2 e
‘vn..‘l
23:1 .

1......

\ u .x 3
er-fifi... "dune
(3.. v. . u inwvnu

 

”$437.

 

 

THFSiS

11529 llllllllllllllllllllllllllllllllllllllllllllllllllllHll

31293 01810 4319

This is to certify that the

thesis entitled

An Examination of Factors Predicting the Reporting
of Sexual Assault in a College Sample.

presented by

Jennifer S. Paul

has been accepted towards fulfillment
of the requirements for

 

 

Ph. D. degree in Clinical Psychology

   
 

: jor professor

Date Tait/'1 lit/992

0.7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

A“

##fi—‘fi 4T-—'W-Ww v—Q-— 04's.

r-‘

v

" " ~fi'f—vfir— ., n.»

 

 

LIBRARY
Michigan State
University

 

 

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.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ma WMpfiS—p.“

AN EXAMINATION OF FACTORS PREDICTING THE REPORTING OF SEXUAL
ASSAULT IN A COLLEGE SAMPLE

By

Jennifer Susan Paul

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Department of Psychology

1 999

ABSTRACT

AN EXAlVIINATION OF FACTORS PREDICTING THE REPORTING OF SEXUAL
ASSAULT IN A COLLEGE SAMPLE

By

Jennifer Susan Paul

Sexual assault is a serious and pervasive problem on college campuses; however,
studies have shown that as few as 28% of women who have been sexually assaulted seek
out help from anyone (Ogeltree, 1993). In order to help sexual assault survivors, barriers
to reporting need to be identified. This study expands upon previous research by
examining a comprehensive model of factors that contribute to women’s decisions
whether or not to tell anyone about their experience of sexual assault. The model
examined the relationships between women’s definitions of their experiences, reporting
behavior, sex-role stereotypes, type of relationship between perpetrator and survivor,
physical injury, type of force, and beliefs about what others would want the women to do
and how they would react.

The sample consisted of 397 college females, ages 17-23. Results indicated that
49% of the total sample had been sexually assaulted; however, 59% of the women who
were assaulted did not label their experience as an assault. Twenty percent of women
who were assaulted told no one and 69% told only one person, while most non-assaulted
women (85%) believed they would tell two or more people. Results also indicated
differences in expectations for outcomes of reporting and who was blamed between

assaulted and non-assaulted women. Structural Equation Modeling revealed that the data

fit the model for non-assaulted women; however, most of the paths were not significant.
In addition, due to low variance in reporting, the model for assaulted women was difficult
to interpret. Despite limited findings with the models, the study revealed several
significant, interesting relationships between the variables. Specifically, for non-
assaulted women significant, positive relationships were found between expectations for
reporting, blame, beliefs about what others would want them to do, and reporting. For
women who have been sexually assaulted, relationships were found between labeling the
incident as an assault and level of injury, type of force, blame and reporting. In addition,
reporting was found to be related to perceived outcomes, injury and normative beliefs
about family members. The findings suggest that colleges could benefit from
implementing campus-wide psychoeducational programs about sexual assault as well as

how to help survivors of sexual assault.

ACKNOWLEDGMENTS

Completing a dissertation requires a lot of effort on the part of not only the author
but also fi'om many others. This process has taught me the importance of seeking out
support for both the logistical nightmares involved in research and graduate school but
also the emotional roller coaster that often occurs. The process of completing my
doctoral degree was smoother and more enjoyable because of the support from my
committee, my chair, my friends and my family. I know that without these supports this
process would have been even more draining and less rewarding.

I would like to thank my committee members, Cris Sullivan, Gary Stollak, and
Alex VonEye for their time, effort, guidance and patience with me throughout this
process. I could not have completed this project without the support and guidance of my
chair, Alytia Levendosky, Ph.D. Our countless meetings and her input throughout this
process were invaluable. Her support for not only the methodological problems I had but
also for stressors in my personal life means a lot to me. She is an excellent role model
and her continued interest in me as both a psychologist and as a person was instrumental
in both my professional and personal development.

Of course, I would never have made it through this process without the support
and love of my family. My parent’s support throughout my entire graduate school
experience was invaluable. The numerous weekends I spent with Nancy, Tige and
Zachary helped me keep my sanity throughout the trials and tribulations of writing a

dissertation. Susie’s ability to listen to me throughout this process and to make me laugh

iv

uncontrollably means more to me than I could ever express. Geoffi'ey’s support and our
wonderful time together in Italy was very special to me.

Graduate school taught me to appreciate the many wonderful fiiends in my life.
Late-night chats and walks with Alison Ward and her never-ending support helped me
remember the larger perspective and always cheered me up. Her support through the
stressful times in my life and through losses I experienced is something for which I will
be forever grateful.

My friendship with Christina Haemmerle which began on my first day of graduate
is truly a blessing from God. Our fiiendship and our crazy times together made graduate
school actually fun! In addition to helping me with the technical aspects of writing a
dissertation she was always there to listen and to put a smile on my face. The laughter,
tears, and secrets we shared will always be with me. I would also like to thank Chelsea
and Suzanna whose craziness and wild spirits reminded me to never forget what is really
important in life!

I would like to thank the women whose stories inspired this project. Last, I would

like to thank the many women who participated in my study and shared their

experiences with me.

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................... ix
LIST OF FIGURES ....................................................................................................... x
INTRODUCTION ......................................................................................................... 1
Definition of Sexual Assault .............................................................................. 2
Prevalence of Sexual Assault ............................................................................. 5
Methodological Problems in Research on the Prevalence of Sexual Assault.... 7
Effects of Social Support for Sexual Assault Survivors .................................... 9
Reporting Sexual Assault ................................................................................... 11
Factors Contributing to Women’s Reporting Behaviors Following
Sexual Assault .................................................................................................... 14
Feminist Control Model of factors influencing reporting .................................. l4
Denial or Minimization of Sexual Assault ............................................ 16
Type of Force ......................................................................................... 18
Physical Injury ....................................................................................... 19
Relationship between Survivor and Offender ........................................ 21
Sex Role Socialization ........................................................................... 22
Attribution of Blame ............................................................................. 24
Costs and Benefits of Telling Someone ................................................. 25
Social-Cognitive Model of Factors Influencing Reporting .............................. 27
Attitude Component ............................................................................. 28
Normative Component ........................................................................... 28
Empirical Evidence of Fishbein’s Model .............................................. 29
Combined Model of Reporting Behavior ......................................................... 33
Conclusions ........................................................................................................ 37
HYPOTHESES .............................................................................................................. 39
METHODS .................................................................................................................... 41
Participants ........................................................................................................ 41
Procedures ......................................................................................................... 42
Measures ............................................................................................................ 43
Demographic Questionnaire .................................................................. 43

Sexual Assault Experiences, Labeling of Incident, and Type of Force. 44
Reporting Behavior, Physical Violence, Intimacy,

Attribution of Blame .............................................................................. 46
Normative Expectations and Motivations to Comply with these

Expectations ........................................................................................... 47
Perceived Outcomes ............................................................................... 48
Sex Role Socialization ......................................................... 50

vi

RESULTS ...................................................................................................................... S3

Frequencies and Means of Variables ................................................................ 53
Reporting ................................................................................................ 60
Blame .................................................................................................... 6O
Expectations for Reporting ................................................................... 61
Label of Incident ................................................................................... 62
Relationship between Survivor and Perpetrator ..................................... 62
Level of Injuries .................................................................................... 62
Relationships between Observed Variables ....................................................... 62
Structural Equation Models ............................................................................... 67
Review of the Measurement Model .................................................................. 69
Sex Role Stereotypes ............................................................................. 7O
Normative Expectations ......................................................................... 70
Perceived Outcomes ............................................................................... 71
Blame ..................................................................................................... 72
Reporting ................................................................................................ 72
Type of Force, Relationship, Violence, and Label of Incident .............. 73
Results of the Overall Model Fit ........................................................................ 74
Model for Non-Assaulted Women .................................................................... 74
Model for Survivors of Sexual Assault ............................................................. 83
Results of the Hypotheses .................................................................................. 74
Structural Model for Non-Assaulted Women ................................................... 78
Structural Model for Survivors of Sexual Assault ............................................ 79
DISCUSSION ............................................................................................................... 81
Characteristics of Sample and Frequencies of the Observed Variables. . 81
Differences between Demographic Variables ................................................... 81
Prevalence of Sexual Assault ............................................................................ 83
Relationships between Variables in Assaulted Women .................................... 88
Correlations between Variables in Non-assaulted Women ................................ 91
Structural model of Factors Affecting Women’s Decisions Whether or not to
Report Sexual Assault (Sample B) ................................................................... 93
Structural model of Factors Affecting Women’s Decisions Whether or not to
Report Sexual Assault (Sample A) .................................................................... 94
Limitations of this Study .................................................................................. 95
Directions for Future Research ........................................................ 97
FIGURES ....................................................................................................................... 102
APPENDIX A ................................................................................................................ 1 13
APPENDIX B ................................................................................................................ 114
APPENDIX C ................................................................................................................ 131

vii

APPENDIX D ................................................................................................................ 133

REFERENCES .............................................................................................................. 1 34

viii

 

 

 

Tal

Ta!
Ass

Ta

LIST OF TABLES

Table 1. Types of Assaults Endorsed and Used as Critical Incidents .......................... 54
Table 2. Crosstabulations of Frequencies of Unwanted Sex Play Experiences and
Characteristics of the Assault ........................................................................................ 56
Table 3. Crosstabulations of Frequencies of Unwanted Sexual Intercourse and
Characteristics of the Assault ........................................................................................ 57
Table 4. Means, Standard Deviations, and Range of Observed Variables in

Assaulted Women (Sample A) ....................................................................................... 58
Table 5. Means, Standard Deviations, and Range of Observed Variables in
Non-assaulted Women (Sample B) ................................................................................ 59
Table 6. Correlations of Observed Variables in Assaulted Women (Sample A) ......... 63

Table 7. Relationship between Force and the other Observed Variables in
Assaulted Women (Sample A) ...................................................................................... 65

Table 8. Correlations of Observed Variables in Non-assaulted Women (Sample B) .. 66

ix

LIST OF FIGURES
Figure l. Fishbein’s Model of Behavioral Intentions .................................................. 104

Figure 2. Reduced Model of factors that influence women’s decisions whether
or not to report sexual assault ........................................................................................ 105

Figure 3. Combined Model of factors that influence women’s decisions whether
or not to report sexual assault ....................................................................................... 106

Figure 4. Model of factors contributing to women’s decisions about how many
people they would tell about a hypothetical sexual assault (Sample B) ........................ 107

Figure 5. Model of factors contributing to women’s decisions about whom they
would report a hypothetical sexual assault to (Sample B) ............................................ 108

Figure 6. Model of factors contributing to women’s decisions about how many
people they told about their experience of sexual assault (Sample A) ......................... 109

Figure 7. Model of factors contributing to women’s decisions about whom they
did report their experience of sexual assault to (Sample A) ......................................... 110

Figure 8. Total model of factors contributing to women’s decisions about whom
they did report their experience of sexual assault to (Sample A) ................................. 111

Figure 9. Model of factors contributing to women’s decisions about how many
people they would tell about a hypothetical sexual assault (Sample B, Final Model) . 112

Figure 10. Model of factors contributing to women’s decisions about whom they
would report a hypothetical sexual assault to (Sample B, Final Model) ....................... 113

Figure 11. Modified Model of factors contributing to women’s decisions about
whom they did report their experience of sexual assault to (Sample A) ....................... 114

INTRODUCTION

Sexual assault is a serious problem on college campuses. National surveys in the
United States and Canada have shown that many women are sexually assaulted while at
college (DeKeseredy & Kelly, 1993; Koss et. al, 1987). Estimates of percentages of
college women who have experienced sexual assault have ranged from 15% to 47%
(Ogeltree, 1993). It is well documented that women who have been raped or sexually
assaulted experience a wide range of psychological symptoms, such as depression
(Resick, Calhoun, Atkeson, & Ellis, 1982; Kilpatrick, Veronen, & Resick, 1979), fear and
anxiety (Calhoun, Atkeson, & Resick, 1982; Kilpatrick, Resick, & Veronen, 1981),
sexual dysfunction (Orlando & Koss, 1983; Norris & F eldman-Summers, 1981) and
problems in social adjustment (Kilpatrick et. a1, 1979; Resick et. a1, 1982). To cope
successfully with the psychological effects of sexual assault survivors often need support,
yet research has shown that many survivors do not seek any kind of support (Koss &
Oros, 1982; Koss, 1985; and Stets & Pirog-Good, 1987). In order to help sexual assault
survivors, barriers to reporting sexual assault in relationships need to be identified. While
prevalence, etiology, and risk factors of sexual assault have been studied, little research
has focused on variables that affect women's decisions about whether or not they will
seek help after being. sexually assaulted.

This literature review and research project explores why women do not talk to people
and/or report their experiences of sexual assault. The primary purpose of this literature
review is to provide a context for the current study examined factors contributing to
women’s decisions to report their sexual assault experiences to others. This literature

review covers four main areas. First, a general overview of the problem of sexual assault

and low reporting rates are provided by defining sexual assault, its prevalence, and
reporting rates. Second, reasons for not reporting are examined from a feminist
perspective. Third, a social cognitive perspective of reasons why women do not report
sexual assault is presented. Finally, flaws in the research are reviewed and the current
study is discussed. The purpose of this study was to test a comprehensive model of
variables related to reporting behavior, using both a feminist and a social-cognitive
perspective to consider a wide variety of factors.

Definition of Sexual Assault

Research in this area has been plagued by problems with changing definitions of
sexual assault. The definition of sexual assault can affect estimation of prevalence rates
and can make it difficult to compare findings across studies. Muehlenhard, Powch,
Phelps, & Giusti (1992) noted that definitions of sexual assault can vary along four
dimensions: 1) the individuals who are specified, 2) the criteria used for determining
nonconsent, 3) the sexual behaviors that are specified, and 4) who decides if sexual
assault has occurred.

Individuals who are specified can vary by gender and age. While men and women
can be survivors and perpetrators of sexual assault, it is most common for women to be
sexually assaulted by men (U .S. Department of Justice, 1997). As a result, this study will
focus on sexual assault of women by men. In addition, assault occurs to individuals of
all ages; however, since different factors may contribute to reporting based on the age of
the individual, this study will focus specifically on one age group. Since research has

shown that college women are at approximately three times greater risk for sexual assault

~‘ ‘~ MT 'J
-"

 

 

 

pt

EX
pr:
Wc
CXt
Val
int:

bel.

OCC

than women in the general population, college students were chosen as the subjects for
this study (Koss, Gidycz, Wisniewski, 1987).

Criteria for establishing nonconsent can rely upon the women’s state of mind (i.e., she
didn’t want to, it was against her will, etc.), verbal or non-verbal signs of resistance (i.e.,
crying, screaming, hitting, etc.), or the use of force by the perpetrator. A woman’s state
of mind is known only by her; therefore, studies which rely upon state of mind criteria
for nonconsent can be problematic depending upon who is asked. Specifically, asking the
perpetrator about the women’s state of mind leaves the open possibility that he has
misinterpreted the event. In fact, Muehlenhard (1988) found that men interpret various
behaviors more sexually than women do. Using the victim’s behavior as a definition of
nonconsent can also be problematic if the range of behavior used in the definition is
limited or relies upon only physical signs of nonconsent. If a woman stated her
unwillingness during the assault but the study asks if she physically fought back her
experience would not be labeled as an assault. In addition to resulting in decreased
prevalence rates, this type of questioning can result in increased self-blame or shame of
woman who did not physically resist. Last, using force as a definition of nonconsent
excludes many cases of non-stereotypical forms of sexual assault. Definitions can also
vary by what behaviors are specified. Specifically, studies can examine unwanted sexual
intercourse only or can include any unwanted sexual behavior. Clearly, the range of
behaviors listed will affect the prevalence rates.

In addition there are different ways to assess who decides whether sexual assault has
occurred. For instance, it can be measured by including only incidences which women

label as an assault, using only women who have reported the incident to the police, using

cases which resulted in legal charges, or using women’s responses to behavioral
descriptions of the legal definitions of assault. With the exception of using behavioral
descriptions, these mechanisms for deciding if an assault occurred are based on subjective
definitions and would not accurately represent the prevalence of sexual assault.

Since sexual assault can be defined in many different ways and these different
definitions affect research findings, it is important for researchers to clarify how they will
define sexual assault. As noted above, it is important to use behavioral descriptions of
assault and to examine a wide variety of unwanted behaviors to gain a more accurate
assessment of the problem of sexual assault. Koss et al. (1987) suggest there are four
types of sexual assault: sexual contact, sexual coercion, attempted rape and rape. This
study used the following definition of sexual assault:

" Sexual contact includes unwanted sex play (fondling, kissing or petting) arising from
the use of menacing verbal pressure, misuse of authority, threats of harm or actual
physical force. Sexual coercion includes unwanted sexual intercourse arising from the
use of menacing verbal pressure or the misuse of authority. Attempted rape includes
attempted unwanted sexual intercourse arising fi'om the use of threats of force, or the use
of drugs or alcohol. Rape includes unwanted sexual intercourse arising from the use of
threats of force and unwanted sex acts (anal or oral intercourse or penetration by objects
other than the penis) arising from the use of threat of force, or the use of drugs or alcohol"
(Koss et al., 1987).

In addition to differences in definitions of sexual assault, researchers have used
different words to refer to women who have experienced sexual assault. Traditionally,
women who have been assaulted are referred to as “victims” of sexual assault. This term
implies powerlessness and can for some people suggest helplessness. Muehlenhard,
Powch, Phelps, & Giusti (1992) suggest using the term survivor to refer to women who

have experienced sexual assault. Since the term survivor is more empowering, women

who have experienced sexual assault will be referred to as survivors in this study.

 

 

‘1
t»-
'—

 

the
ha'
of
be:
am
the
the;
that
atte

sun

3352
com
4%
four
Corn

854

datin

Prevalence of Sexual Assault

Research with college students has shown that sexual assault is a common
phenomenon. In a national survey of 32 institutions of all types of higher education in
the United States, Koss et al. (1987) found that 54% of 3,187 college females reported
having experienced some type of sexual assault since the age of 14. Approximately 15%
of the women reported that their most serious sexual victimization since the age of 14 had
been a completed rape and 12% reported that their most serious victimization was
attempted rape. Eighty-four percent of the rape survivors knew their attacker and 57% of
the rapists were dates. Koss et al. (1987) also asked the women to report whether or not
they had been sexually assaulted in the last year to assess incidence rates. They found
that 207 women had experienced a total of 353 rapes, 323 were survivors of 533
attempted rapes, 366 were survivors of 837 cases of sexual coercion, and 886 were
survivors of 2,024 experiences of unwanted sexual contact in the last year.

Koss et al.(1987) also asked 2,972 men to report about their sexual activities.
Approximately 25% of the men in their sample reported engaging in some form of sexual
assault since the age of 14. Ten percent admitted to having engaged in unwanted sexual
contact, 7% reported using sexual coercion, 3% reported engaging in attempted rape and
4% reported raping someone. When assessing the incidence rates, Koss et. a1 (1987)
found that there were 187 rapes reported by 96 different men, 167 attempted rapes
committed by 105 offenders, 311 episodes of sexual coercion perpetrated by 167 men and
854 unwanted sexual contact incidents committed by 374 men.

Researchers have also examined prevalence rates of sexual assault specifically in

dating relationships. In a study of 330 women at a large southeastern university, Himelein

(1995) found that 29% of students had been sexually victimized in a dating situation
since entering college. Eight percent of the women experienced unwanted sex play
(kissing, fondling, petting), 13% experienced attempted rape or unwanted sexual
intercourse obtained through verbal pressure, and 8% experienced unwanted sexual
intercourse obtained through threat of force, use of force, or administration of alcohol or
drugs. In another study examining the prevalence of sexual assault in dating situations in
a college population, Muehlenhard and Linton (1987) found that 78% of 341 women
reported having experienced some form of unwanted sexual activity, ranging fiom
unwanted kissing to rape, in high school or college. In this study, 15% of the women had
experienced rape. Fifty-seven percent of the 294 male respondents in this study admitted
to engaging in some type of sexual assault with a partner, and 7% admitted to raping a
woman. Muehlenhard and Linton (1987) found similar prevalence rates of rape to Koss
et al. (1987), however their rates of sexual assault were much higher. The discrepancies
in prevalence rates may have been a result of methodological differences between the two
studies. While the overall rates were different, the incidence of rape reported in the two
studies is very similar for both women and men. Muehlenhard and Linton (1987) used a
broader definition of unwanted sexual activity, including seventeen items ranging fi'om
unwanted kissing without tongue contact to sexual intercourse, which may explain the
difference in the overall rate of sexual assault found in these studies.

Sexual assault on college campuses is not a problem unique to American universities.

For instance, Gavey (1991) examined prevalence of sexual assault in a university in New

Zealand. He found that 52% of 327 undergraduate women had experienced some form of

sexual assault and 25% had experienced rape or attempted rape. In addition, the

Canadian National Survey, a sample survey of 3,142 community college and university
students, found that 28% of the women in their sample had been sexually assaulted in the
past year by someone they knew and 11% of the men reported sexually assaulting a
dating partner (DeKerseredy & Kelly, 1996). The study also examined prevalence of
sexual assault since leaving high school and found that 45% of the females stated they
had been sexually assaulted and 20% of the men reported they had sexually assaulted.

In sum, these studies found high prevalence and incidence rates of sexual assault.
Although rates varied across studies, all results indicate that sexual assault among
acquaintances in college and universities is a widespread and serious social problem.
Methodological Problems in Research on the Prevalence of Sexual Assault

Comparing findings across studies on sexual assault can be difficult because of
methodological differences. The following section discusses the various problems that
can affect the interpretation of sexual assault research. As noted earlier, prevalence and
incidence rates of sexual assault vary across studies. This wide range of findings can be
explained in part by differences in methodologies (e. g. interview versus questionnaire)
and wording of the definition of assault.

Discrepancies between the rates of sexual assault experienced by women and the rates
of admitted sexual assault by men are consistent across studies. Koss et al. (1987)
showed that this discrepancy was not likely a result of a small number of men abusing
many women, since the average number of incidents per respondent reported by women
and men were equivalent. They hypothesized that either men were under-reporting the
severity of sexual assault they engaged in or a large number of women in the study were

victimized by men who were not college students. In support of the former, Koss et al.

(1987) cited several studies with convicted rapists, which found that rapists often
perceived their coercive and threatening behavior towards their victims as within the
bounds of mutually consensual sexual activity. Warshaw (1988) offers a different
explanation for the discrepancy between female report of victimization and male report of
sexual assault offenses. He suggests that male perpetrators may realize that sexual assault
is wrong and therefore be hesitant to admit having sexually assaulted anyone. In fact,
some researchers have argued that social desirability may play a key role in shaping male
responses to these questionnaires (DeKerseredy & Kelly, 1993). While this theory has not
been tested with sexual assault offenders, Button and Hemphill (1992) did examine the
role of social desirability in self-report behavior of physical and emotional abuse. They
used the Balanced Inventory of Desirable Responding (Paulhus, 1984) to examine two
components of social desirability: impression management or lying and self-deception or
distorted responses. They found that men’s scores on both the impression management
and self-deception scales negatively correlated with self-report of physical and emotional
abuse.

Another methodological problem in researching sexual assault is the effect that
different types of data collection can have on reported prevalence and incidence rates.
Prevalence rates of rape and attempted rape in the studies previously discussed are
significantly and substantially higher (10 to 15%) than prevalence rates of rape and
attempted rape in official victimization studies (Koss et al., 1987). For instance, the
National Crime Survey, conducted by the Bureau of Justice Statistics (1997), using
telephone surveys found that less than one percent of women (approximately 1 out of

every 625 women) had experienced rape or sexual assault. The National Crime

Vict

3CCC

deft

hart

ofn

rap

int

 

ift

ex;

rap

Clt

Victimization survey uses the word "rape" and asks about sexual "attacks", which may
account for the low prevalence rate. These type of questions assume that the victim
defines her experience as rape or perceives sexual assault as an "attack." In fact, studies
have shown that many women who have had experiences which meet the legal definition
of rape do not label their experiences as rape when asked directly (Koss, 1985; Koss et
al., 1987; Lane & Gwartney-Gibbs, 1985).

Koss labels women who do not identify their experiences as rape as "unacknowledged
rape victims." These women respond affinnatively to a question that describes an
incident meeting the legal definition of rape, but do not respond affirmatively when asked
if they have been "raped." The percentage of survivors who labeled their rape
experiences as a rape have ranged fiom only 27% (Koss ct.al, 1987) to 57% (Koss, 1985).

Wording of questions, men and women's views of what classifies as rape or attempted
rape, and social desirability may all affect the results of studies assessing sexual assault.
Clearly, researchers can not only directly ask individuals about rape. Instead, it is
important for researchers to use behavioral descriptions when inquiring about sexual
assault, in order to get the most accurate findings. Accurate findings are important not
only to the validity of the research but also for the usefulness of the research.
Underestimating the extent of the problem of sexual assault can have negative effects on
sexual assault survivors. For instance, low prevalence rates in studies may lead to less
fimding for prevention and treatment of sexual assault.

Effects of Social Supmrt for Sexual Assault Survivors

It is important that enough services are available to help sexual assault survivors

because research demonstrates that social support can buffer the psychological effects of

H'L-X.

 

 

stress
expet
inten

andc'

they
talki

“'01!

5m;
Studi
Crisi:
bent

Socia

other

net It

bell,

dEIEI

stressful life experiences (Cohen & Willis, 1985), including rape and sexual assault
experiences (Golding et al., 1989; Popiel & Susskind, 1985). Research using both
interview data and questionnaire data demonstrates a relationship between social support
and decreased symptomatology in sexual assault survivors. For instance, Meyer and
Taylor (1986) asked 58 rape survivors what they did that made them feel better and most
survivors mentioned counseling, talking to fiiends and family, making life changes and
taking precautions. Frazier and Burnett (1994) also asked sexual assault survivors what
they did that made them feel better. They found that 46% of the sample reported that
talking about the rape and expressing feelings helped them feel better and 45% of the
women also indicated that getting support from fiiends and family was helpfirl.
Researchers have also examined the relationship between social support and
symptomatology and physical health. For instance, Kimmerling and Calhoun (1994)
studied 115 women between the ages of 15 and 71 who had received services at a rape
crisis center. They found that social support was a moderating variable in the relationship
between physical health and victimization, F (4, 651)=2.48, p<.04, with higher levels of
social support being associated with better physical health following victimization.
Overall, current research supports the hypothesis that women who receive help fi'om
others experience fewer problems associated with the sexual assault than women who do
not receive support. Failure to receive any help may result in prolonged dysfunctional
behavior and problems for the victim (Braswell, 1989) and therefore it is very important
that the reasons for not reporting be examined. In addition, it is important to understand
why women do not report sexual assault, since higher reporting rates may serve as a

deterrent to some potential offenders because they may fear that their behavior would also

10

 

be report
potential
believe t
likely to
to engag
sanction
m

In a:
campus
assaulte
aged l2
law enf
reponip
insane:
Epidem
assaulte
240 (15
aSked u
Pei 3011,
tape Cris
aSSaulte.
and 29%

mangr

be reported. However, if increased reporting occurs only to peers this may lead some
potential offenders to believe their behavior is normal and/or common. If individuals
believe they are likely to be punished for committing sexual assault, they may be less
likely to commit acts of sexual assault. In fact, research has shown that men’s inclination
to engage in sexually assaultive behaviors decreased significantly by threat of formal
sanctions through the university or the police (Bachman, 1993).
_no__gRe rtin Sm Assails

In assessing prevalence of sexual assault, researchers can not rely upon police and
campus reports of sexual assault because many women do not report being sexually
assaulted. The National Crime Victimization Study (1997) of more than 100,000 females
aged 12 and older found that 32% of rape/ sexual assault victimizations were reported to a
law enforcement agency. Studies have also found that there are differential rates of
reporting dependent upon type of person reported to (i.e. police versus family). For
instance, Ulhnan and Siegel (1993) used a subsample fiom the Los Angeles
Epidemiological Catchment Area (ECA) study to examine reporting behavior of sexually
assaulted women in a community sample. They interviewed 1651 women and found that
240 (15%) of the women had been sexually assaulted as adults (age 18 or older). They
asked women if they talked to either a fiiend/relative professional helper (religious
person, physician, and police), or a mental health professional (mental health worker or
rape crisis counselor), after the assault occurred. They found that 67% of women
assaulted by someone they knew told a relative or fiiend, 30% told a professional helper,
and 29% told a mental health professional. Eighty-four percent of women assaulted by a

stranger told a friend or relative, 55% told a professional helper, and 29% told a mental

11

 

 

health ser
to friends
women at

Georg
Mental H
the Nonh
samples 1
abused d
rape. in c
different
definitior
Angeles
lithaVior
found [h
friend or
16% tolc
leader, 5
acquajm

Rese

health service worker. Chi square analyses revealed significant differences on reporting
to fiiends/relatives and professional helper between women assaulted by a stranger and
women assaulted by an acquaintance (chi square=.448 and 3.24 respectively, p<.05).
George, Winfield and Blazer (1992) also used data from the National Institute of
Mental Health Epidemiological Catchment Area (ECA) Program to examine women from
the North Carolina ECA site. They found lower rates of lifetime sexual assault than most
samples previously discussed; they reported that only 6% of the sample had been sexually
abused during their lifetime. The authors point out that 82% of these reports involved
rape, in contrast to 50% of the Los Angeles sample (Sorenson et al., 1987). These
differences suggest that women in the North Carolina sample may have restricted their
definition of sexual assault to more serious forms of assault than did women in the Los
Angeles sample. Despite differences in prevalence rates, they found similar reporting
behaviors to other community samples. Using the anonymous responses, the researchers
found that 31% of their sample told no one about being sexually assaulted, 69% told a
friend or relative, 27% told a psychiatrist or mental health counselor, 20% told the police,
16% told a physician, medical professional, or emergency room staff, 5% told a religious
leader, 5% told a rape crisis center. This study did not differentiate between stranger or
acquaintance rape or between acknowledged or unacknowledged sexual assault survivors.
Research with college populations find that many women do not talk about their
experience with others. One study that examined women's reporting/help—seeking
behaviors after being sexually assaulted by their dating partners (Ogeltree, 1993) found
that 42% (278) of the women had been survivors of sexual coercion in dating situations

while in college. Of the women who had experienced sexual assault, 199 (72%) indicated

12

 

that th
seeker
couns
help f
and 1'
indie:

seekii

speci
assat
30km
unatl
Ctnlt
358211
36km.
cTisis
M05
Sun-i
16m!

unae

amt):

Elli;

that they did not seek help fiom anyone after the incident. Within the group of help
seekers (N=79), 75% (59 women) went to a fiiend for help, 9% (7 women) went to a
counselor or therapist, 6% (5 women) sought help from the police, 4% (3 women) sought
help from a rape crisis center, 4% (3 women) sought help from a parent or other relative
and 1% (1 woman) sought help from a physician or hospital. Results of this study
indicate that most college women who have been survivors of sexual assault are not
seeking potentially valuable help.

Koss (1985) also used a college sample to examine reporting rates and to look
specifically at differences in reporting for acknowledged versus unacknowledged sexual
assault survivors. For instance, Koss (1985) examined unacknowledged and
acknowledged rape survivors using a university sample and found that none of the
unacknowledged rape survivors reported their experience to the police, a rape crisis
center, or a hospital emergency room. In fact, more than half of them did not reveal their
assault to anyone, other than anonymously to the researchers. Eight percent of
acknowledged rape survivors reported their experience to the police; 13% went to a rape
crisis center or hospital emergency room, and 48% did not discuss their rape with anyone.

Most of the rape survivors in this study knew their attacker; 59% of acknowledged
survivors and 100% of unacknowledged survivors reported that at minimum they
recognized the man. Thirty-one percent of acknowledged rape survivors and 76% of
unacknowledged rape survivors were romantically involved with their attackers.

Overall, the literature suggests that in both community and college samples, and
among both acknowledged and unacknowledged rape survivors, many women are not

telling anyone about their experiences of sexual assault. Even fewer women are seeking

13

 

valuabl
doctors
friends
most It
sexuall
unders
barrier
and so.
This st
theorie

WOmCl

valuable help from professionals (police, rape crisis centers, mental health workers,
doctors, etc.). Most of the women who sought help turned to their friends. Although
fiiends may be a valuable source of social support for these women, they may not be the
most knowledgeable source of support. In conclusion, the high percentage of women
sexually assaulted in combination with low reporting rates suggest the need for an
understanding to whom survivors turn for support and why they rarely seek help so that
barriers to seeking help can be addressed. Two main theoretical perspectives, feminist
and social-cognitive, have been used to explain why women do not report sexual assault.
This study will review each of these theories, the empirical evidence to support the
theories, and evaluate the validity of a combination of these models in predicting
women’s help-seeking/reporting behaviors.

Factors’ Contributing t_o Women’s Remrting Behaviors Following Sexual Assault

 

m MM 91’ .Se_xu_&il _A_88_au_lt.

In order to understand why women do not tell anyone about being sexually assaulted,
it is important to review the processes that contribute to the occurrence of sexual assault.
Feminists view sexual assault as an integral part of a patriarchal society and as a social
tradition of male domination and female exploitation (Brownmiller, 197 5, Rose, 1977).
Feminist theory postulates that social norms contribute to stereotyped attitudes about rape
and create a rape-supportive environment (Rose, 1977). These norms contribute to I
acceptance of rape as a part of the continuum of normal sexual behavior. These lead to
the assumption that sexual assault is not pathological in the individual sense but is a

reflection of society’s pathology. In addition to maladaptive norms, power differentials

between men and women are seen as contributing to the occurrence of sexual assault.

l4

 

Sexual a
women (
economi
the high
promote
power d

contribu

“PM
many it
culture I
domina
exagget
1998).
L’sit
rel30rtir.
1995).
Factors
Ofiende
mimini;
Of tellin

”Ch of

Sexual assault is understood as a way in which men as a class maintain control over
women (Brownmiller, 1975). In fact research has found a negative correlation between
economic and social gender inequality and sexual abuse: the less powerful women were,
the higher the rates of rape (Baron, Straus and Jaffee,1987). These gender differences are
promoted by cultural gender roles and expectations. Norms for sexual behavior, and the
power difl‘erential between genders that promote a rape supportive environment also may
contribute to women's hesitancy to report crimes of sexual assault.

Sexual assault is understood to be intrinsic to a system of male supremacy and in
support of this theory, some feminist theorists have pointed to the social acceptance of
many forms of sexual assault and the glorification of sexual violence in the dominant
culture (Herman, 1990). Sexuality is often scene as involving the erotization of male
dominance and female submission, therefore the use of coercion to achieve sex may be an
exaggeration of, but not a departure fi'om society’s norms. (Herman, 1990; Russell,
1998)

Using feminist explanations of sexual assault, researchers have postulated that the low
reporting of sexual assault results from two general processes (Gartner & Macmillan,
1995). The first process involves denial or minimization of the assaultive behavior.
Factors such as stereotypical beliefs about assault, relationship between victim and
offender, sex role socialization and attribution of blame can all contribute to the denial or
minimization of the assault. The second process involves weighing the costs and benefits
of telling anyone. The following section discusses the theory and research findings for

each of these processes.

15

 

c:

53
E.
A

l».

Ferr
tiolenc:
eneoure
This "si
support
demons
the S€Xt
assault
likely t
experie
found t
their er

M:
(1995)
Whfithe
fitting
One pe:
fitment
label t}
to guts

on

Words .

Feminist theory postulates that women tend to and are encouraged to minimize the
violence that they experience from men (Hester, Kelly & Radford, 1996). Women are
encouraged not to speak about their experiences of sexual abuse, coercion, and assault.
This "silencing" is a complex process. In order to speak about and seek out social
support, sexual assault survivors must first be able to name and define it. Research
demonstrates a link between defining a sexual assault experience as abusive and reporting
the sexual assault. A study at Kent State University found that unacknowledged sexual
assault survivors (women who did not label their experience as an assault) were less
likely to tell anyone ( other than anonymously telling the researcher) about their
experience than acknowledged survivors (Koss, 1985). Pirog-Good and Stets (1989) also
found that unacknowledged sexual assault survivors were less likely to tell anyone about
their experience.

Many women do not define their experiences as assaultive. Wiehe and Richards
(1995) studied women who had experienced acquaintance rape. They asked participants
whether or not they defined their own experience of being forced to engage in sexual
activity without their consent with an acquaintance as rape at the time it happened. Sixty-
one percent (143 women) of the sample reported that they did not define their own
experience as rape. Another study found that fifty-eight percent of 206 women failed to
label themselves as rape survivors, despite the fact that they had responded affirmatively
to questions that gave behavioral descriptions of rape (Amick & Calhoun 1987).

One of the most significant aspects of feminist theory and practice has been to find

words to define women's experiences so that these experiences can be discussed and

16

 

 

women
harassm
these be
previou
also ext
defined
by acqu

0th
experie
postula‘
eonsrru
medieir
through
accepra
Ot’erpgt
Sexual i
messag
as abus;
the exp,
genera]
all hold
demOnS

adeSm

women are not silenced. For instance, while words such as domestic violence and sexual
harassment did not always exist, these behaviors have existed for a long time. Naming
these behaviors helps bring into awareness and facilitate discussion of what was
previously invisible and unspeakable (Kelly, 1988). Feminists not only created words but
also expanded the definition of previously existing words. For instance, rape is no longer
defined (at least not in all settings) as the brutal attack by a stranger but also includes rape
by acquaintances and attacks not involving physical force.

Other paths by which women are silenced are the many ways in which women's
experiential knowledge about sexual assault is denied and invalidated. Feminist theory
postulates that men, who are the majority of perpetrators of sexual violence, have
constructed "knowledge" about sexual violence through such mechanisms as law,
medicine, and the media (Hester, Kelly & Radford, 1996). For instance, the media,
through its depiction of women and sexual behavior, can teach society about what is
acceptable and what is abnormal. The portrayal of an unwilling woman who is
overpowered by a passionate man is a popular movie theme and teaches that this type of
sexual interaction is acceptable (Gunn & Minch, 1988). Women may internalize this
message and as a result may have difficulty labeling their own sexual assault experiences
as abusive. Women are less likely to report acts of sexual violence if they do not perceive
the experience as abusive or consider it a sexual assault. Research demonstrates that the
general public (Burt, 1980), the police (Feldman-Summers & Palmer, 1980) and judges
all hold stereotypical views about the crime of rape and sexual assault. Researchers
demonstrate that the general public sees rape and sexual assault as involving an attack in

a deserted public place by a stranger who uses physical force to commit the sexual assault

17

 

K." ”a.‘

 

(Estrich.
women ‘
define ti
Specific.
physical
be less If
stereoty]
tradition
Stereoty}
percenta
alcohol t
the stere
(Orcurt t
cOrtjettu
till be H
Verbally
RCSeartl
Oms et a
factor in
to label t
phl'sica]

Unacknm

(Estrich, 1987; Williams, 1984; Weis & Borges, 1973). Feminist theory postulates that
women whose experiences do not meet this stereotypical definition will be less likely to
define their experiences as assaultive (Orcutt & Faison, 1988; Williams, 1984).
Specifically, when physical force is not used to force a woman to have rape, when no
physical injuries occur and when the perpetrator is someone the woman knows, she will
be less likely to define her experience as assaultive. In addition, the relationship between
stereotypical rape and reporting is hypothesized to be stronger for women who hold
traditional views of women and dating, since they will be more likely to believe
stereotypes about sexual assault (Orcutt & F aison, 1988).

Type of Force. Stereotypical rape involves the use of physical force; however, a large

 

percentage of women are forced to have sex through verbal coercion and by being given
alcohol or drugs. Feminist theory postulates that women whose experiences do not meet
the stereotypical definition will be less likely to define their experiences as assaultive
(Orcutt & Faison, 1988; Williams, 1984). Using this hypothesis it makes sense to
conjecture that women who are physically forced to have sexual contact with someone
will be more likely than women who are given alcohol or drugs, or women who are
verbally coerced to have unwanted sexual contact, to label their experience as an assault.
Research supports a link between type of force and labeling experiences. For instance,
Oros et al. (1980) found that the degree of physical force used by rapists was the main
factor in determining how the women labeled their experience; women were more likely
to label their experience as an assault if they had been subjected to a high degree of
physical force. In addition, Koss (1985) found that acknowledged survivors and

unacknowledged survivors of sexual assault differed in the number of different types of

18

 

 

physit‘
sun'it'
W
affects
demor.
shows
and les
1982).
for the
summit
on drug
beliet'e
Altl
0Oercior
has dem
assault (
directly
as assaul
my,
wWMu
related it

phl’SICall

physical violence they experienced (F (3,137)=118.27, p<.001), with acknowledged
survivors having experienced more types of violence.

While researchers have not specifically examined how the type of coercion used
aflects women’s perceptions of their sexual assault experiences, research has
demonstrated a correlation between type of force and blame. Specifically, research
shows that people place more blame on the woman if she was drunk during the assault
and less blame on the male if he was drunk during the assault (Richardson & Campbell,
1982). In addition, research demonstrates that women place more blame on themselves
for the incident if they have been drinking (Abbey, 1991). This finding lends indirect
support for the hypothesis that women who are forced to have sex while they are drunk or
on drugs will be less likely to label their experience as a sexual assault because they
believe they are responsible for what happened to them.

Although there is little research directly examining the relationship between type of
coercion (physical, verbal, alcohol) and labeling of sexual assault experiences, research
has demonstrated a link between physical force and labeling one’s experience as an
assault (Oros et al., 1980). This study hopes to address this flaw in the research by
directly examining the relationship between type of force and labeling one’s experience
as assaultive.

Phys_ic_al_ Iggy. Since stereotypical rape involves physical injury, feminist theory
would conjecture that the amount of physical injury incurred during an assault will be
related to women’s definitions of their experiences. Specifically, women who are not

physically injured or are only mildly injured (small cuts or bruises) will be less likely

19

 

 

than v
experi
Re
injurie
injury
1988;
sexual
definit
likely t
Goldstt
does Ilt
Study a
z’LSStiult
Res
demom
Sustain
inillt'ies
Signific.
have als
0ffende:
Studies t
if there i

WQmen~

than women who are severely injured (i.e., require medical attention) to label their
experience as an assault.

Researchers have argued that women learn that violent crimes are associated with
injuries and that sexual assault is a violent crime (Gunn & Minch, 1988). Therefore, if no
injury is incurred women may believe that no crime could have occurred (Gunn & Minch,
1988; Skelton & Burkhart, 1980). Research has not examined survivors' perceptions of
sexual assault and physical injury; however, studies have examined men’s and women’s
definitions of sexual assault scenarios. Research shows that people are significantly more
likely to label an incident as rape if the victim is injured (Bourque, 1990; Shotland &
Goldstein, 1983). While studying how the general public labels sexual assault is useful, it
does not imply that actual sexual assault survivors will respond in a similar manner. This
study attempts to address this current flaw in the literature by directly examining sexual
assault survivors’ definitions of their experiences and their relationship to physical injury.

Researchers have examined reporting rates in relation to physical injury. Research
demonstrates that physical injury is related to reporting a sexual assault. Women who
sustain a visible injury are more likely to report (68%) than if they did not suffer visible
injuries (32%) (Gunn & Minch, 1988). Bachman (1998) found that physical injuries
significantly increased the likelihood of reporting rapes to the police. Other researchers
have also found that women were more likely to report if they had been injured or if the
offender had used force (Bachman, 1988; F eldman-Surnmers & Norris, 1984). These
studies did not measure women’s definition of their experiences. As a result, it is unclear
if there is a direct link between reporting and injury or if this relationship is moderated by

women’s definitions of their experiences. Both links will be examined in this study.

20

 

saith in
experie
(Gartnt
traditic
stereot
womer
labelin
breakd
& Mac
experit
than vy
labelin
if they

Re:
label 3
of won
ticrim
raped h
as rape

Nor
nor tell

cornple

Relationship between Survivor an_d Offender. Feminist theory also postulates that

 

with increasing intimacy, there will be decreasing acknowledgment of sexual assault
experiences as an assault and consequently decreased reporting of these experiences
(Gartner & MacMillan, 1995). One reason for this relationship may be because the
traditional rape is believed to occur by a stranger and situations that do not fit the
stereotype will make defining the experience as an assault more difficult. In addition,
women who are assaulted by someone they know have a greater investment in not
labeling the experience as rape (Weis & Borges, 1973) because this might lead to a
breakdown in the relationships and the women may experience a sense of failure (Gartner
& Macmillan, 1995). Women who are raped by strangers may be more able to label the
experience as an assault and view their assaults as more random and less personal events
than women sexually assaulted by people they know. Researchers have theorized that by
labeling an experience as sexual assault, women assaulted by acquaintances may question
if they are competent to ftmction in relationships (Allison & Wrightsman, 1993).

Research supports the theory that women raped by acquaintances are less likely to
label their experience as rape. For instance, Koss et al. (1988) found that the percentage
of women describing their experience as rape decreased with increasing intimacy between
victim and offender; 28% of women raped by a nonromantic partner, 18% of women
raped by casual dates, and 18% of women raped by steady dates labeled their experiences
as rape.

Norms for intimate relationships may also increase the likelihood that the women will
not tell someone about sexual assault. In intimate relationships reporting is even more

complex, since relationships are seen as private and therefore seeking help from others is

21

 

 

58811 t
respo
(Bele:

that te

womer
membt

assault

wOmen
role Soc
mSault ;
Fttrrrinis
roles of
define th
Orcun &
leam abo

A mu
rol

e 500a.

acctturn fc

seen as a violation of the privacy of relationships (Gartner & Macmillan, 1995). The
responsibility for the success of the relationship is often seen as the women's job
(Belenky, Clinchy, Goldberger & Tarule, 1986). As a result, women may be concerned
that telling others will be a sign of failure and will be shameful and embarrassing.

Research demonstrates that women raped by acquaintances are less likely to report
the incident to anyone. Gunn and Minch (1 98 8) found that reporting decreased with
increasing familiarity with the victim. They found that 75% of women assaulted by a
stranger told the police, 47% of those assaulted by someone casually known, 43% of
women assaulted by someone well-known, and 25% of women assaulted by a family
member. Since research strongly supports a link between labeling an incident as an
assault and reporting, this link will be used in the model for this study.

§e_x_ Role Socialization. In addition to learning norms for intimate relationships,
women are also socialized to learn general roles of women and men. This process of sex
role socialization may be related to women’s understanding of what constitutes sexual
assault and consequently how women label their own experiences of sexual assault.
Feminist theory postulates that women who have more narrow or traditional views of the
roles of men and women (i.e. more traditional sex role expectations) will be less likely to
define their experiences as a sexual assault (Check & Malamuth, 1983; Lloyd, 1991;
Orcutt & Faison, 1988). Sex role socialization involves the process by which women
learn about types of behaviors and responsibilities of men and women.

A number of researchers have suggested that rape is the logical extension of the sex-

role socialization process and those assumptions about men, women, and sexuality

account for the rape-supportive nature of American society (Burt, 1980; Koss, 1987;

22

' “ “-

 

 

Sanday.
dating b
situatior
coercion
of norm;
the pursr
that all s
fact, resr
masculir
Brannigz
restttonsi
SUggeste
oriented

Tradj
of 58thal
sfilial or
that these
dating sit
iRUSSdl,

Tradii
“01ml (L
rem" that

(Mahmud

Sanday, 1981). Because traditional sex role expectations teach women norms regarding
dating behavior, the process of sex role socialization may be more influential in dating
situations. These norms may contribute to a rape supportive culture in which the use of
coercion and force to obtain sexual contact is seen, not as deviant, but as within the realm
of normal dating behavior (Check & Malamuth, 1983). In this culture, males are seen as
the pursuers of sexual contact and women are responsible for limit setting. This implies
that all strategies for avoiding sex are feminine and the responsibility of the female. In
fact, research has found that college students stereotype strategies for having sex as
masculine and strategies for avoiding sex as feminine (LaPlante, McCormick, &
Brannigan, 1980). In addition, research has shown that people often hold women
responsible for rape because it is their job to control the situation. Research has also
suggested that men are more likely than women to have dating goals that are sexually
oriented (Alksnis, Desmarais, & Wood, 1996).

Traditional dating norms (Check & Malamuth, 1983) teach men to persist in attempts
of sexual contact, even if the woman resists, since women are supposed to initially resist
sexual overtures (e.g., men may believe it is a token resistance). Theorists have argued
that these traditional norms regarding token resistance contribute to misunderstandings in
dating situations and may lead to the belief that rape is okay under certain conditions
(Russell, 1975; Check & Malamuth, 1983; Koss, Leonard, Beexly & Oros 1985).

Traditional norms appear to teach men that using force to achieve their sexual goals is
normal (Lloyd, 1991). One study supporting this hypothesis found that 35% of men
report that they might force their partner to have sex if they “could get away with it”

(Malamuth, 1981). The fact that approval of use of force to obtain sex is greater in the

23

 

 

context
situatim
(Bridge
h-chrai
a reason
Res:
assault 2
more he
date rap
found a
sexual a
likely to
inditidu
Scale (S;
(Muehla
between
s“PPOrtet
Study.
m
f0t being
Britten, 1
Sexual ass

(Bldggio,

context of a dating situation (first date or long term dating partner) than it is in a stranger
situation also supports the hypothesis that the use of force is part of dating norms
(Bridges & McGrail, 1989; Margolin, Miller, & Moran, 1989). In addition, Bridges &
McGrail (1989) found that men and women are more likely to label miscommunication as
a reason for date rape than for stranger rape.

Research has shown support for the link between definitions of the experience as an
assault and sex-role expectations. Check and Malamuth (1983) found that observers with
more traditional sex-role beliefs were less likely to perceive a favorable reaction of the
date rape survivor than individuals with more egalitarian attitudes. Bridges (1991) also
found a positive, significant relationship between sex-role beliefs and perceptions of
sexual assault experiences, with observers with more traditional sex-roles being more
likely to interpret date-rape situations more favorably . Research has shown that
individuals with more egalitarian attitudes according to the Attitudes Toward Women
Scale (Spence & Helrnreich, 1972) were more likely to label the experience as violent
(Muehlanhard, Friedman & Thomas, 1985; Shotland & Goldstein, 1983). The link
between sex-role socialization and labeling the experience as an assault has been
supported in the literature and therefore will be used in the model being examined in this
study.

Attribution of Blame. Sex role socialization leads some women to feel responsible

 

for being sexually assaulted (Bridges, 1991; Coller & Resick, 1987; Wilson, Faison &
Britton, 1983). Unlike many other crimes, there is a stigma attached to being a victim of
sexual assault because the victim is often considered responsible for the incident

(Biaggio, Brownell, & Watts, 1991). Women have historically been given the

24

 

respor
wome
may I]
fault a
feel re
the as:
tthat l
will re
egalitz
& The
that w.

will be

responsibility of controlling men’s sexual behavior, so unwanted sex is often seen as the
women’s failure to control the man’s sexual behavior (Brownmiller, 1975). A woman
may not label her experience as abusive if she blames herself because she views it as her
fault and not the offender’s fault. Feminist theory postulates that women are taught to
feel responsible for sexual assaults and that feeling responsible decreases the likelihood
the assault will be labeled as a crime (Brownmiller, 1975). If women feel responsible for
what happened they may also believe that others will hold them responsible and therefore
will refrain from telling anyone. Research has shown that individuals with more
egalitarian sex-role attitudes were less likely to blame the victim (Muehlanhard, Friedman
& Thomas, 1985; Shotland & Goldstein, 1983). Research also supports the hypothesis
that women who believe they are partly or totally responsible for being sexually assaulted
will be less likely to tell anyone about their experience. Gunn and Minch (1988) found
that women who blame themselves for the assault (31%) are less likely to report the
incident to the police than are women who felt they shared the blame with the offender
(48%) or women who believe that the assailant was totally responsible (90%). Since
research has suggested a link between blame and reporting and between blame and
labeling the incident as an assault, the relationships between these variables will be
examined in this study.

Costs and Benefits 91‘ Telling Someone.

 

 

If women do consider seeking help from others after sexual assault, a second process
works to discourage this behavior. At this stage, the woman begins to make decisions
about the costs and benefits of seeking help from others (Gartner & Macmillan, 1995).

Survivors are affected by societal reactions to violence against women. A woman's

25

 

decisit
knotyl
the of:
contrc
Womt
to the
peopl-
has sl
talker
anon

R
react
aflyor
negat
term
that r
Polio
“’Om
d6cig
asSat
Wom
matte

relift:

decisions not to report sexual assault to the police or others may be a result of her
knowledge that she may not be taken seriously, may be blamed for the attack, may anger
the offender and increase her chances of being assaulted again, and/or may not be able to
control the legal process (Dobash & Dobash, 1979; Hanmer, Radford, & Stanko, 1989).
Women will examine not only how society in general will react but also how those close
to them (in a sense still a reflection of society) will react. In addition, the reactions of the
people they care about may influence their decision to tell anyone else. In fact, research
has shown that women who received a supportive response from the first person they
talked to about the attacks were more likely to report the assault than those who received
a nonsupportive response (47 versus 21 percent) (Gunn & Minch, 1988).

Research has shown that women who believe that others (police, family, fiiends) will
react negatively to their talking about their assault will be less likely to seek help from
anyone (Gunn & Minch, 1988). Gunn and Minch (1988) found that women who had a
negative(3 8% reported) or neutral view (30% reported) of the police were less likely to
report than those who had a positive opinion of the police (59% reported). They noted
that many of the women who did not report their assault to the police noted that "the
police would not care" or "I did not think I would be believed." Factors that may affect
women's feelings that they will be believed by others have been shown to influence their
decision to report. Bachman (1993) examined reasons why women do not report sexual
assault using data from the National Crime Victimization Study. He found that 39% of
women felt the most important reason for not disclosing abuse was that it was a private
matter, 20% felt that the police would not do anything about it, and 13% were afraid of

reprisal from offender. Jensen and Gutek (1982) found that more than one third of sexual

26

assault survivors in their sample stated that they did not report the incident because they
thought it would be held against them or they would be blamed.

Feminist theory predicts that labeling of sexual assault experiences, type of force,
level of injury sustained during the assault, the relationship between the victim and
offender, sex role socialization, attribution of blame and costs and benefits of telling
others all contribute to women’s decisions to seek help after being sexually assaulted.
While the area is still new, the current research supports the contribution of these factors
to reporting behavior. In addition to the above-described factors predicted by the feminist
theory, social-cognitive theorists have hypothesized several other factors that may also
contribute to reporting behavior. The following section will examine the theory and

empirical evidence for the social-cognitive model.

Social-Cogm'tive Model of Factors Influencing Women’s’ Decisions to Report Sexual;

 

5&1}.

Researchers have used social cognitive explanations to examine factors contributing
to low reporting rates of sexual assault. The main model used is based on F ishbein’s
model of behavioral intentions (See Figure 1). F ishbein (1963) hypothesizes that the
intention to engage in a particular behavior is a function of two variables, an attitude
component and a normative component. The first variable, the attitude component,
consists of the person’s attitude toward performing that behavior. This variable is
hypothesized to be a function of the act’s perceived consequences and the evaluation of
these outcomes (Ajzen & F ishbein, 1972; Feldman-Summers & Ashworth, 1981) of the
model. The second variable is the normative component and is comprised of the sum of

the beliefs about what others expect the individual to do in that situation multiplied by the

27

 

 

indis'iduz
Summer.
predict b
1977). f
experien
intention
Summer:
the valid
assaults l

m
decision ;
and empi
ll'0an rr
unfavora
example,
b€lieved,

 

CW ne

likely to l

 

 

 

individual’s motivation to comply with the norms (Ajzen & F ishbein, 1972; Feldman-
Summers & Ashworth, 1981). Studies show that Fishbein's model has been able to
predict behavioral intentions quite well (Davidson & Jaccard, 1975; Ajzen & Fishbein,
1977). Previous research has examined low reporting rates of women who have
experienced sexual harassment and/or sexual assault using Fishbein's model of behavioral
intentions (F eldman-Summers & Ashworth, 1981; Brooks & Perot, 1991; Feldman-
Summers & Norris, 1984). The following section examines the empirical evidence of
the validity of Fishbein's model as a method of explaining low reporting of sexual
assaults by women.

Attitude Commnent. The attitude component is similar to the feminist rational-
decision making explanation of low reporting rates; however, it offers a more detailed
and empirically validated explanation of reporting behavior. Both theories postulate that
women may decide not to report because they believe the outcomes of reporting would be
unfavorable (F eldman—Summers & Ashworth, 1981; Gartrrer & Macmillan,l995). For
example, women may think that others would react negatively to them, they would not be
believed, police would not do anything about it, they would be blamed or feel ashamed,
or the offender may retaliate against them. In fact, research has shown that women who
expect negative outcomes such as not being believed or not being taken seriously are less
likely to report (Gunn & Minch, 1988; Bachman, 1993).

Normative Commnent. Fishbein's model of behavioral intentions suggests that social
pressure also contributes to low reporting rates of sexual assault. Women may be not

reporting because they are doing what they believe others would want them to do. They

28

may believe that their fiiends, family or significant others would not want them to tell
anyone about the incident (F eldman-Summers & Ashworth, 1981).

Empirical Evidence of Fishbein’s Model. Two studies examined the validity of
F ishbein's two factor model to predict women's decisions whether or not to report sexual
assault to anyone (F eldman-Summers & Ashworth, 1981; Feldman-Summers & Nonis,
1984). F eldman-Summers and Ashworth (1981) used a sample of non-assaulted women
fiom the community and examined the effectiveness of the model in predicting women’s
intention to report or not to report being sexually assaulted. Women were asked what
they would do if they were assaulted. Specifically, women reported their likelihood of
reporting a rape to police, a rape crisis center, a hospital or health clinic, personal doctor,
mental health professional, romantic partner, parents, sibling or other relative, closest
female fiiend and a member of the clergy. They also responded to a set of 24 items of
possible outcomes for each recipient and the evaluation of these outcomes. Women also
reported their perceptions of the expectations of romantic partner, close family members,
and close friends and how much they wanted to do what the person thought they should
do.

To assess the validity of the model to predict intentions, behavioral intentions to
report a rape were regressed onto the perceived outcomes and normative expectations, for
all participants, yielding a regression coefficient for each report recipient. All regression
coefficients were significant at p<.01, suggesting that that this model does account for
some of the variance in expected reporting behavior. The mean R2 was .27, suggesting a
relatively good fit of the model. They found that overall normative expectations and

perceived outcomes both predicted (with about the same weight) behavioral intentions to

29

 

report a r
of the cle
expectati
were detr
behavion
physiciar
addition.
report rec
rape crisi

women.

report a rape. They found that the fit of the model was weakest for reporting to a member
of the clergy or mental health worker were primarily influenced by normative
expectations, whereas intentions to report to a romantic partner or a rape crisis center
were determined primarily by perceived outcomes. They also found that age affected
behavioral intentions, older women were more likely to report to the police, clergy, and
physicians than younger women and less likely to tell parents or female friends. In
addition, intentions to report varied significantly by ethnicity for most of the potential
report recipients. Caucasian women were significantly more likely to tell the police, a
rape crisis center, or a romantic partner than Asian, Afiican-American or Hispanic
women. Afiican-American women were significantly more likely to tell a female fiiend
than were Asian, Hispanic, or Caucasian women. Hispanic women were more likely to
tell a clergy member than the other women. Asian women were significantly less likely
to tell a physician than the other women. Despite these differences, adding ethnicity and
age to the model improved the predictive validity of the model very little (largest increase
was .059 for female fiiend). They also examined the ability of the model to predict single
subject behavioral intentions and found that overall the model worked quite well. In
addition, they found that for 66% of the subjects the beta and r coefficients for the
normative components were larger than those for the perceived outcomes.

Subject by subject analysis also indicated that the type of perceived outcomes that
best predicted behavioral intentions varied according to ethnic group, while the normative
expectations tended to be the same. All ethnic groups were more likely to report if they
felt it “would result in my feeling calm, safe, and better having talked to someone about

the rape”. Other perceived outcomes that significantly predicted reporting varied by

30

ethnic group. For Caucasian women, the second best predictor was if they felt it would
result in them being given a test for pregnancy, VD, etc. For Asian and Hispanic women,
getting adequate medical attention was the second best predictor of reporting. For
African-American women, no other perceived outcomes were consistently able to predict
reporting.

While this study provided some evidence of the validity of Fishbein’s model for
predicting reporting behaviors, it had three major flaws. First, while research on
predicted behavior is useful, it does not necessarily imply that similar factors contribute
to actual reporting for real survivors. Therefore, it is important to examine the validity of
the model for women who have been sexually assaulted. Second, the model was only
tested for likelihood to report rape and failed to examine the validity of the model for a
wider range of sexual assault experiences, such as unwanted fondling, or attempted rape.
Finally, women were asked to respond to a generic question that directly asked if they
were raped what they would do, without specifying any details about the rape and/or the
rapist. Characteristics of the rapist and the offense have been shown to relate to people’s
attitudes about the reporting (e.g. Koss et. a1, 1987). By failing to specify these
characteristics, the authors can not control for these factors. Women may have been
responding to very different questions. For instance, one woman may think of being
raped by an acquaintance who intoxicated her and another may imagine being raped by a
stranger at gunpoint. These different scenarios may affect the ability of the model to
predict reporting and this study was unable to assess this.

One study has examined the validity of Fishbein’s model in predicting reporting

behavior of actual rape survivors (Feldman-Summers & Norris 1984). They recruited

31

subjects by means of public notices in community medical and mental health clinics,
clinics established for rape survivors, local newspapers and public service announcements
on radio and television in the Seattle metropolitan area. They examined factors
contributing to reporting to a social service agency or the police. They found that the best
predictor of group membership was the perceived outcome that reporting to a social
service agency would result in “being given the necessary tests to detect pregnancy, VD,
etc.” Discriminant analyses indicated that women who told someone at a social service
agency agreed more strongly with this statement than woman who did not tell someone
(F =1 1.06, F58, p<.001). In addition, women who received help at a social service
agency were more likely to believe that reporting would make them feel “calm, safe, and
better after having talked to someone about the rape” (F =5.66, F.41, p<.02) and reporters
indicated that family, fiiends and romantic partner wanted them to report to a social
service agency more than those for non-reporters did (F =8.12, r=.50, P<.006). For police
reporting, the strongest predictor of group membership was the social expectation factor.
Women who reported to the police indicated stronger social pressures to report than did
women who did not report (F =21 .53, r=.69, p<.001). Two perceived outcomes also
differed significantly between reporters and non-reporters. Reporters were more likely to
believe that reporting would result in being treated in a positive way (F =10.05, r=-.47,
p<.02) and less likely to believe that it would result in a trial that the victim would have
to testify at (F =5.44, r=.34, p<.02) than women who did not report. While this study did
examine reporting in actual rape survivors, it also failed to examine reporting among
other sexual assault survivors or control of characteristics of the assault and the rapist.

These flaws make it difficult to assess the validity of the model for all types of sexual

32

assault survivors and to control for other factors that contribute to reporting. This study
attempts to address all of these flaws by studying the efficacy of this model to predict
reporting by actual sexual assault survivors, for a wide range of sexual assault and also
assesses other factors related to the assault.

Combined Model of Rep_orting Behavior

To date, no study has examined or proposed a comprehensive model of reporting
behaviors. This study used both feminist and social-cognitive theories to study a
comprehensive model of reporting behaviors in sexual assault survivors (See Figure 2).
The research and theoretical explanations of each factor in the combined model were
extensively discussed earlier in this thesis. The following section will briefly review these
findings and discuss the combined model.

Feminist theory postulates that low reporting rates result from two main processes;
denial or minimization of the behavior and weighing the costs and benefits of telling
anyone (Gartner & Macmillan, 1995). Most of the feminist theoretical explanations and
empirical studies focus on the first process, the denial or minimization of the behavior.
As discussed in this review, the inability to acknowledge the assault as a crime is related
to stereotypical beliefs about rape, sex roles, and intimate relationships.

Researchers speculate that women whose experiences do not meet the stereotypical
rape scenario will be less likely to define their experiences as assaultive (e. g. Orcutt &
Faison, 1988). In fact, research demonstrates that when more force is used and when
injuries are sustained, women and the general public are more likely to label the
experience as abusive (e.g. Orcutt & Fiason, 1988; Gunn & Minch, 1988). Research has

also supported a link between reporting and injuries, with more injuries related to a

33

greater likelihood of reporting. It is unclear if the relationship between injuries and
reporting is direct or if this relationship is moderated by women’s definitions of their
experiences. As a result, this study examined both an indirect and direct path between
injuries and reporting.

Research has not examined the relationship between type of force (physical,
alcohol/drugs, or psychological) and acknowledgment of crime. However, because the
stereotypical rape does not involve the use of alcohol or drugs, or psychological coercion
it makes sense to conjecture that they would be associated with a decreased likelihood of
labeling the experience as abusive. As a result, the combined model contained a link
between type of force and injury and acknowledgment of the experience as an assault.

Stereotypical rape rarely occurs between intimates. This in addition to women’s
greater investment in not labeling the experience as abusive (Weis & Borges, 1973), led
feminist theorists to postulate that level of intimacy would be negatively correlated with
labeling sexual assault experiences as abusive (Gartner & Macmillan, 1995). Indeed,
research demonstrates that women raped by acquaintances are less likely to label their
experience as rape (e. g. Koss et al., 1988). Since current research supports the link
between acknowledgment of sexual assault as abusive and the relationship between the
women and the offender, this factor was tested in this study.

Feminist theory also hypothesizes that women’s sex role socialization affects how
women label their experiences of sexual assault because traditional sex role expectations
teach rape-supportive beliefs (e.g. Check & Malamuth, 1983). Traditional dating norms
teach that men are the pursuers of sexual contact and it is the woman’s responsibility to

avoid sexual contact (Check & Malamuth, 1983). Sexual assault may be seen as within

34

the norm:
sex nomu
1983). In
were mor

& Thoma

 

 

the normal realm of dating behavior. In fact, research has demonstrated that traditional
sex norms are related to acceptance of rape in certain conditions (Check & Malamuth,
1983). In addition, research has shown that individuals with more egalitarian attitudes
were more likely to label sexual assault experiences as violent (Muehlanhard, Friedman,
& Thomas, 1985). Both theoretical arguments and research suggest that there is a
relationship between sex-role socialization and labeling the incident as an assault.
Therefore, the combined model contained a link between sex role stereotypes and
acknowledgment of crime.

Research with sex role socialization indicates that women with more egalitarian
attitudes are less likely to blame the victim (Muehlanhard, Friedman & Thomas, 1985).
Therefore, this model hypothesized a link between sex role socialization and attribution
of blame. In addition to this relationship, research has demonstrated a relationship
between attribution of blame and reporting the incident (Gunn & Minch, 1988). As a
result, the combined model also contained a link between attribution of blame and telling
anyone.

Feminist theory suggests that women do not report sexual assault when they do not
define their experiences as assaultive. Research demonstrates that women who do not
label their experiences as assaultive are less likely to tell anyone about their experiences
(e.g. Koss, 1985). As a result, the combined model will also contain a link between
definition of one’s experiences (acknowledgment of crime) and reporting behavior.

The feminist theory and empirical evidence is less developed for the second half of
the feminist model, the costs and benefits appraisal. However, the research reviewed

previously suggests that this process may also be valid. Social-cognitive theorists have

35

 

also expl
telling o
expectatit
expectatit
decisions
others do
experienc
cognitive
1998). 8;
less likely
found a rt
(TCldmar
The 5
second p1
0f low re
"report:

exPettati

also explained low reporting rates as related to an appraisal of the costs and benefits of
telling others. As discussed earlier, social-cognitive theorists believe that the
expectations of outcomes (i.e., what would happen if they told) in addition to normative
expectations (i.e., what women believe others would want them to do) relate to women’s
decisions to report sexual assault. Women who expect negative outcomes and/or believe
others do not want them to tell are expected to be less likely to report sexual assault
experiences. Research with rape survivors has found support for the validity of the social
cognitive model (F eldman-Summers & Norris, 1984; Gunn & Minch, 1988; Bachman,
1998). Specifically, studies have found that women who expect negative outcomes are
less likely to report sexual assaults (Gunn & Minch, 1988). In addition, studies have
found a relationship between believing others want you to report and actual reporting

(F eldman-Summers & Norris).

The social-cognitive explanation is a more detailed and elaborate explanation of the
second process proposed by feminist researchers. As a result, the social-cognitive model
of low reporting will be added to the feminist model, to result in a comprehensive model
of reporting sexual assault. Specifically, a link between reporting and normative
expectations and outcome expectations was maintained in the combined model.

In sum, the combined model examined all of the associations supported by the
feminist and social-cognitive theories and research. While the literature has examined
many of these factors separately it has failed to comprehensively study factors related to
reporting sexual assault. An assessment of a comprehensive model of reporting

behaviors would be a significant contribution to the field.

36

Conclusions

The problem of sexual assault among college women is a serious and pervasive
problem. Sexual assault is a problem that affects 15 to 47 percent of college women and
has serious mental health consequences (Muehlenhard & Linton, 1987; Ogeltree, 1993).
Despite the fact that sexual assault survivors experience anxiety, depression, sexual
dysfunction and a number of other serious problems (see Resick, 1993), many women do
not seek out any help or tell others about their experiences. Not reporting sexual assault
has many negative consequences. First, if women do not report an accurate
understanding of the prevalence of sexual assault and the impacts of rape on women can
not be obtained. Prevalence rates may affect funding for rape prevention programs and
rape crisis centers. In addition, if women do not tell others they will not be able to get the
support that they need to help them cope with their experience. Last, by not reporting the
offender is not punished. In conclusion, not reporting sexual assault has many negative
consequences for both the survivor and the larger community. Therefore, it is important
to examine the reasons why women do not tell others about their experiences of sexual
assault.

This literature review discussed two main models for reporting and indicated that
three main issues needed further exploration. First, while the literature has examined
many of these factors separately it has failed to comprehensively study factors related to
reporting sexual assault. An assessment of a comprehensive model of reporting
behaviors would be a significant contribution to the field. Second, much of the research
on factors contributing to reporting have focused on the general public and have not

thoroughly examined the relationships with actual assault survivors. Third, many of the

37

factors have been examined only in rape survivors and not with women who have
experienced other types of sexual assault. Examining sexual assault survivors who have
experienced varying levels of abuse will be an important addition to the current research

on reporting behaviors.

38

HYPOTHESES
Hmtheses for Both Assaulted and Non-Assaulted Women

Hymthesis 1. It is hypothesized that blaming oneself for the experience will be
associated with a decreased likelihood of reporting (or expecting to report) the incident to
anyone.

Hypothesis 2. There will be a positive relationship between sex role stereotypes
and reporting; specifically a more egalitarian attitude about women is expected to be
related to an increased chance that the woman will discuss her experience with others.

Hymthesis 3. An association between sex role stereotypes and attribution of
blame is expected; specifically a more egalitarian attitude toward women is hypothesized
to be related to an increased chance that the woman will blame the man more than she
blames herself.

Hymthesis 4. It is hypothesized that there will be an association between
normative expectations and reporting behavior. Women who believe others want them to
talk about their experiences (i.e., high normative belief scores) will be more likely to
report the incident to someone.

Hmthesis 5. It is hypothesized that there is an association between outcome
expectations and reporting behavior. More specifically, it is believed that increasingly
positive expectations of reporting will be associated with greater likelihood of reporting.
Hymtheses for Assaulted Women

Hypgthesis 6. It is expected that there will be an association between level of
physical injury and reporting; specifically, the more severe the injury sustained during the

assault, the greater likelihood the woman reported the incident. This relationship may be

39

direct or may be moderated by the relationship between level of injury and
acknowledgment. Both relationships will be examined in this study.

Hyppthesis 7. An association between type of force and the label of the incident
is expected. Specifically, use of force will be associated with a increased likelihood of
acknowledging the experience as assaultive.

Hypothesis 8. It is hypothesized that there will be an association between the type
of relationship and labeling the experience as an assault. Increased intimacy is expected
to be related to decreased likelihood of acknowledgment of the experience as assaultive.

Hyppthesis 9. It is hypothesized that there will be an association between sex role
stereotypes and the label of the incident. Women who have more egalitarian attitudes
toward women are expected to be more likely to label their experiences as sexual assaults.

Hyppthesis 10. It is hypothesized that there will be an association between blame
and label of incident, with women who blame themselves being less likely to label their
experience as an assault.

Hymthesis 11. It is hypothesized that acknowledgment of the experience as
assaultive will be positively related to reporting the incident to someone.

Hmthesis 12. A reduced model (Figure 3) will be tested for both sexually
assaulted women and non-assaulted women and it is hypothesized that the model fit will
be better for actual assault victims than for women who have not been assaulted.

Overall Fit

Last, the overall fit of each model (reduced for A and B, total for A) will be

examined. If model fit is not excellent, additional correlations may be added, or indirect

links deleted to attempt to improve the overall fit of the model.

40

Part1 r:

of screw
differen
focused
using a
finaneie
sample
compen
reason i
assault 1
Burnam
risk for
Wisniev
serious 2
w'ctrshat
mkfla
for age, 1'
l
mmmYO
Year of 0

inthfilt ft

METHODS

Participapts

The sample consisted of 397 women, ages 17-23, of whom 195 reported a history
of sexual assault (sample A, 49% of total sample), and 202 did not (sample B). Since
different factors may contribute to reporting based on the age of the individual, this study
focused on one age group. A college population was chosen for two main reasons. First,
using a college sample to study a relatively new field can be beneficial because of
financial and time constraints involved in assessing a large number of women. A college
sample can be expected to be more readily accessible than other groups and can be easily
compensated by non-fmancial means for their time. The second and more important
reason is that prior research demonstrates that college is a high-risk period of sexual
assault for women (Patton & Mannison; 1995; Sorenson, Stein, Siegel, Golding &
Bumam, 1987, Warshaw, 1988). College women are at approximately three times greater
risk for sexual assault than women in the general population (Koss, Gidycz, &
Wisniewski, 1987). Research demonstrates that sexual assault on college campuses is a
serious and pervasive problem (Gavey, 1991, Koss et. al., 1987; Ogeltree, 1993;
Warshaw, 1988). Since this study examines a relatively new area of research, the ability
to test a large number of women who have experienced sexual assault, while controlling
for age, is important.

The mean age of the women was 19.3 years old (sd=l .2). The sample consisted
mainly of students in the first two years of college: 44% of the women were in their first
year of college, 27% were in their second year, 16% were in their third year, 11 % were

in their fourth year, and 2% were in their fifth year or beyond. Eighty-three percent of the

41

sample was Caucasian, 11% were Hispanic, 9% were Biracial, 7% were Afiican-
American, 3% were Asian, and 1% were Native-American. Slightly more than half
(55.4%) of the women were in sororities. The average income of the participants’ parents
was in the 60,000 to 69,999 range (sd=20,000). Analyses of differences between samples
on demographic variables revealed one significant difference between the two samples,
with sample A (assaulted women) having significantly fewer women in sororities (43.3%)
than sample B (non-assaulted women) (56.8%, t=2.72, p<.01).
Procedures

Participants were recruited in one of two ways. First, women who were taking
Introductory Psychology classes in the Fall of 1997 and Spring of 1998 were asked to
participate in the study in order to receive research credits that are required for their class.
A large percentage of introductory psychology students are first year students, so in
order to get a more balanced percentage of female participants from each year in college,
sororities were asked to participate. Research indicates that sorority women experience
high levels of sexual victimization; although findings about whether their experiences are
significantly different fi'om nonsorority women are equivocal (Copenhaver & Grauerholz;
Kalof, 1993; Kirpatrick & Kanin, 1957). Since the main purpose of this study was to
study sexual assault victims, it was beneficial to use samples that may have high rates of
sexual assault.

At a monthly sorority council meeting, all of the sororities were asked to consider
participating and were offered a sexual assault workshop that was not contingent upon
participation in this study. Ten sorority representatives agreed to consider participating,

however only 6 sorority presidents agreed to participate.

42

A
State Uni
mmma
30 lIllIlLllt
explainec
In additic
All partit
could atte
Hotlines .
experienc
researche
of help at
assault in
about thei

cOntact th

quesllOHS.

 

All participants answered questionnaires in small testing groups on the Michigan
State University campus or at sorority houses. All participants were given a consent form
prior to answering the questions (see Appendix C). The entire testing took approximately
30 minutes. On completing the assessment, all participants were given a handout that
explained the study and listed resources for survivors of sexual assault (See Appendix D).
In addition, participants enrolled in Introduction to Psychology received research credits.
All participants were informed of times for workshops on Sexual Violence that they
could attend. In addition, all participants were given the numbers of Sexual Assault
Hotlines and Counseling Services in case they wanted to discuss any sexual assault
experiences they may have had. Due to the sensitive nature of these questionnaires, the
researcher was available and prepared to provide support and resources to women in need
of help at the end of each testing session. Several women wanted to talk about sexual
assault in general after completing the questionnaire, however only ten women spoke
about their own experiences. Finally, all participants were instructed that they could
contact the researcher at the Michigan State Psychological Clinic if they had any
questions.
Measures

Demoggphic Questionnaire. This instrument was developed for this study and
consisted of 5 items (See Appendix A, questions 1 through 5). It asked the women to
identify their age, ethnicity, education level, if they were in a sorority, and their parents’
income level. As noted earlier, the only difference in demographic variables between the
samples was the percentage of women in sororities. There were no significant

relationships between demographic variables and number of people told in either sample.

43

tum
mawm
oflender

mDUa

h

 

Within Sample A (assaulted women), there were significant relationships between being
in a sorority and acknowledgment of the incident as an assault and relationship to
offender, with women in sororities more likely to be closer with the perpetrator (tr—2.75,

p<.01) and less likely to acknowledge the experience as an assault (F31, p<.05).

Measure of Sexupl Assault Exmriences, Critical Incideng, Label of Incident, and

 

 

Typp of Force. A modified version of the Sexual Experiences Survey (SES: Koss &
Oros, 1982) was used to measure participants’ experiences with sexual abuse (See
Appendix A, questions 6 through 17). All of the questions were in a yes-no format, asked
about different types of abuse and assessed the type of coercion used (i.e. threat of
physical force, psychological coercion, or use of physical force). A l4-item adapted
version of the SES was used in this study. Twelve of the items used in this study came
fi'om Patton & Mannison’s (1995) adapted version of the SES. They adapted the original
SES to assess a broader range of coercive behaviors within both the category of sexual
intercourse and sex play (fondling, kissing, or petting). This study also used three
additional items. The first item was in the original form and asked directly if the
individual had been raped. This item was used to assess whether women who have
experienced rape perceived their experience as rape. The second additional item asked
women directly if they had been sexually assaulted by anyone. This item was used to
assess if women who were sexually assaulted but not raped defined their experience as an
assault. The third item asked women who had more than one sexual assault experience
to identify which experience had been most upsetting for them (i.e., the critical incident).
Each woman was asked to identify what experience she found most upsetting both as a

way of empowering the woman and to help identify what types of experiences different

44

women
could g
accurat
the iten

about a

particip:
of sexua
assault b

P
their lZ-i
intercour
have beet
Validity a
1987; Rd

1986).

 

K
I
the Origir
Scale Wag
psl'CllOIO
'74 fort

113ng a S‘

Indilldn

women find most upsetting. This method was chosen so that actual woman’s experiences
could guide the research, instead of relying upon research or literature which may not
accurately reflect these women’s experiences. Women were asked to complete the rest of
the items with this incident in mind, so that each women could answer detailed questions
about a specific incident.

This study specifically examined sexual assault perpetrated by men and
participants were informed of this. Women were told they could write about experiences
of sexual assault by a women, however the formal questions referred only to sexual
assault by males and no participants wrote about assault by females.

Patton & Mannison (1995) conducted reliability analysis (Cronbach alpha) on
their 12-item version of the SES and found that sex play items had an alpha of 0.61 and
intercourse items had an alpha of 0.71. No other additional reliability or validity studies
have been conducted on the twelve-item version; however, studies have examined the
validity and reliability of the original version (Gwartney-Gibbs, Stockard, & Bohner,
1987; Koss, 1985; Koss & Gidycz, 1985; Koss, Gidycz, & Wisnieski, 1987; Malamuth,
1986)

K033 and Gidycz (1985) conducted a study assessing the reliability and validity of
the original version of the Sexual Experiences Survey. The internal consistency of the
scale was tested using a sample of 305 women and 143 men enrolled in an introductory
psychology course at a Midwestern university. The internal consistency of the items was
.74 for females and .89 for men (Cronbach alpha). Test-retest reliability was assessed
using a sample of 71 females and 67 males from the introductory psychology classes.

Individuals completed the questionnaire in a group setting on two occasions that were one

45

week apt
regards tt
hide ran;
studies (t
\l'isniesk

R

SES den
66 throug
severe ind
about that
assessed

at the tim
Person. I

relationsh
they blam
self-blam.
rePotting
lh'enrt im
health “.01
abollt the (
they: Suff-c1
that item

 

week apart. The mean item agreement for the two testing session was 93%. With
regards to validity, the Sexual Experiences Survey has been found to be associated with a
wide range of both attitudinal and behavioral measures in laboratory and naturalistic
studies (Gwartney-Gibbs, Stockard, & Bohner, 1987; Koss, 1985; Koss, Gidycz, &
Wisnieski, 1987; Malamuth, 1986).

Remrting Behavior, Level of Injm, Relationship with Assailang, Intimacy, and
Attributipn of Blame. These factors were assessed by additional items at the end of the
SES developed for this study (See Appendix A, questions 18 through 23 and questions
66 through 68). Participants who reported any incident were asked to pick the most
severe incident, tell their age at the time of the incident and answer additional questions
about that experience. First, the relationship between the women and the offender was
assessed by asking the women to identify the nature of their relationship with the person
at the time of the incident (nonromantic fiiend, casually dating, exclusively dating this
person, long term relationship). This was scored so that higher numbers reflected closer
relationships. Second, attribution of blame was assessed by asking women how much
they blamed themselves and the man for the incident (4 point Likert scales). Scores for
self-blame were reverse scored, so higher numbers equaled less self-blame. In addition,
reporting behavior was assessed by asking women to identify who they told [no one,
friend, immediate family member, extended family member, resident assistant, mental
health worker (counselor/flrerapist/rape crisis worker), doctor, police, clergy or other]
about the incident (yes or no). Finally, women were asked the degree of physical injuries
they suffered as a result of the incident, with higher scores equaling more injuries. The

final item has been used by previous researchers (e.g. Parisian, 1990).

46

Vt
play or in
themselst
man used
been assa
(sex play
necessary
women ts
confident
should be
1111111an t
length of
y:
This ques
Rimming.

Contains i

 

panlClpar
assaulted

exlit‘cted

I
”36- Thil

friends. at
Parentsy (1

”items

 

Women who did not report the occurrence of an incident at either level (i.e. sex
play or intercourse) were asked who they would tell, and how much they would blame
themselves or someone else if they engaged in sex play or sexual intercourse because a
man used physical force. Although the main focus of the study is on women who have
been assaulted, women who did not report the occurrence of an incident at either level
(sex play or intercourse) were asked questions for two main reasons. First, this data is
necessary to test the difference in fit of the model for victims and non-victims. Second,
women were filling out the questionnaires in small groups and in order to ensure
confidentiality of their answers, women who have and have not been sexually assaulted
should be completing similar numbers of questions. By asking all women similar
numbers of questions, women who have been assaulted could not be identified by the
length of time they spent answering the questions.

Normative Exp_ectations and Motivations to Comply with these Exmctations.
This questionnaire was developed for this study using the Normative Expectation for
Reporting Questionnaire (Espinosa & Cardoza, 1995) as a foundation. This version
contains 27 questions (Appendix A, questions 24 through 50 and questions 69 to 86). All
participants answered these questions regardless of whether or not they had been sexually
assaulted. The original questionnaire asked women to estimate the level of support they
expected to receive from parents, fiiends, and community if they were to report such a
case. This study examined who people believed that others (i.e., immediate family,
fiiends, and community) would want them to tell about an experience of sexual (i.e.
parents, friends, and community). Women were asked the likelihood that each of three

referents (immediate family, fiiends, and community) would want them to report their

47

experier
recipien‘
mental 1
Women
(4) very

referent'

(normati
commun
family in
clergy m
above .4(
for the CC
Stores f0(
items mu
CtOnbach
commmu'
he Percei
qUCStIOng

Suggested

 

 

Studies. y

 

experience of sexual assault (or for non-victims, future assault) to each of eight report
recipients [friend, immediate family member, extended family member, resident assistant,
mental health worker (counselor/therapist/rape crisis worker), doctor, police, or clergy].
Women were asked to respond using a 4-point Likert scale, ranging from (1) not at all to
(4) very much. In addition, they were asked how much they wanted to comply with each
referent's opinion (4- point Likert Scale).

Exploratory factor analyses (EF A) suggested that there were three factors
(normative beliefs for family, normative beliefs for friends, and normative beliefs for
community). The analyses also suggested that items about whether or not friends and
family members want you to tell a resident advisor, an extended family member, or a
clergy member (items 26, 27, 31, 35, 36, 40) should be deleted as they loaded highly (i.e.
above .40) on two factors. To maintain the consistency of the three factors similar items
for the community referent were deleted (i.e., items 44, 45, and 49). Normative beliefs
scores for family, fiiends and community were compiled by summing the remaining
items multiplied by the individual’s motivation to comply with the each referent.
Cronbach’s alpha reliability for the three normative belief factors (family, fiiend,
community) were .92, .93, and .96, respectively.

Perceived Mares (Eeldman-Summers and Ashworth (1981)). A modified version of
the Perceived Outcomes scale was used (See Appendix A, questions 51 through 65 and
questions 96-111) in this study. F eldman-Summers (personal communication, 1997)
suggested using only the 7 items that were found to relate to intentions to report a rape in
their original study and that have also been found to relate to intentions to report in other

studies. These items include: "would result in my feeling calm, safe and better having

48

talked to someone about the experience", "would result in adequate medical attention",
"would result in my being given the necessary tests to detect pregnancy, VD, etc.”, "a
trial in which I would have to testify", "gathering the necessary evidence that could be
used in court", "nothing being done to help me", "my being treated as an immoral
person", and "retaliation by the offender.” Women were also asked if there was anything
else they thought would happen if they told someone. Women were asked about their
perceptions of these 8 outcomes of reporting to each of eight possible recipients [friend,
immediate family member, extended family member, resident assistant, mental health
worker (counselor/therapist/rape crisis worker), police, or clergy]. In addition, women
were asked to rate how good or bad each of these outcomes is on a 6 point Likert scale
(very bad to very good). The items were developed using open-ended interviews with
about fifty women from four different ethnic groups (African-American, Asian,
Caucasian and Hispanic) to identify possible outcomes women may associate with telling
someone about sexual assault (F eldman-Summers & Ashworth, 1981). No validity or
reliability studies have been done on these items, however using a. sample of 400 women
they were found to be related to potential victims intentions to report a rape (Feldman-
Summers and Ashworth, 1981). A total perceived outcomes score for each report
recipient was computed by multiplying each anticipated outcome response by the
corresponding outcome evaluation response. Higher scores were reflective of more

positive expectations for reporting.

49

(I)
(b
>4
W
O

l

was or
used It
was cr
behay
femin
to 3 I
Egali
Scort

The:

com
itch
pro;

Orig

has
353.
C011.

Usir

Sex Role Socialization.

The Attitudes Toward Women Scale (AWS; Sgnce & Helmreich, 1978) . This
was one of two scales used to assess sex role socialization. It is a 15-item questionnaire
used to assess gender-role beliefs (See Appendix A. questions 112-126 through 76). It
was created to assess individual's beliefs about the responsibilities, privileges and
behaviors in a variety of areas that have historically been classified as masculine or
feminine (Spence & Hahn, 1997). Women are asked how much they agree or disagree (0
to 3 Likert scale) with 15 statements regarding women's rights and roles in society.
Egalitarian items are reverse-scored and all scores are summed to obtain a total score.
Scores can range fi'om 0 to 45, with higher score indicating more egalitarian attitudes.
The original version (Spence & Helmreich, 1972) contained 55 items, several of which
were based on items fi'om a measure of beliefs toward feminism created by Kirkpatrick
(1 93 6). Briefer versions of the scale were created so that the instrument could be
completed in a shorter time (Spence & Hahn, 1997). The brief version consists of 15
items selected from the original version on the merit of their superior psychometric
properties (Spence & Helmreich, 1978). The brief version is highly correlated with the
original version and since its development most researchers in this area have preferred it
(Spence & Hahn, 1997).

Some researchers have suggested that the content of gender-role attitudes might
have changed in the last two decades, making the items on the AWS less useful in
assessing current attitudes towards women's rights and roles. In order to address these
concerns, Spence and Hahn (1997) examined the psychometric properties of the AWS

using a sample from 1992. In addition, over 2000 men and women in four different

50

coho
fiftee
at At
were
who:
won
by 4

3550

and

used

AW
& H
Res
factt
indit
factt
factt
Won
l'ariz

Subs

cohorts (1972, 1976, 1980, and 1992) were compared. All students completed the
fifteen-item form and were enrolled in introductory psychology at the University of Texas
at Austin when they completed the form. They found that the means of the later cohorts
were higher than the earlier cohorts with the exception of women in the 1980 sample,
whose mean was slightly below the women in the 1976 sample. In addition, means for
women were consistently higher than means for males in the same cohort. A 2 (gender)
by 4(cohort) Analysis of variance (AN OVA) indicated that the mean differences
associated with gender (F (1, 2265)=228.40) and with cohort

(F (3, 2265)=173.62) were both significant at p<.0001. The interaction between cohort
and gender was not significant (F (3, 2265)=2.01, p<.10). The mean of the 1992
(x=36.34, sd=6.10) was not significantly different fi'om the mean of the combined sample
used in this study (x=36.54, sd=4.44).

Earlier studies in the 1970's and 1980's had found that the 15 item version of the

AWS was unifactorial (e.g. Smith & Bradley, 1980; Spence & Helmreich, 1978). Spence
& Hahn (1997) found that for the 1992 sample the 15-item version also was unifactorial.
Responses of men and women were analyzed separately using an unrotated principal
factor analyses. For men and women only one factor had an eigenvalue greater than 1.00,
indicating that a single factor solution was appropriate for both groups. For women the
factor had an eigenvalue of 3.81 and for men an eigenvalue of 4.59. With men, the one
factor accounted for 89% of the common variance and 31% of the total variance. For
women, the factor accounted for 86% of the common variance and 25% of the total
variance. Cronbach alphas for both men (.84) and women (.81) in the 1992 sample were

substantial. These results suggest that the AWS is still a reliable and valid measure of

51

attitudes toward women. Reliability analyses for this sample generated a Cronbach’s
alpha of .69.

Sex Role Stereotyping Qpestionnaire (SRSQ;B% 1980). This is a nine-item
questionnaire that was created to assess beliefs about the roles of women (See Appendix
A, questions 126-133). Participants are asked to indicate how much they agree with each
statement using a 7 point Likert Scale. Egalitarian items are reverse-scored and all scores
are summed to obtain a total score. Scores can range from 0 to 63, with higher scores
indicating more traditional attitudes. Burt (1980) studied the reliability of the scale using
598 adults aged 18 and older and found the internal consistency of the items was .80
(Cronbach alpha). Cronbach’s alpha for this study was lower (.59). The mean for the

1980 sample was 37.6 (sd=10.5), while the mean for this sample was 40.25 (sd=6.9).

52

RESULTS

The following section is separated into four main areas. For each area, analyses
will be presented separately for both assaulted and non-assaulted women (Samples A and
B). The first area includes a discussion of the frequencies and means of the observed
variables. This information is provided to inform the reader of the variances in each
construct and the nature and type of sexual assault experiences reported in this sample.
Second, correlations between the observed variables will be reported, as these
correlations are the basis for structural equation modeling. Third, the measurement
models will be reviewed. The results from the measurement model will be incorporated
into the last section, which includes the results of the hypotheses, and the structural
models.

Frequencies and Means of Variables

Frequencies of the types of assaults that women reported are in Table 1. Most of
the women who were assaulted endorsed more than one type of assault. Specifically, in
Sample A , 17% of the women endorsed one assault, 31.4% endorsed two types of
assault, 16% endorsed three, 15% endorsed four, 6% endorsed 5, and 15% endorsed six
or more types of assault. Women were asked to pick the assault that was most upsetting
for them and to complete the rest of the questionnaire about that incident (see Table 1 for
frequencies). Stereotypical views would suggest that rape using physical force would be
the most upsetting experience; however, several women who were raped by physical
force did not choose this as their most upsetting experience. To help identify
characteristics of the different types of sexual assault experiences, cross-tabulations

are presented in Tables 2 and 3. Table 2 reports the frequencies of different unwanted

53

Table 1. Types of Assaults Endorsed and Used as Critical Incidents.

 

 

Type of Sexual Assault
Type of Force Unwanted sex play Unwanted
Sexual
Intercourse
Verbal Coercion
#of women endorsing this item 141 106
# used as critical incident 25 40
Alcohol or Drugs
#of women endorsing this item 143 102
# used as critical incident 33 52
Threatening Physical Force
#of women endorsing this item 27 16
# used as critical incident 3 2
Use of Physical Force
#of women endorsing this item 45 36
12 28

# used as critical incident

 

* "‘ Critical incident refers to the experience that women chose to answer questions 1964
about.

54

SCI

1101

do

56X

ext

acl

Ta

di:

tht

3d

sex play experiences. Notably, the most common occurrences included women who did
not acknowledge their experiences of being sexual assaulted using verbal coercion or by
alcohol and drugs. There were no modal experiences of type of relationship and type of
sexual assault.

Table 3 focuses on experiences of rape Specifically, results indicate that these
experiences occurred in at least 10% of the women. These include women who did not
acknowledge their experiences of being forced to have sex through verbal coercion as
rape; women who did not acknowledge their experience of being raped while under the
influence of alcohol and drugs; and women who were raped while they were under the
influence of alcohol or drugs by someone they were exclusively dating.

Table 4 contains the mean, standard deviation and range of the observed variables
in Sample A. Table 5 lists the mean, standard deviation, and range of the observed
variables in Sample B. The following section reports on this information and includes a
discussion of the frequencies of observed variables.

Remrting: Women who were not assaulted believed they would tell significantly more
people than women who were assaulted actually told (t =-15.84, p<.01). Among women
who were sexually assaulted (Sample A), 20 percent told no one, 69 percent told one
person, 18 percent told two people, 2 percent told three people, 1 percent told four people
and .5 percent told five people. Most women who did talk about their sexual assault
chose to tell a friend (76%) and few women told anyone else. Only 7% told an
immediate family member, 5% told a mental health worker, 2% told their doctor, 1% told
the police, 1% told a clergy member, and 1% told an extended family member. In

addition, 5% told someone with whom their relationship was left unspecified.

55

 

Variable
Acknou
No
Yes
Injury
None
Mild
Mod .
Intimac"
None
Acqt
Non.
Casc
Excl

L011 y

 

My
responc

 

Table 2. Cross-Tabulations of Frequencies of Unwanted Sex Play experiences and
Characteristics of the Assault *

 

 

Type of F orce Used
Verbal Alcohol or Threatening Use of
Coercion Drugs Physical Physical
Force Force
Variable
Acknowledge
No 10% 11% 1% 4%
Yes 3% 6% 1% 2%
Injury
None 12% 15% 1% 3%
Mild 1% 2% 1% 3%
Moderate 1%
Intimacy
None 2%
Acquaintance 3% 1% 1% 1%
Non-romantic Friend 4% 4% 2%
Casually dating 2% 5% 1% 2%
Exclusively Dating 2% 5% 2%
Long-Term Relationship 2% 3% 1% 1%

 

 

*Numbers represent percentages of the entire sample of sexually assaulted women who
responded yes to that cell.

56

Table 3. Cross-Tabulations of Frequencies of Unwanted Sexual Intercourse and
Characteristics of the Assault *

 

 

Type of Force Used
Verbal Alcohol or Threatening Use of
Coercion Drugs Physical Physical
Force Force
Variable
Acknowledge
No 12% 19% 1% 3%
Yes 9% 8% 1% 11%
Injury
None 17% 20% 1% 6%
Mild 4% 7% 1% 7%
Moderate 1% 1%
Intimacy
None 5% 1% 1% 2%
Acquaintance 5% 2% 3%
Non-romantic Friend 6% 5% 1% 4%
Casually dating 1% 4% 1%
Exclusively Dating 4% 11% 5%
Long-term Relationship 5% 1%

 

 

*Numbers represent percentages of the entire sample of sexually assaulted women who

responded yes to that cell.

57

Table 4. Means, standard deviations and range of observed variables in assaulted women

 

 

(Sample A).

Observed Variable Mean Standard Deviation Minimum Maximum
Blame Self 2.60 0.86 1 4
Blame Him 3.43 0.79 1 4
AWS 50.76 4.27 35 59
SRSQ 39.26 6.89 24 56
Injury 1.27 0.49 1 3
Reporting 0.91 0.56 0 2
Norm Friend 51.94 1.45 16 80
Norm Family 52.24 17.31 14 80
Norm Community 40.12 19.94 5 80
Perceived Outcomes 218.32 75.80 58 384

 

58

Table 5. Means, standard deviations and range of observed variables in non-assaulted

women (Sample B).

 

 

Observed Variable Mean Standard Deviation Minimum Maximum
Blame Self 2.58 0.85 1 4

Blame Him 3.43 0.78 1 4

AWS 52.33 4.49 22 38

SRSQ 41.19 6.86 22 56

Label of Incident 0.91 0.55 0 2

Norm Friend 44.88 25.10 8 128

Norm Family 40.99 23.16 8 128

Norm Community 36.01 23.62 8 123
Perceived Outcomes 121.88 60.76 24 343

 

59

In the non-assaulted sample (Sample B), 7% believed they would tell no one, 8%
planned to tell one person, 27% believed they would tell two people, 16% three people,
15% four people, 15% five people, 8% six people, 2% seven people, 2% eight people,
and 2% believed they would tell nine people. Most women (81%). believed they would
tell a fiiend, 52% an immediate family member, 41.5% the police, 35% a doctor, 19.5% a
mental health professional, 9% an extended family member, 7.5% a residential advisor,
and 4% believed they would tell a clergy member. In sum, while most survivors of
sexual assault only told a fiiend, women who were not assaulted believed they would tell
a much larger variety of people.
mpg Women were asked how much they were responsible for, and how much the man
was responsible for the assault. In Sample A, 2% stated the man was not at all
responsible, 13% stated the man was a little responsible, and 26% stated the man was
somewhat responsible. However the majority of women (60%) stated that man was
responsible a lot. Only 12% felt they were not all responsible for what happened, 47%
felt hey were a little responsible, 29% felt they were somewhat responsible and 12% felt
they were responsible a lot for what happened.

In Sample B, 1% believed the man would not be responsible for the rape, 5%
believed he would be somewhat responsible, and unlike Sample A most women (94%)
believed the man would be very responsible for the rape. Of the women in Sample B,
53% believed they would not be at all responsible, 33% believed they would be a little
Responsible, 13% believed they would be somewhat responsible, and 1% believed they

would be very responsible.

60

Exppgtations for Repprting: In Sample A (assaulted women), when asked about possible
expectations for telling someone, 87% believed they would feel calm, safe and better,
37% felt they would get adequate medical attention, 42% felt they would be given the
necessary tests needed to detect for pregnancy, VD, etc., 31% felt a trial would occur and
they would have to testify, 24% felt the necessary evidence that could be used in court
would be obtained, 57% felt they would be treated as an immoral person, 26% felt the
man who did this to them would try to get back at them, and 19% felt other things would
occur. Most women who endorsed the “Other” category wrote in that they were worried
that the person would get mad or not believe them.

In Sample B (non-assaulted women), when asked about possible expectations for
telling someone, 94% believed they would feel calm, safe and better, 93.0% felt they
would get adequate medical attention, 90% felt they would be given the necessary tests
needed to detect for pregnancy, VD, etc., 86% felt a trial would occur and they would
have to testify, 82% believed that telling someone would result in the necessary evidence
that could be used in being obtained, 26.0% felt they would be treated as an immoral
person, 27% felt the man who did this to them would try to get back at them, and 7% felt
other things (e.g., others would get angry and/or try to hurt the man) would occur. It is
interesting to note the qualitative differences between expectations non-assaulted had and
assaulted women had. Specifically, non-assaulted women appear to have stronger beliefs
in the efficacy of their actions, as evidenced by a greater proportion of women believing
telling others would result in the necessary evidence being obtained (82% versus 24) and
they appear to be less likely to believe negative events, such as being treated as an

immoral person will occur (26% versus 57%).

61

MW One very interesting finding in this study is that 60%
of women who reported behavioral descriptions of sexual assault or rape answered no
when asked directly if they had experienced sexual assault or rape. Notably, several
women wrote on the questionnaire and/or verbally noted that their experience was just
“bad” or a “miscommunication” and not an assault.
Relationship between Survivor and Pegpetrator (Sample A only): Ten percent of the abuse
was perpetrated by a stranger, 28% by a causal acquaintances, 14% by a non-romantic
fiiend, 26% by someone the women was casually dating, 13% by someone the women
was exclusively dating and 10% by a partner in a long term relationship.
Level of Injuries (Sample A only): Most women (74%) reported that they were not
physically injured during the assault; however, 24% of the women stated that were mildly
injured and 2% noted that they were moderately injured. It is interesting to note that
several women either wrote on the questionnaire or verbally discussed the emotional
injuries they endured. Unfortunately, emotional distress of the sexual assault victims was
not directly assessed in this study.
Relationships between Observed Variables

Table 6 contains the correlation matrix for the 12 observed variables that were
analyzed for the assaulted women (Sample A), with significant alpha levels marked. Of
particular interest are the correlations between label of incident and the observed
variables found to correlate with perceived outcomes at the p<.01 level and with injury
and normative beliefs about family members at p<.05. Significant correlations were also
found between self-blame and blaming the man (p<.05) ; and between type of

relationship and sex roles, and normative beliefs about the community (p<.05).

62

5V9": nova"...

a:

3 as. :SN. :SN. :Nmm. .E. :8. :80 So. :8. so. Go. 88850 3385
cubesaaco

3 ram. :25. m2. m2. :2. «8. 39. so. 2a.- 2o. acscmcécasz
accuse

3 :58, a. me. as. $0.- as. 2o. *2. So. -eczcm cécécz

3:55

3 5. §. So. So. 5... 2:. so. me. -eaam 9:282

o._ .52. o2. LE. 03. Nvo. mg. 2:. A8 waged
o4 moor 2.3m. Re. a _. than. .33. AD 2822: mo fine..—

so
3 :9- Sq- :2. as; E..- caacuscfic 25

 

3 m8: coo. So. So. 6??

3 :SV. :3. m8. €va<

3 2:. m3; 5 09a

3 .NE. E E: 255

3 Show casm

2 : 2 a w a c m a. m N _ ch~§> Basso

 

.2 caacmv aces» cases a sacs; 8:38 to 882280 g

63

Significant, positive correlations were also found between all three normative belief
variables (all p<.01). Perceived outcomes were found to be significantly correlated with
injury, normative beliefs about family, normative beliefs about fiiends (all p<.01), and
normative beliefs about community members (p<.05).

Chi-square analyses were used to assess the relationships between force and the
other observed variables, since force is a categorical variable. Results of these analyses
are summarized in Table 7. Due to the large variance in scores on the normative beliefs
factors and outcome expectations factor, analyses between these variables and force were
not valid. Test results indicated significant relationships between force and label of
incident, with women who were physically assaulted more likely to acknowledge their
assault and women whose assaults occurred with verbal coercion or the use of alcohol
and drugs less likely to label their experience as an assault (p<.01). Chi-square analyses
also revealed a significant relationship between force and number of people told.
Specifically, women who told two people were more likely to have been physically
assaulted (p<.01). Relationship and force were also found to be related, with most of the
women whose sexual assault experiences involved alcohol and drugs knowing their
assailant (p<.01). A relationship was also found between blame and force, with women
who were physically assaulted placing less blame on themselves than women whose
assaults were less stereotypical. A relationship was found between force and level of
injury, with women who were physically assaulted more likely to be injured.

Table 8 contains the correlation matrix for the 9 observed variables that were
analyzed for the non-assaulted women, with significant alpha levels indicated. It is

interesting to note that reporting correlated strongly with perceived outcomes, normative

64

Table 7. Relationships between Force and the Other Observed Variables in Assaulted

 

 

 

Women (Sample A).

Observed Variable Degrees of Freedom Pearson chi-square
Acknowledge 3 13.96“

AWS 66 80.88

Sex Roles 93 64.36

Blame Self 9 30.84"

Blame Him 9 9.74

Type of Relationship 15 39.24”

Violence 6 28.34"

Reporting 6 31.61"

 

 

**p<.01

65

o ‘lu—fi-VW I Ii

14111 I

.~th\/
QOBOnDOhO 3.0230:an inland

5V9":

 

 

Eye".

6

tax. .32. .32.. 3o. :3. .53.- 5. 388:5 8385
vabasaeco

:5. the. «8. Sq- c8. c8: 2:. seam-cécécz
Erase

3 tea. tam. Sq- «8. m8: N8. seam czaasz
@255

3 2&8. $8. .52. :5. 3:. seem 3:3ch

3 m:. 2:. m2. $2. Cesium

S :5. :SN. «a. $95.

3 $3. ::~. 6095

3 1on O E: has

3 32cm casm

s c m c m N l 8333, Basso

 

.8 scene Sac? cease-82 a escfi> 8530 .8 accusceco flaw-H.

66

mot

blar

outt

bel‘

wit

001

SE

ge

I’C

Va

WC

beliefs about family members, and normative beliefs about friends (all p<.01) and
moderately with self-blame (p<.05). Additional significant correlations included self-
blame and blame him and sex roles ( p<.01); blame him and sex roles and perceived
outcomes (p<.01); sex roles and perceived outcomes (p<.01); sex roles and normative
beliefs about fiiends (p<.05); and normative beliefs about fiiends, family and community
with expectations for reporting (p<.01). Similar to Sample A, significant, positive
correlations were found between all three normative beliefs (p<.01).
Structural Equation Models

Structural equation modeling (SEM) was used to test the hypothesized
relationships in the structural models for both assaulted and non-assaulted women. All
SEM analyses were conducted using the statistical package AMOS. In all analyses,
generalized least squares estimation was employed. Prior to the analyses of the structural
model, decisions regarding missing data and the measurement properties of the scales for
these two samples of data were conducted. The amount of missing data for the non-
assaulted women was minimal (less than 1%). In the group of women who were
assaulted, missing data was minimal for all variables (less than 1%) except for the
expectation variable (16% missing). As a result, mean substitution was used to replace
any missing data in both samples, excluding the expectations data in Sample A because
those missing data appeared not to be random. Some women appeared only to rate how
positive or negative outcomes were if they believed these outcomes would occur and as a
result it was felt that mean substitution would not be suitable for the expectations
variable. Overall, only the responses of 2 participants (both in non-assaulted sample)

were excluded from the analysis because of their difficulty in accurately completing the

67

question

 

problem

For the
Root h‘
measu
8: Hat
for an
of the
variar
for by
than .
functi
.05 to
Prior

the la

 

follox
38831.1
the s(
facto
acce
mea:

first

 

questionnaires. The final N for Sample A was reduced from 195 to 170 because of
problems with the expectations variable. The final N for Sample B was 200.

A variety of Goodness of Fit indices can be examined in determining model fit.
For the purpose of this study, the CIVflN/DF index, Goodness of Fit Index (GFI), and the
Root Mean Square error of approximation (RMSEA) were used. The CMIN/DF is a
measure of the minimum discrepancy firnction divided by the degrees of freedom. Marsh
& Hacevar (1985) reported that different researchers have recommended cut-off values
for acceptable levels, as low as 2 and as high as 5. The GFI is based on a ratio of the sum
of the squared differences between the reproduced and observed matrices to the observed
variances (Schumacker and Lomax, 1996). It indicates the amount of variance accounted
for by the model, with values from .90 to .99 reflecting an excellent fit, and values less
than .85 reflecting a poor fit. The RMSEA is the ratio of the population discrepancy
function to the degrees of freedom, with values from .00 to .05 reflecting an excellent fit,
.05 to .08 reflecting a moderate fit, and values greater than .10 reflecting a poor fit.
Prior to analyzing the model fit, the measurement properties of the scales used to assess
the latent constructs in each of the models was necessary and are reported in the
following section. As in previous sections, all analyses are presented first for non-
assaulted women (Sample A) and then for assaulted women (Sample B). The first part of
the section for each sample contains a description of the goodness of fit criteria for the
factor structure for each construct. Second, for those measures that did not have an
acceptable degree of fit, analyses conducted to improve the factor structure of these
measures are reported. Improvements upon the factor structure of measures was made

first for Sample B. Since the two models were going to be compared, the decision was

68

made to use the same factor structure in both samples if they were adequate. In fact, all

measures, using the same items as Sample B, were found to have adequate fit in Sample

A. These steps led to more accurate measurement models, which in turn were used to

analyze the hypothesized relationships between the constructs in each of the models.
Review of Measurement Model

Measurement Model Smcification for both Samples (Eigtyres 4,5,6,7,8 )-

Sex Role Stereotypes:

Sex Role Stereoggping Questionnaire ( 3%, 1980). The literature reported one
composite scale for this measure (Burt, 1980). All items were coded so that higher
numbers indicated more liberal views. For non-assaulted women (Sample B), when all
items of the Sex-Role Questionnaire were included in the confirmatory factor analysis
(CFA), the overall fit of the measure was good (x2=36.78, df=20, p=.01, GFI=.954,
CMIN/DF=1.84, RMSEA=.07). The CF A results suggest that similar to previous
findings the Sex Role Stereotyping Questionnaire has one factor, which is labeled SRSQ
in the measurement model (Figure 4).

In Sample A one factor (SRSQ) of the Sex Role Stereoflping Questionnaire
t Bu_rta 1980) was used. Confirmatory factor analyses with the assaulted sample also
demonstrated that the overall fit of the measure in Sample A was good (x2=24.74, df=20,
p=.21, GFI=.968, CMIN/DF=1.24, RMSEA=.035).

The Attitude Towards Women Scale (AWS, Spence & Helmreich, 1978). The
AWS was found to have only one factor in previous studies (e.g., Spence & Helmreich,

1978, Spence & Hahn, 1997). All items are scored so that higher scores equal more

69

cgalitar

it‘ll);

therefor

 

 

 

 

egalitarian attitudes. The confirmatory factor analysis (CFA) for this sample (x2=130.90,
df=90, <.01, GFI=.912, CMIN/DF=1.45, RMSEA=.048) indicated a good fit for the data,
therefore one combined score, referred to as AWS, was used for this measure (see Figure
4).

Results of the CF A for Sample A of The Attitude Towards Women Scale AWS

 

Same & Helmreich, 1978) also demonstrated a good fit of the one factor model (labeled
AWS in Figure 5) (x2=109.56, df=77, p<.01, GFI=.919, CMIN/DF=1.42, RMSEA=.047).
It is important to note that all items were scored so that higher scores were indicative of

more liberal beliefs.

Normative Exmctations

Normative Exmctations for Remrting Questionnaire (revised from Espinso &
Cardozp, 1995 ). The foundation of this questionnaire was Ajzen & F ishbein’s (1972)

theory that the normative factor was comprised of the sum of the beliefs about what
others expect them to do multiplied by the individual’s motivation to comply with the
norms. Items are scored so that the higher the score, the more people the woman
believes others would want her to tell. As discussed in the methods section, this revised
questionnaire was used to assess the likelihood that each of three referents (family,
friends, community) would want them to report their experience of sexual assault to each
of nine report recipients (friend, immediate family member, extended family member,
resident advisor, mental health worker, doctor, police, or clergy). It was expected that
there would be three factors, normative beliefs about family, fiiends and community. The
findings of the CF A for Sample B demonstrated a poor fit of this data to the hypothesized

factor structure (x2=592.62, df=249 <.01, GFI=.752, CMIN/DF=2.38, RMSEA=.083).

70

 

As note

factors

deleted

and iii

 

friend.

 

factors

1101111211

criteria

GFI=.f

 

 

but
mean
friend

healtl

mode

 

Indie

adt’lg.

As noted in the methods section, exploratory factor analyses (EFA) suggested three
factors (family, friends, and community) with the elimination of 9 items. Items were
deleted fi'om exploratory factor analyses if items loaded highly on more than one factor
and if item loadings were below .40. These factors are labeled norm-family, norm-
fiiend, and norm-community in the measurement model (see Figure 4).

Confirmatory factor analyses for Sample A for this measure used the same three
factors used in Sample B (normative beliefs for family, normative beliefs for fiiend, and
normative beliefs for community). Results of the CFA for Sample A resulted in fit
criteria that ranged from poor to adequate (x2=222.2,df=85 p<.01,
GFI=.85,CMIN/DF=2.6, RMSEA=.O9). Exploratory factor analyses did not generate a
better fitting model for this sample, therefore this factor structure was used. These factors
are also labeled norm-family, norm-fiiend, and norm-community in the measurement
model (see Figure 5).

Perceived Outcomes

Perceivpd Outcomes-revised (revised from F eldman-Summers & Ashworth

 

£181), No previous validity, reliability or factor analyses have been conducted on this
measure, however it was expected that there would be seven factors (expectations for
fiiend, immediate family, extended family, clergy, police, resident advisor and mental
health worker). Higher scores reflect more positive expectations. CF A analyses for this
sample demonstrated a reasonable, although not entirely adequate fit for the data to this
model (x2=49.34, df=14 <.01, GFI=.929, CMIN/DF=3.53, RMSEA=.113). EF A analyses
indicated that there was only one factor. Expectations for telling clergy, resident

advisors and extended family members were deleted for theoretical reasons. Specifically,

71

fac

lat
ex]
ex]

38

both
blan
and t
blam
itetm

hm

Would

factors such as not living in a dorm, not having extended family members and not being
religious may have contributed to responses and since these extraneous factors could not
be controlled, the items affected by these factors were dropped. The resulting factors are
labeled in the measurement model (Figure 4) as expfiiend (expectations for fiiend),
expmhw (expectations for mental health worker), exppolice (expectations for police), and
expfamily (expectations for family). CFA analyses of the revised questionnaire revealed
a good fit for the data to this structure (GFI=.994, CMIN/DF=1.172, RMSEA=.029). ~—

To maintain consistency, CFA analyses for Sample A (assaulted women) for the

 

perceived outcomes measure used the same items as Sample B. Results of the CF A
demonstrated a reasonable, although not entirely adequate fit for the data to this model
(x2=49.34, df=14 <.01, GFI=.93, CMIN/DF=3.53, RMSEA=.113). EFA analyses
indicated one factor and did not suggest that any items be deleted, therefore the factor
structure was not changed (see Figure 5).
Blpme

Blame Questions. The wording of the questions to assess blame was identical for
both samples. Questions 20 and 21 were used to assess the extent to which women
blamed the man and themselves for the assault and for Sample B these were questions 66
and 67. Higher scores on the question regarding blaming the man are indicative of more
blame, while higher scores on blaming oneself are indicative of less self blame. These
items were labeled Blame Self and Blame Him in the measurement model (Figure 4).
LCM

Remrting. In Figure 4, reporting was measured by how many people a women

would tell (labeled in Figure 4 as tellhow). The score was equal to the number of people

72

 

 

the women expected to tell An additional model (Figure 5) examined relationships
between the variables and reporting (labeled tellwho) as measured by who they
specifically reported to. Items were scored as zero if they would not tell that person and
one if they would tell that person. Similar to the expectations and normative factors,
factors regarding telling an extended family member, clergy, and resident advisor were
not examined as individual factors for theoretical reasons, however they were kept as a
part of the total number of people told. CF A analyses of the report variable, as
measured by who they reported to, revealed an adequate fit for the data to this structure
(Chi square=11.66, df=4, p<.05, GFI=.977, CMIN/DF=2.9, RMSEA=.08).

For assaulted women, reporting was also measured by both how many people a
women did tell (labeled as tellhow in Figure 6) and by whom they did tell (labeled
tellwho in Figure 7). Since the measurement of how many people a women did tell
consisted of one item confirmatory factor analyses was not necessary. To maintain
consistency, the CF A of the reportwho factor for assaulted women used the same
reporting referents as used in Sample B (i.e., fiiend, immediate family member, mental
health worker, police, and doctor). Results of these analyses revealed a very good fit for
the data to this structure (Chi square=5.09, df=5, p>.10, GFI=.990, CMIN/DF=1.02,
RMSEA=.01).

Typg of Force, Relationship, Violence and Label of Incident

Type of Force, Relationship, Violence, and Label of Incident were only measured
in assaulted women. They were used in the total model of factors contributing to
reporting in survivors of sexual assault (see Figure 8). All of these variables were

measured by one item, therefore CFA analysis were not conducted. Type of force is a

73

categorical variable and label of incident was scored as yes or no. Higher relationship
scores represent a more intimate relationship and higher scores on the violence variable
reflect more violence.

The following section reports the results of the overall model fit of the model
separately for assaulted women and non-assaulted women and includes a discussion of
attempts made to improve the model fit in both samples. These analyses also were
conducted using AMOS and the same criterion for goodness of fit, as discussed earlier,
are used. For the non-assaulted women, the model is considered twice, once using the
how many people the person told (i.e.,tellhow) as the reporting variable (Figure 4) and
once using who the person told (i.e., reportwho) as the reporting variable (Figure 5). For
the assaulted women, the model is only considered once using who the women told (i.e.,
reportwho) as the reporting variable (Figure 7) because the extremely limited variance in
number of people actually told, prohibited the model from successfully being analyzed
(i.e., resulted in negative goodness of fit or resulted in error messages because of negative
error variances).

Results of the Overall Model Fit

Model for Non-Assaulted Women

Model for Sample B Using How Many People a Woman Exmcted to Tell as the

Remrting Variable
The chi-square statistic for this model was 107.34 (df=51, p<.001; Figure 4). The

goodness of fit indices suggested that the fit of the model, or how well the observed
variables fit the latent variables, was good, but could be improved (CMIN/DF=2.1,

GFI=.91, RMSEA=.08). One method of attempting to improve the fit of the model is to

74

examine the residuals and to consider adding additional links in the model, particularly
using variables with large residuals (Tabachnick & F idell, 1996). Since both the
expectations and normative factors had large residuals, a correlation between these two
items was added. In fact, analyses yielded a significant, positive relationship between
normative beliefs and expectations (correlation estimate=.428, _z=4.403, p<.01, two tailed
test, see Figure 9). This link suggests that there is a positive relationship believing that
others want you to talk about sexual assault experiences and believing that talking would
have positive outcomes. Correlations between other variables were also examined,
however no additional significant correlations were found. The addition of the
expectations for reporting and normative beliefs correlation did improve the overall fit of
the model (x2=87.86, df=50, p<.01, CMIN/DF=1.76, GFOI=.926, RMSEA=.062).
Analyses comparing the revised model with the original model suggested the revised

model was superior (x2 difference=19.48, df=1, p<.01) and should be used.

Model for Sample B Using Whom the Woman Exmcted to Tell as the Remrting

 

3.1m

For this model, the chi-square statistic was 174.574 (df=100, p<.001; Figure 5)
and the goodness of fit indices suggested that the fit of the model was adequate but
needed improvement (CMIN/DF=1.75, GFI=.89, RMSEA=.06). Similar to the previous
model (Figure 4), initial results yielded high residuals for both the expectations for
reporting and normative beliefs factors, so a correlation between normative beliefs and
expectations for reporting was added to the model. Subsequent analyses revealed a
significant, positive relationship between normative beliefs and expectations for reporting

(correlation estimate=.39, g=3.85, p<.01, two tailed test, see Figure 10). This link

75

suggests that there is a positive relationship believing that others want you to talk about
sexual assault experiences and believing that talking would have positive outcomes. The
addition of the expectations-norms correlation improved the overall fit of the model
(x2=l61.153, df=99, p<.01, CMIN/DF=1.63, GFI=.90, RMSEA=.06. Analyses
comparing the revised model with the original model suggested the revised model was
superior (x2 difference=13.42, df=1, p<.01) and should be accepted.

Model for Survivors of Sexual Assault

Reduced Models for Sample A (Figges 6 and 7)

 

The fit of the reduced model (i.e. Figure 6) using the reporting factor as measured
by how many people were told (i.e. tellhow) was not able to be obtained, although it is
most likely very poor since the model would not run. More specifically, due to limited
variance in the number of people told and negative variances in the model, SEM analyses
could not be conducted. SEM analyses were able to be conducted on the reduced model
using who the person told as the reporting variable (i.e., reportwho factor, Figure 7) and
the chi-square statistic was 149.603 (df=100, p<.01). Goodness of fit indices suggested

that the fit of the model was good (CMIN/DF=1.50, GFI=.89, RMSEA=.05).

Total Model for Remrting Factors in Suyvivom of Sexual Assault (Eigms 8 & 11)

 

Due to an extremely poor fit of the data to the model, as evidenced by difficulty
minimizing and a negative error variances, SEM analyses could not be conducted for this
model. The preliminary analyses indicated that the label of incident variable had a
negative error variance and was contributing to the difficulty running the model. Since
this factor was measured as 0 or 1 and had little variability, it appeared to not be able to

serve as a moderator between other factors and reporting, therefore an additional model

76

(Figure 11) was constructed to examine direct relationships between all variables and
reporting. This model was also unable to minimize. To increase power, additional
analyses attempted to examine the model after deleting the expectations for reporting
factor (since only 170 subjects had complete expectations data), however this model fit
was also too poor to be formally examined. An additional step was taken to try to
examine other factors in relation to reporting. More specifically, all variables in the
reduced model were deleted and that model with only force, relationship, violence, label
of incident and reporting variables was rtrn and generated a non-positive definite
covariance matrix. Unfortunately examination of each of the previous factors alone with
reporting (i.e., force with reporting, violence with reporting, etc.) did not yield any
positive definite covariance matrixes or models that could be analyzed using SEM. The
inability of the additional factors to work appeared to be directly related to the lack of
variance in the main report variable, both who they told and how many people they told.

The following section will discuss the results for the hypotheses for the non-
assaulted women and then for the assaulted women. These analyses were conducted
using the scales discussed previously and represented in Figure 4 and Figure 5. All SEM
analyses were conducted using the statistical package, AMOS. In all analyses the
generalized lease squares method was used. For every hypotheses in the model, results
are presented for the relationship between that variable and number of people told

(tellhow) and who the person told (tellwho).

77

Results of the Hypotheses
Structural Model for Non-Assaulted Women

Hymthesis 1

A significant relationship between blame and how many people were expected to
be told (i.e., reporthow) was not supported (standardized regression weight=.228 , _z_=
1.082 , p>.10 , one-tailed test, see Figure 4). In addition, a direct positive relationship
between blame and who the person would tell (i.e., reportwho) was not supported
(standardized regression weight=.l30, z=.271, p>.10, see Figure 5)
Hymthesis 2

Significant relationships between sex-roles and how many people were expected
to be told (i.e., reporthow, standardized regression weight=.011 , z: .118 , p>.10 , one-
tailed test, see Figure 4) or who would be told (i.e., reportwho, standardized regression
weight=.041 , g= .476, p>.10 , one-tailed test, see Figure 5) variable were not found.
Hymthesis 3

A significant correlation between sex-roles and blame was not supported in either
model (Figure 4, correlation estimate=.153, z=.994, p>10: Figure 5, correlation
estimate=.023, F282, p>.10).
Hymthesis 4

Significant relationships between the normative beliefs variable and the reporting
variable as measured as how many people a woman thought she would tell (reporthow;
standardized regression weight=.029 , _z= .27 , p>.10 , one-tailed test, see Figure 4) or
who participants thought they would tell (i.e., reportwho, standardized regression

weight=.020 , g: .204, p>.10 , one-tailed test, see Figure 5) variable were not found.

78

Hyppthesis 5

As predicted, expectations had a direct, significant, positive, relationship with
whom women believed they would tell about a sexual assault. The relationship was
slightly stronger with how many people a woman planned to tell (i.e., reporthow,
standardized regression weight=.492, '_z=4.39, p<.01, one-tailed test, see Figure 4), then
with who the woman planned to tell (i.e., tellwho variable, standardized regression
weight=.458, _z_=2.02, p<.03, one tailed test, see Figure 5). This finding suggests that

individuals who believe that telling others will have positive effects are more likely to

5'7 "T ‘ _‘7

tell someone about their experiences.
Structural Model for Survivors of Sexual Assault

Hyppthesis 1

A relationship between blame and who was told (i.e., reportwho) was not
supported (standardized regression weight=.1l6, _z_=-064, p>.10, one-tailed test, see
Figure 7) in Sample A.
Hymthesis 2

No significant relationship between sex-roles and who was told (i.e., reportwho)
was found in this sample (standardized regression weight=.066 , _z_= .063 , p>.10 , one-
tailed test, see Figure 7).
Hymthesis 3

Similar to Sample B, a significant correlation between sex-roles and blame was

not supported (correlation estimate=.171, z=.192, p>10).

79

Hymthesis 4

SEM analyses did not support this hypothesis (standardized regression weight=-
-.062 , ;= -.065 , p>.10 , one-tailed test, see Figure 7), suggesting that in this sample there
was not a direct relationship between normative beliefs about reporting and who people
reported to.
Hyp_othesis 5

Unlike with non-assaulted women, a direct, positive relationship between
expectations and telling others about sexual assault experiences was not supported
(standardized regression weight=.036, g=.060, p>.10, one-tailed test, see Figure 7).
Hymtheses 6,7, 8, 9, 10, 11

Due to negative error variances and an extremely poor fit of the data to this model
(Figure 8), these hypotheses could not be examined as the model was unable to be
analyzed (e.g., would not minimize and/or had non-positive definite covariance matrix’s).
Further details regarding models attempted is presented in a future section (i.e., results
from overall models).
Hymthesis 12

Since the model of factors related to number of people told in the assaulted
sample had negative variances and was unable to minimize, it suggests that non-assaulted
women’s responses, which revealed an adequate fit, was the better fitting model of the
two groups. Analyses comparing the models of factors related to who people tell about
sexual assault in both samples, suggested that Sample A was a better fit for the model (x2

difference=24.97, p<.01), however Sample A did not have any direct, significant paths.

80

DISCUSSION

This study examined a comprehensive model of factors that contribute to
women’s decisions whether or not to tell other people about their experience of sexual
assault. This study expanded upon previous research by examining a model using both
feminist and social-cognitive theories. The model examined the relationships between
blame, sex-role stereotypes, type of relationship between perpetrator and survivor,
physical injury, type of force, beliefs about what others would want you to do and how
they would react, women’s definitions of their experiences and reporting behavior. In
addition to individual links between variables, this study also assessed overall frequency
of sexual assault and reporting behavior.
Characteristics of Sample and F regucncies of Observed Variables

Differences between Demogrpphic Variables. In addition to using women in
psychology undergraduate classes, this study also consisted of women in sororities.
Although some research has found higher rates of sexual assault in women in sororities,
in general research in this area has not been conclusive (Kalof, 1993). In this study,
analyses of differences between samples on demographic variables indicated that women
in sororities were less likely to be in the assaulted sample. Many of the sororities in this
study reported that they had participated in educational workshops about sexual assault
and that their national chapters stressed the importance of continuing education in this
area. It may be possible that the differences in sorority membership between samples
was related to the sororities having more information about sexual assault, however the
relationship between sexual assault prevention programs and risk for future assault is

inconclusive (see Lonsway, 1996). An additional explanation could be that the

81

differences are related to the relationship between sororities and fi'aternities. Specifically,
some researchers have suggested that sorority membership might provide some protection
against sexual assault, with the sorority-fraternity relationship guided by norms that limit
the amount of sexual coercion inflicted and may provide a shelter against sexual assault
(Garrett-Gooding & Senter, 1987; O’Sullivan, 1991).

Sorority membership was also the only demographic variable that was
significantly correlated with the observed variables. Specifically, for assaulted women
there were significant relationships between being in a sorority and labeling the incident
as an assault and relationship to offender, with women in sororities more likely to be
closer with the perpetrator and less likely to acknowledge the experience as an assault.
The relationship between sorority membership and knowing the offender may be related
to the fact that sorority women are more likely to know and date fraternity men
(Copenhaver & Grauerholz). Studies have found that compared to other male students
who non-sorority women may be more likely to date, fraternity men are more sexually
aggressive (Garrett-Gooding & Senter, 1987). The relationship between sorority
membership and label of the incident may be mediated by acceptance of rape myths.
Acceptance of rape myths was not assessed in this study, however previous research
suggests that this might be a possible link. Specifically, while Kalof (1993) did not find a
relationship between sorority membership and sex-role stereotypes, she did find that
women in sororities were more likely to accept interpersonal violence and rape myths.
Fisher (1986) found that the acceptance of rape myths and interpersonal violence were

related to acknowledging the experience as an assault. Therefore, it is possible that the

82

relationship between sororities and the label of the incident is not direct and instead is an
indirect relationship involving rape myths.

Prevalence of Sexual Assault. This study found a high prevalence of sexual
assault, with almost half of the total sample (49%) endorsing experiences of sexual
assault or rape. Previous studies have found ranges between 29% to 76% (e.g.,
Hirnmelein, 1995; Koss et.al, 1987; Muehlenhard & Linton, 1987). Notably, this study
was very similar to Koss et.al’s (1987) multi-site study of colleges and universities that
found a rate of 54%. This study’s finding that many of the assaulted women (83%)
reported experiencing more than one type of assault is also consistent with Koss et al.’s
1987 study. Most cases of assault reported in this study differed greatly from the
stereotypical view of sexual assault that has been promoted by the media. Specifically,
most of the women knew their attacker (90%) and were not physically injured in the
assault (74%). The findings are consistent with previous studies which suggest that most
sexual assaults victims do know their attacker and do not involve extreme physical
violence (e.g., Koss et.al, 1987; Muehlenhard & Linton, 1987). It is also important to
note that cross-tabulations did not generate one modal sexual assault scenario and instead
found that women experienced many different types of sexual assault scenarios. These
findings emphasize the importance of asking women about many aspects of their
experiences (i.e., how well they knew the person, level of injury, type of force, etc.) since
these different variables are not consistent and may affect their perception of their
experience and their recovery.

It is also important to note that several women who were raped by physical

violence or the threat of physical violence did not choose this as their most upsetting

83

experience. Unfortunately women were not asked to identify why they found a particular
experience most upsetting; however such variables as how well they knew the person,
their perceptions of their ability to prevent or to have stopped the attack may affect their
decisions. These findings suggest that researchers studying sexual assault need to
consider and explore how women experience during different sexual assault experiences
affects their decisions.

Consistent with previous studies (e.g., Koss et.al, 1987; Wiehe & Richards, 1995),
many women in this study who had experiences that contain the legal elements of sexual
assault did not believe that their experience constituted sexual assault (60%). This does
not mean that their experience was not aversive or even neutral, only that they did not
believe that their experience met their own definition of the crime of sexual assault.
Women may minimize their experiences to protect themselves, since it has been
suggested that the perception of oneself as a victim is aversive to both oneself and others
(Rabinowitz, 1990). Minimization of the experience is a common coping mechanisms in
victims (Janoff-Bulman & Frieze, 1983), so sexual assault victims may not label their
experience as such in an attempt to protect themselves fi'om the aversiveness associated
with being a victim. In addition, not labeling an experience as an assault may be related
to difierences in definitions of assault. The media often teaches that sexual aggression is
normal (Gunn & Minch, 1988). The media often portrays sexual assault and rape
stereotypically, which may lead many people to construe a narrow definition for assault.
Many women appear to be internalizing society’s message and consequently have

difficulty labeling their own sexual assault experiences as assaults.

84

Notably the results indicated that most women who were sexually assaulted told
one person (69%),usually a friend, or they told nobody (20%). This finding is important
because talking about the experience may be the single most important therapeutic
behavior that a victim can do afterward (Davis & Friedman, 1985; Koss, 1988b).
Research has demonstrated that social support can serve as a buffer for the psychological
effects of the experience of sexual assault (Golding et.al., 1989; Popiel & Susskind, 1985)
and it is related to better health and fewer psychological symptoms (Kimerling &
Calhoun, 1994). Social support appears to have both short and long term effects.
Specifically, Burgess & Holmstrom (1978) reported that 45% of sexual assault survivors
with social support felt recovered within months after the incident; whereas none of those
without social support felt recovered in that time frame. They also found that within 4 to
6 years after the assault, 80% of survivors with social support felt they had recovered;
whereas only 47% of those without social support did. Despite the advantages of telling
others, the current study suggests that many women do not tell others and therefore do
not receive the support they need.

Interestingly, the mean number of people told between the two samples was
significantly different, with women who were not assaulted expecting to tell more people
than the average women actually does. This is an important finding as much research on
sexual assault is based on hypothetical situations and this study suggests that there is a
discrepancy between what women believe they would do and what they actually do.
Current theories may not accurately account for changes caused by the trauma of sexual
assault. The trauma of rape is difficult to assimilate and may have a significant impact on

women’s views of themselves and others because it violates basic human assumptions of

85

personal integrity and control and beliefs in a just and safe world (Roth and Lebowitz,
1988). This suggests that more research is needed with actual survivors of sexual assault,
rather than using hypothetical situations, so that a more accurate understanding of
women’s responses to sexual assault can be obtained.

Other differences between the two samples included their expectations about what
would happen if they told someone. Women who were not assaulted were more likely to
believe that telling others would result in getting adequate medical attention, being given
the necessary tests to detect pregnancy, VD, etc., a trial would occur and they would have
to testify, and the necessary evidence that could be used in court would be obtained. This
finding indicates that the non-assaulted women as a group had a stronger belief in the
efficacy of their actions while the assaulted group did not. This could be related to both a
feeling of helplessness and inadequacy engendered by the assault and to actual negative
experiences of reporting. The helplessness many women feel after an assault may alter
their schemas or worldviews and make them less likely to believe in the efficacy of their
actions. In fact, Allison & Wrightsman (1993) reported that a common reaction to scrotal
assault is a feeling of lack of control over one’s life. In addition, women who are actually
assaulted may not believe in the efficacy of reporting because of actual negative
experiences of reporting. In this study, several subjects wrote about negative experiences
of reporting, such as not being believed, and how this stopped them from telling anyone
else. Previous research (Biaggio, Brownell, & Watts, 1991) also found that some support

providers respond in a negative way, such as blaming the woman or making her feel

ashamed.

86

Most women in both groups believed telling someone would result in feeling
calm, safe and not result in the man trying to get back at them. These findings are
surprising given the number of women who told nobody (20%), suggesting that despite a
positive expectation of telling others some women still do not tell others. Perhaps some
women feel that the negative outcomes of reporting outweigh the benefits. Women in
this study were not directly asked whether the negative aspects of reporting outweighed
any possible benefits, therefore this theory can not be tested in this study.

This study also found differences between the samples regarding who women
blame, with women who were not assaulted more likely to place more blame on the man
and less blame on themselves than the women who were assaulted. Notably, differences
among the samples may be confounded by the fact that the hypothetical situation
involved rape using physical force which is more consistent with the stereotype of rape.
This is important to note because previous studies have found a relationship between
blame and force, with women less likely to attribute blame to themselves, and in
hypothetical situations less likely to blame the portrayed victim, if the situation involved
physical force (Koss, 1985; Shotland & Goodstein, 1983). To assess the effects of force
in relation to blame, a post-hoe analysis of differences in blame for women who were
physically forced to have sex and women who answered questions about a hypothetical
rape using physical force was conducted. Notably, there were no significant differences
in self-blame between these groups, however there was a significant difference in
blaming the man (p<.01). Specifically, women who were actually assaulted blamed the

man less. This suggests that the difference between the samples on blaming the man is

87

related in part to a true difference between how women believe they would feel and how
most women actually feel.
Relationships between Variables in Assaulted Women

This study found significant correlations between acknowledgment of assault and
several of the observed variables. Previous research using scenarios found that people are
significantly more likely to label an incident as rape if the victim is injured (Bourque;
Shotland & Goldstein). This study suggests that victim’s perceptions of their own
experiences are also related to level of injury, with women who were injured being more
likely to acknowledge their experience as an assault. This finding is consistent with
feminist theory that suggests that women whose experiences do not meet the stereotypical
definition of assault will be less likely to define their experience as an assault.
Researchers (O’Toole & Shiffrnan, 1997) have suggested that despite recent educational
and public efforts to debunk the myths that surround sexual assault, it is still commonly
believed that women should be able to resist a man successfully unless he uses physical
force. Notably, chi-square tests revealed that women whose sexual assault experience
involved physical force are more likely to acknowledge their experience as an assault
compared with women whose assaults involved alcohol/drugs or verbal coercion.

This study also found support for feminist theory that argues that attribution of
responsibility is related to women’s definition of their experiences. Defining the
experience as an assault was correlated with whom the women blamed, with less self-
blarne and higher levels of blame on the man related to defining the experience as an

assault in this study.

88

It was also hypothesized that there would be a relationship between labeling the
experience as an assault and reporting, since women most likely would not tell others
about an experience that they do not yet perceive as an assault. Similar to previous studies
(Koss, 1985) this study did find that labeling the experience as an assault was positively
correlated with reporting the incident, suggesting that one barrier to reporting an assault is
not defining the experience as an assault. Another barrier to reporting appears to be
people’s expectations about reporting. This study found a positive relationship between
expectations and reporting, with women who expect more positive outcomes being more
likely to report the incident. Notably, a relationship between the label of the incident and
expectations for reporting was also found, with positive expectations for reporting related
to being more likely to label the incident as an assault. Initially, it was unclear if this
relationship was direct or was an indirect relationship involving reporting. Post-hoe
analysis revealed that the relationship between acknowledgment and expectations was
indirect. Specifically, afier partialling out the effects of reporting, the relationship
between acknowledgment and expectations was not significant (p>. 10).

Significant, positive correlations were also found between reporting and perceived
outcomes, injury, and normative beliefs about family members. These results are
consistent with previous findings (e.g., F eldman-Summer & Norris, 1984; F eldman-
Summers & Ashworth, 1981) and suggest that women who perceive positive outcomes
will occur from reporting and believe others want them to tell are more likely to share
their experience with people. In addition, the results suggest that when the experience is
more stereotypical (i.e., results in injury and uses physical force) women are more likely

to tell more people. This may be related in part to actual differences in reactions when

89

women report sexual assault. Specifically, societal institutions such as hospitals and the
legal system may be more likely to believe women who have physical evidence of a
sexual assault. In regards to the legal system, when there is physical evidence the case is
stronger, which may make the entire process of reporting less aversive. Estrich (198 7)
noted that when the accused was a stranger, or armed, or otherwise 'fit the definition of
what society assumed to be the typical rapist, the victim was usually given the benefit of
the doubt regarding nonconsent and women who were severely injured were usually
believed to be credible.

For women whose experiences are more stereotypical they may perceive that
others are likely to be more understanding and/or less judgmental, perhaps because they
are more likely to define their own experience as an assault or place less blame on
themselves. In fact, consistent with previous studies (Check and Malamuth, 1983) this
study found a relationship between type of force used and blaming oneself less, with
more traditional and visible signs of force (i.e., physical instead of verbal coercion)
related to less self-blame. This is consistent with the theory that women are expected to
be able to prevent assault, unless it involves physical force (Calhoun & Atkeson, 1991),
therefore when force or injury do not occur women may blame themselves for “failing” to
stop the assault from occurring. A significant relationship was found between injury and
type of force, with more traditional/visible signs of force related to more injury.

Previous literature (e.g., Bridges, 1991; Friedman & Thomas, 1985, Fisher., 1986)
suggested a relationship between blame and sex-roles, however this study did not confirm
this finding. This may be in part related to the generally limited variance of the sex-role

measures. Notably, in this study the mean of both sex-role measures was higher than the

90

means of previous studies of the measures (e.g., Spence & Hahn, 1997; Burt, 1980),
which suggests a change in sex-role attitudes or at least an understanding of what is
socially acceptable. The current measures may not be able to accurately assess the more
subtle differences in sex-role stereotypes that are relevant for women today. Finally,
unlike previous studies (e.g., Coller & Resick, 1987) a relationship between type of
relationship and acknowledgment and report were not found. Additional analyses re-
coded relationship as known versus stranger and this was also not related to reporting.
This may be related to limited variance, with 90% percent of women knowing their
attacker. Notably, relationship to attacker was correlated with normative beliefs about
fiiends and normative beliefs about the community.
Correlations between Variables in Non-assaulted Women

Consistent with previous studies (e.g., Gunn & Minch, 1988, Feldman-Summers
& Ashworth, 1981) a significant and positive relationship between reporting and
perceived positive outcomes was found for women who were not assaulted. In addition,
similar to previous research (Gunn & Minch, 1988) this study found a relationship
between reporting and blame. Specifically, less self-blame was associated with increased
reporting. This relationship may be related to how they believe others will view them.
Specifically, if women feel less responsible they may be more likely to believe others will
place less blame on them and be more supportive. Similar to Sample A (non-assaulted
women) in this study there was a significant correlation between self-blame and blaming
the man, with blaming oneself more associated with blaming the man less. Both high
self—blame and low blaming the man scores were associated with more traditional sex role

stereotypes. Women with less traditional stereotypes would be expected to be less likely

91

to have traditional definitions of sexual assault and would be expected to not believe that
sexual assault is caused by women’s failure to stop the man’s advances. Therefore, one
would expect as this study and previous research (e.g., Muehlanhard, Freidman &
Thomas, 1985; Shotland & Goodstein, 1983) has found that less traditional sex-roles
would be associated with higher other blame and less self blame.

Women with more egalitarian attitudes appear to have broader definitions of
sexual assault and may be less likely to accept traditional roles of sexual relationships.
More specifically, women with more egalitarian attitudes may be less likely to believe the
women’s role is to stop sexual advances. Therefore they are less likely to blame the
women for “failing to stop the man” when they are assaulted and are more likely to blame
the man who is actually responsible.

Positive perceived outcomes were associated with less traditional sex roles. Aj zen
& F ishbein (1972) stated that the likelihood of reporting is related in part to an
individuals perceived consequences and evaluation of these consequences. Women with
less traditional sex-roles may be more likely to have more accurate/wider definitions of
sexual assault and may therefore believe their experience is an assault and should be
reported. Women’s own reactions most likely are related to how they believe others
would react and therefore if they are less traditional they may be more likely to expect
others to be more interesting in hearing about sexual assault and in general expect more
positive outcomes. Notably, sex-role stereotypes were also found to be related to
normative beliefs about fiiends and family members, with more egalitarian views

associated with stronger beliefs that others would want you to tell. This suggests that

92

women may believe that others share their beliefs about gender roles, so that women who
have less stereotypical beliefs might expect others to hold similar values.
Structural Model of Factors Affectingyomep’s Decisions Whether or not to Report
Sexual Assault (Sample B, non-Assaulted Women)

This study hypothesized that there would be significant relationships between
reporting and norms and between reporting and sex-roles, neither of which were
supported by this study. Although the relationship between blame and reporting was also

not significant in the model it appears to be trend the expected direction. The hypothesis

 

that there would be a significant correlation between sex-roles and blame was not :3
supported using either report variable. This may be related in part to the limited variance
in the blame variable in this sample. The one hypothesis for non-assaulted women that
was supported was the significant, positive, relationship between expectations and
reporting. This finding suggests that women who believe that telling others will have
positive effects are more likely to tell someone about their experiences.

The models for this sample (Figures 4 and 5) both had goodness of fit indices that
suggested that the fit of the models were good but could be improved. For both models,
adding a correlation between the expectations and normative factors significantly
improved the fit of the models, suggesting that this relationship should be included in
future models. This finding suggests that there is a positive relationship between
believing that others want you to talk about sexual assault experiences and believing that
talking would have positive outcomes.

Although the goodness of fit of the model was strong, several of the expected

relationships were not supported by the data. This may be related in part to difficulty

93

with the sensitivity of some measures, such as the sex-role stereotype measures. In
addition, limited variance within some of the measures may also contribute to non-
significant findings. These problems are more pronounced with assaulted women and are

discussed in more depth in the following section.

8mm Model of F actors Affecting Women’s Decisions Whether or not to Remrt

Sexual Assault (Sample A, Assaulted Women). None of the hypothesized relationships

were supported in this sample. Caution must be taken in interpreting this finding, as the
reporting variable had limited variance, therefore these findings are not reliable.
Structural equation modeling assumes that variables are distributed normally and
violation of this assumption imposes limitations and creates problems in the analyses of
results. In this study, the reporting variable in assaulted women was not normally
distributed and had very little variance. Nonnormality often creates higher Type I error
rates and analyses that fail to converge or result in a solution that is incorrect. In the
sample of assaulted women this occurred because the reporting variable had extremely
little variance which resulted in negative error variances and an inability for the full
model to converge. The fit of the reduced model (i.e., Figure 6) using the reporting factor
(i.e., tellhow) was also not able to be obtained, although it is most likely very poor since
the model would not run. More specifically, due to limited variance in the tellhow factor
and negative variances in the model, SEM analyses could not be conducted. Although
obtaining additional subjects might have enabled the model to run, if similar frequencies
were found in other women, the reporting variable would have been very skewed.

Since the model of factors related to number of people told in the assaulted

sample had negative variances and was unable to minimize, it suggests that non-assaulted

94

 

women’s responses, which revealed an adequate fit, was the better fitting model of the
two groups. In addition, although analyses comparing the models of factors related to
whom people tell about sexual assault in both samples, suggested that Sample A was a
better fit for the model, this finding has little empirical validity since there were no direct,
significant paths in Sample A.

This study was unable to examine the fit of the total model (Figure 8) because the
model would not minimize. The problem appears to be related to the lack of variance in i .-

the main report variable, both whom they told and how many people they told. Since the

complete model could not minimize, the validity of the theory postulated in this study

 

could not be assessed. Correlational findings discussed earlier lend support for the
validity of the basic tenants of the theory. It is important to note that these findings do
not suggest the theory is invalid only that it can not be tested using traditional data
collection and statistical procedures.
Limitations of this Study

As noted earlier, structural equation modeling assumes that variables are
distributed normally. Violations of this assumption in this study, in regards to the
reporting variable resulted in a model that can not be validly interpreted and at times
would not converge. Since reporting was a central focus of this study, the limited
variance had a significant impact on the interpretation of the models. Correlations are not
as affected by limited variance as SEM. Thus, in interpreting the data it is helpful to
examine the correlations (Table 6) as several of the expected relations that were not

Significant in the models were supported by significant correlations.

95

It is also important to consider the concept of measurement error in understanding
some of the results, in particular, the measurement of sex-role stereotypes. There were
not any recently developed measures of sex-role stereotypes and this issue was not able to
be successfully resolved. Notably, in this study there was not a relationship between sex-
role attitudes and reporting, in contrast to many previous studies. The fact that this study
did not find a relationship may be an artifact of the weaknesses of the sex-role measures
used. The mean scores of those measures in this study were significantly different than
the means of the measures from previous studies (e.g., Burt, 1980; Spence & Helmreich,
1978). Given these differences in means on sex-role measures, perhaps future research
should examine the usefulness of these tools and work on creating measures that more
accurately depict the subtle sex-role stereotypes that are still pervasive in today’s society.

An additional limitation is that this study only assessed heterosexual assault of
women by men. Waterman, Dawson and Bologna (1989) found that sexual assault is a
problem for a considerable number of homosexuals, with lesbians having higher rates of
sexual assault experience than gay men (e.g., Waterman, Dawson, & Bologna, 1989).
There is no published research on factors affecting reporting same-sex sexual assaults.
There may be additional factors related to reporting in this group, particularly related to
the widespread homophobia in many areas of society. In addition, it is unclear how
sexual orientation may relate to reporting, as factors such as blame have been found to be
related to sexual orientation. Ford, Liwag-McLamb & Foley (1998) found that women
were perceived to be more at fault if they were heterosexual.

The generalizability of the sample is also limited in that it examined only college

women, and therefore important factors such as age and educational level were not

96

assessed. A college sample was chosen because of the reported high rates of sexual
assault in this age group (e.g., Koss et.al, 1987; Muehlenhard & Linton, 1987), however
future studies should examine reporting in other samples because college campuses are
distinctly different than many aspects of society. For instance, variables such as the
influence of Greek life and generally high rates of alcohol use on college campuses need
to be considered before these findings should be generalized. However, since 67 % of
female high school graduates in America attend at least one year of college, this study has
significance for a large part of society for a certain point in their lives (U .S. Department
of Commerce, 1998).

In addition, due to methodological problems this study was unable to accurately
examine the relationships between these variables and the time since the assault. Future
studies may wish to examine how women’s perceptions of the experience change over
time. Some women may consider themselves assaulted immediately after the event and
later decide they were not assaulted, given societal pressures and others’ opinions to the
contrary. On the other hand, some women may not initially define the experience as an
assault but may do so later. Future studies may wish to examine how women’s
perceptions of the experience change over time.

Directions for Future Research

This study showed that despite the devastating effects of sexual assault, few
women tell anyone other than a friend and therefore do not receive services for sexual
assault related distress from mental health professionals. Although the model for
survivors of sexual assault did not minimize, examination of the correlations suggest a

relationship between reporting and acknowledging the experience as an assault and type

97

of force and injury. These findings suggest that many women still define sexual assault
more narrowly than the legal term and still feel uncomfortable discussing non-
stereotypical assaults with others. Studies that examine the process by which women
come to explain and define their experiences of sexual assault are needed. In addition, it
appears that one way of intervening is by helping women to believe that sexual assault
can take many different forms and to increase their comfort in talking about it. In
addition to continuing to provide counseling services and anonymous hotlines for
women, resources may also be wisely spent on sexual assault educational programs that
focus on helping females understand what constitutes sexual assault and to be more
comfortable talking about it. Furthermore given this study’s finding that most women
only tell a fiiend, it is important that through programs (i.e., lectures, workshOps) or
books that women and men learn how to respond to someone’s disclosure to sexual
assault in a supportive way. Books about how to support someone who is sexually
assaulted should be more readily available and discussed. Given that most women in this
study only told a fiiend, it is important that fiiends have the information available to them
about how to support a survivor of sexual assault. To help students understand rape and
how to help survivors of rape colleges should consider implementing campus wide
psychoeducational programs about rape and survivors of rape.

Results from this study point to several possible directions for future research,
including expanding upon the existing models. Much of the research on sexual assault
examines hypothetical situations and as this study suggests there is a large discrepancy
between what women believe they would do and what most women actually do.

Longitudinal studies to examine how sexual assault experiences change women’s

98

perceptions and schemas would be useful. It seems that current theories that focus on
reporting are not accurately assessing the traumatic effects of sexual assault on women’s
schemas about how they feel and how others will behave.

This study suggests that traditional data collection and statistical procedures are
not able to assess models of reporting behavior. It is important for researchers to consider
that reporting may often be a dichotomous variable when they are considering different
methodological and statistical procedures. In addition, the wide variety of sexual assault
scenarios women experienced suggest that close-ended questions may lead to the
omission of important variables. Close-ended questions do not allow the researcher or
clinician to gain an accurate understanding of an individual woman’s experience.
Research using interviews and other types of narrative reports may be more useful.

Exploratory research using interviews with survivors of sexual assault is needed
to learn about what variables they believe contributed to their decisions whether or not to
share their experience with others. Several of the women in this study wrote about or
spoke about other variables such as level of psychological trauma and how the first
person they told reacted. Several women in this study noted that although they had not
been physically injured, they had been psychologically injured and future studies should
examine how that relates to reporting. Given that many women only told a fiiend, it
might be helpful to examine how their friend reacted. Perhaps women only tell their best
fiiends because their friends are very supportive and after receiving support they do not
feel they need to tell anyone else. Another hypothesis is that women are discouraged by
their fiiends fiom telling others which may be related to how well the perpetrator is

known by their friend. While this study did examine the relationship between the women

99

and the perpetrator, it did not examine how much the man was a part of their social circle
and how peer pressure/peer influence could affect their decision whether or not to report
the experience.

An additional variable that could be examined is shame, as shame may make it
difficult to report sexual assault. Women may view the assault as a sign of failure and a
source of embarrassment and shame. In fact, one study found that women often list
shame and embarrassment as one reason they did not report the assault (Gartner &
Macmillan, 1995).

This study was conducted not only to assess relationships between the observed
variables but also to identify possible areas of intervention for survivors of sexual assault
and to develop additional ideas about how to help improve the likelihood that women will
seek help for sexual assault. Since correlations were found between reporting and
acknowledging the experience, physical injury and force, providing information to
women about sexual assault through educational programs discussed earlier would be one
way of increasing reporting.

This study suggests that many women blame themselves to some extent for the
assault. For some women self-blame may serve a purpose, in that it allows women to
make sense of what happened to them (i.e., it was their fault), despite the fact that blame
is not associated with good adjustment (Abbey, 1987; Meyer & Taylor, 1986). Future
studies should examine how blame may be used to meet a need (i.e., maintain belief in a
“just world”) and how support providers can help women meet these needs in a less

destructive manner. In sum, not only can future research expand upon the models tested

100

in this study, but the results can also be interpreted so that they can be utilized in applied
settings, helping with prevention and in intervening with survivors of sexual assault.

In addition to examining why assaulted women do not report their experiences, it
is also imperative for future studies to examine what types of help and/or responses
sexual assault survivors want. Perhaps one reason why women are not seeking support is
that their initial attempt (i.e., telling a friend) was not successful or they feel the type of
support they need is not available. Studies that examine what types of support women
desire are needed because in order to increase the likelihood that women will report, it is
important to ensure that the types of support they desire are available.

In conclusion, this study examined a model of factors affecting women’s
decisions to talk about sexual assault. Although the results of this study should be
considered preliminary, they do provide suggestions for future research in the prevention
of sexual assault and intervention with survivors of sexual assault. Specifically, this
study suggests that additional research on factors related to reporting for actual survivors
of sexual assault and an examination of attempts to increase reporting are needed and

would be a valuable area for future research.

lOl

 

 

33835 _BoEEom

     

 

>388 8
nouguoz

 

 

 

825980 ”Sana-82

 

/ ”—050 MO

maoufioomxm

 

 

 

 

aou§3>m
088:5

 

«eccafioo 339‘s

 

 

88880
333.com

 

 

 

access @333 c 322 £35m .l 2:5

102

      

vacuum
265502

 

meowfiooaxm
080890

 

 

SEM-snow

 

25.5

.3383 388m tone 8 «on .8 3533 8038c @5895 3 wquanmbuoo E38.“ mo .038 383% .N 3:me

103

coco—o5

         

«dog-.35
no 383

 

quasars—om
.«o 09C-

 

      

meow-889nm

mo_om-xom
08850

DEG—m

.8 338mm 338 toga 2:03 .65 £3 Bone meofioov $8895 8 mean—£580 E88“ me $88 Ego-_- .m cam-m

104

fi @ fl

_ 253-832 _ $§§o9§oz_ _ bmfifl-§oz _

. >88-un

 

  

 

 

wcozfioomxm
2:850

32:3

.2 52%.? E mm 38 nowagofipu .m 2953 “333 35x8 3386093
a Son“ :8 339$ >05 23% >58 26: Soon Eommmoov £580? 8 wqunflbaoo 380$ .«o 332 .v 853

         

105

4 4 4_

waoEAEoZL _basfifio?§oz_ >_Eam-§oz

. 8.
cm 4c.

20:88qu
08830

 

       

 

wcodem
oozoqAFm @

Vb.

83¢-me

 

 

   

   
  

“Vania—om
\. ~.
0235.2an

 

o;

.2 5383‘ E a 38 :ofimFan—ad £959 8 “383
333. 30305093 a tone 2:95 >06 Eon? Sons muofifiov @8825 3 wgaflbcoo £88m mo 6on .m onE

106

4 4 4

_‘uaoEAEoZ 4 T338800-§oz_ :Enfléfioz _

 

       

mcoufiooqu
oEooSO

32:3
o§_m.:om

..§a§s s 0325 .< 2959 =54 433. a.
38598 hofi 53m 28 ~35 Bacon baa Boa Scan 9838“. $8803 8 95:55:00 E38.“ .«o 682 .0 832m

107

4 4 _4

EoEéboZ __bE:EEoo-:bo£ baaméboz _

hm.

 

 

hm.

mcousnvomxm
08880

   
 
 

«a

.9 H4
floaom

@ g . 46¢-me 3. % a

. a. E @

ésfié

  

o _ 255.2% 9
camaaom e

o§_m

mo.

.9308 REH< 2959 3 2:83
13:8 33052;: a tone 2:25 >2: 80:3 “:35 £838: 98:83 8 mafia—3:8 muoaofl mo 3on .5 8:me

108

 

 

253-88 Z _ _bgfifioo-§o~m_ fizfiafigoz

 

 

   
     
          

028

age—3050.4
QEmaosm—om

.«o ugh mfimnouflom

8%
833

 
   

80205
mo Bang

 

tomom

  

mo—OMIXDW

  

mcoufiooaxm
2:830

44-4 .\ E
15$-me _ _ oozoqum _ oEmEfiom

.2 “fig—5&3. 3 mm 068 833559: J. 2959 8 gamma Hana
mo 855qu :2: tone.— Ev >05 on? Bonn maommmoov $5803 8 $5358: flouofi mo _ouoE _Soh .uw an

        

 

 

 

109

4_ 4 _4

flucoum-§oz _bgfifiooéqu— b.8mmupcoz _

 

 

        
          

EHESm

3038896—
288:0

>6:on
ofifimfiom

.9802 Rafi .m 295% 2:88 338 30305093
a Son: :8 3:03 >05 030a baa Bo: “:3: 80653 95895 3 mafia—5:8 380$ mo Eco: .m 3:me

110

_4 4 _4

205-582 __bE:EEoo-:boZ_ 3:88-88 Z _

mm. in
on.

maoufloomxm
08830

 

         
   

00:03-:xm @

.2088 :8: .m 2:8di 3 €88
388m 33358.3 a tone: 2:03 :05 80:3 «:3: £838: 9:080? 8 wag—£580 888% Mo 3:02 .2 PSwE

111

 

j warm-€02 1— _EEEEOo-Equ— _ bmfifl-Eoz

9\

      

 

    

 

088

    

:Emcocflom
.3 09C-

 

:Emcouflom
Ommm

moan-Em

 

£5333me
2:830

l.\_

_ :55- axm ~02??me

    

MERE-bow

icfifl- mxm

 

 

2 52%,? a a $8 8:432:94 .< 2:59
8 3:88 Exam :88 2:03 :2: on? Son: £5660: 9:0an 8 953580 €28.“— mo 3:2: 35:02 ”2 8:9:

  
     

112

Abbreviation
AWS

SRSQ
Exp-police
Exp-friend
Exp-mhw
Exp-family
Imfamily
Norm-Family
Norm-Friend
Norm-Community
Tellhow

Tellwho

APPENDIX A

List of Variable Names and their Abbreviations

Variable Name

Attitudes Towards Women Scale

Sex Role Stereotypes Questionnaire
Expectations for Reporting to Police
Expectations for Reporting to Friend
Expectations for Reporting to Mental Health Worker
Expectations for Reporting to Family
Immediate Family

Normative Beliefs about Family Members
Normative Beliefs about Friends
Normative Beliefs about Community
Number of People Told

Who was Reported to

113

APPENDIX B

Demographic Information: Below are a few questions about your background. For each
question, please identify the correct answer on the scantron sheet.

1. How old are you?
A. 17 years old or younger
B. 18 years old
C. 19 years old
D. 20 years old
E. 21 years old
F. 22 years old
G. 23 years old or older

2. What is your ethnicity?
A. African-American
B. Asian
Caucasian
Hispanic
Native American
Other (please specify)

FWPO

 

3. What year are you in college?
A. First year
B. Second year
C. Third year
D. Fourth Year
E. Fifth Year
F. Sixth year or beyond

4. Are you in a sorority?
A. Yes
B. No

5. What is your parents combined income level?

A. Less than 10,000
B. 10,000 to 19,999
C. 20,000 to 29,999
D. 30,000 to 39,999
E. 40,000 to 49,999
F. 50,000 to 59,999
G. 60,000 to 69,999
H. 70,000 to 79,999
1. 80,000 or higher

114

 

Sexual Experiences Questionnaire--

Directions: Below are questions about your sexual experiences with men. Since this
study is specifically examining heterosexual experiences, please answer the following
questions about your experiences with men only. If you have had similar experiences
with women, please feel free to write in on the questionnaire or on the blank pages
provided about these experiences.

In these questions the legal definition of sexual intercourse should be used. That
is, sexual intercourse is defined as "penetration, no matter how slight, of one person by
another, ejaculation is not required". In addition, it includes contact between the sex
organs of one person and the mouth or anus of another. Penetration of the vagina or anus
may occur with any object. At this time, I would like to remind you that your
participation is voluntary and any of the questions that are objectionable to you can be
skipped. I would also like to remind you that all questions are confidential and there is
no way that anyone can link your answers to your name.

Please indicate whether you have experienced any of the following by circling the
appropriate answer.

6. Have you ever been sexually assaulted by a man?
A. No B. Yes

7. Have you ever been raped by a man?
A. No B. Yes

 

Did you ever engage in sex play (fondling, kissing, or petting but NOT intercourse) when
you didn't want to because:

8. You were overwhelmed by his continual arguments and
pressure?

A. No B. Yes

9. He threatened to end the relationship?
A. No B. Yes

10. You were under the influence of drugs or alcohol?
A. No B. Yes

1 1. He threatened to use physical force (twisting your arm, holding you down, etc.)?
A. No B. Yes

12. He used physical force (twisting your arm, holding you down, etc.)?
A. No B. Yes

115

Did you ever engage in intercourse (oral, vaginal or anal) when you didn't want to
because:

13. You were overwhelmed by his continual arguments and pressure?
A. No B. Yes

14. He threatened to end the relationship?
A. No B. Yes

15. You were under the influence of drugs or alcohol?
A. No B. Yes

16. He threatened to use physical force (twisting your arm, holding you down, etc.)?
A. No B. Yes

17. He used physical force (twisting your arm, holding you down, etc.)?
A. No B. Yes ‘

** Now I would like you to review your answers to the previous questions (questions 6
though 17). If you answered yes to ANY of questions, please answer questions 18 to 65.
If you answered no to ALL of the questions, please skip questions 18 to 65 and go to
question 66.

18. Questions 19 through 65 ask about more questions about feelings, thoughts and
reactions your sexual experiences previously asked about. The questions ask you about
only one incident. Please pick the incident from questions 8 through 15 that was most
upsetting for you and answer questions 19 through 65 with that incident in mind. Please
identify below which incident was most upsetting for you.

A. Question 8 or 9

B. Question 10

C. Question 11

D. Question 12

E. Question 13 or 14

F. Question 15

G. Question 16

H. Question 17

*** How old where you when this happened? (please write in your
answer)

 

116

19. What was your relationship like with this person?
A. Did not know the person
B. Causal Acquaintance
C. Nonromantic friend
D. Casually dating
E. Exclusively dating this person
F. Long term relationship

20. How much was the man responsible for what happened?
A. Not at all
B. A little
C. Somewhat
D. A lot

21. How much were you responsible for what happened?
A. Not at all
B. A little
C. Somewhat
D. A lot

22. Who did you tell about the incident (check all that apply)?
A. No one
B. A friend
C. An immediate family member (sibling or parent)
D. An extended family member (aunt, uncle, grandparent, etc.)
E. A resident advisor or sorority house advisor
F. A mental health worker (counselor/therapist/rape crisis worker)
G. A doctor or nurse
H. The police
I. Clergy
J. Other

23. Indicate below how severe the injuries you sustained as a result of the experience
reported above (question 13) were.
A. No injury or pain sustained
B. Mild injury (includes redness, lingering pain or soreness, small cuts, lumps or
bruises, etc.)
C. Moderate injury (includes cuts requiring stitches, black eye, extensive bruising
and swelling, fractured nose or finger, strained muscles or joints, visit to
emergency room or doctor, etc.)
D. Severe injury (includes required hospitalization, broken arm or leg, permanent
disability, unconsciousness, required surgery, termination of a pregnancy, knife or
gunshot wound, etc.)

117

Questions 24-67 ask about who you think other people would want you to tell your
experience to and your expectations of what would happen if you told other people about
your experience. If you have told someone about your experience, please try to think
about how you felt before you told someone and what at that time you thought people
wanted you to do and what you thought might happen.

If you did tell someone and would like to share what their response(s) was (were),
please do so below in the space below. However, you would prefer not to write about
how others responded, just skip to the next page.

118

 

"' “How much do you think your parents or other family members would want you to
report your experience to: (answer each question using the following scale):

A. Not at all

B. A little bit

C. Somewhat

D. A lot

24. a Friend

25. an immediate family member

26. an extended family member

27. a resident assistant

28. A mental health worker (counselor/therapist/rape crisis worker)
29. A doctor or nurse

30. The police

31. The clergy

32. How much do you want to comply with your family’s wishes about who you should
tell about your experience.

A. Not at all C. Somewhat

B. A little bit D. A lot

"*How much do you think your friends would want you to report your experience to:
(answer each question using the following scale):

A. Not at all

B. A little bit

C. Somewhat

D. A lot

33. a Friend

34. an immediate family member

35. an extended family member

36. a resident assistant

37. A mental health worker (counselor/therapist/rape crisis worker)
38. A doctor or nurse

39. The police

40. The clergy

41. How much do you want to comply with your friends wishes about who you should
tell about your experience.

A. Not at all C. Somewhat

B. A little bit D. A lot

119

'"' *How much do you think the community you live in would want you to report your
experience to: (answer each question using the following scale):

A. Not at all

B. A little bit

C. Somewhat

D. A lot

42. a Friend

43. an immediate family member

44. an extended family member

45. a resident assistant

46. A mental health worker (counselor/therapist/rape crisis worker)
47. A doctor or nurse

48. The police

49. The clergy

50. How much do you want to comply with your community’s wishes about who you
should tell about your experience.

A. Not at all C. Somewhat

B. A little bit D. A lot

"The following questions are also referring to the incident you discussed in question .

51. What did you think would happen if you told a fiiend about what happened to you?
Please check all that apply:

A. I would feel calm, safe and better

B. I would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

52. What did you think would happen if you told an immediate family member (sibling,
parent) about what happened to you? Please check all that apply

A. I would feel calm, safe and better

B. I would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

120

 

53. What did you think would happen if you told an extended family member (aunt,
uncle, grandparent, etc.) about what happened to you? Please check all that apply

A. I would feel calm, safe and better

B. I would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

 

54. What did you think would happen if you told a resident advisor about what happened
to you? Please check all that apply

A. I would feel calm, safe and better

B. 1 would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

 

55. What did you think would happen if you told a mental health worker (counselor,
therapist, rape crisis worker) about what happened to you? Please check all that apply

A. I would feel calm, safe and better

B. I would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

 

56. What did you think would happen if you told the police about what happened to you?
Please check all that apply

A. I would feel calm, safe and better

B. I would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would fly to get back at me

H. Any other reasons

 

121

57. What did you think would happen if you told a clergy member about what happened
to you? Please check all that apply

A. I would feel calm, safe and better

B. I would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

** Please indicate how good or bad you think the following possible outcomes of telling
someone about your experience are. Please use the following scale:

A. Very bad D. A little good
B. Somewhat bad E. Somewhat good
C. A little bad F. Very good

59. I would feel calm, safe and better

60. I would get adequate medical attention

61. I would be given the necessary tests to detect pregnancy, VD, etc.

62. A trial would occur and I would have to testify

63. The necessary evidence that could be used in court would be obtained.
64. I would be treated as an immoral person

65. The man who did this to me would try to get back at me

** If you answered questions 18 through 65 skip to Question 112** If you did not
answer questions 18 through 65 please answer questions 66 to the end.

122

 

These questions ask about how you would feel and what you would do if you had sex
with a man even though you did not want to because he used physical force (holding you
down, twisting your arm, etc.).

66. How much would you think the man was responsible for what happened?
A. Not at all
B. A little
C. Somewhat
D. A lot

67. How much would you think you were responsible for what happened?
A. Not at all
B. A little
C. Somewhat
D. A lot

68. Who would you tell about the incident (check all that apply)?
A. No one
B. A friend
C. An immediate family member (sibling or parent)
D. An extended family member (aunt, uncle, grandparent, etc.)
E. A resident advisor or sorority house advisor
F. A mental health worker (counselor/therapist/rape crisis worker)
G. A doctor or nurse
H. The police
I. Clergy
J. Other

"*How much do you think your parents or other family members would want you to
report this experience to: (answer each question using the following scale-

A. Not at all C. Somewhat

B. A little bit D. A lot

69. a friend

70. an immediate family member

71. an extended family member

72. a resident assistant

73. A mental health worker (counselor/firempist/rape crisis worker)
74. A doctor or nurse

75. The police

76. The clergy

123

77. How much do you want to comply with your family’s wishes about who you should
tell about your experience.

A. Not at all C. Somewhat

B. A little bit D. A lot

"*How much do you think your fiiends would want you to report your experience to:
(answer each question using the following scale):

A. Not at all

B. A little bit

C. Somewhat

D. A lot

78. a Friend

79. an immediate family member

80. an extended family member

81. a resident assistant

82. A mental health worker (counselor/therapist/rape crisis worker)
83. A doctor or nurse

84. The police

85. The clergy

86. How much do you want to comply with your friends wishes about who you should
tell about your experience.

A. Not at all C. Somewhat

B. A little bit D. A lot

"*How much do you think the community you live in would want you to report your
experience to: (answer each question using the following scale):

A. Not at all

B. A little bit

C. Somewhat

D. A lot

87. a Friend

88. an immediate family member

89. an extended family member

90. a resident assistant

91. A mental health worker (counselor/therapist/rape crisis worker)
92. A doctor or nurse

93. The police

94. The clergy

124

 

95. How much do you want to comply with your community’s wishes about who you
should tell about your experience.

A. Not at all C. Somewhat

B. A little bit D. A lot

The following questions also ask you to think about what would you would do if you had
sex with a man, even though you did not want to, because he used physical force.

96. What did you think would happen if you told a fiiend about what happened to you?
Please check all that apply

A. I would feel calm, safe and better

B. I would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

 

97. How much do you want to comply with your families wishes about who you would
tell about this experience to.

A. Not at all C. Somewhat

B. A little bit D. A lot

98. What did you think would happen if you told a friend about what happened to you?
Please check all that apply

A. I would feel calm, safe and better

B. I would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

 

125

99. What did you think would happen if you told an immediate family member (sibling,
parent) about what happened to you? Please check all that apply

A. I would feel calm, safe and better

B. I would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and 1 would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

 

100. What did you think would happen if you told an extended family member (aunt,
uncle, grandparent, etc.) about what happened to you? Please check all that apply

A. I would feel calm, safe and better

B. I would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

 

101 . What did you think would happen if you told a resident advisor about what
happened to you? Please check all that apply

A. I would feel calm, safe and better

B. I would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

 

102. What did you think would happen if you told a mental health worker (counselor,
therapist, rape crisis worker) about what happened to you? Please check all that apply

A. I would feel calm, safe and better

B. 1 would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

 

126

 

103. What did you think would happen if you told the police about what happened to
you? Please check all that apply

A. I would feel calm, safe and better

B. I would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

 

104. What did you think would happen if you told a clergy member about what happened
to you? Please check all that apply

A. I would feel calm, safe and better

B. I would get adequate medical attention

C. I would be given the necessary tests to detect pregnancy, VD, etc.,

D. A trial would occur and I would have to testify

E. The necessary evidence that could be used in court would be obtained.

F. I would be treated as an immoral person

G. The man who did this to me would try to get back at me

H. Any other reasons

 

** Please indicate how good or bad you think the following possible outcomes of telling
someone about having sex when you did not want to because a man used physical force.
Please use the following scale:

A. Very bad D. A little good
B. Somewhat bad E. Somewhat good
C. A little bad F. Very good

105. I would feel calm, safe and better

106. I would get adequate medical attention

107. I would be given the necessary tests to detect pregnancy, VD, etc.,
108. A trial would occur and I would have to testify

109. The necessary evidence that could be used in court would be obtained.
110. I would be treated as an immoral person

1 l 1. The man who did this to me would try to get back at me

Please go to the next page

127

Please answer the following questions about some of your attitudes. Please check
whether you agree strongly, agree mildly, disagree mildly, disagree strongly.

112. Swearing and obscenity are more repulsive in the speech of a woman than a man.
A. Agree Strongly C. Disagree Mildly
B. Agree Mildly D. Disagree Strongly

113. Under modern economic conditions, with women active outside the home, men
should share in household tasks such as washing dishes and doing laundry.

 

A. Agree Strongly C. Disagree Mildly
B. Agree Mildly D. Disagree Strongly
P
114. It is insulting to women to have the "obey" clause still in
the marriage service.
A. Agree Strongly C. Disagree Mildly
B. Agree Mildly D. Disagree Strongly
115. A woman should be as free as a man to propose marriage. i
A. Agree Strongly C. Disagree Mildly
B. Agree Mildly D. Disagree Strongly

116. Women should worry less about their rights and more about becoming good wives
and mothers.

A. Agree Strongly C. Disagree Mildly

B. Agree Mildly D. Disagree Strongly

117. Women earning as much as their dates should bear equally the expense when they
go out together.

A. Agree Strongly C. Disagree Mildly

B. Agree Mildly D. Disagree Strongly

118. Women should assume their rightful place in business and all the professions along
with men.

A. Agree Strongly C. Disagree Mildly

B. Agree Mildly D. Disagree Strongly

119. A woman should not expect to go to exactly the same places or to have quite the
same freedom of action as a man.

A. Agree Strongly C. Disagree Mildly

B. Agree Mildly D. Disagree Strongly

128

120. Sons in a family should be given more encouragement to go to college than

daughters.

A. Agree Strongly C. Disagree Mildly

B. Agree Mildly D. Disagree Strongly

121. It is ridiculous for a woman to run a locomotive and for a man to darn socks.

A. Agree Strongly C. Disagree Mildly

B. Agree Mildly D. Disagree Strongly

122. In general, the father should have greater authority than the mother in the bringing
up of children.

A. Agree Strongly C. Disagree Mildly

B. Agree Mildly D. Disagree Strongly

123. The intellectual leadership of a community should be largely in the hands of men
A. Agree Strongly C. Disagree Mildly

B. Agree Mildly D. Disagree Strongly

124. Economic and social freedom is worth far more to women than acceptance of the
ideal femininity, which has been set up by men.

A. Agree Strongly C. Disagree Mildly

B. Agree Mildly D. Disagree Strongly

125. There are may jobs in which men should be given preference over women in being
hired or promoted.

A. Agree Strongly C. Disagree Mildly

B. Agree Mildly D. Disagree Strongly

126. Women should be given equal opportunity with men for apprenticeship in the
various trades.

A. Agree Strongly C. Disagree Mildly

B. Agree Mildly D. Disagree Strongly

127. A man should fight when the woman he’s with is insulted by another man.
A. Agree A lot E. Disagree A little
B. Agree Somewhat F. Disagree Somewhat
C. Agree A little G. Disagree A lot

D. Neither Agree nor Disagree

128. It is acceptable for the woman to pay for the date.

A. Agree A lot E. Disagree A little
B. Agree Somewhat F. Disagree Somewhat
C. Agree A little G. Disagree A lot

D. Neither Agree nor Disagree

129

129. A woman should be a virgin when she marries.

A. Agree A lot E. Disagree A little
B. Agree Somewhat F. Disagree Somewhat
C. Agree A little G. Disagree A lot

D. Neither Agree nor Disagree

130. There is something wrong with a woman who doesn’t want to marry and raise a
family.

 

A. Agree A lot E. Disagree A little

B. Agree Somewhat F. Disagree Somewhat

C. Agree A little G. Disagree A lot

D. Neither Agree nor Disagree ‘4
131. A wife should never contradict her husband in public.

A. Agree A lot E. Disagree A little

B. Agree Somewhat F. Disagree Somewhat

C. Agree A little G. Disagree A lot

D. Neither Agree nor Disagree 5
132. It is better for a woman to use her feminine charm to get what she wants rather than
ask for it outright.

A. Agree A lot E. Disagree A little

B. Agree Somewhat F. Disagree Somewhat

C. Agree A little G. Disagree A lot

D. Neither Agree nor Disagree

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

A. Agree A lot E. Disagree A little
B. Agree Somewhat F. Disagree Somewhat
C. Agree A little G. Disagree A lot

D. Neither Agree nor Disagree

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

A. Agree A lot E. Disagree A little
B. Agree Somewhat F. Disagree Somewhat
C. Agree A little G. Disagree A lot

D. Neither Agree nor Disagree

130

APPENDIX C

DEPARTMENTAL RESEARCH CONSENT FORM
Michigan State University

You are invited to participate in a study of college women’s sexual experiences.
This study examines what types of experiences women have had and who, if anyone, they
tell about their experiences. Participation involves filling out an anonymous
questionnaire. It includes questions about your sexual experiences, who you have talked
to about these experiences and your views toward women and men. All of your answers
to these questions will be kept completely anonymous and confidential. Neither your
name nor any other identifying information appears on the questionnaire to ensure
anonymity. No one will ever be able to identify which questionnaire you completed.

Participation in this study is completely voluntary. Some of the questions are
personal, and perhaps, upsetting events that have occurred in your life. If you decide to
participate, you are free to skip any question that you would rather not answer without
being penalized. Also, if you want to stop at any time, you may do so without being
penalized. If you are enrolled in a Psychology class that accepts research credits, you will
receive 1 research credit for participating in this study.

If you have any questions about this study, or if you want to discuss any feelings
or reactions that you have related to the questionnaire or specific experiences it focuses
on, you can reach me, Jennifer Paul, at the Michigan State University Psychological
Clinic, 355-9564. You can also reach me through email at pauljenn@pilot.msu.edu.

If you are willing to participate in this study, please read the rest of this page and
then begin the questionnaires. Thank you.

1. I have freely consented to take part in a scientific study being conducted by: Jennifer
Paul, M.A. Under the supervision of : Altyia Levendosky, Ph.D.

2. I have read and understood the information explaining this study in the above
paragraph-

3. I understand that my participation in this study is completely voluntary and that I am
free to skip any questions that I do not wish to answer or discontinue my participation at
any time without being penalized.

4. My participation in the study will take approximately 30 minutes.
5. I understand that my answers to the questions will be confidential and no one,
including the researcher, will be able to link my responses to my name or any other

identifiable information.

6. I understand that there may be no direct benefits to me as a result of my participation in
this study.

131

7. I understand that I can request additional information about this study or finding from
this study afier my participation is completed.

8. I am at least 18 years old.

9. I understand that by completing these questionnaires and returning them to the
investigator that I have consented to participate in this research study.

132

APPENDIX D
ABOUT THIS STUDY...

You have participated in a research study that is examining factors that may
contribute to women’s decision wether or not to report being sexually assaulted.
Research has shown that many women do not tell others about their experiences of sexual
assault. This study examined how women’s definition of sexual assault, their views
about roles of men and women, their expectations of telling someone about sexual
assault, and what they believe other people would want them to do all contribute to their
decision to tell or not tell anyone about being sexually assaulted. Research has shown
that for many victims, seeking support can help them cope with some of the
psychological symptoms they may be experiencing as a result of being sexually assaulted.
I am interested in why women decide whether they will talk to someone about being
sexually assaulted so that barriers to telling others can be decreased. This research may
also be helpful in identifying both preventive programs and more effective ways of
helping sexual assault victims. If you are interested in the findings of this study, please
feel free to contact the principal investigator (Jennifer Paul, M.A.,phone 355-9564; email:
pauljenn@pilot.msu.edu). The results should be available in May or June of 1998.
Please do not share this information with anyone who has not completed the study, as it
may affect future participant’s responses.

At the beginning of the study, you were invited to attend a workshop on women’s
issues. The workshop will focus specifically on how to prevent sexual assault and what
you can do if you are sexually assaulted.

If you have been sexually assaulted and would like to talk to someone about your
experience, here are some resources that are available:

MSU Counseling Center
Student Services Building 355-8270
Olin Health Center 355-2310
MSU Sexual Assault Program 355-8270 (Business Hours)
24 hours 372-8460 (Crisis Line)
The Listening Ear 337-1717
24 hours (telephone only)
Community Mental Health 346-8300
Emergency Services 372-8460

If you have any questions or comments regarding this research, please feel free to contact
me at any time. Thank you for participating.

133

REFERENCES

Amick, A.E. & Calhoun, KS (1987). Resistance to Sexual Aggression:

 

Personality, Attitudinal, and Situational Factors. Archives of Sexual Behavior 16 153-
163.

Allison, J.A. & Wrightsman, LS. (1993). Rape: The misunderstood crime.
Newbury Park, CA: Sage.

Alksnis, C., Desmarais, S. & Wood, E. (1996). Gender differences in scripts for

different types of dates. Sex Roles 34 321-336.

 

Atkeson, B.M., Calhoun, K.S., Resick, P.A., & Ellis, EM. (1982). Survivors of
rape: Repeated assessment of depressive symptoms. Journal of Consulting and Clinical
Psychology, 50, 96-102.

Ajzen, I. & Fishbein, M. (1972). Attitudes and normative beliefs as factors
influencing behavioral intentions. Journal of Personalig and Social Psychology, 21, 1-9.

Bachman, R. (1993). Predicting the reporting of rape victimizations: Have rape

reforms made a difference? Criminal Justice and Behavior 20, 254-270.

 

Belenky, M.F., Clinchy, B.M., Goldberger, N.R., & Tarule, J .M. ( 1986).

Women’s ways of knowing: the development of self, voig, and mind. Basicbooks: New
York. ‘

Biaggio, M.K., Brownell, A., & Watts, BL. (1991). Reporting and Seeking

 

Support by Victims of Sexual Offenses. Journal of Offender Rehabilitation 17 33-42.
Braswell, Michael C. Correctional treatment and the human spirit: A focus on

relationship. Federal Probation 53, 49-60.

 

134

Bridges, J .S. (1991). Perceptions of date and stranger rape: A difference in sex

role expectations and rape-supportive beliefs. Sex Roles 24 291-307.

 

Bridges, J .S. & McGrail, CA. (1989). Attributions of responsibility for date and

stranger rape. Sex Roles 21 273-286.

 

Brooks, L. & Perot, AR. (1991). Reporting sexual harassment: exploring a

predictive model. Psychology of Women Quarterly, 15, 31-47.
Bourque, LB. (1989). Defining Ram. Durham, NC: Duke University Press.

Braswell, L. (1989). Quest for resmt: A guide to healing for survivors of ram.
Ventura, CA: Pathfinder.

Brownmiller, S. (1975). Against our will: Men, women and rape. New York:
Simon & Shuster.

Burt, MR. (1980). Cultural myths and supports for rape. Journal of Personalifl
and Social Pachology, 38, 217-230.

Calhoun, K.S., Atkeson, B.M., & Resick, P.A.(1982). A longitudinal examination
of fear reactions in survivors of rape. Journal of Consulting and Clinical Psychology, 29,
665-661.

Check, J .V.P. & Malamuth, NM. (1983). Sex role stereotyping and reactions to

depictions of stranger versus acquaintance rape. Journal of Personality and Social

Psychology, 45, 344-356.
Cohen, S. & Willis, TA. (1985). Stress, social support and the buffering

hypothesis. Psychological Bulletin, 96 310-357.

135

 

Coller, S.A. & Resick. RA. (1987). Women's attributions of responsibility for

date rape: The influence of empathy and sex-role stereotyping. Violence and Survivors

 

2, 115-125.
Copenhaver, S. & Grauerholz, E. (1991). Sexual victimization among sorority
women: exploring the link between sexual violence and institutional practices. §C_X_

Roles 24 31-41.

 

Davidson, A. & Jaccard, J. (1975). Population psychology: A new look at an old
problem. Journal of Personality and Social Psychology, 31, 1073-1082.

DeKeseredy, W. & Kelly, K. (1993). The incidence and prevalence of woman
abuse in Canadian university and college dating relationships. Canadian Journal of
Sociology,18, 137-159.

DeKeseredy , W. & Kelly, K. (1996). Sexual abuse in Canadian university and
college dating relationships: The contribution of male peer support. Journal of Family
Violence, 10, 41-53.

Dobash, R.E. & Dobash, RP. (1979). Violence against wives: A case against
patriarchy. New York: Free Press

Dutton, R.E. & Hemphill, K.J. (1992). Patterns of socially desirable responding

among perpetrators and survivors of wife assault. Violence and Survivors 7 29-39.

 

Espinosa, C. & Cardoza, D. (1995). An examination of factors predicting the
reporting of sexual harassment.
Estrich, S. (1987). Real ram. Cambridge, MA: Harvard University Press.

Feldman-Summers, S. & Ashworth, CD. (1981). Factors related to intentions to

report a rape. Journal of Social Issues, 3714), 53-70.

136

Feldman-Summers, S. & Norris, J. (1984). Differences between rape survivors
who report and those who do not report to a public agency. Journal of Applied Social
Psychology, 14, 562-573.

Feldman-Summers, S. & Palmer, G. (1980). Factors related to intentions to report

a rape. Journal of Social Issues 37 53-70.

 

Fishbein, M. (1963). An investigation of the relationship between beliefs about

an object and the attitude toward that object. Human Relations 16 233-240.

 

F iske, S.T., & Taylor, SE. (1991). Social Cogr_u'tion. New York: McGraw-Hill,
Inc.

Ford, T.M., Liwag-McLamb, M.G., Foley, LA. (1998). Perceptions of rape
based on sex and sexual orientation of victim. Journal of Social Behavior and

Personality, 13, 253-262.

Frazier, P.A. & Burnett, J .W. (1994). Immediate coping strategies among rape
survivors. Journal of Counseling and Development, 72, 633-639.

Gartner, R. & Macmillan, R. (1995). The effect of victim-offender relationship
on reporting crimes of violence against women. Canadian Journal of Criminology, July,
393-429.

Gavey, N. (1991). Sexual victimization among New Zealand University students.

Journal of Consulting and Clinical Psychology, 59, 464-466.
George, L.K., Winfield, 1., & Blazer, D.G. (1992). Sociocultural factors in sexual

assault: Comparison of two representative samples of women. Journal of Social Issues

 

48105-125.

137

 

 

Golding, J .M., Stein, J.A., Siegel, J .M., Bumam, M.A. & Stein, J.A. (1989).
Social support following sexual assault. Journal of Communig Psychology, 17, 92-107.

Gunn, R. & Minch, C. (1988). Sexual assault: The dilemma of disclosure the

 

question of conviction. University of Manitoba Press.
Gwartney-Gibbs, P.A., Stockard, J ., & Bohmers, S. (1987). Learning courtship
aggression: The influence of parents, peers, and personal experiences. Family

Relations 36 276-282.

 

Hanmer, J ., Radford, J ., & Stanko, EA. (1989). Women, mlicing and male

 

violence: International mmctives. Routledge: London.

Herman, Judith Lewis. Sex offenders: A feminist pgrsmctive. [Chapter] Marshall,
William L. (Ed), Laws, D. R.(Ed), et al. (1990). Handbook of sexual assault: Issues,
theories, and treatment of the offender. Applied clinical psychology. (pp. 177-193). New
York, NY, USA: Plenum Press. xvii, 405 pp.

Hester, M., Kelly, L., & Radford, J. (1996). Womep, violence, and male mwer:
feminist activism, research, and practice. Open University Press: Philadelphia.

Himelein, M.J., Vogel, R.E., & Wachowiak, D.G. (1994). Nonconsensual sexual
experiences in precollege women: Prevalence and Risk Factors. Journal of Counseling
and Development,72, 411-415.

Jensen, I.W. & Gutek, BA. (1982). Attributions and Assignment of
responsibility in sexual harassment. Journal of Social Issues, 38 (4), 121-136.

Kalof, L. (1993). Rape-supportive attitudes and sexual victimization experiences

of sorority and nonsorority Women. Sex Roles 19 767-777.

 

138

Kelly, L. (1988). Surviving sexual violence. Minneapolis: University of
Minnesota Press.

Kilpatrick, D.G., Veronen, L.J. & Resick, PA. (1979). The aftermath of rape:
Recent empirical findings. American Journal of Orthopsychiag, 49, 658-669.

Kilpatrick, D.G., Resick, P.A. &. Veronen, L.J. (1981). Effects of a rape

experience: A longitudinal study. Journal of Social Issues 37 105-122.

 

Kimmerling, R. & Calhoun, KS. (1994). Somatic symptoms, social support, and
treatment seeking among sexual assault survivors. Journal of Consulting and Clinical
Psychology, 62(2),333-340.

Kirkpatrick, C. (193 6). The construction of a belief-pattem scale for measuring
attitudes toward feminism. Journal of Social Psychology, 7, 421-437.

Kirkpatrick, C. & Karrin, E. (1957). Male sexual aggression on a university

campus. American Sociological Review, 22, 52-58.

Koss, MP. (1985). The hidden rape victim: Personality, attitudinal, and
situational characteristics. Psychology of Women Quarterly, 12, 1-24.

Koss, M.P., Gidycz, C., & Wiskniewski, 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., Leonard, K.E., Beezley, D.A., & Oros, C.J. (1985). Nonstranger
sexual aggression: A discriminant analysis of the psychological characteristics of

undetected offenders. Sex Roles 12 981-992.

 

139

 

Koss, M.P. & Oros, C. (1982). Sexual experiences survey: A research instrument
investigation sexual aggression and victimization in a national sample of students in
higher education. Journal of Consulting and Clinical Psychology, 50, 455-457.

Kourany, J .A., Sterba, J .P., & Tong, R (1999). Feminist Philosophies, 2nd

Edition. Upper Saddle River, New Jersey: Prentice Hall.

 

Lane, K.E. & Gwartney-Gibbs, RA. (1985). Violence in the context of dating and

sex. Journal of Family Issues, 6, 45-59.
LaPlante, M. N., McCormick, N. & Brannigan, G.G. (1980). Living the sexual
script: College students' views of influence in sexual encounters. The Journal of Sex

Research, 16(4), 338-355.

Lloyd, SA. (1991). The darkside of courtship: Violence and sexual exploitation.

Family Relations, 40, 14-20.

Malamuth, NM. (1981). Rape Proclivity among males. Journal of Social Issues

 

fl, 138-57.

Malamuth, NM (1986). Predictors of naturalistic sexual aggression. Journal of

Personalifl and Social Psychology, 50, 953-962.

Margolin, L., Miller, M. & Moran, P. (1989). When a kiss is not just a kiss:

Relating violations of consent in kissing to rape myth acceptance. Sex Roles 20 231-

 

243.

Meyer, C.B., & Taylor, SE. (1986). Adjustment to Rape. Journal of Personality

and Social Psychology, 50, 1226-1234.
Muehlenhard, CL. (1988). Misinterpreted dating behaviors and the risk of date

rape. Journal of Social and Clinical Psychology,6, 20—37.

140

 

1i

 

 

Muehlanhard, C.L., Friedman, D.E., & Thomas, OM. (1985). Is date rape

justifiable? Psychology of Women Merly, 9, 297-310.

Muehlenhard, C.L. & Linton, M.A. (1987). Date rape and sexual aggression in

dating situations: Incidence and risk factors. Journal of Cormseling Psychology, 34, 186-

196.

Muehlenhard, C.L., Powch, 1.G., Phelps, J .,L., & Giusti, L.M. (1992). Definitions

of Rape: Scientific and Political Implications. Joumpl of Social Issues 48 23-44.

 

Norris, J. & Feldman-Summers, S. (1981). Factors related to the psychological
impacts of rape on the victim. Journal of Abnormal Psychology, 90, 562-567.
Ogeltree, RJ. (1993). Sexual coercion experience and help-seeking behavior of

college women. Journal of American College Healtl_r, 41, 149-152.

Orcutt, J .D. & Faison, R (1988). Sex-Role attitude change and reporting of rape
victimization, 1973-1985. The Sociological Merly, 29(4), 589-604.

Orlando, J .A. & Koss, M. (1983). The effect of sexual victimization on sexual
satisfaction: A study of the negative association hypothesis. Journal of Abnormal
Psychology, 90, 562-567.

Paulhus, D. (1980). Control of social desirability in personality inventories:
principal-factor deletion. Journal of Research in Personaligy,15, 383-388.

Parisian, J .A. (1990). Toward an Understanding of abuse in the dating
relationships of college students: The relevance of the feminist/social control model.
Michigan State University.

Patton, W. & Mannison, M. (1995). Sexual coercion in high school dating. Sex

Roles 33 447-457.

 

141

 

 

Pirog-Good, M.A. & Stets, A. (1989). inflow
Praeger: New York.

Popiel, D.A. & Susskind, EC. (1985). The impact of rape: Social support as a
moderator of stress. American Journal of Communig Psychology, 13, 645-676.

Resick, P.A., Calhoun, K.S., Atkeson, B.M., & Ellis, EM. (1982). Social
adjustment in survivors of sexual assault. Journal of Consulting and Clinical Psychology,
42, 705-712.

Rose, V.M. (1977). Rape as a social problem: A byproduct of the feminist

movement. Social Problems 25 75-89.

 

Roth, S. & Lebowitz, L. (1988). The experience of sexual trauma. Journal of

Traumatic Stress 1 79-107.

 

 

Russell, DE. (1975). The pplitics of rape: The victims mrsmctive. New York:

Stein and Day.

Russell, DE. (1998). Dangerous relationships: Pornography, misogpy, and ram.

Thousand Oaks, CA:Sage Publications, Inc.

 

Sanday, RR (1981). The socio-cultural context of rape. Journal of Social Issues
3_7, 5-27.

Shotland, L., & Goldstein (1983). Just because she doesn't want to doesn't mean
it's rape: An experimentally based causal model of the perception of rape in a dating
situation. Social Psychology Merly, 446, 220-232.

Skelton, C.A. & Burkhart, RB. (1980). Sexual assault: Determinants of victim

disclosure. Criminal Justice and Behavior 7 229-236.

 

142

 

Smith, R.L. & Bradley, N. (1980). In defense of the Attitudes Toward Women
Scale: An affirmation of validity and reliability. Psychological Remus, 47, 511-522.
Sorenson, S.B., Stein, J.A., Siegel, J .M., Golding, J.M., & Bumam, M.A. (1987).

The prevalence of adults’ sexual assault: The Los Angeles epidemiological catchment

area project. American Journal of Epidemiology, 126, 1154-1164.

Spence, J .T. & Helmreich, R.L. (1978). Masculinig and Femininity: Their
Psychological dimensions, correlates, and antecedents. Austin, TX: University of Texas
Press.

Spence, J .T. & Hahn, ED. (1997). The Attitudes Toward Women Scale and
attitude change in college students. Psychology of Women Quarterly, 21 , 17-33.

Stets, J .E. & Pirog-Good, M.A. (1987). Violence in dating relationships. M
Psychology Qu_a;terly, 50, 237-246.

Tong, R. (1989). Feminist Thought: A Comprehensive Introduction. London:

Routledge.

Ullman, S.E. & Siegel, J .M. (1993). Victim-offender relationship and sexual
assault. Violence and Survivors, 8(2), 121-133.

United States Department of Justice (1997). Bureau of Justice Statistics
sourcebook of criminal justice statistics-1995. The Hindelang Criminal Justice Research
Center, Albany, NY.

Waterman, C.K., Dawson, J .L., & Bologna, M.J. (1989). Sexual coercion in gay
male and lesbian relationships: Predictors and implications for support services. J_o__um_al,

of Sex Research 26 118-124.

 

Warshaw, R. (1988). I never called it ram. New York: Harper & Row.

143

 

 

Weis, K., & Borges, SS. (1973). Victimology and rape: The case of the
legitimate victim. Issues in Criminology, 8, 71-115.
Williams, LS. (1984). The Classic Rape: When do Survivors Report? Social

Problems, 31(4), 459-467.
Wiehe, V.R. & Richards, AL. (1995). Intimate Betrayal: Understanding and

Resmnding to the Trauma of Acquaintance Ram. Sage Publications.
Wilson, K., Faison, R., & Britton, GM. (1983). Cultural Aspects of male sex

aggression. Deviant Behavior 4 241-255.

 

144