * .50* *
Acceptance of interpersonal violence -.01 -.02
Acceptance of general violence .19* .21 *
Hostility toward women .10 .1 1
Sex role stereotyping .00 -.01
Severity of sexual aggression .15* .19*
Model 3
Rape myth acceptance (pre-test) .63** .54* *
Acceptance of interpersonal violence .00 .00
Acceptance of general violence .15* .17*
Hostility toward women .07 .08
Sex role stereotyping .02 .03
Severity of sexual aggression .14* .17*
Win orientation -.15 -. 10
Year participated in collegiate sports .10 .12
Goal orientation .09 .04
Contact or non-contact sport participation .16 .09
Competitiveness .04 .02

Model 1: R2 = .448
Model 2: R2 .528 AR2=.080
Model 3: R2 = .551 AR2= .023

** p<.00
*p<.05

99

Program Utilization

 

General Program Usefulness

Approximately seventy-five percent (n ranged from 93 to 95) of the respondents
answered the three closed-ended utilization questions at the end of the post-test, while
slightly fewer respondents answered the open-ended items (n ranged from 55 (46% of
respondents) to 75 (62% of respondents)).

Individual-utilization. Even though sixty-percent of respondents stated that they
would either somewhat (44%) or very much (16%) use the information presented in the
workshop, forty-percent of respondents stated that they would not very much (21%) or
not much at all (19%) use the information presented in the workshop. The mean for this
question was 2.44, SD = .98 (four-item scale, ranging from 1 = very much to 4 = not
much at all) (n = 94, 76% of respondents).

Factors determining individual-utilization. When asked to explain their reasoning

 

for using or not using information from the workshop, respondents indicated a variety of
reasons (n = 63). The highest percentage of responses reflected whether or not the
respondents felt the program offered new information (see Table 21).

Team utilization. Interestingly, respondents reported about the same percentage

 

for individual (self) utilization as they did for the extent to which they thought that their
teammates would use the information presented at the workshop. Fifty-eight percent of
respondents stated that their teammates would either somewhat (48%) or very much

(10%) use the information presented in the workshop. Forty-two percent of respondents

stated that their teammates would not very much (24%) or not much at all (18%) use the

100

information presented in the workshop. The mean for this question was 2.51, SD = .90

(four-item scale, ranging from 1 = very much to 4 = not much at all) (n = 93, 74%).

Table 21

Factors Determining Individual Utilization (n = 63)

 

 

 

 

 

Will use the information Will use the
“not very much” or “not information
much at all” “very much” or
n=44a “somewhat”
n=19a
1. No new information/common
sense 66%
(n = 29)
2. These situations do not happen
often 16%
(n = 7)
3. Would behave in the ways
suggested by the program 9%
anyways (n= 4)
4. Would not behave in the ways
suggested by the program 7%
(n = 3)

 

 

,/,l I via .0 -. -/ .
’ r 1' fl 1 . 4'
/ /

 
  

 
  

 

 

 
 
 
 
 
 
    

 

5. New information 79%
(n = 15)

6. These situations happen often 21%
(n = 4)

 

 

 

percent 15 broken 1nto two categories ofresponse wrllusemformatlon or‘wnll not use information .

Each category totals 100%.

Factors determining team utilization. When asked why they thought that their

 

teammates would or would not utilize the information provided in the workshop,

respondents indicated a variety of reasons (see Table 22) (n = 55). It is interesting to note

that three of the four categories of response given for why a team would not use the

101

information refer to inherent factors within the team, rather than factors relating directly
to the program (one of the four factors given). Yet, for responses indicating that the team
would use the information, one of the two reasons given referred to inherent factors

within the team, while the other referred to the program itself.

Table 22

Factors Determining Team Utilization (n = 55)

 

 

 

 

 

Team will use the information Team will use the
“not very much” or “not much information
at all” “very much or
n=36a somewhat”
n=1 921
1. Lack of attentiveness or
caring for program 43%
material (n = 16)
2. Information was “common 26%
sense” (n = 9)
3. Team is inherently bad 17%
(n = 6)
4. Team lacks intelligence 14%
(n = 5)

 

 

 
   

 
 
 

    

    

.' ‘ ,4 .. .
// 'f
/’ . ’.-/g/

 

 

 

5. New information 53%
(n = 10)

6. Teammates are inherently 47%
good (n = 9)

 

a . . . “ . . . ,, “ . . . ,,
percent 15 broken mto two categories of response w1l1use1nformatlon or Wlllnot use lnformatlon . Each category
totals 100%.

Confrontation With A Teammate

 

Findings suggest that respondents felt the workshop would help them most with
their ability to confront a teammate. Overall, respondents reported that the information
presented in the workshop would help them confront a teammate in certain situations in

which he was doing something with a woman that they thought was wrong (11 = 95,

102

76%). Forty-four percent of respondents stated that the information presented in the
workshop might help them in certain situations, while seventeen percent stated that the
workshop would help them in most situations. Almost 12% of respondents stated that the
information presented in the workshop would help them in all situations, while 10%
stated that the information presented would not help them at all, in any situation. Fifteen
percent of respondents stated that they would have already stood up to a teammate before
the intervention.

Factors determining confrontation with a teammate. The seven categories of

responses that were coded for this open-ended item are listed in Table 23 (n = 75).

Table 23

Factors Influencing Respondents’ Decision to Confront a Teammate (n = 75*)

 

 

 

 

 

 

 

 

 

Factor Percent coded “yes”8
1. Cues from the woman (n = 19) 25
2. Situational factors (11 = 15) 20
3. Protecting teammate (n = 13) 17
4. Personal morals — it’s just wrong (n = 12) 17
5. Relationship with teammate (n = 12) 16
6. Relationship with woman (n = 9) l3
7. Cues from teammate (n = 8) 9

Total table percent sums to over 100% because respondents could list more than one factor.
*n = 75 responses, but some responses listed multiple factors. The total number of factors given was n = 93.

103

Chapter 4

DISCUSSION

The current study was the first to not only evaluate the effectiveness of a sexual
assault education program targeting male collegiate athletes, but also the first to explore
factors specifically related to the sports experience that may influence a program’s
effectiveness. Researchers have suggested that collegiate athletes have special
characteristics that make them more prone to sexually aggress (Crosset et al., 1996), and
worthy of a tailored prevention program format which reflects their special life
experiences (Katz, 1995). The findings of this study addressed these issues, and reflect
many of the challenges that are faced when delivering a sexual assault prevention
program to a specific population like male collegiate athletes. Although some of the
findings of this study were disappointing (and at times downright disturbing), they offer
valuable information pertaining to the planning and implementation of sexual assault
prevention programs targeting male student athletes.

The core finding that the program was unsuccessful in attaining its three specific
goals of 1) reducing men’s rape myth acceptance, 2) reducing men’s likelihood of
sexually aggressing, and 3) increasing men’s likelihood of confronting, was
disappointing. Even more disturbing was the reverse effect seen with men’s increased
likelihood of sexually aggressing and lessened likelihood of confronting after attending

the workshop. Although there are many factors to consider when interpreting these

104

 

findings, these results reflect the difficulty in targeting male collegiate athletes with
sexual assault prevention programming. Open-ended responses from participants helped
to clarify areas of improvement for the development and implementation of this type of
educational workshop. Even so, the apparent lack of effectiveness of such programming
needs to be seriously considered when efforts are made to develop future programs.

One of the attitudes most commonly targeted for change via workshops such as
the current one is participants’ acceptance of rape myths. It is interesting to note that
even on such a seemingly basic concept to rape prevention education, no positive change
was found for the current study. One of the reasons for this lack of significance may
have been the scale’s limited range seen through the low acceptance of rape myths at
both pre and post-test. It is possible that the scale was designed to be applicable for an
audience with a wider range of attitudes about rape myths than those exhibited by male
collegiate athletes. Indeed, the original scale psychometrics were obtained from an
undergraduate non-intercollegiate athlete population consisting of both males and
females in psychology classes (Payne, et al., 1999). It is possible that although a general
undergraduate population is sometimes viewed as a narrowly defined population,
attitudes of people within that population may consist of a wider range than an even more
specific population like male collegiate athletes. A more appropriate scale for an
audience of male collegiate athletes would have captured more of a range of their
attitudes.

Nevertheless, although mean differences between test administrations on
participants’ rape myth acceptance (the first outcome measure) were deemed non-

significant, it is worthwhile to note the response differences between pre and post-test. It

105

 

 

is interesting that more people were undecided about their adherence to rape myths after
the intervention than before the intervention. At first it might be construed that an
increase in this percentage is a positive outcome, given that being undecided about rape
myths is at least better than agreeing with them. However, the percent of agreement with
rape myths actually stayed relatively equal between pre and post-test (1% for both pre
and post-test). This indicates a rather negative outcome, in that some participants
actually moved from disagreeing with rape myths to being undecided about their
agreement with them. It is important to note, though, that participants started the
intervention largely unsupportive of rape myths (pre scale mean = 2.52, SD = .76; 7-point
scale; 1 = not at all agree, 7 = very much agree), so there was not much room for change
in attitude. One of the goals of the intervention was to reduce support for rape myths, yet
the level of reduction may have been limited by participants’ reported low acceptance of
rape myths before the intervention (at pre-test). The myths and facts exercise that
incorporated rape myths was introductory in nature, and mainly intended for an audience
who had never been presented with rape myths or topics related to sexual assault before.
For reasons discussed later in this section, it is possible that a more in—depth discussion of
rape myths was necessary for creating attitudinal change that would have been reflected
in the outcome measures used for this study.

The findings that were the most disappointing, however, were the possible
“boomerang,” or rebound effects that the program seemed to have on participants’
likelihood of sexually aggressing and likelihood of confronting (the two remaining

outcome measures). Despite the fact that there was no significant change in participants’

106

 

 

 

acceptance of rape myths, participants were slightly more likely to sexually aggress and
less likely to confront after participating in the workshop.

The potential boomerang effect seen with participants’ likelihood of sexually
aggressing and likelihood of confronting is not unique to this study. Berg et a1 (1993)
conducted an intervention divided into two groups of men: one group that heard a
woman’s story of being assaulted, and another that heard a man’s story of being
assaulted. Both groups reported no significant change in victim empathy, while the
group that heard the woman’s story actually reported a greater likelihood of sexual
aggression than before hearing the story. Although Berg et al (1993) hypothesized that
men’s lack of empathy may be due to their lack of connection to the issue of sexual
assault; the rebounding effect may in fact be due to other reasons.

It is possible that giving participants information they already feel they know or
do not want to hear has a threshold of acceptance for the participant, which results in the
dismissal of information that pushes participants past their threshold. Persuasion
researchers have studied this effect and have incorporated it into social judgment theory
(O’Keefe, 1990). Social judgment theory asserts that people have issue-specific latitudes
of acceptance (the positions that the respondent finds acceptable), latitudes of rejection
(the positions that the respondent finds unacceptable), and latitudes of noncommitment
(the positions that the respondent neither accepts nor rejects). A person’s attitude change
(or adherence to persuasive information like that which was provided in the current
study’s program) depends on where they fall on these three latitudes. For example, if
information about sexual assault prevention falls into a participant’s latitude of

acceptance, that person will more likely be persuaded by the information (they already

107

 

agree with the position being advocated). On the other hand, if a person (pre-workshop)
does not agree with being educated about sexual assault prevention, they will be harder to
persuade. O’Keefe (1990) explains how the concept of attitudinal latitudes may
contribute to the boomerang effect seen in the current study:
“A communication that is perceived to advocate a position that falls in the latitude
of acceptance or the latitude of noncommitment will produce attitude change in
the advocated direction (that is, in the direction sought by the message), but a
communication that is perceived to advocate a position that falls in the latitude of

rejection will produce no attitude change, or perhaps “boomerang” attitude
change (that is, change in the direction opposite that advocated by the message)”.

(p. 37)

Additionally, if a person is highly personally involved with the topic (i.e. “ego-
involved”) the size of their latitude of rejection will be extremely large, while their
latitude of acceptance and noncommitment will be small. This is mainly due to the fact
that the more involved someone is with an issue, the more opinionated they tend to be —
they have thought about the issue and have formed strong Opinions about it, so they are
less likely to want to change those opinions and more likely to be oppositional to
contradictory or new information. At first, it may seem that since some men have
traditionally not considered sexual assault an issue that is relevant to their lives (i.e., it’s a
“woman’s issue”), their level of involvement should be low. If their involvement were
low, we would expect to see a larger latitude of acceptance and a smaller latitude of
rejection. However, it is very possible that at least some of the athletes who attended the
current study’s program were actually highly involved in the topic of sexual assault
prevention. This is not to say that they were supportive of sexual assault prevention
itself, but rather involved in their attendance at the program and the fact that they felt

unfairly targeted for a prevention intervention. This would support the possible

108

 

 

 

boomerang effects seen with two of the outcome measures. As a result of respondents
being more personally involved with the intervention, they possibly had a limited latitude
of acceptance and a high latitude of rejection.

It is also probably not just coincidence that the potential boomerang effect was
seen on the two outcome measures that were obviously related to a person’s personal
involvement with sexual assault prevention (confronting a perpetrator and actually
sexually aggressing). Although rape myths (the other outcome measure) are undoubtedly
related to sexual assault (Fonow et al., 1992; Koss & Gaines, 1993), they are more clearly
connected to people’s opinions about the issue of sexual assault, and not as obviously
related to a participant’s behavior or personal decisions. This may be one reason why
rape myth acceptance did not endure a boomerang effect - participants were less “ego-
involved” when just asked to provide their opinions. The more personal an outcome
measure was, the higher possibility that it provoked the participant to feel “on guard” or
personally attacked, which would involve them more with the issue of being unfairly
targeted for the intervention. Indeed, the attitude seen with many of the current
program’s participants seemed to be one of skepticism and reluctance to attend. In
candid conversations the researcher had with participants after workshops, some men
actually still questioned why they had to be there and were visibly upset that they had to
attend the program. A few participants asked if non-athletes on campus had to go
through the same program, explaining that they felt the athletes were being unfairly
targeted. Even the name of the program was changed to try and cover it’s targeting of
athletes, and to hopefully prevent their feeling unfairly targeted (the word athlete was not

included in the program name when it was delivered to participants).

109

Some of the results from the utilization questions at the end of the program
evaluation reflected some of the resentment that participants felt for having to attend.
Respondents claimed that the main reason they (as individuals) would not utilize the
program information was because they had already “heard the information before,”
indicating that the possibility that participants had a low latitude of acceptance for the
information was prominent. The fact that “already knowing the information” was the
second main reason why participants felt their teammates would not use the information,
preceded by their teammates not caring about the program material or wanting to pay
attention, further supports the possibility that participants rebounded because the
information given in the program fell directly into their latitude of rejection.
Unfortunately, only 46% of respondents chose to answer the open-ended question
relating to why their teammates would or would not use the program information, and
only 52% of respondents answered the question relating to why they as individuals would
or would not utilize the program information. As a result, the interpretation of these
results must be taken with caution. It is possible that a higher response rate would be
necessary to capture the complexity of reasoning behind why people think their
teammates or themselves would use information from a program such as this.

The interpretation of these findings is not to say that programs should avoid
delivering new information to people with high latitudes of rejection, or low latitudes of
acceptance. Rather, solutions offered by the persuasion literature advocate for more
formative evaluation to help tailor the delivery of information at a rate that is acceptable

(and therefore less likely to boomerang) for the audience (O’Keefe, 1990). In other

110

words, it is essential to know where your audience is coming from before trying to
change where they are going.

Lonsway (1996) stated in her review of rape prevention programs that program
developers need to look to the field of persuasion research and tap into the strides that
persuasion researchers have made with program message effectiveness. The findings
from this study not only support Lonsway’s assertion, but also add more information on
which to base the assertion. Namely, not only do persuasion theories need to be used in
the development of program content and delivery, but also a more in-depth investigation
of the targeted audience needs to inform the program developers in their choice of
persuasive theories and rate of information delivery. Programs targeting specific
populations (such as male collegiate athletes) who may have multiple resistances to the
program, are in special need of clearly pre-defining the attitudes of the target audience to
optimize the persuasive effectiveness of the program. Although most exploration of
target audience attitudes is done through surveys, as discussed later in this section,
involving students athletes themselves in the development and implementation of the
program would be another way of ensuring that the program is appropriately addressing
issues that are salient to a male collegiate athlete’s life.

Despite the value of understanding the function of boomerang effects and how to
avoid them, it is still important to review some of the implications of the results found for
respondents’ likelihood of sexually aggressing. Participants’ responses to the pre-test and
post-test measures of this outcome tell a more detailed story than that allowed by only
reporting (and using) the scale mean. A participant’s likelihood of sexually aggressing is

also important to interpret in the context of how it was measured. An increase in score

111

(and therefore “likelihood of sexually aggressing”), while not ideal, may not be as dire as
it seems upon first glance. While a score of 1 indicated that a person would never do the
aggressive behavior, and a score of 5 indicated that a person would always engage in the
aggressive behavior, a score of 3 indicated that he was unsure of whether or not he would
be sexually aggressive. The mean score for participants’ likelihood of sexually
aggressing increased slightly (1 .21 at pre-test to 1.31 at post-test) indicating that
participants were more likely to sexually aggress after the intervention. Interestingly, the
only category of response that increased at post-test was “undecided”, while more
extreme responses (scores of 4 “sometimes” or 5 “always”) remained consistent.

Although uncertainty was not the goal of the workshop, nor is it a particularly
desirable response, it may be viewed as more positive than if someone says they rarely
will sexually aggress. Even if someone is rarely sexually aggressing, they still are
sexually aggressing, whereas if they are unsure as to whether they would or not, they are
considering possibly not doing it. In addition, the scores for this scale must be viewed in
the context of the limited number of people who stated they would sexually aggress at
pre-test. Since most people were already reporting they would never or rarely sexually
aggress, there was a much narrower gap to close than if more people had reported at pre-
test that they would sometimes or always sexually aggress.

It is also important to note that the skewness of the likelihood of sexually
aggressing variable at pre-test may have complicated multivariate analyses. Considering
that at pre-test the majority of respondents answered that they would “never” do any of
the sexually abusive behaviors (scale mean = 1.15), there was little room for positive

change at post-test due to the workshop.

112

The open-ended and closed-ended utilization questions added interesting insight
into the possibility of group-level attitude and/or behavior change. As mentioned earlier,
one of the most important factors in determining whether or not a member of a group will
take the minority position and resist the majority pressure is if they feel confident and
informed enough to take that action. It was encouraging that 86% of respondents stated
that the information presented in the workshop would help them confront a teammate if
that teammate was making a woman do something sexual that she did not want to do.
Even though some respondents did not mention a teammate’s actions as a factor worthy
of confrontation, the fact that some people did (and some were morally opposed to sexual
assault), lent some hope that these latter people would be the ones to take a dissenting
minority position that would start group-level change. Interactive discussions that
incorporated realistic scenarios in which peer pressure is used by teammates were
purposefully incorporated to increase men’s ability to stand up to their teammates.

It was encouraging to see that although attitude change was harder to come by,
giving participants the right information about sexual assault and peer confrontation was
something that helped most men feel more able to confront their teammates. Indeed,
most of the participants who stated that they would use the information presented in the
workshop said that the reason it would help them was due to new or useful information
that the program offered. Unfortunately, due to the lack of people who chose to answer
this Open-ended question (52% of respondents) the number of people who said that the
program offered new or useful information was only fifteen people. These results must
be interpreted with caution, as a higher response rate must be attained to determine the

strength of any findings such as the ones presented here. Nevertheless, it is interesting to

113

consider that the factors related to participants’ confrontation of a teammate might stand
in the way of any action being taken by participants who end up actually being in a
situation like one of those described in the workshop.

Two additional utilization-focused items aid in the discussion of what factors
participants felt would influence their decision to confront a teammate. Eighty-six
percent of respondents reported that they thought the information presented in the
workshop would help them confront a teammate. Yet, this high percentage stands in
conceptual contrast to the fact that all but one of the factors respondents stated would
influence their decision to confront their teammate were not discussed in the workshop.
For example, the factor given most frequently was the woman’s behavior or cues from
the woman, and the factor least given was their teammate’s behavior or cues from the
teammate. Their teammate’s behavior was the only factor that was at all advocated in the
intervention, as part of the goal of helping participants learn how to confront a teammate.
This discrepancy is concerning, given the fact that participants reported the program
information was useful, yet the factors they considered most important when deciding to
confront a teammate had little to do with the information from the workshop. Essentially,
the information about confronting a teammate provided by the workshop will be useful to
them when they decide that the factor most important in a particular situation is related to
their teammate’s behavior. Although a high percentage of respondents indicated that the
information from the program would be useful, the likelihood that it would actually be
put to use needs to be viewed in the context of the limited teammate-centered factors
participants felt indicated that a situation was worthy of confrontation. It is important to

consider these somewhat contradictory findings in light of the number of people who

114

 

 

chose to answer the open-ended question that captured what factors participants would
consider when deciding to confront a teammate. It is possible that clarifying information
relating to the actual use of program information would have been more feasible if more
than just over half (62%) of respondents had answered the question.

One of the reasons for the lack of desirable responses to the confrontation
questions may be related to the fact that the scenario exercise did not often culminate into
a right or wrong set of answers. The program developers and facilitators had wanted to
engage the participants in a discussion about each scenario, and in doing so, discussed
each person’s options and consequences around certain decisions for action. However,
after participants gave their ideas for behavioral options, the facilitators did not
specifically say that the decision was a wrong one, or clearly state what the right decision
would be. The intention of this approach was to emphasize to participants that each
person has multiple decisions to make in a situation and each decision has a good or bad
consequence, depending on the situation. It is up to the participant to decide on his own
what consequence is best based on each unique situation. This approach was modeled
after the Mentors in Violence Prevention Project, which has taken a similar stance:
“When we discuss with the young men their options for intervention in the various
scenarios, we are careful not to choose for them the “best” option; that choice is for each
person to make based on a unique set of circumstances” (Katz, 1995, p.168). Yet, the
lack of explicitly labeling each decision as “good” or “bad” may have prevented
participants from considering certain decisions more desirable than others. More clarity
and a well-stated consistent program stance on certain options given by the audience

would most likely help to more clearly advocate one decision over another and be more

115

effective at desirable attitude change among participants. This finding is particularly
salient in light of the fact that the current study is the first to evaluate this approach for
presumed effectiveness. Although the current program was significantly shorter in length
than the Mentors in Violence Prevention Project (one hour or multiple sessions,
respectively), it is worth further exploration into whether concise directional messages
are needed to have an attitudinal impact for shorter programs.

Another interesting finding that emphasizes the importance of team relations to
athletes is the fact that the third largest factor in participants’ decision to confront a
teammate was to protect their teammate, while not one person mentioned that protecting
the woman was a factor. It could be construed that the respondents who stated that a
factor in confrontation was their moral opposition to sexual assault could have been
considering women’s best interests in being against sexual assault. However, the fact
that many respondents would act on behalf of their teammate’s best interests, yet no one
specifically mentioned acting on behalf of the woman’s best interests, shows the strong
influence that peer relations have in a team setting. It is for this reason that the current
workshop focused on empowering members of a team to confront and sometimes use
peer relations as a primary component to create group-level change within the team.

However, as reflected by the lack of desired significant findings for this study, it
is doubtful that much peer interaction would take place as a result of the workshop
(resulting in a lack of group-level change). One of the best ways to influence such a
peer-focused group is to use peer facilitators in the development and delivery of the
workshop itself (Lonsway et al., 1998; Mahlstedt & Jacobson, 2000; Smith & Welchans,

2000; Vickio et al., 1999). Unfortunately, student-athletes were not used in the program

116

 

development for the current workshop, nor were they used as facilitators. It is likely that
the use of student athletes in the development stages of the workshop may have tailored
the workshop in a more effective way, since student athletes are most familiar with what
information is relevant to their lives. Peer involvement would not only aid in better
curtailing the program’s delivery and content; peer facilitators would help bring the
attention and focus of participants back to the program. Merely by their visibility, peer
facilitators would help to convince participants that the program material was relevant
and worthwhile to their lives, and may increase the chances that individuals would be
more likely to take the chance of being a minority dissenter and help create group-level
change.

Other programs, such as the Mentors in Violence Prevention Project (Katz, 1995),
the Fraternity Violence Education Project (Mahlstedt & Jacobson, 2000), and the First
Step Peer Education Project (Smith & Welchans, 2000) strongly advocate the use of peer
facilitators. Although research on the effectiveness of such programs in reducing rape-
supportive attitudes is relatively scant, there is support that peer facilitation is an effective
method of intervention for short-term attitude change (Foubert & Marriot, 1997; Heppner
et al., 1995; Smith & Welchans, 2000). Not only do peer-facilitated programs offer a
“safe” environment for men to discuss their honest feelings about the issue of sexual
assault, it is possible that athletes involved in peer facilitation may actually gain a sense
of “ownership” of the program and become more heavily involved in the marketing and
delivery of the program to their fellow student athletes. Incorporating the use of student
athletes into the design and implementation of a program targeting student athletes is of

utmost importance and may indeed result in a much different (and positively significant)

117

outcome than seen in the current study which neglected to utilize current student athletes
at all stages of implementation.

The concept of using student athletes to “sell” the program to other student
athletes is related to the larger issue of ensuring that participants see how the program is
relevant to their own lives and experiences. After examining participants’ open-ended
responses to the question of why they felt they would or would not use the information
presented in the workshop, it was apparent that people thought they were either
previously well educated about the topics presented (“already knew the program
information, information was common knowledge”) or were previously uneducated about
the topics presented (“program offered new/useful information”).

One issue that is also related to participant’s latitude of acceptance, is that
collegiate athletes often receive a high amount of programming about how to behave
appropriately through “life skills” programming on topics such as dangers of drug use,
how to stay off of academic probation, and relationship issues. It was not surprising that
some of the players felt that they had “already heard this stuff before” if they were
considering this program just one of many that they were presented during the course of
one semester. Indeed, the program was scheduled and promoted by the same student
athlete academic support center that provides other “life skills” trainings. If participants
perceived the program as nothing new or useful, it is highly likely that the information
given would have fallen directly into their level of rejection, thereby reducing program
effectiveness.

This underscores the importance that future programming efforts targeting

athletes take care to market their programs as being different than other “life skills”

118

trainings throughout the year. Having the participants know that the program they are
attending is not just another “lecture” on how to stay out of trouble might help them to
pay attention and come into the program expecting to learn something unique (which
may also function to increase their latitude of acceptance). Again, it is important to note
that the main reason people indicated that their teammates would not use the information
presented in the workshop was that they thought their team “did not pay attention during
the program or did not care about the program material.” Although it is improbable that
the marketing for one workshop could work to undo years of sexist comments that make
men feel that sexual assault is just a “woman’s issue,” and therefore not worthy of their
attention, having the program be viewed as unique and useful to the players (and not just
“targeting” them) should help to keep their attention and increase their level of
acceptance for the information presented.

Although program marketing and scheduling may seem like a peripheral concept
for a sexual assault prevention program, it is of utmost importance when programs target
a particular population of people (such as student athletes) who have an unequal amount
of power and privilege compared to the rest of the general population. There are many
reasons why a program needs to essentially be “sold” to male collegiate athletes. One
reason (as discussed above) is that many male collegiate athletes feel unfairly targeted as
“guilty” because they are asked to attend a program about sexual assault prevention.
Although the issue that some athletes may have felt negatively targeted was taken
seriously by the current program’s developers (seen through change in program title, and
the use of a video and discussion throughout as to why they were there), it seems as

though the scope of this problem is almost too extreme for one program to conquer.

119

However, there are ways to increase the interest of participants and decrease the
amount they feel unfairly targeted. Aside from peer involvement in the planning and
implementation of the program, one helpful tool in marketing a program is coach
involvement in the planning and implementation of the program. Unfortunately, coach
involvement was severely lacking in the current study’s program development, which
undoubtedly impacted the reluctance of many of the program participants to participate in
the workshop. When coaches are involved in the planning of a program, they may not
only offer needed insight into the players’ lives which helps frame program content and
scheduling (e. g. scheduling the workshop when teams are out of season and have more
time), they may also help influence whether or not players feel justified in going to the
program. Even though collegiate athletes have elevated power over most domains of the
college campus, they rarely have power over their coaches who decide the fate of their
athletic career on a daily basis (Benedict, 1998). Having coaches involved with and
openly supportive of the program helps set the standard for player involvement and
accountability. Although coaches were not involved in the planning of the current
program, coach involvement at the trainings themselves obviously played a role in
players’ willingness to participate. The three sessions that had the highest attendance and
best audience participation were swimming, football, and basketball. For all three
trainings, the head coach was present at either the beginning or the end of the program,
and in some cases even introduced the program to the team. It was clear through the
introduction, as well as simply the coaches’ presence at the training, that the coach felt

the program was important and worthy of the team’s attention.

120

The strength of the effect of coach involvement is even more pronounced when
considering that men’s football and basketball are the two most high profile teams on
campus, often resulting in a lack of player accountability (Benedict, 1998). It was
amazing to not only get some players from these teams to the trainings, but most players
from these teams to the trainings. This was indeed a clear sign that coach involvement
with and visibility to the players increased players’ accountability to the prevention
program. It is highly likely that only through increased player and coach involvement
will participant accountability for and attendance at a prevention program be increased.

The findings related to which characteristics of the athlete (sport-specific or non
sport-specific) predicted post-intervention change are interesting to consider. There are
many sport-specific factors that have been suggested to predict the effectiveness of a
sexual assault prevention program targeting male collegiate athletes (e. g.
competitiveness, contact vs. non-contact sport participation, number of years participated
in intercollegiate sports). It is interesting to note that none of the sport specific variables
measured in this study contributed a significant amount of unique variance to any of the
three outcome variables. Three of the variables (win orientation, competitiveness, and
goal orientation) were used to identify if there was a component of an athlete’s sport-
related personality that would be more closely associated with his attitude change due to
participation in the workshop. The other two sport variables (contact vs. non-contact and
number of years participated in intercollegiate athletics) are both components of the
extra-individual sport experience that have been suggested in the literature to have a
possible relationship with an athlete’s sexual aggression (Benedict, 1998; Boeringer,

1996; K035 and Gaines, 1993).

121

However, neither extra-individual characteristics of the sports experience, nor
intra—individual personality factors played a significant role in athletes’ attitude change
due to participation in the workshop. Rather, significant factors included non sport-
related variables that are commonly used in the field of sexual assault research.
Participants’ acceptance of general violence as well as their severity of prior sexual
aggression both significantly predicted change in participants’ rape myth acceptance due
to the workshOp. It is unclear why participants’ acceptance of general violence, and not
their acceptance of interpersonal violence, was significantly related to the outcome of the
workshop. One would think that statements endorsing general violence (e. g. “any
prisoner deserves to be mistreated by other prisoners in jail”) would be less related to
attitudes around sexual assault, than statements endorsing interpersonal violence (e. g. “it
is alright for a partner to hit the other if they flirt with others”). It is possible that there is
a more complex relationship between participants’ attitudes about sexual assault and their
acceptance of general and interpersonal violence that was not captured within the scope
of the analyses used in this study.

It was not surprising that participants’ level of prior sexual aggression
significantly predicted participants’ post-intervention change in rape myth acceptance. It
is likely that the more severely a man sexually aggresses, the more convoluted and
extreme his perceptions of sexual assault and rape victims are, hence less of a chance that
his attitudes about rape myths would be affected by a one-hour workshop. This would
also serve to further expand participants’ latitude of rejection, which reduces the chances

of attitude change due to the workshop.

122

It was not surprising either, that the only variable that significantly predicted
partieipants’ post-intervention change in likelihood of confronting was participants’
severity of prior sexual aggression. For the same reasons listed above, attitude change
would be more difficult if a person had perpetrated some level of sexual aggression and
had in any way started to justify or ignore his actions. People who have more severe
levels of sexual aggression may also be less likely to recognize situations that are
potentially abusive, so educating them about the warning signs of abuse and how to
intervene could possibly take longer than the short period of time allotted for the current
program.

Initially, the lack of significant sport-specific predictors related to post-
intervention change does not support the inferences that have been made suggesting that
athletes have specific factors that are associated with their level of sexual assault.
Although the addition of sport specific variables to each of the three regression equations
did not significantly predict post—intervention change, it is important to recognize two
reasons for why these sport—specific variables were not found to be significant. One
reason may be their interrelatedness to other non sport-specific variables that were found
to be significantly related to change in attitude at post-test. It is possible that involvement
in sports may less overtly affect a man’s scores on other attitudinal factors related to
program effectiveness (e. g. severity of sexual aggression). For example, it has been
found that a man’s hostility toward women is significantly related to his severity of
sexual aggression (Koss & Gaines, 1993). Men’s sports are largely known. to foster an
anti-female sentiment, and be socially degrading toward women, female sexuality, and

anything feminine (Crosset et al., 1996; Nelson, 1994). It is this overt opposition to

123

anything feminine in men’s sports that may increase some male athlete’s hostility toward
women. A further investigation of this relationship is warranted, however, because not
all male athletes are hostile toward women. This relationship, as well as the
relationships between other general attitudinal and sport-specific factors could only be
explored with the use of a non-athlete comparison group, which was not included in the
current study. However, more work needs to be done to determine the effects of sport-
specific characteristics on general attitudinal variables that are related to sexual assault
and the effectiveness of sexual assault prevention programs. Before researchers conclude
that sports involvement (or the culture) is indicative of sexual assault, the missing link
between sports involvement and general attitudes related to sexual assault needs to be
clarified.

In addition to the above needed conceptual clarifications, a second reason for the
lack of sport-specific variable significance in the prediction of post-intervention
attitudinal change may be the possible limitation of testing intra-individual sport
variables (e. g. competitiveness) with a specifically defined group such as intercollegiate
athletes. A restriction of scale range seemed to be prevalent as seen through the lack of
variability on each of the sport scales (most participants scored high on all three of the
scales) used in the current study. This was not surprising, when taking into account how
devoted and focused on sports one must be to compete at the intercollegiate level.
Nevertheless, this lack of variability may explain to some extent the lack of relationship
that each of these three sport variables had with the criterion variables. The scales used
to measure the three attitudinal sport-specific characteristics (i.e. competitiveness, win

orientation, goal orientation) may have been psychometrically unsound for a population

124

as specifically defined as male collegiate athletes. Not only were the sport orientation
scale psychometrics obtained from the general student body at a large university, the
scale itself was not specifically designed for intercollegiate athletes (Gill & Deeter,
1988). Because the developers of the scale intended for it to capture a person’s sport
achievement orientation both in noncompetitive and competitive sports involvement, it
may have offered too narrow a range of responses for the level of sports achievement
seen at the intercollegiate level. It is possible that (as seen in the current study) male
collegiate athletes hold extreme enough attitudes for each of the sport scales used that the
full range of respondents’ attitudes about sport involvement were not captured. A scale
that is able to discern between varying levels of sport-specific attitudes within the
collegiate athlete population would be necessary to more accurately test whether sport-
specific characteristics truly account for any differences in the prediction of attitude
change after attending a sexual assault prevention workshop.

As a whole, the lack of sport-specific characteristics that were related to the
effectiveness of a program targeting athletes emphasizes two points. First, that
prevention programs targeting athletes need to target men’s general attitudes about
women and sexual assault, rather than just sport-specific characteristics of the athletes.
Although the experience of being an athlete may contribute to attitudinal factors related
to sexual assault prevention program effectiveness, the underlying general attitudinal
factors (e.g. acceptance of general violence) are still the most salient contributors to a
program’s effectiveness. This is not to say that program content should not be tailored to
the specific experiences of a collegiate athlete. Rather, programs need to focus overall

goals of attitude change toward general attitudes about women and sexual assault (e. g.

125

hostility toward women), and incorporate the athletic experience into the details of the
workshop (e.g. through examples given in exercises). For example, the scenarios used in
exercises should be relevant to the life of an athlete (e. g. including discussions about
“groupies”). Yet the desired attitudinal change should be focused on reducing rape
myths and decreasing the likelihood that the participant may choose to sexually assault a
person in the future. The fact that the audience consists of male collegiate athletes should
guide the development of examples and situations discussed, yet the goal of attitude
change for the workshop should not differ drastically from any prevention program
targeting college men, which usually aims to reduce acceptance of rape myths, likelihood
of sexually aggressing, etc.

The fact that involvement in sports may contribute to general attitudinal factors
related to sexual assault leads to the second point taken from the lack of sport-specific
predictors. Prevention program efforts that target student-athletes may serve as the bare
minimum required to actually change a culture that may be supportive of sexual assault.
Although the current program aimed to change the athlete peer group culture by
increasing the chance that participants would hold teammates accountable for their
actions and help break the protection of the male peer environment, it is apparent that
players’ decisions for intervening are based on a myriad of different reasons.
Considering that one of the most mentioned reasons for confronting a teammate was to
protect him from “doing something that would get him in trouble,” it is obvious the
weight that outside punitive actions and player accountability from outside of the peer
group of a team has on players’ decisions to take action. Although it would be ethically

desirable for male collegiate athletes to want to take steps to prevent sexual assault

126

because of the heinous and destructive nature of the crime, the fact that sexual assault is a
crime may take precedent to actually create change in the behavior of this group’s culture
as a whole.

The role of administrators, coaches, and the general public in holding collegiate
athletes more accountable for their off-the-field behavior may be the critical link that is
missing from the prevention of sexual assault within this group culture (Benedict, 1998).
As seen with the current study, the individual characteristics of being an athlete may not
be related to the effectiveness of prevention programming, but the experience of being an
athlete and how accountability among players is handled may indeed be related to the
effectiveness of any prevention efforts. It has been found that non-athlete collegiate men
are not always given the same punitive leniency as some collegiate male athletes
(Benedict, 1998), which indicates that the prevention of sexual assault seems to be more
linked to changing the way the system of male collegiate sports treats athletes, rather than
solely focusing on changing the athletes themselves.

It is much easier to point the finger of blame at individuals, because change at the
individual level is seen as simpler and more cost-effective. Changing athletes’ individual
attitudes, which in turn may change some team norms, is a start. Yet the level of
intervention needs to transcend the individual-level of athletes and reach the systemic-
level in which they operate that allows sexual assault and lack of accountability for
offenses to occur. The fact that coach involvement greatly increases the chance that
players are accountable for merely attending and participating in a workshop, indicates
the strength with which changes made to the world outside of the athlete may enhance

sexual assault prevention efforts. If sexual assault truly is a cultural phenomenon as

127

Sanday (1990) asserted years ago, it is necessary to explore areas outside of even group-
level characteristics which are a determinant of intra-individual student athlete
characteristics, and look to change systematic factors such as coaches, athletic
administrators, apparel and equipment sponsors, fans, and campus administration. As
Benedict (1998, p. xiii) concluded, “the problem of rape by athletes is much more a result
of their celebrity than their athleticism.” We must actually strive to change the culture in
which male collegiate athletes exist in order to effectively change their behavior within

that culture.

Measurement Design Limitations and Suggestions

 

As a final note, it is important to note the limitations of the particular
measurement design used in this study. Self-report questionnaires may have resulted. in a
degree of social desirability for some respondents. This may be reflected in respondents’
extremely low pre-test scores for certain outcome measures (e. g. likelihood of sexually
aggressing). As stated above, participants knew they were attending a program about
relationship violence, so it is possible that they may have initially endorsed low levels of
sexual aggression, as well as other attitudes that are seen as socially undesirable to the
majority American culture (e. g. hostility toward women). Additionally, to more
accurately interpret program evaluation results like those found in the current study, it
would be necessary to compare all results to a comparison (or control) group who did not
receive the workshop. By using a control group, issues such as social desirability may be

more easily identified, as a control group that does not know about or attend a

128

relationship violence intervention may possibly have lower social desirability than the
target group who receives the intervention.

Finally, it is also important to strive to conduct pre-post tests at time points that
are separate from the program being evaluated (e. g. one week or more apart from the
program). Conducting immediate pre and post-tests could possibly have an effect on
both the pre and post-tests scores (including social desirability). It is also important to
incorporate follow-up testing to measure any rebounding effects that may occur after the
intervention, as well as to capture any general attrition of attitudinal change. The
planning committee and evaluator of the current study negotiated with numerous
individuals responsible for scheduling the program yet were unable to arrive at a
compromise for conducting pre and post-tests apart from the workshop. Scheduling the
workshop itself was also quite challenging, due to conflicting team schedules and lack of
coach and player interest. It was seen as too time consuming and almost logistically
impossible to schedule times for team members to complete pre or post-tests at a time
apart from the time scheduled for the workshop. The challenge of scheduling evaluation
time further emphasizes the importance of building collaborative relationships with
stakeholders in the athletic department, including players, coaches, and administrators.
Only when those people who are directly affected by programming (e. g. players, coaches
and administrators) value the importance of sexual assault prevention workshops, will
they build time into their busy schedules that allows for more effective measurement

designs to be possible.

129

LIST OF REFERENCES

130

LIST OF REFERENCES

Baker, P. (1992). Maintaining male power: Why heterosexual men use pornography. In
C. Itzen (Ed.), Pornography: Women, violence, and civil liberties (pp. 124-144).
New York: Oxford University Press.

Baron, L., & Strauss, M. (1989). Four theories of rape. New Haven, CT: Yale University
Press.

 

Benedict, J ., & Klein, A. (1997). Arrest and conviction rates for athletes accused of
sexual assault. Sociology of Sport, 14, 86-94.

 

Benedict, J. (1998). Athletes and acquaintance rape. Thousand Oaks, CA: Sage
Publications.

 

Berg, D. R., Lonsway, K. A., & Fitzgerald, L. F. (1999). Rape prevention education for
men: The effectiveness of empathy induction techniques. Journal of College
Student Development, 40(3), 219-234.

 

 

Berger, N.-M. (1994). An exploration of the effectiveness of an acquaintance rape
prevention program designed for male intercollegiate athletes. Dissertation
Abstracts International: Section B: the Sciences and Engineering, 54(9-B), 4907.

 

Berkowitz, A. (1992). College men as perpetrators of acquaintance rape and sexual
assault: A review of recent research. Journal of American College Health, 40(4),
175-181.

 

Berkowitz, A. D., Burkhart, B. R., & Bourg, S. E. (1994). Research on college men and
rape. In A. D. Berkowitz (Ed), Men and rape: Theory, research, and prevention
programs in higher education (Vol. 65, pp. 3-20). San Francisco: Jossey-Bass.

 

 

Boeringer, S. B., Shehan, C., & Akers, R. ( 1991). Social contexts and social learning in
sexual coercion and aggression: Assessing the contribution of fraternity
membership. Family Relations, 40, 58-64.

 

Boeringer, SB. (1996). Influences of Fraternity Membership, Athletics, and Male Living
Arrangements on Sexual Aggression. Violence Against Women, 2(2), 134-147.

 

Boeringer, S.-B. (1999). Associations of rape-supportive attitudes with fraternal and
athletic participation. Violence Against Women, 5(1), 81-90.

 

Bohmer, C., & Parrot, A. (1. 993). Sexual assault on campus: The problem and the
solution. New York: Lexington.

 

Borden, L. A., Karr, S. K., & Caldwell-Colbert, A. (1988). Effects of a university rape

131

prevention program on attitudes and empathy toward rape. Journal of College
Student Development, 29, 132-136.

 

 

Breitenbecher, K.H. (2000). Sexual assault on college campuses: Is an ounce of
prevention enough? Applied and Preventive Psycholggy, 9, 23-52.

 

Brownmiller, S. (1975). Against our will: Men , women and rape. New York: Fawcett
Columbine.

 

Bryson, L. (1987)-Sport and maintenance of masculine hegemony. Women's Studies
International F orung 10, 349-360.

 

 

Burt, M. R. (1980). Cultural myths and support for rape. Journal of Personality and
Social Psychology, 38, 217-230.

 

 

Caron, S.L., Carter, D.B., & Brightman, LA. (1985). Sex role orientation and attitudes

toward women: differences among college athletes. Perceptual and Motor Skills,
61, 803-806.

 

Caron, S.L. (1993). Athletes as rape-awareness educators: Athletes for Sexual
Responsibility. Journal of American College Health, 41(6), 275-276.

 

Caron, S.L., Halteman, W.A., & Stacy, C. (1997). Athletes and rape: Is there a
connection? Perceptual and Motor Skills, 85(3, Pt 2), 1379-1393.

 

Carroll, K., & Jackson, TL. (1996). Rape education and prevention training. In Jackson,
Thomas L. (Eds), Acquaintance rape: Assessment, treatment, and prevention
(pp. 177-213). Sarasota, FL, USA: Professional Resource Press/Professional
Resource Exchange, Inc.

 

Check, J. V. P., Malamuth, N. M., Elias, B., & Barton, S. (1985, April). On hostile
ground. Psychology Today, 56-61.

 

Clark, RD, and Maass, A. (1988). A social categorization in minority influence: The
case of homosexuality. European Journal of Social Psychology, 18, 347-364.

 

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 and Social Issues, 19(2), 126-140.

 

Crosset, T.-W., Ptacek, J ., McDonald, M.-A., & Benedict, J .-R. (1996). Male Student-
Athletes and Violence against Women: A Survey of Campus Judicial Affairs
Offices. Violence Against Women, 2(2), 163-179.

 

Curry, T.-J. (1991). Fraternal Bonding in the Locker Room: A Profeminist Analysis of
Talk about Competition and Women. Sociolggy of Sport Journal, 8(2), 119-135.

 

Dallager, C., & Rosen, L. A. (1993). Effects of a human sexuality course on attitudes
toward rape and violence. Journal of Sex Education and Therapy, 19, 193-199.

 

DeKeseredy, W.-S., Schwartz, M.-D., & Tait, K. (1993). Sexual assault and stranger
aggression on a Canadian university campus. Sex Roles, 28(5-6), 263-277.

 

Dershowitz, A. (1994, August 6). When women cry rape - falsely. Boston Herald, pp. 13.

 

Dworkin, A. (1993). I want a twenty-four hour truce during which there is no rape. In E.
Buchwald (Ed), Transforming a Rape Culture (pp. 13-22). Minneapolis, MN:
Milkweed Editions.

 

Ellis, A. L., O'Sullivan, C. S., & Sowards, B. A. (1992). The impact of contemplated
exposure to a survivor of rape on attitudes toward rape. Journal of Applied Social
Psychology, 22, 889-895.

 

 

Ezkenazi, G. (1990, June 3). Athletic aggression and sexual assault. The New York
Times pp. p.A27.

 

 

Feltey, K.-M., Ainslie, J .-J ., & Geib, A. (1991). Sexual coercion attitudes among high
school students: The influence of gender and rape education. Youth and Society,
_2_3_(2), 229-250.

 

F ischoff, B., Furby, L., & Morgan, M. (1987). Rape prevention: A typology of strategies.
Journal of Interpersonal Violence, 2(3), 292-308.

 

Fonow, M. M., Richardson, L., & Wemmerus, V. A. (1992). Feminist rape education:
Does it work? Gender and Society, 6, 108-121.

 

F oubert, J .-D., & McEwen, M.-K. (1998). An all-male rape prevention peer education
program: Decreasing fraternity men's behavioral intent to rape. Journal of
College Student Development, 39(6), 548-556.

 

Frintner, M.-P., & Rubinson, L. (1993). Acquaintance rape: The influence of alcohol,
fraternity membership, and sports team membership. Journal of Sex Education
and Therapy, 19(4), 272-284.

 

 

Funk, R. E. (1993). Stopping rape: A challenge for men. Philadelphia: New Society
Publications.

 

Furby, L., Fischhoff, B., & Morgan, M. (1991). Rape Prevention and Self-Defense: At
What Price? Women's Studies International Forum, 14, 1-2.

Gibson, P.—R. (1992). An intervention designed to modify attitudes toward acquaintance
rape in college students. Dissertation Abstracts International, 53(2-B).

 

133

Gilbert, B. J ., Heesacker, M., & Gannon, L. J. (1991). Changing men's sexual aggression-
supportive attitudes: A psychoeducational intervention. Journal of Counseling
Psychology, 38, 197-203.

 

 

Gilbert N. (1992b, May). The phantom epidemic of sexual assault. The Public Interest,
103, 54-65.

 

Gill, D. L., & Deeter, T. E. (1988). Development of the sport orientation questionnaire.
Research @arterly for Exercise and Sport,59(3), 191-202.

 

Gray, M. D., Lesser, D., Quinn, E., & Bounds, C. (1990). The effectiveness of
personalizing acquaintance rape prevention: Programs on perception of

vulnerability and on reducing risk-taking behavior. Journal of College Student
Development, 31, 217-220.

 

 

Hackman, J .R. (1976). Group influences on individuals. In Dunnette, M.D. (Ed),
Handbook of industrial and organizational psychology (pp. 1455-1525).
Chicago, IL : Rand McNally.

 

Hanson, K.-A., & Gidycz, C.-A. (1993). Evaluation of a sexual assault prevention
program. Journal of Consulting and Clinical Psychology, 61(6), 1046-1052.

 

Helmreich, R., Spence, J ., & Wilhelm, J. (1981). A psychometric analysis of the Personal
Attributes Questionnaire. Sex Roles, 7, 1097-1108.

 

Heppner, M.-J., Humphrey, C.-F., Hillenbrand-Gunn, T.-L., & DeBord, K.-A. (1995).
The differential effects of rape prevention programming on attitudes, behavior,
and knowledge. Journal of Counseling Psychology, 42(4), 508-518.

 

Heppner, M.-J., Neville, H.-A., Smith, K., Kivlighan, D.-M., Jr., & Gershuny, B.-S.
(1999). Examining immediate and long-term efficacy of rape prevention
programming with racially diverse college men. Journal of Counseling
Psychology, 46(1), 16-26.

 

 

Humphrey, S.E. & Kahn, AS. (2000). Fraternities, athletic teams, and rape: Importance
of identification with a risky group. Journal of Interpersonal Violence, 15(12),
1 3 13-1322.

Kang, L., Gill, D., Acevedo, E., Deeter, T. (1990). Competitiveness orientations among
athletes and nonathletes in Taiwan. International Journal of Sports Psychology,
21, 146-157.

Kanin, E. (1957). Male aggression in dating-courtship violence. American Journal of
Sociologyé3, 191-204.

 

 

134

Katz, J. (1995). Reconstructing Masculinity in the Locker Room: The Mentors in
Violence Prevention Project. Harvard Educational Review, 65(2), 163-174.

 

Kidd, B. (1990). The men's cultural center: Sports and the dynamic of women's
oppression/men's repression. In M. A. Messner & D. F. Sabo (Eds), Sport, men2
and the gender order: Critical feminist perspectives (pp. 31-42). Champaign, IL:
Human Kinetics.

 

Kimmel, M.S., Messner, M.A., & Benedict, J .R. (1989). Athletes and Acquaintance
Rape. In M.S. Kimmel, M.A. Messner (Eds), Men's lives. New York, NY,
USA: Macmillan Publishing Co, Inc.

Koss, M.P., & Gidycz, CA. (1985). Sexual experiences survey; Reliability and validity.
Journal of Consulting and Clinical Psychology, 53, 422-423.

 

Koss, M.P., Gidycz, C.A., & Wisniewski, N. (1987). The scope of rape: Incidence and
prevalence of sexual aggression and victimization in a national sample of higher
education students. Journal of Consulting and Clinical Psychology, 55, 162-170.

 

Koss, M.P., & Harvey, MR. (1991). The rape victim: Clinical and community
interventions (2nd ed). Newbury Park, CA, USA: Sage Publications, Inc.

 

 

Koss, M.P., & Gaines, J .A. (1993). The Prediction of Sexual Aggression by Alcohol Use,
Athletic Participation, and Fraternity Affiliation. Journal of Interpersonal
Violence 8(1), 94-108.

 

 

Koss, M.-P., & Cleveland, H.-H., III. (1996). Athletic Participation, Fraternity
Membership, and Date Rape: The Question Remains-Self-Selection or Different
Causal Processes? Violence Against Women, 2(2), 180-190.

 

Krulewitz, J. E., & Kahn, A. S. (1983). Preferences for rape reduction strategies.
Psychology of Women Quarterly, 7(4), 301-3 12.

 

Lackie, L., & de-Man, A.-F. (1997). Correlates of Sexual Aggression among Male
University Students. Sex Roles, 37, 5-6.

 

Lee, L. A. (1987). Rape prevention: Experiential training for men. Journal of Counseling
and Development, 66, 100-101.

 

 

Lenihan, G. O., Rawlins, M. E., Eberly, C. G., Buckley, B., & Masters, B. (1992).
Gender differences in rape supportive attitudes before and after a date rape
education intervention. Journal of College Student Development, 33, 331-338.

 

Lenihan, G.-O., & Rawlins, M.-E. (1994). Rape supportive attitudes among Greek
students before and after a date rape prevention program. Journal of College
Student Development, 35(6), 450-455.

 

 

135

 

 

Lonsway, K. A., & Fitzgerald, L. F. (1994). Rape myths: In review. Pgsychology of
Women Quarterly, 18, 133-164.

 

 

Lonsway, K.-A. (1996). Preventing acquaintance rape through education: What do we
know? Psychology of Women Quarterly, 20(2), 229-265.

 

Lonsway, K. A., Klaw, E. L., Berg, D. R., Waldo, C. R., Kothari, C., Mazurek, C. J ., &
Hegeman, K. E. (1998). Beyond "No means no": Outcomes of an intensive
program to train peer facilitators for campus acquaintance rape education.
Journal of Interpersonal Violence, 13(1), 73-92.

 

MacKinnon, C. (1987). Feminism Unmodified: Discourses on life and law. Boston:
Harvard University Press.

 

MacKinnon, C. (1993). Only Words. Cambridge, MA: Harvard University Press.

Mahlstedt, D., & Jacobson, S. (2000). Innovative sexual assault prevention programs for
men: Fraternity Violence Education Project. Unpublished presentation given at
the National Sexual Assault Prevention Conference, May 16-19, 2000.

Makepeace, J. M. (1986). Gender differences in courtship violence victimization. Family
Relations, 35, 383-388.

 

Malamuth, N. M., Sockloskie, R. J ., Koss, M. P., & Tanaka, J. S. (1991). Characteristics
of aggressors against women: Testing a model using a national sample of college
students. Journal of Consulting and Clinical Psychology, 59, 670-681.

 

Mann, C. A., Hecht, M. L., & Valentine, K. B. (1988). Performance in a social context:
Date rape versus date right. Central States Speech Journal, 3/4, 269-280.

 

Martin, P. Y., & Hummer, R. A. (1989). Fraternities and rape on campus. Gender and
Society, 3, 457—473.

Matabuti, H. (1993). On Becoming Anti-Rapist. In E. Buchwald (Ed), Transforming a
Rape Culture (pp. 165-178). Minneapolis, MN: Milkweed Editions.

 

 

Messner, M. A. (1992). Power at play: Sports and the problem of masculinity. Boston:
Beacon Press.

 

Messner, M. A., & Sabo, D. F. (Eds). (1994). Sex, violence, and power in sports:
Rethinking masculinity. Freedom,CA: The Crossing Press.

 

 

Moskovici, S. (1980). Studies in social influence: V. Minority influence and conversion
behavior in a perceptual task. J oumal-of-Experimental-Social-Psychology. Vol
16(3): 270-282

 

136

Muehlenhard, C., & Linton, M. (1987). Date rape and sexual aggression in dating
situations: Incidence and risk factors. Journal of Counseling Psychology, 6, 20-
31.

 

Nemeth, C. J. (1995). Dissent as driving cognition, attitudes, and judgments. Social
Cognition, 13 (3), 273-291.

Nemeth, OJ. and Chiles, C. (1988). Modeling courage: The role of dissent in fostering
independence. European Journal of Social Psychology, 18, 275-280.

 

Nemeth, C. J. (1986). Differential contributions of majority and minority influence.
Psychological Review, 93 (1), 23-32.

 

Neimark, J. (1991, ). Out of bounds: The truth about athletes and rape. Mademoiselle, 97,
196-199, 244-246.

 

National Victim Center (1992). Rape in America: A Report to the Nation . Charleston,
SC: Report prepared by the Crime Victims Research and Treatment Center,
Medical University of South Carolina.

 

Nelson, M. B. (1994). The stronger women get, the more men love football: Sexism and
the American culture of sports. New York: Avon Books.

 

 

Northeastern University Student Handbook. (1999)

 

O’Keefe, DJ. (1990). Persuasion: Theory and research. Thousand Oaks: Sage
Publications.

 

Parrot, A., Cummings, N., Marchell, T.C., & Hofher, J. (1994). A rape awareness and
prevention model for male athletes. Journal of American College Health, 42(4),
179-184.

 

Payne, D. L., Lonsway, K. A., & Fitzgerald, L. F. (1999). Rape myth acceptance:
Exploration of its structure and its measurement using the Illinois rape myth
acceptance scale. Journal of Research in Personality, 33, 27-68.

 

Patton, MO. (1990). Qualitative evaluation and research. methods. Newbury Park: Sage
Publications.

 

Pinzone-Glover, H.-A., Gidycz, C.-A., & Jacobs, C. (1998). An acquaintance rape
prevention program: Effects on attitudes toward women, rape-related attitudes,
and perceptions of rape scenarios. Psychology of Women Quarterly, 22(4), 605-
621.

 

Proto-Campise, L., Belknap, J ., & Wooldredge, J. (1998). High school students'

137

 

adherence to rape myths and the effectiveness of high school rape-awareness
programs. Violence Against Women, 4(3), 308-328.

 

Rickert, V. I., & Wiemann, C. M. (1998). Date rape among adolescents and young adults.
Journal of Pediatric and Adolescent Gynecology, 11(4), 167-75.

 

Ring, T. E., & Kilmartin, C. (1992). Man to man about rape: A rape prevention program
for men. Journal of College Student Developmentg33, 82-84.

 

Rivera, G., & Regoli, R. (1987). Sexual victimization experiences of sorority women.
Sociology and Social Research, 12, 39-42.

 

Sanday, P. R. (1990). Fraternity Gang Rape: Sexgbrotherhood, and privilege on campus.
New York: New York University Press.

 

Sanday, P. R. (1996). Rape-prone versus rape-free campus cultures. Violence Against
Women 2, 191-208.

 

 

Schaeffer, A., & Nelson, E. (1993). Rape supportive attitudes: Effects of on-campus
residence and education. Journal of College Student Development, 34, 175-179.

 

Schewe, P., & O'Donohue, W. (1993). Rape prevention: Methodological problems and
new directions. Clinical Psychology Review, 13(7), 667-682.

 

Schewe, P.-A., & O'Donohue, W. (1996). Rape prevention with high-risk males: Short-
term outcome of two interventions. Archives of Sexual Behavior, 25(5), 455-
471.

 

Schewe, P.A., & O'Donohue, W. (1998). Psychometrics of the Rape Conformity
Assessment and other measures: Implications for rape prevention. Sexual Abuse:
Journal of Research and Treatment, 10(2), 97-112.

 

 

Schmidt, B., & Peter, A. (1996). The Petze Project: Working with teachers on the
prevention of sexual violence against girls and boys in Germany. Women's
Studies International Forum, 19(4), 395-407.

 

Schwartz, M.D., & Nogrady, CA. (1996). Fraternity membership, rape myths, and sexual
aggression on a college campus. Violence Against Womeng(2), 148-162.

 

Schwartz, M.D., & DeKeseredy, W.S. (1997). Sexual assault on the college campus: The
role of male p_eer support. Thousand Oaks, CA: Sage Publications.

 

Schwartz, MD. (1997). Researching sexual violence against women: Methodological and
personal perspectives. Thousand Oaks, CA: Sage Publications.

 

 

Smith, P., & Welchans, S. (2000). Peer education: Does focusing on male responsibility

138

change sexual assault attitudes? Violence Against Women, 6 (11), 1255-1268.

 

Sornmers, OH. (1994). Who stole feminism? New York, NY: Touchstone.

 

Spence, J. T., Helmreich, R. L., & Stapp, J. (1974). The Personal Attributes
Questionnaire: A measure of sex-role stereotypes and masculinity-femininity.
J SAS Catalgg of Selected Documents in Psychology, 4(127).

 

Stanko, EA. (1990). Everyday violence: How women and men experience sexual and
physical danger. London: Pandora.

 

 

Tabachnick, B. & F idell, L. (2000). Using Multivariate Statistics. Needham Heights,
MA: Allyn & Bacon.

 

Thorne, B. (1993). Gender may: Girls and boys in school. New Brunswick, N.J.: Rutgers
University Press.

 

Thorner-Finch, R. (1992). Ending the silence: The origins and treatment of male violence
against women. Toronto: University of Toronto Press.

 

 

Toufexis, A. (1990, Aug. 6). Sex and the sporting life. Time, 9(No. 243), 76-77.

Turner, SP. (1996). Male students and date rape related attitudes: A comparative study of
athletes, greeks, and independents. Dissertation Abstracts International: Section
B: the Sciences and Engineering, 56(11-B), 6471.

 

 

Vanburen-Hay, BE. (1999). Unpublished training materials from the Sexual Assault
Program at Michigan State University.

Vickio, C.J., Hoffman, B.A., Yarris, E. (1999). Combating sexual offenses on the college
campus: Keys to success. Journal of American College Health, 47, 283-286.

 

Volpato, C., Maass, A., Mucchi-Faina, A., and Vitti, E. (1990). Minority influence:
Social categorization. European Journal of Social Psychology, 20, 119-132.

 

Whitson, D. (1990). Sport and the social construction of masculinity. In M. A. Messner &
D. F. Sabo (Eds), Sport, men, and the gender order: Critical feminist
perspectives (pp. 19-30). Champaign, IL: Human Kinetics.

 

 

Yeater, E.A. & O’Donohue, W. (1999). Sexual assault prevention programs: Current
issues, future directions, and the potential efficacy of interventions with women.
Clinical Psychology Review, 19, 7, 739-771.

 

139

APPENDICES

140

APPENDIX A

Pre-test Questionnaire:

Consent Form

Sport Orientation Questionnaire

Sex Role Stereotyping Scale

Sexual Experiences Survey

Likelihood of Sexually Aggressing
Likelihood of Confronting
Illinois Rape Myth Acceptance Scale — Short Form
Hostility Toward Women Scale
Attitudes Toward Violence Scale
Thank You Form

141

_ The following pages contain questions about your attitudes and

opinions about a variety of issues. Your responses are completely w
; confidential. Please DO NOT write your name on any of the sheets in
this packet.

_' Your participation in this survey is completely voluntary, and you may '
' refuse to participate at any time. You may refuse to respond at any
time to particular questions without penalty. You indicate your
voluntary agreement to participate by completing and returning this
questionnaire. You will have 15 minutes to complete this packet.
Some of the questions are highly personal, and in the case that any
of the topics presented upsets or offends you, a list of helpful phone
numbers and resources has been included in your folder. You may
also contact the project investigator (Wendi Siebold) if you have any
questions or concerns about this survey at: 353-5015.

 

Thank you for your time, and please remember that all of your
responses are strictly confidential.

 

When you have completed your packet, place it in the manila
envelope and hold onto it until the end of the workshop — do not
show your answers to anyone else or write your name on any of
the pages in the packet!

Thai/tree again. {or l/leLp’Ll/tg!

Please turn to the next page :>

142

Sport Orientation Questionnaire

Please circle the number of years you have participated as an intercollegiate athlete at MSU
(including redshirt years):

1 2 3 4 5 more than 5

The following statements describe reactions to sport situations. We want to know
how you usually feel about sports and competition. Read each statement and circle
the letter that indicates how much you agree or disagree with each statement on the
scale: A, B, C, D, E. There are no right or wrong answers; simply answer as you
honestly feel. Do not spend too much time on any one statement. Remember,
choose the letter that describes how you usually feel about sports and competition.

Strongly Slightly Neither Slightly Strongly

agree agree agree nor disagree disagree
disagree

1. I am a determined competitor. A B C D E
2. Winning is important. A B C D E
3. I am a competitive person. A B C D E
4. I set goals for myself A B C D E

when I compete.
5. I try my hardest to win. A B C D E
6. Scoring more points than my

opponent is very important

to me. A B C D E
7. I look forward to competing. A B C D E
8. I am most competitive when

I try to achieve personal goals. A B C D E
9. I enjoy competing against

others. A B C D E
10. I hate to lose. A B C D E
11. I thrive on competition. A B C D E

143

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

I try hardest when I have

a specific goal. E
My goal is to be the best

athlete possible. E
The only time I am satisfied

is when I win. E
I want to be successful

in sports. E
Performing to the best of

my ability is very important

to me. E
I work hard to be successful

in sports. E
Losing upsets me. E
The best test of my ability

is competing against others. E
Reaching personal

performance goals is very

important to me. E
I look forward to the

opportunity to test my skills

in competition. E
I have the most fun when I win. E
I perform my best when I am

competing against an

opponent. E
The best way to determine

my ability is to set a goal and

try to reach it. E
I want to be the best every

time I compete. A B C D E

Please turn to the next page :>

144

Sex Role Stereotyping Scale

 

Please use the following scale to indicate how strongly you agree with each of these
statements. Write the number of your choice on the line to the left of the statement.

1 2 3 4 5 6 7
Not at all Mostly do Somewhat Undecided Somewhat Mostly Very
agree not agree do not agree agree much
agree agree
1. A man should fight when the woman he’s with is insulted by another man.
2. It is acceptable for the woman to pay for the date.
3. A woman should be a virgin when she marries.
4. There is something wrong with a woman who doesn’t want to marry and

raise a family.
5. A wife should never contradict her husband in public.

6. It is better for a woman to use her feminine charm to get what she wants
rather than ask for it outright.

7. It is acceptable for a woman to have a career, but marriage and family
should come first.

8. It looks worse for a woman to be drunk than for a man to be drunk.

9. There is nothing wrong with a woman going to a bar alone.

Please turn to the next page :>

145

Sexual Experiences Survey

 

Please indicate how frequently you have engaged in each of the listed behaviors. Please
remember that all of your responses are strictly confidential.

 

The response codes indicate:
N = Never 1 = Once 2 = Twice S = Sometimes O = Often

Since you were 14 years old

1. Made “cat-calls”, whistled, or yelled at a
woman on the street. N 1 2 S O

2. Touched a woman’s buttocks, breasts, or
genital area against her wishes. N 1 2 S O

3. Attempted sexual intercourse with a woman
when she didn’t want to by overwhelming her
with continual argument and pressure. N 1 2 S O

4. Attempted sexual intercourse with a woman
when she didn’t want to by giving her more
alcohol or drugs than she could handle. N 1 2 S O

5. Attempted sexual intercourse with a woman by
threatening to use force. N 1 2 S O

6. Attempted sexual intercourse with a woman
when she didn’t want to by using some degree
of force. N 1 2 S O

7. Had sexual intercourse with a woman when she
didn’t want to by overwhelming her with continual
argument and pressure. N 1 2 S O

8. Had sexual intercourse with a woman when she
didn’t want to by giving her more alcohol or drugs
than she could handle. N 1 2 S O

9. Had sexual intercourse with a woman by threatening
to use force. N 1 2 S O

10. Had sexual intercourse with a woman when she
didn’t want to by using some degree of force. N 1 2 S O

146

1 I. Stood in line to have sex with a “party girl.”
12. Had sexual intercourse with a “groupie.”
13. Stood in line to have sex with a “groupie.”

14. Had sexual intercourse with a woman who

protested, but who really wanted to have sex.

147

280

280

280

\ln

 

Likelihood of Sexually Aggressing
(italicized statements are filler items)

 

Please use the following scale to determine your responses to the question below:

1 2 3 4 5
never rarely unsure sometimes always

If you could be assured that no one would know, how often would you engage in each of
the following behaviors:

 

Group sex
__ Sex with someone else besides your steady boy/girlfriend
__ Oral sex
__ Force a woman to do something sexual she doesn’t want to do
__ Bondage
_ Anal sex
__ Talk a woman into doing something sexual she doesn’t want to do
Same-sex sex
Force a woman to have sex when she doesn’t want to
Use alcohol in hopes of having a woman give in to something sexual she wouldn’t
give in to if not drinking

148

Likelihood of Confronting

 

 

 

People do different things under different circumstances. What do you think you
would do in the following situations?

 

Please write the number to the left of the situation:

do nothing

wait for a more convenient time to talk to the woman

wait for a more convenient time to talk with the man

interrupt the situation to ask if everything is alright

none of the above — please write what you would do next to the situation

Ul-F-UJNl—t
I

You see a teammate kissing a woman who is pushing him away, and the woman is
someone you do not know.

You know a man is making a woman do something sexual that she does not want
to do, and the man is someone you do not know.

You know a teammate is making a woman do something sexual that she does not
want to do, and the woman is a good friend of yours.

You see a teammate kissing a woman who is pushing him away and the woman is a
good friend of yours.

You know a teammate is making a woman do something sexual that she does not
want to do, and the woman is someone you do not know.

You see a man kissing a woman who is pushing him away and the man is someone
you do not know.

Please turn to the next page :>

149

 

 

Illinois Rape Myth Acceptance Scale — Short Form

Please write on the blank to the left of each question the number from the scale below
that corresponds to how strongly you agree with each of the statements given.

1 2 3 4 5 6 7
Not at all Mostly do Somewhat Undecided Somewhat Mostly Very
agree not agree do not agree agree much
agree agree
1. __ If a woman is raped while she is drunk, she is at least somewhat responsible for
letting things get out
of control.
2. _ Although most women wouldn’t admit it, they generally find being physically
forced into sex a real
“turn on.”

DJ

If a woman is willing to “make out” with a guy, then it’s no big deal if he goes
a little further and has
sex with her.

4. __ Many women secretly desire to be raped.

5. __ Most rapists are not caught by the police.

6. _ If a woman doesn’t physically fight back, you can’t really say that it was rape.
7. __ Men from nice middle-class homes almost never rape.

8. _ Rape accusations are often used as a way of getting back at men.

9. __ All women should have access to self-defense classes.

10. __ It is usually only women who dress suggestively who are raped.

11. __ If the rapist doesn’t have a weapon, you really can’t call it rape.

12. __ Rape is unlikely to happen in the woman’s own familiar neighborhood.
l3. __ Women tend to exaggerate how much rape affects them.

14. __ A lot of women lead a man on and then they cry rape.

150

15.

16.

17.

18.

19.

20.

It is preferable that a female police officer conduct the questioning when a
woman reports a rape.

A woman who “teases” men deserves anything that might happen.

When women are raped, it’s often because the way they said “no” was not
clear.

Men don’t usually intend to force sex on a woman, but sometimes they get too
sexually carried away.

A woman who dresses in skimpy clothes should not be surprised if a man tries
to force her to have sex.

Rape happens when a man’s sex drive gets out of control.

Please turn to the next page :>

151

Hostility Toward Women Scale

Please use the following scale to indicate how strongly you agree with each of these
statements. Write the number of your choice on the line to the left of the statement.

1 2 3 4 5 6 7
Not at all Mostly do Somewhat Undecided Somewhat Mostly Very
agree not agree do not agree agree much
agree agree
1. I feel that many times women flirt with men just to tease them or
hurt them.
2. I believe that most women tell the truth.

I usually find myself agreeing with women.

U)

4. __ I think that most women would lie just to get ahead.

5. __ Generally, it is safer not to trust women.

6. __ When it comes down to it, a lot of women are deceitful.
7. _ I am easily angered by women.

8. __ I am sure I get a raw deal from the women in my life.
9. _ Sometimes women bother me just by being around.

10. __ Women are responsible for most of my troubles.

Please turn to the next page :>

152

Attitudes Toward Violence Scale
(randomized order of Interpersonal and General)

Please use the following scale to indicate how strongly you agree with each of these

statements.

1
Not at all
agree

2.

6.

7.

8.

9.

10.

11.

12.

13.

14.
15.

16.

Write the number of your choice on the line to the left of the statement.

2 3 4 5 6 7
Mostly do Somewhat Undecided Somewhat Mostly Very
not agree do not agree agree much

agree agree

Violent crimes should be punished violently.

The death penalty should be palt of every penal code.

A child’s habitual disobedience should be punished physically.
Any nation should be ready with a strong military at all times.

Punishing children physically when they deserve it will make them
responsible and mature adults.

The manufacture of weapons is necessary.
It is alright for a partner to hit the other if they are unfaithful.

The government should send armed soldiers to control violent university
riots.

An adult should whip a child for breaking the law.

It is alright for a partner to hit the other if they flirt with others.

Our country has the right to protect its borders forcefully.

Any prisoner deserves to be mistreated by other prisoners in jail.

Giving mischievous children a quick slap is the best way to quickly end
trouble.

Children should be spanked for temper tantrums.
It is alright for a partner to slap the other if insulted or ridiculed.

Young children who refuse to obey should be whipped.

153

17. __ War is often necessary.

18. _ It is alright for a partner to slap the other’s face if challenged.

19. __ Our country should be aggressive with its military internationally.
20. _ Killing of civilians should be accepted as an unavoidable part of war.

154

Please turn to the next page :>

 

 

Thank you

for completing the questionnaire packet! Please put it back
into the envelope you got it from and place the envelope at
your seat until the end of the workshop.

 

155

 

APPENDIX B

Post-test Questionnaire

Consent Form
Likelihood of Sexually Aggressing
Likelihood of Confronting
Illinois Rape Myth Acceptance Scale — Short Form
Group Utilization Questions
Thank You Form

156

1. -;--~.-.- untrue-“"0! 1.- --. nun-m. “— -.

Similar to the packet you filled out before the workshop, this packet
contains questions about your attitudes and opinions about a variety
of issues. Remember, your responses are completely confidential. ,'
Please DO NOT write your name on any of the sheets in this packet.

Your participation in this survey is completely voluntary, and you may
refuse to participate at any time. You may refuse to respond at any ;
'_ time to particular questions without penalty. You indicate your 3
: voluntary agreement to participate by completing and returning this ’
questionnaire. You will have 10 minutes to complete this packet. .-
Some of the questions are highly personal, and in the case that any

of the topics presented upsets or offends you, a list of helpful phone
numbers and resources has been included in your folder. You may
also contact the project investigator (Wendi Siebold) if you have any
questions or concerns about this survey at: 353-5015.

 

   

When you have completed your packet, place it in the manila
envelope with your other questionnaire packet and give the
whole envelope to the people at the door. Remember, don’t put
your name anywhere on the pages or the envelope!

 

Thames again/t for heLle/tgl

Please turn to the next page :>

157

Likelihood of Sexually Aggressing
(italicized statements are filler items)

 

Please use the following scale to determine your responses to the question below:

1 2 3 4 5
never rarely unsure sometimes always

 

If you could be assured that no one would know, how often would you engage in each of
the following behaviors:

Group sex
Sex with someone else besides your steady boy/girlfriend
Oral sex
_ Force a woman to do something sexual she doesn’t want to do
__ Bondage
Anal sex
Talk a woman into doing something sexual she doesn’t want to do
Same-sex sex
Force a woman to have sex when she doesn’t want to
Use alcohol in hopes of having a woman give in to something sexual she wouldn’t
give in to if not drinking

158

 

, 1,11,.

 

Likelihood of Confronting

 

 

 

People do different things under different circumstances. What do you think you
would do in the following situations?

 

Please write the number to the left of the situation:

do nothing

wait for a more convenient time to talk to the woman

wait for a more convenient time to talk with the man

interrupt the situation to ask if everything is alright

none of the above — please write what you would do next to the situation

UI-[AUJNl—t
I

You see a teammate kissing a woman who is pushing him away, and the woman is
someone you do not know.

You know a man is making a woman do something sexual that she does not want
to do, and the man is someone you do not know.

You know a teammate is making a woman do something sexual that she does not
want to do, and the woman is a good friend of yours.

You see a teammate kissing a woman who is pushing him away and the woman is a
good friend of yours.

You know a teammate is making a woman do something sexual that she does not
want to do, and the woman is someone you do not know.

You see a man kissing a woman who is pushing him away and the man is someone
you do not know.

Please turn to the next page :>

159

 

Illinois Rape Myth Acceptance Scale - Short Form

Please write on the blank to the left of each question the number from the scale below
that corresponds to how strongly you agree with each of the statements given.

1 2 3 4 5 6 7
Not at all Mostly do Somewhat Undecided Somewhat Mostly Very
agree not agree do not agree agree much
agree agree
1. If a woman is raped while she is drunk, she is at least somewhat responsible for

letting things get out of control.

2. Although most women wouldn’t admit it, they generally find being physically
forced into sex a real “turn on.”

3. If a woman is willing to “make out” with a guy, then it’s no big deal if he goes
a little further and has sex with her.

4. __ Many women secretly desire to be raped.

5. _ Most rapists are not caught by the police.

6. _ If a woman doesn’t physically fight back, you can’t really say that it was rape.
7. _ Men from nice middle-class homes almost never rape.

8. __ Rape accusations are often used as a way of getting back at men.

9. _ All women should have access to self-defense classes.

10. __ It is usually only women who dress suggestively who are raped.

l 1. __ If the rapist doesn’t have a weapon, you really can’t call it rape.

12. __ Rape is unlikely to happen in the woman’s own familiar neighborhood.
13. _ Women tend to exaggerate how much rape affects them.

14. _ A lot of women lead a man on and then they cry rape.

15. _ It is preferable that a female police officer conduct the questioning when a

woman reports a rape.

160

16.

17.

18.

19.

20.

A woman who “teases” men deserves anything that might happen.

When women are raped, it’s often because the way they said “no” was not

clear.

Men don’t usually intend to force sex on a woman, but sometimes they get too
sexually carried away.

A woman who dresses in skimpy clothes should not be surprised if a man tries
to force her to have sex.

Rape happens when a man’s sex drive gets out of control.

 

Please turn to the next page :>

161

Utilization Questions

 

What factors would influence your decision to confront a teammate who was
making a woman do something sexual that she does not want to do?

 

 

 

How much do you think the information you learned in this workshop would help
you stand up to a teammate if he was doing something with a woman that you
thought was wrong? (Please check one)

 

Cl Won’t help me at all, in any situation

Cl Might help me in certain situations

Cl Will help me in most situations

Cl Will help me in all situations

Cl I already would have stood up to a teammate in any situation

If you answered that it won’t help you, briefly explain why you don’t think it will help:

 

 

 

How much do you think you will use some of the information that was presented in
this workshop?

Cl Very much
CI Somewhat
CI Not very much
CI Not much at all

What makes you say that?

 

 

162

How much do you think members of your team will use some of the information
that was presented in this workshop?

 

Cl Very much
C] Somewhat
Cl Not very much
CI Not much at all

What makes you say that?

 

 

 

 

Please turn to the next page :>

163

 

 

 

Thank you {or participating lVL
the worleshop todaw and filling out the
questionnaires! Your answers will help l/tS
determine how useful this worl’eshop is to
studel/lts, and leal/l/l. what we CMVL do to
Vidal/ea it more evg’ogable all/1.01 helpful to 501x.

Please hand in your envelope with both of
the questionnaires inside as you walk out
the door.

 

If 501A have Ell/La questions or concerns
about MIA/5 of the questions 501A answered,
please contact WCI/LOll Siebold at 353—5015.

Thanks again!

 

164

 

 

 

APPENDIX C

Table of Sexual Assault Prevention Programs Targeting College Students

165

Table 24

Sexual Assault Prevention Programs Targeting College Students”2

 

 

 

Notes and Key (Breitenbecher, 2000)

Many authors report main effects for gender on the instruments used in sexual assault prevention studies.
Effects for gender are not included in this table, unless such effects represent interactions between gender
and experimental condition. In addition, effects described as “marginally significant” or “significant
trends” are not included. In the sample column, numbers in parentheses indicate the number of participants
who completed the follow-up phase(s) of the investigation, if this number was reported as differing from
the original sample size. An equal sign (=) indicates nonsignificant differences between groups. B =
experimental group. C = control group. Tx = treatment. Sig = statistically significant. AR = acquaintance
rape. SR = stranger rape. Wk(s) = weeks(s). Mo(s) = month(s). Min = minutes. Hr(s) = hour(s). Pre =
pretest. Post = posttest. F/u = follow-up assessment. DV(s) = dependent variable(s).

 

Key to scale abbreviations is included at the end of the table

 

 

 

 

 

 

 

 

 

 

 

 

Author(s) Sample Intervention(s) Evaluation Evaluation
Design Results
Berg, et al. 54 men E1: empathy Pretest, - LSA loaded on
(1999) induction intervention single factor
(audiotape of male 10 days later, - E1 > E2 and E2 > C
survivor + factual 2-wk f/u (undesirable effect)
info.) - AIV, ASB, IRMA,
E2 = empathy RE, and ARE
induction loaded on single
(audiotape of factor. At f/u, E1 =
female survivor + E2 = C
factual info.)
Borden et al. 100 men E = factual info., Pretest, 4-wk - ATR and RE: at f/u,
(1988) and legal terms, f/u E = C
women characteristics of
rapists, rape
trauma syndrome,
prevention (45
min)
C = no tx
Dallager & 145 men E = human Pretest, - RMA: E < C
Rosen (1993) and sexuality course posttest at - AIV: E = C
women end of
C = education semester
course
Ellis et al. 151 men Questions asking Question - RMA, MAR, ASB,
(1992) and participants to either and AIV combined
women consider a preceded (E) to form composite
situation in which or followed score
166

 

 

 

 

 

a friend/relative
disclosed that she
had been assaulted

(C) other
measures

- Emen>Cmen

(undesirable effect)

 

 

 

 

Fonow et al. 582 men Intervention was Pretest, 3-wk - RMA: at posttest,
(1992) and either a live or posttest pretested groups
women videotaped scored lower than
program unpretested. Also,
addressing myths, all experimental
prevalence, rape as groups scored lower
act of violence, than control groups.
rape as community Live and video
issue (25 min) interventions were
E1 = pre, live, post equally effective.
E2 = pre, video, RB, ASB, and SRS:
post posttest means not
C. = pre, post reported.
E3 = live, post
E4 = video, post
C2 = post
Foubert & 155 men E = Video in E1 = pretest, RMA: at post, E1 =
McEwen which man being intervention, E2. Also, E1 = C.
(1998) raped is described, immediate LSA: at post, E. =
how to help posttest E2 = C.
survivor, sexual E2 = SMCRP: was sig.
communication, intervention, Negatively
confront sexism immediate correlated with post
and abuse of posttest RMA and LSA for
women (60 min) C = posttest E1 and E2.
C = no tx
Gilbert at al. 75 (53) E = intervention Pretest, AIV, ASB, RMA,
(1991) men based on ELM, intervention and SRS combined
dramatic 1-2 wks later, into composite
presentation, immediate change score: E > C
negative posttest, 1- Phone call: E

 

 

intrapsychic and
social
consequences
associated with
AIV, RMA, ASB,
and male
dominance
ideology (60 min)
C = no tx

 

mo f/u phone
call

 

listened to more of
call and made more
favorable comments
than C. No sig
difference w/respeet
to willingness to
volunteer time

NC: 1 of 3 items
measuring state
motivation, I of 2
items measuring
ability, and 1 of 2

 

167

 

 

items measuring
thought favorability
were sig correlated
w/change scores

 

 

 

 

Gray et al. 70 E = personalized Pretest, BIDB: pre-post
(1990) women intervention (e. g. intervention, change for E > C
local statistics) immediate Perceptions of
featuring role posttest vulnerability: pre-
playing, myths, post change for E >
dating behaviors, C only when
and sexual married women
communication excluded from
C = sample
nonpersonalized
intervention
featuring role
playing, myths,
dating behaviors,
and sexual
communication
Hanson & 360 E = myths, videos Pretest, DB: at post, E < C
Gidycz (346) of AR and intervention, SAA: at post, E < C
(1993) women protective 9-wk f/u SC: at post, E = C
behaviors, SV: at post, E < C
prevention (60 for women without
min) histories of assault
C = no tx prior to intervention
Heppner, 294 E1 = ELM-based Pretest, ELMQ: E] > E2 > C
Humphrey et (258, intervention intervention TL: E1 > E2 > C
al. ( 1995) 133) featuring 5-7 days RMA: only
men and interactive, later, difference between
women dramatic immediate groups was E2 men
presentation of AR posttest, 5— < C men. E1 and E2
and protective wk f/u, 4- showed rebounding
behaviors, sex role mos f/u, 5- at f/u. Rebound
socialization (90 mos + lwk pattern for E1 = E2.
min) f/u CCC: For men, E; >

 

 

E2 = didactic,
video of survivors
(90 min)

C = stress
management

 

 

E2 > C. For women,
E1 = E2 = C.
Willingness to
volunteer, time
thinking, time
talking, and #
people talked to: E
> E2 and C.

- Fee increase and

 

168

 

 

willingness to
recommend: E1 = E2
= C.

 

 

 

 

 

Heppner, 119 (57) E1 = “colorblind” Pretest, 1-wk - Factor analysis of
Neville et al. men intervention posttest, 5- RMA, SIARA, SA,
(1999) E2 = culturally mos posttest LSA, and SVAWS-
specific for SV yielded 1 factor
African American (rejection of rape)
participants - Rejection of rape:
C = not described ANOVA showed
only sig effect was
low high low
(rebounding)
pattern
- Cluster analysis of
flu data indicated 3
groups:
deteriorating,
rebounding, and
improving.
Participants in
improving group
were more likely to
be in either E1 or E2
than C
- ELMQ: African
Americans in E2 >
than other
participants
Katz (1995) male None None
athletes
Lee (1987) 24 men E = myths, E1 = pretest, - ATR-L: at post, E1
account of male intervention, = E2. Also, E1
survivor, guided immediate showed sig pre-post
fantasy (imagine posttest change
observing E2 =
roommate intervention,
engaging in immediate
coercive sex) (2 posttest
hI)
Lenihan et al. 821 E = video of AR E1 = pretest, Paired t-tests:
(1992) (445) effects on men and intervention - Women in E1 and
men and women, several days C1 showed sig pre-
women characteristics of later, posttest post change on

 

 

rapists, cultural
reasons, reasons

 

1 mos later
C1 = pretest,

 

ASB, SCNS, and
AIV

 

169

 

 

 

 

 

why AR not posttest - Women in E1 only

identified as rape, E2 = showed sig pre-post

and female intervention, change on RMA

survivor posttest 1 - Men in E1 and C1

describing effects mos later did not show sig

on her (50 min) C2 = posttest pre-post changes on
any scale

ANOVA:

- RMA: E1 and E2 <
C1 and C2

- ASB, SCNS, and
AIV: E1 and E2 = C1
and C2

- ASB: 2-way
interaction between
pretest and
experimental
condition (f/u
analyses not
reported)

- AIV: 3-way
interaction between
gender, pretest, and
experimental
condition (f/u
analyses not
reported)

Lonsway et 99 (92, E = semester long Precourse - IRMA: at post, E <
al. (1998) 55) men course that trains assessment C. At f/u, E< C.
and students to (E also - AHSB: at post,
women facilitate rape repond to E<C. At f/u, E = C.
education for peers videotaped - ATP: at post, E>C.
C = semester long sexual At f/u, E = C.
human sexuality conflict), - Responses to
course postcourse videotaped sexual
assessment conflict: E more
(B also likely(than at
respond to pretest) to use direct
videotaped verbal resistance. E
sexual less likely to use
conflict), 2- indirect verbal
yr f/u resistance, indirect

 

 

 

 

physical resistance,
and internal
monitoring. No

 

170

 

 

 

 

change in use of

 

 

 

 

 

 

 

 

 

direct physical
resistance.
Mann et al. 92 men E1 = dramatic Intervention, Post
( 1988) and presentation + immediate - SAS: E1 > E3 = C
women discussion (30 posttest, 5- - SRAS: E1 = E2 = E3
min) wk f/u = C
E2 = dramatic
presentation (15 F/u
min) - SAS: main effect
E3 = discussion for condition sig
(15 min) (pattern of means
C = no tx not reported)

- SAS: interaction
between condition
and previous SA,
but authors warn
about small cell
sizes. Among those
with low SA, E1 >
E3 and C

- SRAS.E1= E2 = E3
= C

Parrot et al. male None None
(1994) athletes
Pinzone- 152 men E = AR prevention Pretest, - RE: E showed
Glover et al. and program, myths, intervention greater pre-post
(1998) women characteristics and 1 wk later, 1 change than C
behaviors of wk f/u - RMA: at f/u E = C
rapists, prevention - ATW: Among men,
(60 min) pre-post change for
C = sexually E > C. Among
transmitted disease women, pre-post
prevention change for E = C
program — Responses to AR
scenarios: At f/u,
men in E were
more likely to
identify scenario as
rape than men in
C. At f/u, among
women E = C.
Ring & Men, E = destructive No formal Men reported greater
Kilmartin number aspects of men’s evaluation willingness to
(1992) not sex role was participate because of
reported socialization, film, conducted non-threatening

 

l71

 

 

experiential
exercises to
increase awareness
of objectification
of men’s and
women’s bodies
and media’s
victimization of
women

nature of program,
feeling less
defensive, and greater
awareness about
destructive aspects of
male socialization.

 

 

 

 

 

 

 

Schaeffer & 160 men Specific program Quasi- - RMA: E = C
Nelson not tested or experimental - ATW: E = C
(1993) described.

E = men who had

previously

attended a rape

education program

C = men who had

not
Schewe & 74 E1 = video of Pretest, - RCA: E1 = E2 = C
O’Donohue “high- victim intervention, - AIV: E1 and E2
(1996) risk” empathy/outcome 2-wk f/u showed sig pre-

men expectancy post changes

program (50 min) - ASB: E2 showed

E2 = video of rape sig pre-post

supportive changes

cognitions - RMA: E2 showed

program (50 min) sig pre-post

C = no tx changes

- ASA: E1 and E2

showed sig pre-
post changes.

 

l

Portions of this table and table format were adapted from Breitenbecher (2000).

2 This table includes only programs that were discussed within this paper, and is not exhaustive of all
programs in the literature. For a full table of programs in the literature up to the year 2000, please refer to
Breitenbecher (2000).

172

 

 

 

Key to scale abbreviations (reprinted from Breitenbecher, 2000)

AHSB = Adversarial Heterosexual Sexual Beliefs Scale (Lonsway & Fitzgerald, 1995)

AIV = Acceptance of Interpersonal Violence (Burt, 1980)

ARE = Acquaintance Rape Empathy Scale (Berg etal., 1999)

ASA = Attraction to Sexual Aggression Scale (Malamuth, 1989)

ASB = Adversarial Sexual Beliefs (Burt, 1980)

ATP = Attitudes Toward Feminism and the Women’s Movement Scale (Fassinger, 1994)

ATR = Attitudes Toward Rape Scale (Field, 1978)

ATR-L = Attitudes Toward Rape Scale (Lee, 1987)

ATR-R = Attitudes Toward Rape Scale — Revised (Harrison et al., 1991)

ATSI = Attitudes Toward Sexuality Inventory (Patton & Mannison, 1993)

ATW = Attitudes Toward Women Scale (Spence et al., 1973)

BIDB = Items assessing behavioral intent to avoid high-risk dating behaviors (Gray et al., 1990)

CCC = Comprehension of Consent/Coercion Measure (Gibson & Humphrey, 1993)

DB = Dating Behavior Survey (Hanson & Gidycz, 1993)

DRAS = Date Rape Attitudes Survey (Holcomb etal., 1993)

DSPARS = Dating Self-Protection Against Rape Scale (Moore & Waterman, 1999)

ELMQ = Elaboration Likelihood Model Questionnaire (Heppner et al., 1995)

FDR = Forcible Date Rape Scale (Giarusso, 1979)

IRI = Interpersonal Reactivity Index (Davis, 1980)

IRMA = Illinois Rape Myth Acceptance Scale (Payne et al., 1999)

LSA = Likelihood of Sexual Aggression, including any modification or combination of items from the
Likelihood to Rape Scale (Briere & Malamuth, 1983; Malamuth, 1981) or other items devised by the
author(s) to measure this construct.

MAR = Myths About Rape (Costin & Schwarz, 1987)

NC = Need for Cognition Scale (Cacioppo & Petty, 1982)

RB = Attribution of Rape Blame (Ward & Resick, 1979, cited in Resick & Jackson, 1981)

RCA = Rape Conformity Assessment (Schewe & O’Donohue, 1993b)

RE = Rape Empathy Scale (Deitz et al., 1982)

RMA = Rape Myth Acceptance Scale (Burt, 1980)

SA = Sexual Aggression, as measured by the Sexual Experiences Survey (Koss & Oros, 1982, Koss et al.,
1987)

SAA = Sexual Assault Awareness Survey (Hanson & Gidycz, 1993)

SAKS = Sexual Assault Knowledge Survey (Breitenbecher & Scarce, 1999)

SAS = Sexual Attitude Scale (Meuhlenhard & Felts, 1987)

SC = Sexual Communication Survey (originally developed by Hanson & Gidycz, 1993, and later revised by
Breitenbecher & Gidycz, 1998)

SCNS = Sexual Conservatism (Burt, 1980)

SIARA = Scale for the Identification of Acquaintance Rape Attitudes (Humphrey, 1996)

SRAS = Simple Rathus Assertiveness Schedule (McCormick, 1986)

SRF = Speaker Rating Form (Heppner etal., 1995, adapted from Barak and LaCrosse, 1975)

SMCRP = State Measure of Central Route Processing developed by Gilbert et al., 1991, based on Petty &
Cacioppo, 1986)

SV = Sexual victimization, as measured by the Sexual Experiences Survey (Koss & Oros, 1982; Koss et al.,
1987)

SVAWS-SV = Sexual Violence Subscale of the Severity of Violence Against Women Scale (Marshall,
1992)

TL = Thought Listening (Heppner, et al., 1995, adapted from Heppner et al., 1988)

 

173

 

 

 

 

 

to”.
.:
teams?
25:...
..