JUSTIFYING INJUSTICE: HOW THE CRIMINAL JUSTICE SYSTEM
EXPLAINS ITS RESPONSE TO SEXUAL ASSAULT
By
Jessica Shaw

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Psychology—Doctor of Psychology
2014

ABSTRACT
JUSTIFYING INJUSTICE: HOW THE CRIMINAL JUSTICE SYSTEM
EXPLAINS ITS RESPONSE TO SEXUAL ASSAULT
By
Jessica Shaw

Sexual assault is a pervasive crime leading to myriad detrimental physical and
psychological health problems for the victim. Post-assault, the victim may choose to report the
crime to the criminal justice system which consists of two interrelated, yet distinct stages: the
investigation stage, carried out by law enforcement, and the prosecution stage, carried out by the
prosecutor’s office. However, in practice, most sexual assaults do not transition from the
investigation stage to the prosecution stage and instead fall out of the criminal justice system
process while under the purview of law enforcement. Prior literature suggests that law
enforcement personnel conduct less-than-thorough investigations and do not refer cases to the
prosecutor’s office because they endorse rape myths concerning what qualifies as ‘real’ rape and
who can be raped. Prior research has only documented law enforcement personnel’s attitudes
towards rape via questionnaires and has not examined how rape myths are observed in sexual
assault case investigations.
Guided by social dominance theory, the current project first documented the extent and
types of rape myths observed in police records of sexual assault case investigations (i.e., Study
1). Then, the current project examined the relationships between the different types of rape
myths endorsed with the investigative steps completed, outcomes, and other factors of the sexual
assault investigations (i.e., Study 2) so as to determine if rape myths, are used by law
enforcement personnel to justify their action and inaction in sexual assault case investigations.

The police records of N=248 sexual assault cases corresponding to a random sample of
unsubmitted sexual assault evidence collection kits (SAKs) found in a Detroit police property
storage facility in 2009 were examined as they consisted of cases that had been subjected to the
less-than-thorough investigation, as indicated by each cases’ corresponding unsubmitted SAK.
Qualitative methods, including directed and conventional content analysis, were used to
document the extent and types of rape myths in police records (i.e., Study 1). Quantitative
methods, specifically path analysis, were then used to analyze the relationships between the
different types of rape myths, investigatory effort, and case outcomes within the context of
specific social identity factors (i.e., sex, race, and age) of the victim and perpetrator, and the
number of perpetrators (i.e., Study 2).
Findings reveal that law enforcement personnel drew upon traditional rape myths
regarding what qualifies as ‘real’ rape and who can be raped in order to justify their response to
sexual assault. Or particular interest was the identification of a new strategy used by police, in
which they blamed the victim for law enforcement personnel’s inaction in sexual assault case
investigations. Implications for implementing policy and practice change within the criminal
justice system to improve criminal justice outcomes are discussed, as well as directions for future
research.

ACKNOWLEDGEMENTS

It has been quite the journey, and one that I did not go alone. There are many people that
deserve recognition for their continued support, love, and inspiration along the way.
First and foremost, I would like to thank my friends, family and partner. Each of you
have somehow shaped the person I am today (for better or worse…) and therefore have made an
invaluable contribution to my work. In many ways, this is our achievement. Thank you for
keeping me grounded.
To my committee members, Debi Cain, Rebecca Campbell, Bill Davidson, and Angie
Kennedy. I always thought of each one of our meetings as such a privilege as I was able to get
together several brilliant minds in one room and convince them to dedicate just a bit of their time
to thinking about my research—what an honor. Thank you for your feedback, guidance, and
genuine interest. I have become a stronger researcher and community partner as a result of our
interactions.
To Becki Campbell, wow. It happened. You shaped and facilitated my graduate
experience, allowing me to develop into the scholar I am today. And, it looks like I’m going pro
too. Thank you. I am honored to make the transition from calling you my advisor, to my
colleague and friend.
To Debi Cain and your brilliant colleagues at the Michigan Domestic and Sexual
Violence Prevention and Treatment Board, I am thankful to have had the opportunity to work
with you on a wide range of projects over the years. I also want to thank you for your willingness
to share and entrust me with the 400 Project data, making this research possible.

iv

To Ross Wantland, a bit of throwback. You inspired me to do this work. When we met at
Illinois, I was just doing this community stuff ‘on the side.’ Our conversation as to if we can ever
end rape (and I don’t know if you even remember this…) changed my life trajectory. Wherever
my career path takes me, you will be an integral part of its foundation.
Finally, I would like to thank Lilah Clevey, Kara England, and Khyrah Simpson who sat
beside me and read literally hundreds of detailed accounts of sexual assault. Your commitment to
this work and to the survivors was inspiring. I can’t wait to see what you all do next.

v

TABLE OF CONTENTS

LIST OF TABLES……..…………………………………………………………………...…….ix
LIST OF FIGURES……..….…………………………………………………………………......x
INTRODUCTION………………………………………………………………….……………..1
THE PREVALENCE AND IMPACT OF SEXUAL ASSAULT…………………………..…….6
The Prevalence of Sexual Assault…………………………………………………………6
The Impact of Sexual Assault...…………………………………………………………...9
SYSTEMS’ RESPONSES TO SEXUAL ASSAULT……………………………….…………..12
The Medical System Response……………..……………………………………………12
The Anticipated Criminal Justice System Response………………………………….....13
The Actual Criminal Justice System Response………………………………………….17
Descriptive research on the response to sexual assault.………………………….17
Descriptive research on the response to other crimes..…………………………..19
Predictive research on the response to sexual assault………………………........22
Summary of the existing literature…………………………………………...…..25
SOCIAL DOMINANCE THEORY………………………………………………….………….26
An Introduction to Social Dominance Theory…………………………………………...26
Social Dominance Theory and the Criminal Justice System Response to Sexual
Assault……………………………………………………………………………………29
THE CURRENT STUDY…………………………………………………………………….….37
Study 1…………………………………………………………………………………...41
Study 2…………………………………………………………………………………...44
STUDY 1: DOCUMENTING LEGITIMIZING MYTHS…………………………………..…..49
Method…………………………………………………………………………………………...49
Sample…………………………………….……………………………………………..49
Preparing Case Files for Coding…………………………………………………………50
Coder training………………………..…….………..…………………………...............50
Directed Content Analysis Coding Procedures…………………………………………..52
Directed Content Analysis Codes…………………………………………………..……54
Conventional Content Analysis Coding Procedures……………………………………..56
Conventional Content Analysis Codes…………………………………………………..58
vi

Results……………………………………………………………………………………………59
Circumstantial Legitimizing Myths……………………..…………………….................64
Victim is lying…………………………………………………………................66
Victim is not injured……………………………………………………………..69
Victim consented………………………………………………………...............70
Victim is not upset………………………………………………………….……74
Victim didn’t act like a victim afterwards……………………………………….75
Characterological Legitimizing Myths……………………..……………………............77
Victim is a regular drug user……………………………………………………..79
Victim is a sex worker…………………………………………………...............81
Victim has “done this before”……………………………………………………83
Victim is “mental”……………………………………………………………….84
Victim is promiscuous…………………………………………………...............86
Victim is not credible…………………………………………………………….87
Investigatory Blame Legitimizing Myths………………………………………………..90
Victim is uncooperative………………………………………………………….91
Victim doesn’t have enough information………………………………………..94
Victim has no phone/address for contact………………………………...............96
Victim or case is weak………………………………………………...................97
Summary…………………………………………………………………………………99
STUDY 2: EXAMINING RELATIONSHIPS BETWEEN LEGITIMIZING MYTHS
AND OUTCOMES……………………………………………………………………………..101
Method…………………...……………………………………………………………………..101
Sample……..……………………………………………………………………………101
Preparing Case Files for Coding……………………………………………..................101
Coder Training………………………………………………………………………….101
Coding Procedures……………………………………………………………………...102
Measures………………………………………………………………………..............102
Data Analysis…………………………………………………………………………...108
Rationale for path analysis and an exploratory analytic approach……………..108
Identifying the final model…..……………………………………………….....111
Results………………………………………………………………………………………......114
DISCUSSION…………………………………………………………………………………..129
Study 1: Documenting Legitimizing Myths……………………………………………131
Study 2: Examining Relationships Between Legitimizing Myths and Outcomes……...135
The role of legitimizing myths in predicting investigatory effort and case
outcomes………………………………………………………………………..136
The influence of victim and perpetrator social identity factors………………...138
vii

The relationship between investigatory effort and case outcomes……………..143
Limitations……………………………………………………………………………...145
Implications for Policy and Practice……………………………………………………148
Changing apparent structures via policy change……………………………….149
Changes deep structures via training……...……………………………………154
Implications for Future Research……………………………………………………….156
Conclusion………………………………………………………………………….......159
APPENDICES……………………………………………………………………………….....161
Appendix A. Coding Instructions………………………………………………………162
Appendix B. Coding Scheme for Directed Content Analysis of Legitimizing Myths
(Study 1)………………………………………………………………………………...171
Appendix C. Codebook for Investigative Steps, Case Outcomes, and Social Identity
Factors (Study 2)………………………………………………………………………..176
Appendix D. Final Codes (Study 2)…………………………………………………….182
REFERENCES………………………………………………..………………………………..187

viii

LIST OF TABLES

Table 1: Crime and arrest estimates for violent crimes (2010)…………………………………..20
Table 2: Legitimizing myths descriptives……………………………………………...………...63
Table 3: Count totals of each type of legitimizing myth…………………………………….......64
Table 4: Circumstantial legitimizing myths and coding scheme………………………...………65
Table 5: Characterological legitimizing myths and coding scheme……………………………..78
Table 6: Investigatory blame legitimizing myths and coding scheme…………………………...91
Table 7: Investigative steps summed to produce “Number of Investigative steps” variable…...105
Table 8: Regression coefficients from the final model…………………………………………125
Table 9: Alignment between circumstantial and characterological legitimizing myths in current
project with the prior literature…………………………………………………………………132
Table 10. Directed content analysis codes……………………………………………………...172
Table 11. Investigative steps, case outcomes, and social identity factors codebook…………...177
Table 12. Final codes…………………………………………………………………………...183

ix

LIST OF FIGURES

Figure 1: Multi-step Process for Law enforcement investigating sexual assault offenses…...….14
Figure 2: Simplified schematic of social dominance theory………………………………….….29
Figure 3: Rape myths, case outcomes and investigatory effort, and social identity factors
of the victim and perpetrator in the SDT framework………………………………..…………...34
Figure 4: Relationships between rape myths, case outcomes and investigatory effort, and
social identity factors of the victim and perpetrator as examined by the empirical literature…...36
Figure 5: Current study’s conceptual model……………………………………………………..44
Figure 6: Examples of the conservative coding scheme in action……………………………….53
Figure 7: Example of a discrete line of text identified during conventional content analysis…...57
Figure 8: Simple schematic of legitimizing myths typology…………………………………….61
Figure 9: Investigator’s scene sheet documenting “victim is lying”……………………………67
Figure 10: Progress notes documenting “victim is lying”………………………………………68
Figure 11: Fax cover sheet documenting “victim is lying”……………………………...............68
Figure 12: Initial police report documenting “victim is not injured”……………………………70
Figure 13: Initial police report documenting “victim consented”……………………………….72
Figure 14: Progress notes documenting “victim consented”…………………………….............72
Figure 15: Initial report documenting “victim is not upset”……………………………………..74
Figure 16: Inter-office memorandum documenting “victim didn’t act like a victim afterwards”.75
Figure 17: Additional inter-office memorandum notes for case 157…………………………….76
Figure 18: Case file notes documenting “victim is a regular drug user”………………………...80
Figure 19: Additional case file notes for case 167……………………………………………….81

x

Figure 20: Investigator’s scene sheet documenting “victim is a sex worker”…………………...82
Figure 21: Progress notes documenting “victim has “done this before””…………………….....84
Figure 22: Initial police report documenting “victim is “mental””……………………………...85
Figure 23: Initial police report documenting “victim is promiscuous”………………………….87
Figure 24: Progress notes documenting “victim is not credible”………………………………..89
Figure 25: Progress notes documenting “victim is uncooperative”……………………………...93
Figure 26: Progress notes documenting “victim doesn’t have enough information”……………96
Figure 27. Progress notes documenting “victim has no phone/address for contact”……………97
Figure 28. Progress notes documenting “victim or case is weak”………………………………99
Figure 29: Current study’s statistical model……………………………………………….…...110
Figure 30: Revised statistical model for analysis………………………………………………112
Figure 31: Models 1 and 2 tested via path analysis…………………………………………….117
Figure 32: Models 3-5 tested via path analysis…………………………………………............121
Figure 33: Models 6 tested via path analysis………………………………………...................123
Figure 34: Final model with regression coefficients……………………………………………124
Figure 35: A reprint of Figure 5…………………………………………………………...........135
Figure 36: The conceptual model illustrating confirmed relationships only…………………...145

xi

INTRODUCTION
Sexual assault is a pervasive crime, as epidemiological data suggest that one-in-five
women will be raped in her lifetime (M. C. Black et al., 2011). This traumatic event can lead to
myriad psychological and physical health problems for victims1 and places a financial burden on
society at large (e.g., see S. Martin, Macy, & Young, 2011). Fortunately, there are systems in
place to provide resources and services to survivors post-assault, including the criminal justice
system, the medical system, the mental health system, and rape crisis centers. The focus of the
current project was the criminal justice system, as it offers victims the opportunity to report the
assault and pursue prosecution of the offender. Through these means, this system can provide
justice to survivors and safety to the general public by holding perpetrators responsible for their
crime.
However, rape prosecution is a complex process (e.g., see Bouffard, 2000; Campbell,
2008; P. Y. Martin, 2005) consisting of two interrelated, yet distinct stages: the investigation,
carried out by law enforcement personnel, and prosecution, carried out by the prosecutor’s
office. In most jurisdictions, law enforcement personnel are responsible for facilitating the
transition of sexual assault cases from the investigation stage to the prosecution stage by
referring the case to the prosecutor’s office for the consideration of charges against the identified
perpetrator. The majority of sexual assault cases, however, does not make this transition and
instead fall out of the criminal justice system process while under the purview of police
personnel. Indeed, prior research examining rates of sexual assault case attrition has found 7393% of reported sexual assault cases are never prosecuted (Campbell et al., 2014; see Lonsway
& Archambault, 2012 for a review.)
1

The terms ‘victim’ and ‘survivor’ will be used interchangeably throughout this document to reflect that sexual
assault is a violent crime that takes tremendous strength and courage to survive. The term patient will also be used
within the context of the medical system. (see Campbell & Townsend, 2011).

1

Furthermore, prior literature has documented that law enforcement personnel conduct
less-than-thorough investigations prior to the (non-) referral to the prosecutor’s office. For
example, police often don’t process key crime scene evidence from sexual assault cases, most
notably the “rape kit” or sexual assault kit (SAK). A SAK is collected by a health care
practitioner post-assault during the course of a medical forensic exam. The SAK houses potential
forensic evidence of the assault collected from the victim’s body. Law enforcement personnel are
supposed to pick up the SAK from the medical facility and submit it to the crime lab for analysis,
so that its contents may be used in potential prosecution of the offender. SAK contents may also
help law enforcement personnel with the investigatory process, prior to prosecution, as DNA in
the SAK may identify an unknown assailant. However, many SAKs never make it to the crime
lab for analysis, as law enforcement personnel fail to submit them. The problem of unsubmitted
SAKs has garnered national attention as stockpiles of unsubmitted SAKs have been discovered
in jurisdictions across the county (see Human Rights Watch, 2009; Human Rights Watch, 2010;
Rubin, 2009; Strom & Hickman, 2010; Tofte, 2009), providing a tangible symbol of police not
completing all steps of their investigation.
Prior literature on the criminal justice system response to sexual assault begs the question
as to why law enforcement personnel behave in this way. Specifically, why do law enforcement
personnel fail to facilitate sexual assault case transitions from the investigation stage to the
prosecution stage (i.e., refer the case to the prosecutor and provide a perpetrator for charging),
and why do they conduct a less-than-thorough investigation leading up to the (non-) referral?
Social dominance theory (SDT) provides an integrated conceptual framework for analyzing law
enforcement personnel’s response to sexual assault cases (Sidanius, Liu, Shaw, & Pratto, 1994;
Sidanius & Pratto, 2001, 2011). SDT comes from the field of sociology and is a comprehensive

2

theory for understanding and explaining individual and institutional acts of discrimination—
actions, policies, rules, and roles that support the unequal allocation of social value across groups
(e.g., some groups have good housing, health, and access to resources whereas other groups have
relatively poor housing, health, and limited access to resources). SDT explains that individual
and institutional acts of discrimination are justified through the use of legitimizing myths—
shared ideologies that justify acts by appealing to morality and intellect (Sidanius et al., 1994;
Sidanius & Pratto, 2001).
Within the SDT theoretical framework, law enforcement personnel’s response to sexual
assault could be considered an act of institutional discrimination. This suggests, therefore, that
police rely on legitimizing myths to justify their actions and inaction. Prior literature suggests
that legitimizing myths that minimize rape (i.e., rape myths) are commonly endorsed by police,
and may be used to justify their response to sexual assault (Brown & King, 1998; Campbell &
Johnson, 1997; Edwards, Turchik, Dardis, Reynolds, & Gidycz, 2011; Feldman-Summers &
Palmer, 1980; Field, 1978; Galton, 1975; Gylys & McNamara, 1996; LaFree, 1989; Maddox,
Lee, & Barker, 2012; Page, 2008a, 2008b, 2010; Pratto, Sidanius, Stallworth, & Malle, 1994;
Sidanius & Pratto, 2001). However, previous studies have only documented police attitudes
towards rape via questionnaires and have not examined how rape myths are observed in sexual
assault case investigations. As a result, the police response to sexual assault is assumed to be
rooted in rape myth endorsement, but to date, there have been no empirical studies that have
observed these beliefs in actual case work (i.e., the steps completed during an investigation, like
submitting the SAK to the crime lab for analysis, and case referral to the prosecutor) and if law
enforcement personnel provide justification (i.e., other types of legitimizing myths) beyond
traditional rape myths for their response to sexual assault.

3

SDT was used as a guiding theoretical framework to design the current project examining
the police response to sexual assault cases, which consisted of two studies. The purpose of Study
1 was to explore if and how legitimizing myths are observed in police investigations of sexual
assault cases. The purpose of Study 2 was to examine empirically if the different types of
legitimizing myths documented in Study 1 predicted law enforcement personnel’s responses to
sexual assault cases (i.e., investigatory effort and case outcome). The police records of 248
sexual assault cases corresponding to a random sample of unsubmitted SAKs found in a Detroit
police property storage facility were the sample for the current project. This specific sample was
selected as it consisted of cases that had been subjected to a less-than-thorough investigation, as
indicated by each case’s corresponding unsubmitted SAK. Qualitative methods, including
directed and conventional content analysis, were used to document the extent and types of
legitimizing myths in police records (i.e., Study 1). Quantitative methods, specifically path
analysis, were then used to analyze the relationships between the different types of legitimizing
myths, investigatory effort, and case outcomes within the context of specific social identity of the
victim and perpetrator factors (i.e., sex, race, and age), and the number of perpetrators (i.e.,
Study 2). Specifically, the aims of the two studies were to:


Document the extent and type of legitimizing myths (e.g., rape myths) in police
records of sexual assault case investigations (i.e., Study 1), and



Examine the relationships between the different types of legitimizing myths,
investigatory effort, and case outcomes of sexual assault cases, within the context
of specific social identity factors of the victim and perpetrator (i.e., sex, race, and
age), and the number of perpetrators (i.e., Study 2).

4

To set the stage for the current project, I will begin with a literature review on the extent
and impact of sexual assault and the current systems of services in place for victims. The
literature review will then focus on the criminal justice system response, as it is the venue for the
current project. I will describe the multi-stage criminal justice system process, with particular
emphasis on the first stage of this process: the sexual assault case investigation overseen by the
police. I will detail the different steps that could and should be completed during the course of a
sexual assault case investigation, and contrast this process to what most frequently happens in
practice. I will then introduce social dominance theory as the theoretical framework used to
understand and examine law enforcement personnel’s response to sexual assault in the current
project, detailing how this theory was used to design the two-study approach. Rationale,
methods, results, and implications for the current project will follow.

5

THE PREVALENCE AND IMPACT OF SEXUAL ASSAULT
The Prevalence of Sexual Assault
Although there is variability across state and federal statutes as to how rape is legally
defined, it is generally described as any unwanted act of oral, vaginal, or anal penetration by
body parts or objects using force, threat of force, or while the victim in incapacitated or unable to
give consent (National Institute of Justice, 2010). Sexual assault refers to a range of unwanted
sexual activity, including contact and noncontact, up to and including rape (Campbell, 2008).
The current project included cases of sexual assault involving penetration (i.e., rape) and cases of
non-penetration (e.g., ‘fondling’), and the subsequent criminal justice system response.
Accordingly, the current literature review will cover rape, specifically, and sexual assault,
generally.
In 1996, the National Institute of Justice and the National Center of Injury Prevention and
Control, Centers for Disease Control and Prevention (CDC) jointly sponsored the National
Violence Against Women Survey (NVAWS) in order to measure the extent of violence against
women across the United States (Tjaden & Thoennes, 2000, 2006). Tjaden and Thoennes (2000,
2006) used random digit dialing to produce a nationally random sample of men and women, then
asked participants five behaviorally specific questions to screen for rape victimization (see
Tjaden & Thoennes, 2006, p. 10 for specific screening questions). Findings from the NVAWS
(2000, 2006) revealed that rape was far more common than might have been imagined: one-insix women were raped in their lifetime with 300,000 women raped in the last year. This study
provided empirical data documenting the pervasive nature of rape and sexual assault in the
United States. However, these rates were likely underestimates as the NVAWS (2000, 2006) did
not include drug and alcohol-facilitated rape, in which the perpetrator deliberately gave the

6

victim drugs or alcohol without her permission, or incapacitated rape, in which the victim was
raped after voluntarily taking drugs or alcohol. Furthermore, because the study used random digit
dialing to secure its sample, “those who are homeless or live in institutions, group facilities, or
residences without telephones” were necessarily excluded (Tjaden & Thoennes, 2006, p. iv).
A decade later, The National Crime Victims Research and Treatment Center from the
National Institute of Justice funded another study to revisit the scope of the problem of violence
against women on a national scale (Kilpatrick, Resnick, Ruggiero, Conoscenti, & McCauley,
2007). This time, drug- and alcohol-facilitated rape as well as incapacitated rape were accounted
for in the study. Again, a random sample representative of the United States was secured via
random digit dialing and participants were asked about their experiences as rape victims.
Kilpatrick and colleagues (2007) found that nearly one-in-five women had been raped in their
lifetime and over one million women had been raped in the past year; these rates were
considerably higher than the one-in-six rate of lifetime prevalence and annual estimate of
300,000 rapes documented by the NVAWS a decade earlier (2000, 2006). The inclusion of drugfacilitated and incapacitated rape may have partially accounted for the discrepancy across the
two studies, with the more recent study providing a more comprehensive and accurate snapshot
of the scope of the problem. However, Kilpatrick and colleagues (2007) did not include
attempted rape, even though it is considered a crime in many jurisdictions and can have a
significant impact on the victim’s health and well-being (Perilloux, Duntley, & Buss, 2012).
Additionally, like the NVAWS a decade earlier, the use of random digit dialing on home
telephones necessarily excluded specific populations (e.g., homeless persons, individuals living
within institutions, persons without home phones).

7

In 2010, the National Intimate Partner and Sexual Violence Survey was launched by the
CDC’s National Center for Injury Prevention and Control with support from the National
Institute of Justice and Department of Defense (M. C. Black et al., 2011). This study again
attempted to document the extent of the problem of violence against women in the United States.
The study replicated the one-in-five lifetime prevalence rate of rape found by Kilpatrick and
colleagues (2007). However, it documented that 1.3 million women were raped in the past year,
up from one million according to the Kilpatrick study (2007). Black and colleagues (2011) recent
study may provide the most accurate figures to date as their study significantly improved upon
the previous national studies in two key ways. First, this study included drug-facilitated and
incapacitated rape, unlike the NVASW (2000, 2006) and attempted rape, unlike the study by
Kilpatrick and colleagues (2007). This more inclusive definition of rape provides a more
complete understanding of the scope of violence against women in the United States. Second,
Black and colleagues (2011) included cell phones in addition to home phones in their random
digit dialing to secure a nationally representative sample. This allowed the researchers to access
individuals without landlines, which is increasingly important as home phones continue to
become obsolete and replaced by cell phones (Svensson, 2013).
Still, these rates are likely underestimates of the true scope of sexual violence for two
reasons: non-coverage bias and non-response bias. Regarding non-coverage bias, all of these
studies still exclude individuals without phones and it is very likely this group of people is
systematically different from persons with phones. For example, this approach would likely
exclude lower-income individuals and folks living on the street. Additionally, in regards to nonresponse bias, these studies do not include individuals who decline to participate in the survey
and it is possible that non-participants are systematically different than participants. Non-

8

participants may have a differential history of sexual violence which affects their decision to
participate. Therefore while it is possible that the actual rates of sexual assault are higher among
the general population than is documented in the most recent and improved data by Black and
colleagues (2011), these current rates still illustrate that rape is not an uncommon event (see
Raphael & Logan, 2011 for a more thorough discussion of the reporting rates of sexual assault).
Recent research also confirms the gendered nature of sexual assault. Black et al. (2011)
found that 93% of individuals reporting attempted or completed rape were women (21,840,000
US women compared to 1,581,000 US men), with nearly one-in-five US women reporting rape
in their lifetime, compared to one in 71 US men (2011). These numbers corroborate earlier
empirical work (Basile, Chen, Black, & Saltzman, 2007; Kilpatrick et al., 2007; Tjaden &
Thoennes, 2000, 2006) and evidence that more than nine-times-out-of-ten, the victim of rape will
be a woman. Conversely, the vast majority of perpetrators are male: 98.1% of female victims and
93.3% of male victims report only male perpetrators2 (M. C. Black et al., 2011).
Black and colleagues (2011) also documented variation in rape prevalence based on race
and ethnicity. Higher rates of rape victimization are reported among multiracial non-Hispanic
women (33.5%), followed by American Indian and Alaskan Native women (26.9%). The one-infive statistic applies to both Black (22.0%) and White non-Hispanic (18.8%) women with the
lowest rates of reported rape victimization among Hispanic women (14.6%).
The Impact of Sexual Assault
The prevalence of sexual assault is particularly alarming given the wide range of
detrimental outcomes that can result from the assault (see S. Martin et al., 2011). Immediately
post-assault, victims may exhibit heightened levels of distress and experience shock, confusion,
2

Given the gendered nature of sexual assault, female pronouns may be used to refer to rape victims while male
pronouns may used to refer to rape suspects and perpetrators throughout this document. This in no way negates the
experiences of male rape survivors or any sexual crimes perpetrated by female offenders.

9

shame, self-blame, fear, agitation, emotional detachment and social withdrawal (Basile & Smith,
2011; Breitenbecher, 2006; Bryant-Davis, Chung, & Tillman, 2009; Herman, 1992; Jordan,
Campbell, & Follingstad, 2010; Ullman, Filipas, Townsend, & Starzynski, 2006; Yuan, Koss, &
Stone, 2006). Additionally, victims may have acute physical injuries that require attention and be
at risk for unwanted pregnancy or sexually transmitted infections (Basile & Smith, 2011; ElMouehly, 2004; S. Martin & Macy, 2009; S. Martin et al., 2011; Resnick, Acierno, Holmes,
Dammeyer, & Kilpatrick, 2000; Tjaden & Thoennes, 2000).
Early psychological distress may later manifest as post-traumatic stress disorder (PTSD),
obsessive-compulsive disorder, anxiety, depression, substance abuse, poor self-esteem, stress,
somatic complaints, and suicidality, which are more common among survivors compared to
those who have not been sexually assaulted (e.g., see Bonomi, Anderson, Rivara, & Thompson,
2007; Bryant-Davis et al., 2009; Campbell, Dworkin, & Cabral, 2009; Hanson et al., 2008;
Jordan et al., 2010; S. Martin & Macy, 2009; S. Martin et al., 2011; Palm & Follette, 2008;
Zinzow et al., 2010). In addition to these mental health problems, survivors are more likely to
have poorer perceptions of their health and to report a greater number of physical and behavioral
health symptoms including gynecologic and reproductive health problems, gastrointestinal
symptoms, trouble sleeping, frequent nightmares, and engagement in risky sexual behavior
(Amstadter, McCauley, Ruggiero, Resnick, & Kilpatrick, 2010; Campbell, Lichty, Sturza, &
Raja, 2006; Cloutier, Martin, & Poole, 2002; Clum, Nishith, & Resick, 2001; Gidycz,
Orchowski, King, & Rick, 2008; Golding, 1999; Jordan et al., 2010; Kimerling & Calhoun,
1994; S. Martin & Macy, 2009; S. Martin et al., 2011; Palm & Follette, 2008; Resnick, Acierno,
et al., 2000; Skinner et al., 2000).

10

The impact of rape extends beyond these individual-level psychological and physical
health outcomes; it can also be measured at community levels by considering its cost to society.
Recent modeling has estimated that each rape costs nearly $450,000 in 2008 US dollars by
taking into account the costs to the victim, to the justice system, loss in offender productivity,
and collateral costs (e.g., prevention expenditures for personal security, avoidant behaviors to
safeguard against victimization, insurance costs, welfare programs, etc.) (DeLisi et al., 2010). If
we take these estimates and apply them to the prevalence rates provide by Black and colleagues
(i.e., 1.3 million women reported being raped over a 12 month period), the annual cost of rape
was $585 billion in 2008 US dollars. Taken together, rape is no longer an event that affects only
those involved, but is a serious crime that presents a public health and economic issue for society
at large.

11

SYSTEMS’ RESPONSES TO SEXUAL ASSAULT
One of every five women is raped in her lifetime and may experience a wide range of
detrimental effects as a result of the traumatic experience. Therefore, it is important to
understand the services available to victims following a sexual assault and their utilization of
these services3. Post-assault, victims may choose to access the medical system for health-related
services and a medical forensic exam, and/or the criminal justice system to report the sexual
assault and pursue criminal prosecution. A brief review of the healthcare services available to
victims through the medical system will be followed by a more in-depth examination of the
criminal justice system response to sexual assault, and more specifically the police response, as it
is the context for the current project.
The Medical System Response
If victims choose to access the medical system post-assault, a number of healthcare and
medical forensic services are available. In terms of healthcare services, medical care providers
can detect and care for injuries; screen and provide preventative treatments (prophylaxis) for
sexually transmitted infections (STIs), including HIV; and test for pregnancy and provide
emergency contraception (Campbell, 2008; Department of Justice, 2013). These services attend
to the patients’ health needs and concerns, and are one of the primary reasons cited by victims as
to why they seek medical services following a sexual assault (Campbell, Bybee, Ford, &
Patterson, 2009; Department of Justice, 2013; Du Mont, White, & McGregor, 2009; Kilpatrick et
al., 2007; Resnick, Holmes, et al., 2000). In addition to attending to physical health needs,
medical care professionals can also address survivors’ psychological needs and concerns. Indeed,

3

This review will focus on the medical and legal systems as these two systems often work together closely in the
process of investigating sexual assaults, which is the focus of the proposed study. Campbell (2008) noted that
victims’ contact with the mental health system and rape crisis centers often comes much later in victims’ recoveries.

12

many victims report that they seek out medical services to better understand and validate their
experience (Campbell, Bybee, Ford, et al., 2009).
Medical care providers can also conduct a medical forensic examination to attend to the
needs of the criminal justice system via evidence collection with a sexual assault forensic exam
kit (SAK) (Department of Justice, 2013). Specifically, this process includes a complete head-totoe physical examination; the documentation and photographs of any injuries; the collection of
the patient’s clothing; the collection of any foreign material on the patient’s body (i.e., blood,
dried sections, loose hairs, and other debris); head and pubic hair combing and plucking; a visual
assessment of the genitals for trauma; the collection of vaginal, penile, anal, oral, and body
swabs corresponding to points of contact with the perpetrator; collection of the patient’s blood or
saliva as a referent sample; and blood or urine samples for drug analysis (Department of Justice,
2013). The evidence collected during the medical forensic examination can aid in the
investigation and subsequent prosecution of the sexual assault case. Specifically, analysis of the
collected samples may reveal the identity of an unknown assailant and/or corroborate the
victim’s story with physical evidence. The legal role of the medical forensic examination, and
specifically the sexual assault forensic exam kit, will be revisited in more detail in the review of
the criminal justice system below.
The Anticipated Criminal Justice System Response
Following a sexual assault, victims may choose to access the criminal justice system to
report the rape and later pursue prosecution of the offender via a two-stage process: case
investigation, overseen by law enforcement personnel, and case prosecution, overseen by the
prosecutor’s office. Victims’ first contact with a representative from the criminal justice system
during the investigation stage will typically be with patrol officers who will take the initial report

13

(Campbell, 2008; P. Y. Martin, 2005; Patterson & Campbell, 2010). This initiates a multi-step
investigative process for law enforcement, as illustrated in Figure 1.4

Figure 1. Multi-step process for law enforcement investigating sexual assault offenses

Patrol officer
takes initial
report

Process scene

Transport
victim to
medical facility

Maintain
progress notes

Case assigned
to detective

Submit SAK to
crime lab

Conduct
Investigation

Refer case to
prosecutor

Arrest suspect

4

It is important to note, however, that the police response to sexual assault, in practice, is complex and frequently
progresses in a nonlinear fashion in which investigatory decision-making and action is complicated and impacted by
a multitude of factors. For example, victim involvement is one key factor that can affect how a sexual assault
investigation progresses. Victim involvement may vary across cases or across investigative steps taken in a single
case (e.g., the victim may be very involved at the beginning of the investigation, and then has no further
communication with the investigators). If a victim is not perceived to be involved in the case, law enforcement
personnel may not conduct as thorough of an investigation; alternatively, if law enforcement personnel are perceived
to not be putting as much effort into a case, the victim may choose not to be involved. The description and
accompanying schematic provided here should be interpreted as simplified representations of this process.

14

If the report is taken at the scene of the crime, the patrol officers may call evidence
technicians to process the scene (the evidence technicians make take photographs at the scene of
the crime and/or collect evidence). The patrol officers on the scene may also canvass the area in
an attempt to locate and take statements from any witnesses. The patrol officers will frequently
aid in securing transport for the victim to a medical facility to receive a medical forensic exam,
complete with forensic evidence collection (see The Medical System Response above) that could
be used in later prosecution. Alternatively, many victims first report to the medical system for
medical care before calling police. In this situation, police would arrive at the medical facility to
take the initial report. Regardless of which system victims access first, if they consent to a
medical forensic exam complete with evidence collection via the SAK, victims may choose to
sign a medical release form authorizing their medical records and sexual assault kit to be
released to law enforcement personnel. If this document is signed, law enforcement personnel
are supposed to submit the SAK to the crime lab for forensic analysis, in addition to retaining
copies of the medical report.
Following the initial report, crime scene processing, and transport of the victim to a
medical facility for a medical forensic exam, a detective is typically assigned to the case
(Campbell, 2008; P. Y. Martin, 2005; Patterson & Campbell, 2010). The detective maintains
progress notes to document the different steps he/she takes in the investigation. Steps taken are
specific to each case and may include attaining additional witness statements, a statement from
the victim (if not previously attained by the patrol officer), bringing in suspects for an interview,
and facilitating a suspect photo or in-person lineup to give the victim an opportunity to identify
the offender.5 After conducting a thorough investigation, the detectives decide if they will refer

5

These different actions take place during the “conduct investigation” step in Figure 1.

15

the case to the prosecutor, requesting a warrant for the suspect’s arrest6. This facilitates the
transition of the case from the investigation stage to the prosecution stage.
After charges have been filed, the prosecutor’s office may elect to stay or dismiss the
charges depending on the specific details of the case and if the prosecutor believes there is
probable cause. If the charges are not dismissed, the case may end in a plea bargain in which
case the defendant pleads guilty to a specified charge in exchange for an agreed upon lower
sentence. Alternatively, the case may go to trial and result in a conviction of the offender or in an
acquittal if the defendant is found not guilty. The convicted offender is then sentenced for his
crime.
There are potential benefits to victims for engaging in this multi-stage complex process.
By reporting the rape, the victim may become eligible for victim compensation funds that can
pay for medical expenses, lost wages due to missed work, psychological counseling or treatment,
and other related services or assistance received in relation to the assault (National Association
of Crime Victim Compensation Boards, n.d.). If the perpetrator is held responsible for his actions
through the criminal justice system process, the victim may feel as though justice was served.
When perpetrators are incarcerated for their actions, victims may feel that they are safer and that
their communities are safer as well (RAINN, 2009). Finally, even if the perpetrator is not found
responsible for the assault, victims may still feel a sense of justice if victims believe police did
all that they could to solve the case (i.e., procedural justice) (Elliott, Thomas, & Ogloff, 2012).
However, a large, multidisciplinary literature suggests that for a great many rape victims, these
benefits are rarely realized.

6

In some cases, special circumstance calls for an arrest of the suspect prior to a warrant being issued by the
prosecutor. The prosecutor then issues a retroactive warrant.

16

The Actual Criminal Justice System Response
Descriptive research on the response to sexual assault. How the criminal justice
system is supposed to respond to reported sexual assaults is not in fact how it typically responds
(e.g., see Bouffard, 2000; Campbell, Bybee, Patterson, & Dworkin, 2009; Campbell, Townsend,
Bybee, Shaw, & Markowitz, 2012; Campbell et al., 2014; Frohmann, 1997; LaFree, 1981;
Lonsway & Archambault, 2012; Rose & Randall, 1982; Tasca, Rodriguez, Spohn, & Koss,
2013). In a recent review of this literature, Lonsway and Archambault (2012) summarized the
empirical evidence for this “justice gap:” for every 100 rapes, 5-20 are reported to law
enforcement, 0.4-5.4 are prosecuted, 0.2-5.8 result in a conviction, and only 0.2-2.8 result in
incarceration of the offender. Assuming a sexual assault is reported to law enforcement, the
greatest point of attrition is from the initial report to prosecution: 73-93% of reported sexual
assault cases are never prosecuted (Lonsway & Archambault, 2012).
This is due in large part to the majority of reported sexual assaults not being referred by
law enforcement to the prosecutor’s office for the issuing of charges, or no charges being filed
by the prosecutor’s office after a referral. A recent evaluation found that of 1,465 sexual assault
cases reported over 15 years across 6 sites, 86% were never referred to the prosecutor or charged
(Campbell, Townsend, et al., 2012; Campbell et al., 2014). This rate was surprisingly consistent
across the six communities given that they differed in terms of their geographic location and
population served (i.e., urban, rural and mid-sized). Furthermore, this rate may even be better
than what would be found in other communities as all participating sites in the study had
established multidisciplinary sexual assault response teams. The anticipated two-stage criminal
justice system response, consisting of an investigation stage with law enforcement personnel and

17

a prosecution stage with the prosecutor’s office, becomes a one-stage response from law
enforcement personnel in practice.
In addition to documentation that the majority of cases never transition to the second
stage of prosecution (i.e., cases are not referred onto the prosecutor and an arrestee is not
provided), there is also evidence that cases receive a less-than-thorough investigation while
under the purview of law enforcement. Specifically, a large proportion of the SAKs containing
potential forensic evidence to aid in prosecution are never submitted to the crime lab for analysis.
Stockpiles of unsubmitted SAKs have been found in jurisdictions across the country (Human
Rights Watch, 2009, 2010; Rubin, 2009; Strom & Hickman, 2010; Tofte, 2009). In March of
2009, Human Rights Watch reported on the 12,669 unprocessed medical forensic exam kits in
police storage facilities in the Los Angeles Police Department, the Los Angeles County Sheriff’s
Department, and 47 independent police departments in Los Angeles County. Similarly, Human
Rights Watch reported on 10,000 unprocessed medical forensic exam kits found in Detroit police
storage facilities (Tofte, 2009). In 2010, Human Rights Watch released a report revealing that
only 1,474 of 7,494 medical forensic exam kits booked into evidence since 1995 in Illinois were
confirmed as tested. This means that up to 80% of sexual assault forensic exam kits had never
been examined in Illinois. Other communities across the country are showing similar trends. For
example, Patterson and Campbell (2012) and Shaw and Campbell (2013) found that that over
40% of adult and adolescent SAKs, respectively, collected in a sexual assault nurse examiner
(SANE) program were not submitted to the crime lab for analysis. These unsubmitted SAKs
serve as a tangible symbol of the lack of response to rape within the criminal justice system.
The criminal justice system response to sexual assault, and more specifically what takes
place while the case is under the purview of the police, is described as a “justice gap” when

18

referring to the high rates of case attrition, and “justice denied” when referring to the high rate of
unsubmitted SAKs across jurisdictions (Lonsway & Archambault, 2012; Strom & Hickman,
2010). These patterns of injustice define the criminal justice system response to sexual assault
and have serious implications for victims that may negate any potential benefits of reporting the
assault (see The Anticipated Criminal Justice System Response above).
Research documents that when victims turn to the criminal justice system post-assault,
legal personnel respond in a cold or impersonal manner, lack empathy, express disbelief, blame
the victim for the assault, and even deny services (Campbell, 2005, 2008; Campbell & Raja,
2005; Logan, Evans, Stevenson, & Jordan, 2005; Madigan & Gamble, 1991; P. Y. Martin, 2005;
P. Y. Martin & Powell, 1994). Termed, “secondary victimization” or the “second rape,” the
majority of rape victims have these types of experiences when interacting with criminal justice
system personnel and they are not without significant consequence (Campbell, 2005; Campbell
& Raja, 2005; Campbell, Wasco, Ahrens, Sefl, & Barnes, 2001; P. Y. Martin & Powell, 1994).
These negative interactions extend and exacerbate the trauma of the assault. Victims report
feeling bad about themselves, violated, depressed, anxious, and to blame for the assault from
their interaction with system personnel (Campbell, 2005; Campbell & Raja, 2005). Research has
found that these negative interactions are then associated with more severe posttraumatic stress
symptoms as well as poorer physical and psychological health overall (Campbell et al., 2001;
Filipas & Ullman, 2001; Ullman, 2010; Ullman, Filipas, Townsend, & Starzynski, 2007). The
criminal justice system is supposed to intervene and help the victim, but more often than not
hurts them.
Descriptive research on the response to other crimes. Taken together, the body of
research on the criminal justice system response to sexual assault presents a negligent system,

19

frequently resulting in injustice that hurts the victim more than it helps, which raises the
question: is the criminal justice system response to sexual assault and experiences of sexual
assault victims different from other crimes? Given that the majority of sexual assault cases are
subject to attrition while under law enforcement personnel’s purview, the question can be asked
more specifically—do law enforcement personnel respond to sexual assault and treat sexual
assault victims differently than they do other crimes?
To explore this, it is essential to include only cases to which law enforcement personnel
have an opportunity to respond—sexual assault cases and other violent crimes that are reported
to police.7 There are no comprehensive social science studies that have compared the attrition
rates of rape to other violent crimes from the time the case is initially reported to law
enforcement through prosecution (Daly & Bouhours, 2010). However, data estimates from the
Federal Bureau of Investigation’s (FBI) Uniform Crime Reporting (UCR) Statistics provide
insight. Table 1 presents the estimated crime rates and arrests for violent crime across the United
States in 2010.

Table 1. Crime and arrest estimates for violent crime (2010)8
Type of
Violent Crime
Murder
Aggravated Assault
Robbery
Forcible Rape

Estimated Number
of Offenses
14,722
781,844
369,089
85,593

7

Estimated Number
of Arrests
11,201
408,488
112,300
20,088

Ratio of
Arrests to Offenses
0.761
0.522
0.304
0.235

It is important to acknowledge the differential rates of reporting across violent crimes. Rates of reporting for
violent crimes like robbery, aggravated assault, and simple assault vary (Bosick, Rennison, Gover, & Dodge, 2012;
Hart & Rennison, 2003). However they are consistently higher than rates of reporting for sexual assault, as only 520% of all sexual assaults are reported to police (Bosick et al., 2012; Hart & Rennison, 2003; Lonsway &
Archambault, 2012).
8
This table was created from US Department of Justice Federal Bureau of Investigation (March 2010) and US
Department of Justice Federal Bureau of Investigation (2010).

20

As can be seen, the ratio of estimated arrests to offenses for forcible rape is 0.235 and is
considerably lower than all other types of violent crime; robbery is 0.304, aggravated assault is
0.522 and murder is 0.761 (see US Department of Justice Federal Bureau of Investigation, 2010,
March 2010). These data alone provide evidence of a differential police response to sexual
assault as compared to other crimes. However, the considerably lower ratio of estimated arrests
to offenses for forcible rape is likely even more pronounced than is indicated in these data as the
2010 estimates of forcible rape only included “the carnal knowledge of a female forcibly and
against her will” (US Department of Justice Federal Bureau of Investigation, January 2009). This
means that male victims, drug- or alcohol-facilitated victims, victims with disabilities, and minor
victims are necessarily excluded in these estimates (Archambault & Lonsway, May 2012). If
these victims were included, the estimates of assaults would increase, likely further widening the
gap between the number of assaults and arrests for rape. Additionally, social science research has
documented that law enforcement personnel are more likely to deem sexual assault cases as
unfounded as compared to other crimes (Lonsway, 2010), more likely to “solve” (i.e., clear by
arrest or exception) a rape if it co-occurs with another crime (i.e., robbery, burglary, or property
theft) as compared to rapes that do not have co-occurring crimes (Addington & Rennison, 2008),
and more likely to submit forensic evidence for unsolved homicides as compared to unsolved
rapes (Strom & Hickman, 2010). These findings further solidify how police respond differently
to sexual assault as compared to other crimes in that they are less likely to believe victims of
rape, to solve rape cases, and to submit forensic evidence from rape cases for analysis.
There also appear to be differences in how crime victims perceive their experiences with
law enforcement personnel and how they are affected by those experiences. Whereas the
majority of sexual assault victims report secondary victimization as a result of their interactions

21

with the criminal justice system (Campbell, 2005; Campbell & Raja, 2005; Campbell et al.,
2001; P. Y. Martin & Powell, 1994), most victims of violent crime in general report that they
feel as though they were treated fairly (Wemmers, 2013). Furthermore, when crime victims do
have negative experiences with the criminal justice system (i.e., disrespectful treatment from the
police, denying voice, and poor quality investigations), those experiences have a greater negative
impact on the psychological outcomes (i.e., a negative impact on victims’ ability to cope with the
crime, self-esteem, optimistic outlook, trust in the legal system, and faith in a just world) of
sexual assault victims as compared to non-sexual assault victims (Laxminarayan, 2012).
Predictive research on the response to sexual assault. Across all reported sexual
assaults, we see consistently high rates of case attrition, unsubmitted SAKs and secondary
victimization. However, it is possible for there to be variance between cases in how they are
investigated and how they progress through the criminal justice system. Prior research has
examined which sexual assault cases are more likely to receive additional investigational effort
and progress further in the criminal justice system as compared to others. This body of work has
explored what characteristics of the victim, perpetrator, and assault can predict a range of
investigative steps from the case’s initial entry into the criminal justice system through law
enforcement personnel’s referral to the prosecutor (i.e., during the investigation stage).
Victim and suspect characteristics are generally static, exist prior to the assault, and
provide information on the social identity of the victim and suspect (i.e., age, disability status,
and race/ethnicity). Victim age and disability status have been found to impact police action and
decision making in sexual assault cases. Specifically, law enforcement personnel are more likely
to view the case as “suspicious” and to classify the case as unfounded, and less likely to classify
it as a rape, think that charges should be filed, and refer the case to the prosecutor if the victim is

22

a teenager or young adult, or is disabled (Campbell, Greeson, Bybee, & Fehler-Cabral, 2012;
Campbell, Greeson, Bybee, Kennedy, & Patterson, 2011; Heenan & Murray, 2006; Kelly,
Lovett, & Regan, 2005; Kerstetter, 1990; LaFree, 1981; Rose & Randall, 1982; Schuller &
Stewart, 2000; Spohn & Spears, 1996; Triggs, Mossman, Jordan, & Kingi, 2009).
Previous research on the influence of the victim’s race on sexual assault investigations
and final case outcomes is mixed. Some studies have found no effect of race on the investigation
and processing of the case (Kerstetter, 1990); other studies have found non-White victims are not
taken as seriously, are more likely to be unfounded by police, and that their cases are less likely
to be prosecuted (D. Black, 1978; Frohmann, 1997; LaFree, 1981; Reiss, 1971; Rose & Randall,
1982; Smith & Klein, 1983; Wriggins, 1983); still, other studies have found that cases with nonWhite victims are more likely to have a suspect identified (though not arrested), less likely to be
unfounded, and more likely for their rape kit to be submitted to the crime lab for analysis
(Bryden & Lengnick, 1997; Horney & Spohn, 1996; Shaw & Campbell, 2013). These discrepant
findings may be due to the notion that it is actually the race of the perpetrator that has a greater
impact on law enforcement decision making as compared to the race of the victim, as was found
in early research (LaFree, 1981). Unfortunately, the vast majority of the research has focused on
characteristics of the victim as opposed to characteristics of the suspect.
While characteristics of the victim and suspect are considered to be static and exist prior
to the assault (e.g., the victim had a disability before the assault happened), assault characteristics
emerge in the assault context (e.g., victim injuries sustained during the assault); several
characteristics have been linked to investigatory effort and if the case transitions from the
investigation state to the prosecution stage (i.e., if the case is referred to the prosecutor’s office
and there is an associated arrest). Specifically, the nature of the relationship between the victim

23

and perpetrator(s), the number of perpetrators, tactics employed by the perpetrator(s), and
injuries sustained by the victim have been found to impact the investigation and case outcomes.
Specifically, if the perpetrator is not known to the victim (i.e., stranger sexual assault), the case is
more likely to be thoroughly investigated and less likely to be unfounded by law enforcement
(Bachman, 1998; Kerstetter, 1990; LaFree, 1989; Spohn & Spears, 1996). However, the
perpetrator is less likely to be arrested and the case is less likely to be referred to the prosecutor
(Campbell, Bybee, Ford, et al., 2009). These findings may not be contradictory; stranger sexual
assault cases may be more readily investigated and believed whereas identification of a suspect,
leading to a referral to the prosecutor and their arrest, may be easier in cases with a known
offender. The number of perpetrators has also been found to impact SAK submission to the
crime lab for analysis in that cases with a single perpetrator are more likely to be submitted for
analysis as compared to cases with multiple perpetrators (i.e., “gang rape”) (Shaw & Campbell,
2013). The perpetrator’s use of a weapon has been found to increase the likelihood of the
encounter being classified as a rape and a suspect arrest in some studies (Kerstetter, 1990;
LaFree, 1981). Additionally, law enforcement is more likely to consider the offense “legitimate”
or “prosecutorial,” question a suspect, submit the SAK to the crime lab for analysis and the case
is more likely to move forward to prosecution if there was full penetration and if the victim
sustained injuries from the assault (Campbell, Bybee, Ford, et al., 2009; Frazier & Haney, 1996;
Patterson & Campbell, 2012; Rose & Randall, 1982). Finally, if the victim has a history of drug
or alcohol use, or used these substances prior to the sexual assault, law enforcement is less likely
to classify the assault as a rape and more likely to deem it unfounded, less likely to arrest a
suspect, less likely to think the suspect should be charged, and the case is less likely to progress
to prosecution (Campbell, Bybee, Ford, et al., 2009; Heenan & Murray, 2006; Kelly et al., 2005;

24

Kerstetter, 1990; LaFree, 1981; Rose & Randall, 1982; Schuller & Stewart, 2000; Triggs et al.,
2009).
Summary of the existing literature. In summary, the literature to-date has helped shed
light on the “what”, “who,” “when,” “when,” and “where” regarding the criminal justice system
response to sexual assault. Prior descriptive literature provided the “what”—law enforcement
personnel’s negligent response to sexual assault cases. Previous studies have documented that
sexual assault cases are less likely to have an associated arrest, to be believed, to be solved, and
to have their associated evidence submitted for analysis. Prior predictive literature provides the
“who,” “when,” and “where”—victim, perpetrator, and assault characteristics have been found to
predict how much effort is put into investigating a sexual assault case and the likelihood that the
case will transition from the initial investigation stage to the prosecution stage within the
criminal justice system process (i.e., a referral to the prosecutor’s office and an arrest).
However, there has not been much work to-date on they “why” and “how” of the criminal
justice system response to sexual assault. Why does the criminal justice system respond in this
way? More specifically, how do law enforcement personnel explain their less-than-thorough
response to sexual assault cases and the variation in investigatory effort and case progression
between cases? The “why” and “how” are crucial for if we are able to identify specific
mechanisms used by law enforcement personnel to explain why some cases are investigated and
progress through the criminal justice system while others do not, we can target those
mechanisms with change interventions and attempt to improve the criminal justice system
response to sexual assault.

25

SOCIAL DOMINANCE THEORY
Social dominance theory (SDT) (Sidanius & Pratto, 2001, 2011) is one framework for
identifying and understanding the mechanisms that support the current criminal justice system
response, and more specifically law enforcement personnel’s response, to sexual assault. By
focusing on the specific mechanisms, this framework can assist in conceptualizing and studying
the “why” and “how.” In this section, a brief overview of social dominance theory will be
provided, including its key assumptions and suppositions. This discussion will focus on SDT’s
ability to explain system level phenomena, as the current study is focused on the criminal justice
system. This general introduction and explanation of SDT will be followed by its specific
application to the investigation stage of criminal justice system response to sexual assault.
An Introduction to Social Dominance Theory
SDT is based on the idea that societies are organized in hierarchies based on group
memberships (i.e., group-based social hierarchies) (Sidanius & Pratto, 2001). Within these
hierarchies, groups at the top (i.e., dominant groups) have a disproportionate share of the good
things in life (i.e., positive social value), like good health, good housing, access to resources and
powerful roles, whereas groups at the bottom (i.e., subordinate groups) have a disproportionate
share of the bad things in life (i.e., negative social value) like relatively poor health, poor
housing, limited access to resources, and menial roles. Within this theory, forces operate at the
individual, interpersonal, and system-wide level so as to maintain (or sometimes fight against)
these group-based social hierarchies. Individual level forces refer to individuals’ attributes or
actions; interpersonal forces refer to interactions or juxtapositions between individuals with
different group memberships; and system-wide forces refer to attributes or actions of whole
institutions or societies. Indeed, an orientation towards these multiple levels of analysis is what

26

defines SDT from other theories of prejudice, discrimination, and stereotypes (e.g., conflict
theory, social identity theory, self-categorization theory) (Pratto, Sidanius, & Levin, 2006).
SDT (Sidanius & Pratto, 2001, 2011) contends that if we want to understand acts of
discrimination, prejudice, and oppression, we must recognize that group-based social hierarchies
exist. In other words, individual social value is not just based on individual attributes. Instead,
individual social value is dependent on membership in different social groups. These groups (i.e.,
group-based social dominance hierarchies) can come in three forms: age-based hierarchies in
which older persons have more positive social value as compared to younger persons; genderbased hierarchies in which men have more positive social value as compared to women; and
arbitrary-set hierarchies, in which social groups are defined in a variety of ways depending on
the specific social context. For example, race-based social hierarchies are a prominent form of
arbitrary-set hierarchies throughout history in the United States whereas class-based social
hierarchies are more prominent in European history (Sidanius & Pratto, 2001).
These different group-based social hierarchies are supported or alternatively combated
via a variety of different forces. These forces include specific roles (e.g., the role of a police
officer, human rights advocate, or charity worker; see Sidanius, Pratto, Sinclair, & Van Laar,
1996 for an in-depth discussion of roles), institutions (e.g., educational, religious, and
government institutions), and belief structures (e.g., values, norms, and stereotypes) (Sidanius et
al., 1994; Sidanius & Pratto, 2001, 2011) . Forces that support these hierarchies are termed
hierarchy-enforcing forces whereas forces that combat these hierarchies are termed hierarchyattenuating forces. Hierarchy-enhancing and –attenuating forces operate at individual,
interpersonal, and system-wide levels.

27

There are two hierarchy-enhancing forces operating at the system-wide level of SDT:
institutional discrimination and legitimizing myths. Institutional discrimination includes
institutional rules, procedures, and actions that result in a disproportionate allocation of social
value among the dominant and subordinate groups (Sidanius & Pratto, 2001). Legitimizing
myths are shared ideologies that are frequently embedded in a culture and are drawn upon to
explain how institutional discrimination, among other forms of discrimination, prejudice, and
oppression, are fair, legitimate, natural, or even necessary (Sidanius & Pratto, 2001).
Legitimizing myths can take many forms including attitudes, values, beliefs, or stereotypes and
frequently appeal to morality or intellect. Additionally, they may be based on some truth, or be
completely false. Among this variability, what unites legitimizing myths is their ability to justify
action and inaction in a social system so as to maintain (in the case of hierarchy-enhancing
legitimizing myths) or challenge (as in the case of hierarchy-attenuating legitimizing myths)
current group-based social hierarchies.
A final key component of SDT in understanding the mechanistic relationship between
legitimizing myths and institutional discrimination is context. Unlike legitimizing myths and
institutional discrimination, context operates on the intergroup level in SDT and is more
specifically referred to as unequal intergroup context (Sidanius & Pratto, 2011). This refers to
any setting in which individuals occupying different group memberships interact with one
another (e.g., a young teenager interacts with a middle-aged adult). SDT explains that this
intergroup context is unequal as it brings up stereotypes between the interacting groups and
histories of past intergroup conflicts, perceived threats, and beliefs in separate identities. As a
result, this context gives way to and provokes prejudicial and discriminatory behavior that
ultimately reinforces group-based social hierarchies (Sidanius & Pratto, 2011). The unequal

28

intergroup context impacts both institutional discrimination and legitimizing myths at the
system-wide level. Figure 2 below provides a simplified schematic of SDT that illustrates the
relationships between the system-wide components of institutional discrimination and
legitimizing myths with the interpersonal component of unequal intergroup context. SDT’s
theoretical model identifies legitimizing myths as the key mechanisms that allow for institutional
discrimination within unequal intergroup context.

Figure 2. Simplified schematic of social dominance theory9

Unequal
Intergroup
Context

Legitimizing
Myths

Institutional
Discrimination

Social Dominance Theory and the Criminal Justice System Response to Sexual Assault
Social dominance theory has long been considered a useful theoretical model for studying
the criminal justice system; one of the original theorists credits his own interactions with the
criminal justice system as the “building blocks of SDT” (Sidanius & Pratto, 2011, p. 6). Within

9

Adapted from Sidanius and Pratto (2001, p. 40).

29

this theoretical framework, “the criminal justice system is one of the most important hierarchyenhancing social institutions” (Sidanius et al., 1994, p. 340). SDT notes that whereas criminal
justice system policies and rhetoric purport “equality before the law,” its practices do not uphold
this fundamental principle (Sidanius et al., 1994). Instead, criminal justice system actions
frequently constitute institutional discrimination.
This disconnect between the policies and rhetoric of the criminal justice system, and its
institutional practices, can be applied readily to the criminal justice system response to sexual
assault. Lisak (2008) presents the paradox:
In the hierarchy of violent crimes, as measured by sentencing guidelines, rape
typically ranks only second to homicide, and in some cases ranks even
higher…such sentencing structures serve as a message from the community: “we
view rape as an extremely serious crime.” At the same time, however, the number
of rapes that are actually prosecuted is a tiny fraction of the number committed in
any year…among those [sexual assaults] that are reported, attrition at various
levels [of the criminal justice system] dramatically reduces the number of actual
prosecutions. Ultimately, only a tiny handful of rapists ever serve time for rape, a
shocking outcome given that we view rape as close kin to murder in the taxonomy
of violent crime. (p. 1)
The high rates of sexual assault case attrition, as highlighted by Lisak (2008) and prevalence of
unsubmitted SAKs nationwide could be considered concrete manifestations of institutional
discrimination within the SDT framework.
Second, to continue the application of SDT to the investigation stage of the criminal
justice system response, there must also be evidence of unequal intergroup context (see Figure

30

2). This too has been discussed in the literature. Predictive studies have examined the impact of
social identity factors of the victim and the suspect on investigational effort and case outcomes.
Factors such as the age and race of the victim and perpetrator have been found to impact how
cases progress through the criminal justice system and SAK submission to the crime lab for
analysis (D. Black, 1978; Bryden & Lengnick, 1997; Campbell, Bybee, Ford, et al., 2009;
Campbell, Bybee, Patterson, et al., 2009; Frohmann, 1997; Heenan & Murray, 2006; Horney &
Spohn, 1996; Kelly et al., 2005; Kerstetter, 1990; LaFree, 1981; Reiss, 1971; Rose & Randall,
1982; Schuller & Stewart, 2000; Shaw & Campbell, 2013; Smith & Klein, 1983; Triggs et al.,
2009; Wriggins, 1983) (see The Actual Criminal Justice System Response above).
The third component of SDT needed to understand the investigation stage of the criminal
justice system response to sexual assault is the supporting mechanism of legitimizing myths.
Rape myths, specifically, have been identified as one set of legitimizing myths in the United
States that operate to justify or excuse related discrimination, prejudice, and oppression
(Edwards et al., 2011; Pratto et al., 1994; Sidanius & Pratto, 2001, pp. 85-87). The concept of
rape myths were originally introduced in the 1970s in order to explain a set of predominately
false beliefs about how and why sexual violence is perpetrated against women (for a review, see
Edwards et al., 2011). Rape myths frequently define who rapes and who can be raped, minimize
or rationalize the assault, assign blame to the victim for being raped, and excuse the rapist of all
responsibility. The key question is if law enforcement personnel are evoking rape myths as a
means to justify, morally or intellectually, their pervasive negligent response to sexual assault.
To date, the literature examining rape myth acceptance among law enforcement
personnel, and the general public, has most commonly used the Rape Myth Acceptance Scale
(RMAS) (Burt, 1980) and the Illinois Rape Myth Acceptance Scale (IRMAS) (Payne, Lonsway,

31

& Fitzgerald, 1999), though a variety of other measures exist and are utilized (see Edwards et al.,
2011; Lonsway & Fitzgerald, 1995) These scales include items like, “rape mainly occurs on the
“bad” side of town,” “if the rapist doesn’t have a weapon, you really can’t call it rape,” “a
woman who dresses in skimpy clothes should not be surprised if a man tries to force her to have
sex,” and “men don’t usually intend to force sex on a woman, but sometimes they get too
sexually carried away” (Burt, 1980; Lonsway & Fitzgerald, 1995; Payne et al., 1999).
Law enforcement personnel’s overall endorsement of rape myths has been extensively
studied in research over many years (Brown & King, 1998; Campbell & Johnson, 1997; Edwards
et al., 2011; Feldman-Summers & Palmer, 1980; Field, 1978; Galton, 1975; Gylys & McNamara,
1996; LaFree, 1989; Maddox et al., 2012; Page, 2008a, 2008b, 2010). While the specific rape
myths have shifted in that many law enforcement personnel no longer believe that, for example,
all victims are promiscuous, have bad reputations, and secretly desire to be raped, there are new
“modern attitudes” in which police officers tend to discount the experiences of some survivors
including married women reporting marital rape or women engaged in sex work who have been
assaulted (Page, 2008a, 2008b, 2010). In a recent study of rape myth acceptance among police
officers, Page (2010) found that even when police officers did not endorse items on the rape
myth acceptance scale, six percent provided pejorative and sexist remarks in a write-in comment
section at the end of the survey. Many of the comments targeted the researcher asking if she had
been raped, if she was gay, and if she hated men. Comments also attempted to discredit the
research itself by citing its female agenda and describing it as worthless. Other comments
focused on allegations of rape and what qualifies as a sexual assault. For example, one officer
wrote in that a problem with rape investigations are “victims/prostitute who do not get paid for
their services. This is not rape” (Page, 2010, p. 27). So while law enforcement personnel may not

32

have endorsed rape myths on the close-ended quantitative scale, some exhibited rape myth
endorsement with their write-in comments. This literature suggests that rape myths may be
acting as legitimizing myths in the criminal justice system response to sexual assault and,
furthermore, that law enforcement personnel may provide additional justifications for their
response to sexual assault beyond what may be expected given the rape myth literature.
The existing literature on the investigative stage of the criminal justice system response
to sexual assault can be aligned with the system-wide and intergroup component in SDT: current
research on rape myth endorsement among police aligns with legitimizing myths; research
documenting minimal investigational effort (e.g., unsubmitted SAKs) and low case outcomes
(i.e., high case attrition) across sexual assault cases align with institutional discrimination; and
differential investigational effort and case outcomes between sexual assault cases depending on
specific identity factors of the victim and suspect align with unequal intergroup context. Figure 3
provides an augmented version of the social dominance theory schematic, filling in these specific
components in their appropriate domain.

33

Figure 3. Rape myths, investigatory effort and case outcomes, and social identity factors of the
victim and perpetrator in the SDT framework

Victim and Perpetrator Social
Identity Factors
(Unequal Intergroup Context)

Investigatory Effort and
Case Outcomes
(Institutional Discrimination)

Rape Myths
(Legitimizing Myths)

However, the current literature base on the criminal justice system response to sexual
assault has not tested explicitly the relationships between rape myths as legitimizing myths and
the other constructs in Figure 3 (i.e., victim and perpetrator social identify factors as unequal
intergroup context and investigatory effort and case outcomes as institutional discrimination).
SDT predicts a relationship between rape myth endorsement and investigatory effort and case
outcomes, but this relationship has yet to be tested empirically. Specifically, prior literature on
rape myth endorsement among police has only measured their attitudes towards rape via a
variety of rape myth scales and has not examined how these attitudes towards rape are observed
during the course of a police investigation. For example, an officer may note in his/her case
records that the victim did not have any visible injuries and appeared rather ‘put together’ for just
being raped (i.e., endorses rape myths regarding what the victim should look like—effectively
minimizing the rape). In doing so, he/she provides justification for his/her lack of investigational

34

effort on the case (e.g., he/she may not have canvassed the area, obtained a witness statement, or
questioned a potential suspect) and perhaps its premature closure (i.e., closed prior to referral to
the prosecutor or an arrest of a suspect).
However, it is also possible that police attitudes towards rape cannot be observed in this
manner and no explanation is provided on individual cases for their lack of investigational effort
or case progression. Because no studies have documented how rape myths and other legitimizing
myths can be observed in police records of individual sexual assault cases, it has not been
possible to test empirically their relationship with the investigatory effort put into the case (as
indicated by investigational steps completed; see Figure 1) and the case outcome. Furthermore,
because rape myth and other legitimizing myth endorsement has not been investigated in this
way (i.e., observations of the endorsement of rape myths and other legitimizing myths on a caseby-case basis), it has also not been possible to examine empirically the relationship between the
social identity factors of the victim and suspect with the observations of rape myth and other
legitimizing myth endorsements on a specific case. Figure 4 provides a further augmented
version of the social dominance theory schematic. In this figure, the domains and relationships
that have yet to be tested empirically are presented as dashed lines.

35

Figure 4. Relationships between rape myths, case outcomes and investigatory effort, and social
identity factors of the victim and perpetrator as examined by the empirical literature
Victim and Perpetrator Social
Identity Factors
(Unequal Intergroup Context)

Investigatory Effort and
Case Outcomes
(Institutional Discrimination)

Rape Myths
(Legitimizing Myths)

In examining the investigative stage of the criminal justice system response to sexual
assault through the SDT lens (as presented in Figure 4), it becomes evident that research needs to
document how rape myths, and other legitimizing myths not previously identified in the
literature, are observed in police records on a case-by-case basis in order to examine its
relationship with investigatory effort and case outcomes, and with social identity factors of the
victim and perpetrator. In the language of SDT, this would allow for an examination of the
relationship between legitimizing myths (e.g., rape myths) and institutional discrimination (i.e.,
investigatory effort and case outcomes), as impacted by the unequal intergroup context (i.e.,
specific social identity factors the victim and perpetrator). Addressing these key empirical gaps
in the literature is critical because if we can identify how law enforcement personnel justify their
response to sexual assault in police records (e.g., via rape myth endorsement), change efforts that
target these justifications can be developed in order to improve the criminal justice system
response to sexual assault.

36

THE CURRENT STUDY
Sexual assault is a gendered crime perpetrated primarily by men against women (M. C.
Black et al., 2011). Post-assault, victims may choose to report the assault to the criminal justice
system; however, very few of these reported cases will continue on in the criminal justice system
process (Lonsway & Archambault, 2012); what is supposed to be a two-stage process involving
an investigation stage and a prosecution stage many times becomes a one-stage investigative
response in practice. The problem of sexual assault case attrition is to be expected from a social
dominance theory perspective, as the criminal justice system is a hierarchy-enhancing institution
and is expected to exhibit institutional discrimination (Sidanius et al., 1994; Sidanius & Pratto,
2001). The stockpiles of unsubmitted sexual assault kits in jurisdictions across the country
(Human Rights Watch, 2009, 2010; Strom & Hickman, 2010) are tangible symbols of the overall
lack of response to sexual assault within the criminal justice system, and more specifically
among law enforcement personnel. Varied investigational effort has also been documented
between sexual assault cases based on the unequal intergroup context of the race and age of the
victim and perpetrator (e.g., see Campbell et al., 2011; Frohmann, 1997; LaFree, 1981; Rose &
Randall, 1982; Shaw & Campbell, 2013; Wriggins, 1983).
The institutional discrimination exercised by the criminal justice system, and more
specifically law enforcement personnel (i.e., minimal investigational effort and high case
attrition), operates on a system-wide level to support group-based social hierarchies (Sidanius &
Pratto, 2001, 2011). Also operating on this system-wide level are legitimizing myths. SDT
explains that legitimizing myths are used to justify and provide moral and intellectual legitimacy
for actions taken by the criminal justice system, and more specifically law enforcement
personnel (Sidanius et al., 1994). Rape myths have been identified as one set of legitimizing

37

myths in the United States and can facilitate the action and inaction taken by police within the
context of sexual assault (Edwards et al., 2011; Lonsway & Fitzgerald, 1995; Pratto et al., 1994;
Sidanius & Pratto, 2001). To date, rape myth acceptance among law enforcement personnel has
only been examined via quantitative rape myth acceptance scales and has effectively measured
attitudes. No study has yet documented observations of rape myth endorsement in the regular
day-to-day operations of law enforcement personnel, or attempted to document other ways in
which law enforcement personnel justify their response to sexual assault, and then link these
endorsements to investigatory effort and case outcomes on specific cases. In other words, while
prior literature has illuminated the “what,” “who,” “when,” and “where” in regards to the
investigation stage of the criminal justice system response to sexual assault, no prior studies have
provided the “why” and “how.”
Figure 4 on page 36 provides a simplified schematic of SDT focusing on its system-wide
components (i.e., legitimizing myths and institutional discrimination), in consideration of the
interpersonal component of unequal intergroup context, and how each of these components
interfaces with the criminal justice system response to sexual assault. The dashed lines in figure
4 highlight the relationships we would expect via SDT, but that have yet to be examined
empirically. Examination of these relationships requires a two-step process.
It is first necessary to examine if and how law enforcement personnel’s attitudes towards
rape are observed in their sexual assault case investigations. Prior literature has relied upon rape
myth acceptance scales to provide a measure of law enforcement personnel’s attitudes towards
rape. Page (2010) documented rape myth endorsement in law enforcement personnel’s write-in
comments on a survey. Additionally, law enforcement personnel have “fundamental control”
over what “facts” are included in a case (see Fisher, 1993, p. 3). Therefore, it is possible that

38

attitudes towards rape are observable in law enforcement personnel’s police records of sexual
assault cases—they record statements in their notes that indicate their endorsement of rape
myths, as well as provide other justifications for their response to sexual assault. Police records
have not been examined in this way in prior literature. Therefore, this is a necessary first step.
Second, the observations of legitimizing myths documented in Study 1 need to be
examined to see if they predict investigatory effort and case outcomes, within the context of
specific victim and perpetrator social identity factors. Prior literature has recorded and examined
relationships between investigatory effort, case outcomes, and victim and perpetrator social
identity factors. However, because observations of legitimizing myths in sexual assault case
records have never been documented, their relationships to these other variables have not been
assessed. Therefore, the second step is to record the investigatory effort, case outcomes, and
victim and perpetrator information alongside observations of legitimizing myth endorsements to
test empirically if law enforcement personnel rely on legitimizing myths to justify their response
to sexual assault, or if attitudes towards rape cannot be observed in this way.
To explore these ideas, it was necessary to access sexual assault police reports in order to
code for these variables. Given that the focus of the current project was the negligent criminal
justice system response to sexual assault during the investigation stage, it was necessary to use a
sample that had been subject to this inadequate response. As previously discussed, there have
been stockpiles of unsubmitted SAKs found in police property storage facilities in jurisdictions
across the country (see Strom & Hickman, 2010). These unsubmitted SAKs serve as a recent
tangible symbol of the inadequate response to sexual assault; their corresponding police records,
therefore, provide the ideal sampling frame.

39

Detroit was one such jurisdiction with a large number of unsubmitted SAKs in policy
property. In 2009, representatives from local police, state police, and the prosecutor’s office were
touring a remote police property storage facility in Detroit to discuss how to best manage the
high volume of evidence in police custody. During the tour, an assistant prosecutor noticed
dozens of storage boxes that turned out to be holding approximately 10,000 SAKs. The assistant
prosecutor informed the elected Prosecutor of what was seen in the property storage facility. The
Michigan Domestic and Sexual Violence Prevention and Treatment Board (MDSVPTB) soon
thereafter learned what had happened and assembled an interdisciplinary team of prosecutors,
law enforcement, medical professionals, and community-based victim advocates10 to review and
discuss how to respond to the problem. MDSVPTB was able to secure federal funds from the
Department of Justice, Office of Violence Against Women, to fund testing of a random sample
of 400 of these SAKs and sustain the interdisciplinary team to review all case files corresponding
to each of these 400 kits to determine potential investigatory and prosecutorial action to be taken
on each case (i.e., The 400 Project).
Whereas the stockpile of unsubmitted kits in Detroit was not the first of its kind, the 400
Project was an unprecedented undertaking in that never before had a jurisdiction responded to
such a discovery in a systematic, empirically-informed way. The 400 Project informed the
resource requirements and potential outcomes of testing all of the unsubmitted kits (Pierce &
Zhang, 2011), how to approach, notify, and interact with victims after so many years had passed

10

In general, victim advocates promote victims’ rights and ensure that their needs are given priority. Advocates may
be systems-based or community-based. Systems-based advocates work within a specific organization or system and
provide services to victims currently interacting with that organization/system and may be limited in their ability to
provide confidential services. For example, many law enforcement agencies have systems-based advocates that
work with victims while they are interacting with the criminal justice system. Community-based advocates work in
independent organizations that are not housed within another organization or system. Unlike systems-based
advocates, community-based advocates typically provide confidential services and work with victims regardless of
if they choose to engage with other systems (e.g., report the crime to law enforcement personnel) (Office for Victims
of Crime, n.d.)

40

since their assault, and the potential legal viability of cases that have been shelved for a number
of years.
With the permission and approval of MDSVPTB and the Department of Justice, Office of
Violence Against Women, the corresponding case files for each of the 400 randomly selected
SAKs from the 400 Project were used as the data source for the current study. Police records did
not exist or existed but could not be located for 136 SAKs, 14 SAKs (and their corresponding
police records when available) were not for a sex crime 11, and two SAKs corresponded to sexual
assaults that occurred outside of Detroit’s jurisdiction. This provided a final sample size of
N=248 case files. Having attained access to this sample, a multi-study was designed that would
first examine if and how police attitudes towards rape are observed in their written records of
sexual assault investigations, and then test empirically if these observations are used to justify
their response (i.e., investigational effort and case outcomes). Each study, along with its specific
aims, is described below.
Study 1
The purpose of Study 1 was to document the extent and types of legitimizing myths (e.g.,
rape myths) in police records of sexual assault case investigations. Rape myths have been
identified as one form of legitimizing myths used to justify or excuse related discrimination,
prejudice, and oppression (Edwards et al., 2011; Pratto et al., 1994; Sidanius & Pratto, 2001, pp.
85-87), such as the criminal justice system response to sexual assault. To date, rape myth
endorsement has usually been measured with attitude surveys, such as the Rape Myth
Acceptance Scale (Burt, 1980) and the Illinois Rape Myth Acceptance Scale (Payne et al., 1999)

11

Some of the SAKs found in the Detroit police storage facility were associated with non-sex crimes (e.g.,
homicide); cases associated with these SAKs are not included in the current study. For example, after the SAK was
retrieved and opened at the crime lab, it was discovered that its contents were not related to a sexual assault and the
box had simply been used as a container to hold evidence for other types of crimes.

41

(see Social Dominance Theory and the Criminal Justice System Response to Sexual Assault
above). These scales are consistently used to measure attitudes towards rape and are
administered as a survey to research participants.
In 2013, Kettrey developed a coding scheme for identifying these same constructs (i.e.,
deductive coding) in narrative and archival text. To build a codebook, Kettrey (2013) reviewed
the items in the Rape Myth Scale (Lonsway & Fitzgerald, 1995) and produced 16 myths with
corresponding operationalizations. However, Page (2009; 2010a; 2010b) investigated attitudes
towards rape among law enforcement personnel and found that there were new, “modern
attitudes” about who rapes, who can be raped, and what qualifies as ‘real’ rape. Furthermore,
Page found that even when law enforcement personnel did not endorse rape myths on the rape
myth acceptance scale, they provided write-in comments that SDT would likely classify as
legitimizing myths as they were beliefs, values, and stereotypes that could be used to justify
discrimination, prejudice and/or oppression (e.g., law enforcement personnel wrote in that the
researcher was only doing this work because she had been raped, was gay, and/or hated men).
Therefore, in order to capture the full spectrum of manifestations of legitimizing myths in police
records of sexual assault cases, it was necessary to identify and code for more traditional rape
myths, as operationalized by Kettrey (2013), as well as identify and theme other potential
legitimizing myths that could be used to justify institutional discrimination, prejudice, and/or
oppression.
The first aim of Study 1 was to identify if legitimizing myths are observable in police
records of sexual assault case investigations. Prior literature has only assessed attitudes towards
rape among law enforcement personnel. The first aim of Study 1 was to examine if these
attitudes are observable in the written police records of sexual assault investigations. To do this,

42

Kettrey’s (2013) operational definitions for rape myths were adapted for Study 1 and police
reports were reviewed and coded (i.e., directed content analysis) for whether the text included
references to (1) if the victim fought back; (2) exaggerated or lied; (3) consented to any part of
the sex act or had previous sexual relations with the offender; (4) had bruises or marks; (5) was
upset; (6) was a sex worker; or (7) was drunk or high when interacting with law enforcement.
However, when Page (2010) assessed for rape myth endorsement among law enforcement
personnel, she found that participants wrote in additional, unsolicited comments on the survey
that were not detected when screening for the more traditional rape myths. Similarly, there could
have been different manifestations of legitimizing myths in police records, beyond what was
included in Kettrey’s operationalizations. Therefore, Study 1 also included a phase of
conventional content analysis to allow for different manifestations of legitimizing myths to
emerge from the data.
The second aim of Study 1 was to document the extent and types of legitimizing myth
endorsement in police records of sexual assault case investigations. After determining if
legitimizing myths were observable in police records of sexual assault case investigations, it was
necessary to examine how frequently they occurred (i.e., extent of legitimizing myth
endorsement) and a legitimizing myth typology (i.e., types of legitimizing myth endorsement).
The second aim of Study 2 intended to categorize the observations of legitimizing myths into
higher-order conceptually defined categories and describe their frequency across cases (i.e., how
many cases included each of the different types of legitimizing myths) and within cases (i.e.,
how many of each type of legitimizing myths were endorsed in a single case).

43

Study 2
The purpose of Study 2 was to examine the relationships between the different types of
legitimizing myths, investigational effort, and case outcomes in sexual assault cases, within the
context of specific social identity factors of the victim and perpetrator (i.e., sex, race, and age)
and the number of perpetrators. Figure 4 on page 36 provides a simplified schematic of SDT;
Figure 5, below, provides the conceptual model for Study 2 and is slightly different from Figure
4 in that investigatory effort and case outcomes (i.e., institutional discrimination) have now been
split into two separate variables.

Figure 5. Current Study’s Conceptual Model

Victim and Perpetrator
Social Identity Factors
(Unequal Intergroup Context)
Box
4

Investigatory Effort
(Institutional Discrimination)
Rape Myths and Other
Justifications
(Legitimizing Myths)

Box
2

Box
1

Case Outcomes
(Institutional
Discrimination)
Box
3

While possibly related, investigatory effort and case outcomes were conceptualized and
assessed in the current study as two separate constructs. There can be varying levels of

44

investigational effort (i.e., investigative steps completed, see Figure 1 on page 13) across sexual
assault cases, and the level of investigational effort put into a single case may or may not be
related to the case’s outcome while under the purview of law enforcement (i.e., if the case results
in an arrest and a referral to the prosecutor). For example, two cases may have the same
investigational effort put into them (i.e., investigative steps completed) with one case moving
onto prosecution and the other closed because the investigators were unable to locate the suspect.
Indeed, if law enforcement personnel consistently conduct thorough investigations prior to
deciding a case outcome, we would expect to see the same investigational effort across all cases
(i.e., investigative steps completed) and a varied range of case outcomes. The relationship
between these two variables, and to the other variables in the model, was tested empirically.
Empirical examination of all relationships in Figure 5 was done in an exploratory
fashion—specific hypotheses for each relationship were not delineated prior to analysis. The
rationale for this approach is described in the section, “Rationale for path analysis and an
exploratory analytic approach” on page 108. In short, prior literature suggested which constructs
and variables should be included in the conceptual model. However, prior literature had not
examined empirically the relationships between included variables. That is, it was possible that
legitimizing myths were used by law enforcement to explain their response to sexual assault
cases. Alternatively, law enforcement may not have felt compelled to provide a justification and
bypassed it entirely. Accordingly, Study 2 intended to examine the relationships among the
variables in Figure 5 via path analysis in an exploratory fashion, without specifying directional
hypothesis to be tested. The key relationships to be tested are described in the study aims below.
The first aim of Study 2 was to examine the relationships between the different types of
legitimizing myths, as identified in Study 1, and investigational effort and case outcomes. While

45

the overall criminal justice system response to sexual assault is stunted (see Lonsway &
Archambault, 2012) and all the cases in this study were “shelved,” there may have been variation
across individual cases in terms of how they were investigated and if the case lead to an arrest
and/or referral to the prosecutor, given previous research (e.g., see Campbell, Greeson, Bybee,
Kennedy, & Patterson, 2011; Frohmann, 1997; LaFree, 1981; Rose & Randall, 1982; Shaw &
Campbell, 2013; Wriggins, 1983). It was then possible that the investigatory effort and case
outcomes were predicted by legitimizing myth endorsements in the police file. To examine this
relationship, investigational effort and case outcomes were recorded across the 248 cases files in
the sample. To document investigational effort, the case files were each coded for the specific
investigational steps taken in the case (e.g., if the officer obtained a victim’s statement,
canvassed the area, interrogated a suspect, etc.) and summed to yield a total number of
investigative steps. To document case outcomes, the case files were each coded for if the case
resulted in (1) an arrest and a referral to the prosecutor; (2) no arrest and a referral to the
prosecutor; (3) an arrest and no referral to the prosecutor; or (4) no arrest and no referral to the
prosecutor.
This set of possible outcomes was selected as it presents the different possible outcomes
of the case while still under the purview of law enforcement. A case that ends with an arrest and
a referral to the prosecutor is the only way for a sexual assault case to continue on in the criminal
justice system. Aim1 then examined the relationships between different types of legitimizing
myths (from Study 1), investigational effort, and case outcomes (see the arrows between Boxes 1
and 2 and between Boxes 1 and 3 in Figure 5).
The second aim of Study 2 was to examine the relationship between investigational effort
and case outcomes. If a thorough investigation was done on every case prior to determining its

46

case outcome while under the purview of law enforcement, there should be no relationship
between the number of investigative steps completed on the case and the case outcome (i.e., if
the case resulted in an arrest and a referral to the prosecutor). For example, if a case is
determined to be unfounded (and there was no associated arrest or referral to the prosecutor), this
should be decided after the case has been thoroughly investigated. If a case is referred to the
prosecutor, this should occur after a thorough investigation. However, SDT would suggest that
law enforcement personnel may prematurely decide case outcomes that support the status quo
prior to and therefore influencing the investigative effort they put into a case. This relationship
will be examined empirically with no specific directional hypothesis (see the arrow between
Boxes 2 and 3 in Figure 5).
The third aim of Study 2 was to examine the impact of specific social identity factors of
the victim and perpetrator (i.e., sex, race, and age) and the number of perpetrators on the types of
legitimizing myths endorsed, investigational effort, and case outcomes. SDT posits that at the
intergroup level, contexts that involve interactions across groups (i.e., intergroup contexts) afford
for prejudicial and discriminatory behavior (Sidanius & Pratto, 2011). Sexual assaults occur in
intergroup contexts in which individuals of different sexes, races, and ages interact with one
another (e.g., an African-American 24 year-old male assaults an African-American 32 year-old
female). As such, the specific context of each case (i.e., the sex, race, and age of the victim and
perpetrator) could affect endorsement of the different types of legitimizing myths, investigatory
effort, and/or the case outcome while under the purview of law enforcement. All of these
potential relationships will be empirically tested with no specific directional hypotheses (see the
arrows between Box 4 and Boxes 1-3 in Figure 5).

47

It is important to note that nearly one quarter of the cases in the current sample involved
multiple perpetrators. However, the level of detail regarding the race, and particularly the age, of
additional perpetrators was inconsistent. This made it difficult to code consistently for and
analyze the specific social identity factors for the additional perpetrators and would have resulted
in a greater degree of missing data. Therefore, the specific social identity factors for the first
perpetrator listed in the sexual assault report were recorded and used for analysis examining the
influence of specific social identity factors of the perpetrator on other variables in the model. An
additional binary variable indicating whether or not the case has a single perpetrator or multiple
perpetrators was created and incorporated into the analysis so as to not have a complete loss of
data on additional perpetrators involved in the assault.

48

STUDY 1: DOCUMENTING LEGITIMIZING MYTHS
Method
Sample
Study 1 examined the case records corresponding to 400 sexual assault kits (SAKs)
randomly selected from the population of 10,559 SAKs counted in the Detroit police storage
facility (as of 2009). The random selection of the 400 SAKs was conducted by Michigan State
University Center for Statistical Training and Consulting (Pierce & Zhang, 2011). All case files
(e.g., police records, medical records, investigator and prosecutor notes) associated with each of
the 400 SAKs were collected, compiled, and maintained by the Michigan Domestic and Sexual
Violence Prevention and Treatment Board (MDSVPTB). Police records did not exist or could
not be located for 136 SAKs, 14 SAKs (and their corresponding police records when available)
were not for a sex crime12, and two SAKs corresponded to sexual assaults that occurred outside
of Detroit’s jurisdiction. This yielded a final sample size of N=248 case files.13 MDSVPTB, with
permission of their project funder, the Office of Violence Against Women, U.S. Department of
Justice, authorized use of these data for the current study both in their raw form (the original case
files) and with a data file that contains information on the sex, age, and racial identity of the
victim and perpetrator(s).

12

Some of the SAKs found in the Detroit police storage facility were associated with non-sex crimes (e.g.,
homicide); cases associated with these SAKs are not included in the current study. For example, after the SAK was
retrieved and opened at the crime lab, it was discovered that its contents were not related to a sexual assault and the
box had simply been used as a container to hold evidence for other types of crimes.
13
There are no data or information to suggest that the 136 cases eliminated from the current project due to missing
police records were systematically different from the 248 cases included in the project. The responding police
department moved locations (and their case files) at least six times over the thirty years that the cases accumulated
and maintained paper records during this time. This likely resulting in the misplacement or loss of police files.
Therefore, the cases excluded from the current project are assumed to be missing at random rather than
systematically missing.

49

Preparing Case Files for Coding
This project utilized archival data and required no direct contact with participants,
however the files did contain identifying and sensitive participant information. To protect
participant privacy, all data were maintained electronically on password protected computers and
only the investigators had access to the data. Coders were only able to gain access to the data on
the password protected computers with the supervision of the project director. All cases were
assigned a unique identifier for project purposes and all data coding was completed
electronically (i.e., hard copies of the raw and coded data will not be maintained).
Identifying information was redacted as soon as possible after cases had been coded. Victim
names were replaced with “Victim 1,” “Victim 2,” etc. A similar scheme was used for suspects,
bystanders, friends, and any other names that appeared in the case records. To redact these
names, the “redaction” tool was used in Adobe Acrobat 9 Pro. The investigator selected
“advanced” in the toolbar, went down to “redaction” and selected “redaction properties.” On the
“appearance” tab, custom text was entered according to the naming scheme described above
(e.g., “Victim 1,” “Suspect 1,” etc.). Each redaction was applied permanently removing the
original text. It was replaced with a black box with the custom text overlayed. The redacted PDF
was saved and the original intact file was deleted and completely removed from all computers.
Redaction tools in Adobe Acrobat 9 Pro are designed to permanently delete sensitive data (See
http://www.adobe.com/products/acrobat/pdf-redaction.html). All procedures were approved by
the Michigan State University IRB.
Coder Training
Case files included in the study were coded by four coders (i.e., the project director and
three undergraduate research assistants) for the presence of legitimizing myths. Researching rape

50

is emotionally challenging work (Campbell, 2002) and prior literature on training interviewers to
work with victims of violence recommends a two-stage approach in which interviewers are first
trained on violence against women itself, then trained on how to interview (Campbell, Adams,
Wasco, Ahrens, & Sefl, 2009). While the coders for Study 1 were not working directly with
victims of violence, they were exposed to victims’ experiences of violence across dozens of
police reports. Accordingly, coder training followed the same two-stage training model. As is
common in this type of research (see Campbell, Adams, et al., 2009), the project director was
responsible for all coder training and employed group discussion methods.
During the first stage of training, coders read multiple articles on the dynamics of sexual
assault, the criminal justice response to sexual assault, self-care, and systems of oppression (e.g.,
Adams, Bell, & Griffin, 2007; M. C. Black et al., 2011; Lonsway & Archambault, 2012; Pharr,
1997). These specific topics were selected intentionally to prepare coders for engaging in this
emotionally-challenging research and to provide them with necessary resources and information
should they be triggered by the data; the readings did not include information about specific rape
myths or legitimizing myths so as to prevent priming effects. Coders completed assigned
reading, then discussed the material with the research team, as modeled in previous research (see
Campbell, Adams, et al., 2009). Readings were discussed at weekly research team meetings.
During the second stage of training, coders were trained on how to code the police reports
using directed content analysis (Hsieh & Shannon, 2005) for legitimizing myths that law
enforcement personnel used to justify their response to sexual assault. Directed content analysis
relies on existing theory or prior research for the design of an initial coding scheme, which is
then used to validate or extend conceptually the preexisting theoretical framework (Hsieh &
Shannon, 2005). In this study, prior research on rape myths was used to develop the coding

51

scheme. Specifically, Study 1 adapted Kettrey’s (2013) operationalizations for identifying rape
myths in narrative and archival text, based on existing attitude measures.
Directed Content Analysis Coding Procedures
Coders completed institutional review board (IRB) training and signed a confidentiality
agreement before viewing (e.g., for training purposes) or coding any data. Additionally, all
coders received additional training on ethics and the importance of protecting privacy and
confidentiality within the context of this study (e.g., these are criminal cases that are legally
viable). The project director supplied coders with coding instructions and the codebook defining
the initial set of a priori codes (see Appendix A for the coding instructions provided to all coders
and Appendix B for the coding scheme used for directed content analysis). This document was
reviewed with all coders to ensure they understood each operationalization and how to code the
data accurately and consistently. Of key importance during training was employing mechanisms
to protect against demand characteristics. Specifically, the coding scheme was conservative in
that the coder must have been able to identify the discrete line of text where the specific
legitimizing myth was endorsed by the police officer/investigator. If a specific line of text could
not be identified, it could not be coded. For example, Figure 6 shows two samples of text from
police records. Figure 6a shows a discrete line of text from a case file (case 184) that was coded
for “victim exaggerates or lies” (see Directed Content Analysis Codes below). The law
enforcement officer records that the “compl [complainant] story [ineligible] fictatious.” Figure
6b shows how the idea of the code “victim exaggerates or lies” was perhaps implied in a case file
(case 308) when the law enforcement officer recorded that the victim “stated that this was the
16th time being rapped,” but it was not directly stated. Therefore, the example provided in Figure
6b was not coded as “victim exaggerates or lies.”

52

Figure 6. Examples of the conservative coding scheme in action
Figure 6a. Example of a discrete line of text for “victim exaggerates or lies” (coded)

“Compl [complainant] story [ineligible] fictatious”
Figure 6b. Example of an implied line of text for “victim exaggerates or lies” (not coded)

“Compl stated that this was the 16th time being rapped”

All coders (i.e., the three undergraduate research assistants and project director) used the
codebook to code N = 6 randomly selected police reports as a group. Four additional cases were
then randomly selected and each coder coded two cases individually (i.e., each case will be
double coded). Codes were compared across coders to ensure consistency and accuracy.
Specifically, kappa was calculated across these 4 cases to examine inter-coder reliability with a
desired kappa score of 0.75 (Bakeman & Gottman, 1986; Fleiss, 1971; Pett, 1997). The first four
double-coded cases yielded a kappa score of 0.46. Disagreements between coders were discussed
among all coders until consensus was reached. Modifications were made to the codebook to
improve clarity and remove ambiguity before randomly selecting and double coding another four
cases. The second round of double coded cases resulted in a kappa score of 0.86, bringing the
overall training kappa score to 0.72 (the total kappa score among the eight double coded cases).
Again, disagreements between coders were discussed and the codebook was modified as

53

necessary and another four cases were randomly selected to be double coded. The third round of
doubled coded cases yielded a kappa score of 0.85, bringing the overall training kappa score to
0.76 (the total kappa score among the twelve double coded cases).
The training kappa score of at least 0.75 was achieved after twelve cases were double
coded, and initial training ended. A total of 18 cases had been coded for the a priori set of
legitimizing myths (see Appendix B) during training, leaving 230 cases to be coded. It was
essential that inter-rater reliability and consistency continued to be monitored. As such, thirty
percent (N = 69) of the remaining uncoded cases were randomly selected to be double coded. A
new kappa score (i.e., did not include the kappa calculated during the training) was calculated
after ten percent of the double coded files had been coded (N = 7). This continued throughout the
coding process to ensure that the total kappa score never fell below 0.75. Indeed, the total kappa
score for legitimizing myths at the end of coding was 0.81.
Directed Content Analysis Codes
Case files included in the study were coded by 4 researchers (i.e., the project director and
3 undergraduate research assistants) for the presence of a pre-determined set of legitimizing
myths. The a priori coding scheme was adapted from Kettrey’s operationalization of rape myths
(2013) and included seven codes (see Appendix B for the codebook). The seven codes were:
1. Victim didn’t fight back: This code referred to notes in police records that stated the
victim did not fight back, scream, or try to run away. For example, this included notes
that the victim had the opportunity to leave the scene of the assault, but chose not to
do so.
2. Victim exaggerates or lies: This code referred to notes in the police records that
stated the victim was exaggerating or lying, as well as statements that questioned the

54

victim’s story. For example, this included notes that the victim’s story didn’t seem to
line up, changed as it was retold, or didn’t seem plausible.
3. Victim consented: This code referred to notes in the police records that the victim
consented to engage in consensual sexual activity with perpetrator(s) at some point
during the assault (and perhaps then changed his/her mind) or on previous occasions.
For example, this included notes in reports documenting that the victim consented to
consensual sex with one perpetrator, and then was raped by several other perpetrators.
4. No bruises or marks: This code referred to notes in the police records that the victim
was not visibly injured during the assault. For example, this included notes that the
victim did not have bruises, marks, injuries, or otherwise appear disheveled. This
code also included comments about the victim’s clothing and outward appearance
(e.g., commenting that the victim’s shoes looked too clean for having been raped).
5. Not upset enough: This code referred to notes in the police records that the victim
did not appear upset or distraught, seemed distracted, or exhibited emotions that
would not be unexpected given the situation. For example, this included notes that the
victim was not crying or did not seem to be at all concerned with having just been
raped.
6. Victim was a sex worker: This code referred to notes in the police records that the
victim was a prostitute, ‘worked the streets,’ or that the rape was actually a
miscommunication related to the exchange of money for sex. For example, this
included notes that the rape was a ‘deal gone bad’ or a ‘trick gone bad.’
7. Victim is drunk/high: This code referred to notes in the police records that the
victim was drunk or high when interacting with law enforcement, or was a regular

55

drug user. For example, this included notes that the victim smelled of alcohol while
being interviewed, was high upon questioning, or was known to be a regular drug
user.
Overall, this coding scheme was successful in identifying observations of rape myths that
have been previously assessed and operationalized based on attitude assessments and
questionnaires (see Kettrey, 2013). However, while coding, all coders identified cases in which
law enforcement personnel offered other justifications for their response to sexual assault that
were not represented adequately in the list of preconceived codes. The project director consulted
with the project’s principal investigator (PI) and determined that the a priori coding scheme used
for directed content analysis may not have been sufficiently comprehensive in its ability to
identify the many different ways that law enforcement personnel justify their response to sexual
assault. This prompted a second phase of coding using conventional content analysis.
Conventional Content Analysis Coding Procedures
Conventional content analysis is an approach generally used when existing theory or
prior research on a phenomenon is somewhat limited (Hsieh & Shannon, 2005). In conventional
content analysis, researchers do not rely on an a priori coding scheme, but instead allow the
categories and codes to emerge from the data. This approach requires that the researchers
immerse themselves in the data in order for new insights to emerge; this frequently involves
reading all data repeatedly (Hsieh & Shannon, 2005; Kondracki & Wellman, 2002).
Accordingly, this task was allocated to a single coder—the project director—so that she would
be able to completely immerse herself in the data. During the second phase of coding, the project
director read and reread all police case files identifying statements that provided a direct answer
to the question, “why did law enforcement personnel respond to the case in this way?” Discrete

56

lines in the police files that answered this question were highlighted. For example, Figure 7
shows an example of text from a case file (case 345) in which law enforcement personnel
explain, in a discrete line of text, why they responded to the case in the way that they did.
Specifically, they noted that the victim was “uncooperative/hostile.”

Figure 7. Example of a discrete line of text identified during conventional content analysis

“C [circumstance]: Writes at scene attempted to t/t [talk to] comp., comp became
hostile/uncooperative during questioning. Writers again tried to obtain
information w/witness nurse [name redacted] in the room. Comp. again became
hostile/un-cooperative during questioning for information.”

In accordance with conventional content analysis, exact lines in police files that answered
this question were highlighted and the project director noted her initial thoughts and analyses
(Hsieh & Shannon, 2005). As this process continued, similarities between highlighted sections
(i.e., codes) were noted and labels emerged that were reflective of multiple codes. Many of these
labels came directly from the text (e.g., victim “is mental”). After all of the case records were
reviewed and coded, the project director consulted with the PI to ensure that the identified codes
were exhaustive and mutually exclusive. As stipulated by conventional content analysis,

57

definitions for each code were developed and examples were identified from the data (see
Conventional Content Analysis Codes and Results below).
Conventional Content Analysis Codes
Conventional content analysis resulted in the identification of nine new codes:
1. Victim didn’t act like a victim afterwards: This code referred to statements that
indicated the victim didn’t act like a victim afterwards and included notes that the
victim’s behavior immediately following the assault suggested that the assault didn’t
really happen as it didn’t match what might be expected given the circumstance.
2. Victim has “done this before:” This code referred to statements that the victim had
“done this before” and included notes that the victim had previously been assaulted,
reported an assault, and/or had reported an assault and then didn’t help in pursuing
investigation of the case.
3. Victim is “mental:” This code referred to statements that the victim was “mental”
and included notes that the victim was “mental” or had a mental illness.
4. Victim is promiscuous: This code referred to statements that the victim was
promiscuous and included notes that the victim was known to sleep with many men,
be sexually active, or even to have had prior abortions.
5. Victim is not credible: This code referred to statements that the victim was a
characterological liar and was not credible and included notes that the victim cannot
be trusted or is known to lie.
6. Victim is uncooperative: This code referred to statements that the victim was
uncooperative and included notes that the victim was hostile or intentionally
withholding information.

58

7. Victim doesn’t have enough information: This code referred to statements that the
victim didn’t have enough information. This did not refer to statements in which the
victim was intentionally withholding information, but rather that he/she just didn’t
know enough about the assault. For example, that the victim didn’t know the name of
the perpetrator or where the perpetrator could be found.
8. Victim has no phone/address for contact: This new code referred to statements that
the victim was not able to be contacted for follow-up during the investigation as
he/she had no phone or address and included notes that law enforcement personnel
were unable to contact the victim.
9. Victim or case is weak: This code referred to statements that the victim or case was
weak and also included notes that the victim was incompetent and wouldn’t make a
good witness.

Results
A total of fifteen codes were delineated during directed and conventional content
analysis. Six of the seven codes included in the a priori list of codes for directed content analysis
appeared in the case files (see Directed Content Analysis Codes above); “victim didn’t fight
back” was not endorsed in any of the case files and thus was not included in the final list of
codes presented here. An additional nine codes were identified via conventional content analysis,
yielding a total of fifteen different sub-categories of legitimizing myths that law enforcement
personnel used to justify their response to sexual assault. The identification of these fifteen
different legitimizing myths met the first aim of Study 1, to identify if legitimizing myths were
observable in police records of sexual assault case investigations.

59

To satisfy Aim 2 of Study 1, the fifteen different categories of legitimizing myths were
conceptually grouped into three different general types: circumstantial legitimizing myths,
characterological legitimizing myths, and investigatory blame legitimizing myths. Figure 8
provides a simple schematic of the distinctions and defining features of the legitimizing myths
typology. To build this typology, legitimizing myths were grouped based on their purpose. Prior
literature on rape myth endorsement suggests that rape myths are frequently used to minimize or
discount the rape, and that this is done by defining what rape is supposed to look like or what
qualifies as ‘real’ rape and who can be raped (Edwards et al., 2011; Kettrey, 2013). The fifteen
legitimizing myths were categorized first as to if their purpose appeared to be to minimize the
rape (i.e., circumstantial and characterological legitimizing myths) or not (i.e., investigatory
blame legitimizing myths). Legitimizing myths that appeared to minimize the rape were then
further categorized into legitimizing myths that minimized rape based on what qualifies as ‘real’
rape (i.e., circumstantial legitimizing myths) and who can be raped (i.e., characterological
legitimizing myths).

60

Figure 8. Simple schematic of legitimizing myths typology

Legitimizing Myths
Typology

Purpose of
Legitimizing Myth

Type & Definition
of Legitimizing
Myth

To minimize/discount
the rape

Circumstantial
Legitimizing Myths:
Suggest rape didn't
happen based on
specific circumstances

Characterological
Legitimizing Myths:
Suggest rape didn't
happen based on
specific characteristics

To blame the victim for
a less than through
investigation

Investigatory Blame
Legitimizing Myths:
Blame the victim for a
less-than-through
investigation

The first type of legitimizing myth was the circumstantial legitimizing myth.
Circumstantial legitimizing myths referred to statements that suggest the sexual assault did not
occur (i.e., minimize or discount the rape) based on specific circumstances of the sexual assault
or ways in which the victim presented to law enforcement (i.e., based on what rape is supposed
to look like). That is, it wasn’t rape because the victim was lying about it, wasn’t injured,
consented, wasn’t upset enough, or didn’t act like a victim afterwards. The defining
characteristic of circumstantial legitimizing myths was that they were limited to statements about
the circumstances of the assault or the victim.
The second type of legitimizing myth was the characterological legitimizing myth.
Characterological legitimizing myths referred to statements that suggest the sexual assault did
not occur based on specific characteristics of the victim (i.e., based on who can be raped). In
other words, it wasn’t really rape because the victim was a drug addict, was a sex worker, had
“done this before,” was “mental,” was promiscuous, or was just not credible; these groups of
61

people can’t really be raped. The defining characteristic of characterological legitimizing myths
was that they were limited to statements about the character of the victim.
The third type of legitimizing myth was the investigatory blame legitimizing myth.
Investigatory blame legitimizing myths blamed the victim for the fact that police conducted a
less-than-thorough investigation. To be clear, these legitimizing myths did not blame the victim
for the rape, or necessarily suggest that the rape did not happen; they blamed the victim for the
investigation not advancing as far as it might have been able to otherwise because the victim was
not willing or able to participate in the process. These legitimizing myths suggested that the case
proceeded as it did because the victim was not cooperating, did not provide enough information,
was not able to be contacted, or was a “weak” victim that would not hold up well during trial.
Table 2 lists all of the legitimizing myths in their respective categories along with how many
cases included each legitimizing myth endorsement. Table 3 provides count data for the total
number of circumstantial, characterological, and investigatory blame legitimizing myths
endorsed on each case.

62

Table 2. Legitimizing myth descriptives
Legitimizing Myth

N (% of total cases)

CIRCUMSTANTIAL MYTHS
Victim is lying
Victim is not injured
Victim consented
Victim is not upset
Victim didn’t act like a victim afterwards
TOTAL
CHARACTEROLOGICAL MYTHS
Victim is a regular drug user
Victim is a sex worker
Victim has “done this before”
Victim is “mental”
Victim is promiscuous
Victim is not credible
TOTAL
INVESTIGATORY BLAME MYTHS
Victim is uncooperative
Victim doesn’t have enough information
Victim has no phone/address for contact
Victim or case is weak
TOTAL
TOTAL OF ANY LM

63

30 (12.10%)
22 (8.87%)
20 (8.06%)
13 (5.24%)
4 (1.61%)
63 (25.40%)
14 (5.65%)
13 (5.24%)
9 (3.63%
8 (3.23%)
6 (2.42%)
6 (2.41%)
42 (16.94%)
72 (29.03%)
24 (9.68%)
16 (6.45%)
5 (2.02%)
102 (41.13%)
141 (56.85%)

Table 3. Count totals of each type of legitimizing myth
Number of
N (% of total cases)
Legitimizing
Myths Endorsed
CIRCUMSTANTIAL MYTHS
(0-5 possible)
0
185 (74.6%)
1
44 (17.7%)
2
14 (5.6%)
3
3 (1.2%)
4
2 (0.8%)
5
0 (0.0%)
CHARACTEROLOGICAL MYTHS
(0-6 POSSIBLE)
0
206 (83.1%)
1
31 (12.5%)
2
8 (3.2%)
3
3 (1.2%)
4
0 (0.0%)
5
0 (0.0%)
6
0 (0.0%)
INVESTIGATORY BLAME MYTHS
(0-4 POSSIBLE)
0
206 (58.9%)
1
88 (35.5%)
2
13 (5.2%)
3
1 (0.4%)
4
0 (0.0%)

Circumstantial legitimizing myths
Circumstantial legitimizing myths referred to statements that suggested the sexual assault
did not occur based on specific circumstances of the sexual assault or ways in which the victim
presented to law enforcement (i.e., based on what rape is supposed to look like). Of the 248 cases
included in the study, 63 (25.4%) had at least one circumstantial legitimizing myth endorsed. Of
the 63 cases with at least one circumstantial legitimizing myth endorsed, 44 (69.8%) cases had
only one circumstantial legitimizing myth endorsed, fourteen (22.2%) cases had two
circumstantial legitimizing myths endorsed, three cases (4.8%) had three circumstantial
64

legitimizing myths endorsed, and two cases (3.2%) had four circumstantial legitimizing myths
endorsed (see Table 3 on page 59 for count data). There were five different sub-categories of
circumstantial legitimizing myths. These categories are listed in Table 4 along with how they
were coded.

Table 4. Circumstantial legitimizing myths and coding scheme
Circumstantial Legitimizing Myth
Victim is lying

Victim is not injured

Victim consented

Victim is not upset

Victim didn’t act like a victim
afterwards

Coding Scheme
0 = Records did not note that the victim is exaggerating,
lying, and does not call into question the victim’s story
(e.g., if it “lines up” or seems plausible)
1 = Records noted that the victim is exaggerating, lying,
or calls into question the victim’s story (e.g., if it “lines
up” or seems plausible)
0 = Records did not note that the victim did not have
bruises, marks, injuries, or appeared disheveled
1 = Records noted that the victim did not have bruises,
marks, injuries, or appeared disheveled
0 = Records did not make any mention of consent or
noted that the victim did not consent to the sexual activity
1 = Records noted that the victim consented to part or all
of the sexual activity with the perpetrator on this
occasion, or on previous occasions
0 = Records did not make any mention of the victim’s
emotional demeanor, or noted that the victim was upset,
distraught, or exhibited emotions that would be expected
given the circumstance
1 = Records noted that the victim did not appear upset or
distraught, seemed distracted, or exhibited emotions that
would be unexpected given the circumstance
0 = Records did not make mention of how the victim’s
actions following the assault were unexpected given the
circumstance
1 = Records noted that the victim’s actions following the
assault were unexpected given the circumstance

65

Victim is lying. The first category of and most frequently endorsed circumstantial
legitimizing myth was “victim is lying” and includes codes in which the records noted that the
victim was exaggerating, lying, or otherwise called into question the victim’s story. Thirty of the
248 cases included in the study (12.1%) had endorsement of this legitimizing myth.
Case 313 provides an example of how this myth was endorsed in police records. Based
on the police files, case 313 involved a 14 year-old Black female who was raped by a 45 year-old
Black male on her way to school. While walking to school, the victim noticed a man walking
towards her on the opposite side of the street. The man tried to hide behind a tree, and then
approached the victim asking her for the time. He then pulled out a gun, grabbed the victim’s
arm, and forced her to a vacant lot across the street. The perpetrator wrapped a scarf around the
victim’s eyes and told her to get on the ground and put her arms over her head. The perpetrator
then raped the victim. After the assault, the perpetrator told the victim that he would watch her
on her way to school. The victim walked to a nearby bus stop, boarded the bus, and told the bus
driver that she had just been raped. The victim asked the driver for a transfer and got off the bus
to wait for the transfer. While at the transfer bus stop, the victim ran into a classmate, told the
classmate that she had just been raped, and started crying. The classmate held the victim and
started crying too as the victim told the classmate that a man with a gun came up to her, told her
to put scarf over her eyes, and raped her. This interaction was relayed to the police by the
classmate.
Police records for case 313 provide evidence that law enforcement believed the victim to
be lying about the sexual assault. Figures 9-11 provide excerpts from the police files. Figure 9 is
an excerpt from the investigator’s scene sheet where the law enforcement officer first indicated
that he/she believed the victim was lying. The law enforcement officer then reiterated this belief

66

on other documents in the police file, including the progress notes and a fax cover sheet (see
Figures 10-11).

Figure 9. Investigator’s scene sheet documenting “victim is lying”

“Case is very weak. Compl’s clothing and shoes are very clean. Her shoes aren’t
dirty and should be. Bus driver that she says she told says no girl came on his bus
and stated she was raped. He further does not believe her. States she wants to go
back to her moms. This case is very weak. None of her account can be verified.”

67

Figure 10. Progress notes documenting “victim is lying”

“PROGRESS NOTES: canvass and tech took photos of area/appears to be
unfounded//compl lied on first told bus driver.”

Figure 11. Fax cover sheet documenting “victim is lying”

“Lie’s have already been uncovered and confirmed…Appears to be a false report
story has not been checking out as complainant states.”

68

Case 313 was classified as unfounded by police and closed; the case file shows no record
of an arrest or that the case was referred to the prosecutor. Upon discovery of the unsubmitted
rape kits in Detroit, the rape kit from case 313 was tested. The DNA results from the rape kit
linked to several other sexual assaults, suggesting a serial rapist.
Victim is not injured. The second most frequently endorsed circumstantial legitimizing
myth was “victim is not injured.” Twenty-two of the 248 case files included in the study (8.9%)
had endorsement of this legitimizing myth. This legitimizing myth referred to the physical
appearance of the victim and included any notes that the victim did not have bruises, marks,
injuries, or appeared disheveled.14
Case 212 provides an example of how this myth was endorsed in police records. Based
on the police files, case 212 involved a 33 year-old Black female raped by two 28 year-old Black
males. After being locked out of her apartment, the victim started to walk to a store. A vehicle
pulled up, asked where she was going, and offered to give her a ride to the store. The perpetrator
then drove to a house and told the victim that he “needed to run in for a minute.” The victim and
perpetrator went into the house. The perpetrator then raped her. A second perpetrator entered the
room and also raped the victim. After the assault, the victim told the second perpetrator that she
was pregnant and asked to use the bathroom. The second perpetrator allowed her to use the
bathroom, but would not allow her to put on her underwear or pants. The victim then was able to
exit the premises (naked from the waist down) and knocked on a neighbor’s door, asking them to
14

It is important to note that the police reports in the current sample included a checkbox option where law
enforcement personnel were to indicate if the victim was injured or not. A check in the ‘no injury’ box was not
counted as endorsement of “victim is not injured” for two reasons: (1) many times, the checkbox marked
contradicted notes in the report, suggesting that the checkbox was not used appropriately (i.e., law enforcement
personnel may have checked boxes out of routine or habit instead of being sure to check the appropriate box); (2) if
a report is filled out completely, there will always be a box checked indicating victim injury or non-injury. A law
enforcement officer may check ‘no injury’ as there are no visible injuries on the victim and go on to conduct a
thorough investigation; the check in the ‘no injury’ box is not because they are using the lack of victim injury to
suggest the rape did not occur, but because they are filling out the report completely, providing answers to all
prompts.

69

call the police. The neighbor called the police, but would not allow the victim to come inside.
The victim then walked to her friend’s house and was subsequently transported to the hospital by
a medic unit. Police records noted that the victim did not appear to be injured or disheveled from
the assault. Figure 12 provides an excerpt from the initial police report.

Figure 12. Initial police report documenting “victim is not injured”

“O [observation]: Victim crying. Victim had no signs of visible injuries. Victim’s
leopard shirt did not appear to be torn.”

Case 212 was classified as unfounded by police and closed; the case file shows no record
of an arrest or a referral to the prosecutor. The SAK associated with case 212 was tested upon
discovery of the unsubmitted rape kits in 2009. Male DNA was isolated in the sample, though it
has not been linked to an offender.
Victim consented. The third most frequently occurring circumstantial legitimizing myth
was “victim consented” and referred to statements in police records that the victim consented to
part or all of the sexual activity with the perpetrator at the time of the assault, or on previous
occasions. Twenty of the 248 case files in the sample (8.1%) had endorsement of this
circumstantial legitimizing myth.
Case 111 provides an example of how this myth was endorsed in police records. Based
on the police files, case 111 involved a 55 year-old White female who was raped by a 32 year70

old Black male. The victim met the perpetrator outside of a supermarket as the perpetrator sat
drinking beer. The victim asked the perpetrator for $1.00 for cigarettes. The perpetrator shared
his beer with the victim and walked with her to a nearby convenience store where he purchased
her cigarettes. The victim and perpetrator continued walking and passed a vacant building; the
perpetrator forced the victim inside. Once inside, the perpetrator demanded sex from the victim
and punched the victim in the face. The victim told the perpetrator that she did not want to be
penetrated and asked if she could perform oral sex instead. The perpetrator agreed and the victim
performed oral sex on the perpetrator. The perpetrator then forced anal and vaginal sex on the
victim. After the assault, the perpetrator fled the scene and the victim went to a nearby restaurant
where employees called 911. Law enforcement personnel arrived on the scene where they
observed the victim’s face to be swollen and bloody. Law enforcement also noted that the
victim’s “R shoulder and compls crotch area were covered in blood.” The victim was conveyed
to a hospital by a medic unit.
Law enforcement personnel talked to witnesses at the convenience store where the
perpetrator and victim purchased cigarettes. Based on a description of the perpetrator from the
complainant and witnesses, police identified a man on the street matching the description and
arrested him. Upon entering the patrol vehicle, the arrestee stated that he had just gotten out of
prison for rape. The progress notes in the case file confirmed this, noting that the arrestee had
previously served 14 years for criminal sexual conduct. The arrestee was held in custody for 48
hours, during which time the victim was still in the hospital and unable to identify the suspect.
After 48 hours, the arrestee was released as no charges had yet been filed. Several days later, the
victim was not able to identify the suspect given a photo lineup. The police officers noted in the
progress notes that the “compl. is mental she will not be able to id she would notr be able to

71

testify clearly.” Law enforcement personnel then cleared the case based on information and
belief that the person they had arrested was the perpetrator. Figures 13-14 provide excerpts from
the initial police report and the progress notes in which law enforcement recorded that the victim
consented to performing oral sex on the perpetrator.

Figure 13. Initial police report documenting “victim consented”

“Compl accepted $1.00 and cigarettes from perp, went into vacant dwelling,
performed fellatio on perp (consentual). Perp then beat compl & forced sodomy
and sexual intercourse.”

Figure 14. Progress notes documenting “victim consented”

“Circumstance: Compl. met perp on [location redacted] accepted 1.00 and 2
cigarettes from the perp. went to service station [location redacted] then both
walked to vacant house at unkn location where compl. willingly performed oral
sex on the perp. that the perp. then forced intercourse and sodomy. also struck the
compl. in the face with his fists. Compl. is mentally deficient.”

72

The case file shows no record of a referral to the prosecutor or charges being filed against
the perpetrator while he was being detained by law enforcement personnel, or at any time after
his release. Upon testing of the SAK many years later, DNA contained in the kit matched to the
arrestee.
Victim is not upset. The fourth most frequently occurring circumstantial legitimizing
myth was “victim is not upset” and referred to statements in the police records that the victim did
not appear upset or distraught, seemed distracted, or exhibited emotions that would be
unexpected given the circumstance. Thirteen of the 248 cases in the study (5.2%) had an
endorsement of this legitimizing myth.
Case 251 provides an example of how this myth was endorsed in police records. Based
on the police files, case 251 involved a 23 year-old Black female who was raped by two Black
males, ages 23 and 21 years old. After leaving the food stamp office with her baby, the victim
was approached by 2 black males in a vehicle. The first perpetrator jumped out of the van and
pushed the victim inside. The perpetrators drove to a vacant building where they held the victim
there and raped her over three consecutive days. After three days, the perpetrators released the
victim. The father of the victim’s baby reported to police that the victim had a “crack” problem
and was known to leave for 3-5 days at a time. The father of the victim’s baby also stated that he
did not believe the victim. It was at this point in the police report that law enforcement personnel
noted the victim’s emotional affect. In the progress notes, law enforcement personnel described
the victim’s story as “very unusual” and stated that the victim seemed “more concerned about
food stamps that were taken than about [the] assault.” In the initial police report, law
enforcement personnel noted that the victim is not as upset as she ought to be, given the assault.
Figure 15 provides an excerpt from police records.

73

Figure 15. Initial report documenting “victim is not upset”

“Compl did not appear scared or distraught & was not crying considering the
circumstances”

There was no record of an arrest or a referral to the prosecutor in the case file. The SAK
corresponding to Case 251 was analyzed upon discovery of the kits in 2009; there was not
enough male DNA in the SAK to produce a DNA profile.
Victim didn’t act like a victim afterwards. The least frequently endorsed circumstantial
legitimizing myth was “victim didn’t act like a victim afterwards” and referred to notes in the
police records that indicate that the victim’s actions following the assault were unexpected given
the circumstance. Four of the 248 case files in the study (1.6%) had an endorsement of “victim
didn’t act like a victim afterwards.”
Case 157 provides an example of how this myth was endorsed in police records. Based
on the police files case 157 involved a 15 year-old Black female who was raped by a 25-30 yearold Black male. The victim was talking on the phone when the perpetrator pulled up in a vehicle
next to the victim. The perpetrator exited the vehicle and stuck what appeared to be a gun into
the victim, instructing her to hang up the phone. The perpetrator ordered the victim into the
vehicle, told the victim to look at the ground, and drove into an alley. The perpetrator got out of
the vehicle and pulled the victim out by her hair, threw her to the ground and raped her.
74

Following the assault, the victim and perpetrator heard voices. The perpetrator got up and the
victim fled the scene. The victim then ran to a phone and called for a cab. The cab drove the
victim home and the victim’s mother took her to the hospital. Police records document that law
enforcement personnel did not believe that the victim acted as would be expected following an
assault. Figure 16 provides an excerpt from an inter-office memorandum sent from one law
enforcement officer to another.
Figure 16. Inter-office memorandum documenting “victim didn’t act like a victim afterwards”

“The compl. took a cab home after the alleged sex assault! She called a cab; and
waited for it! She did not call home; did not ask for help while waiting for a cab,
did not dial 911. The compl has not told her mother the story as of my interview.
When I began to question why a cab and not home? Why wait for cab and not ask
for help at the store while waiting the tears began to flow; and the attitude set in.

”
75

In addition to the notes above, the inter-office memorandum noted disbelief in the
victim’s story, that the victim was not upset, and that there was no visible injury. As showcased
in Figure 17, this portion of the inter-office memorandum also wished the officer-in-charge of
the case, “good luck.”

Figure 17. Additional inter-office memorandum notes for case 157

“This complt is deep! She tells this story. No-tears none!!! The times are off. I
talked with the dr all he found was a little white discharge no trauma!!! Who can
figure it!  Good luck.”

76

The police file for Case 157 had no record of an arrest or referral to the prosecutor. The
SAK associated with Case 157 was tested many years after the assault; no male DNA was
identified in the SAK samples.
Characterological Legitimizing Myths
Characterological legitimizing myths referred to statements that suggested the sexual
assault did not occur based on specific characteristics of the victim (i.e., based on who can be
raped). Characterological legitimizing myths did not pertain to specific factors of the assault, but
instead to static characteristics of the victim that preceded the assault. Of the 248 cases included
in the study, 42 (16.9%) had at least one characterological legitimizing myth endorsed. Of the 42
cases with at least one characterological legitimizing myth endorsed, 31 (73.8%) cases has only
one characterological legitimizing myth endorsed, eight (19.0%) cases had two characterological
legitimizing myths endorsed, and three cases (7.1%) had three characterological legitimizing
myths endorsed (see Table 3 on page 59 for count data). There were six different sub-categories
of characterological legitimizing myths. These categories are listed in Table 5 along with how
they were coded.

77

Table 5. Characterological legitimizing myths and coding scheme
Characterological Legitimizing Myth
Victim is a regular drug user

Victim is a sex worker

Victim has “done this before”

Victim is “mental”

Victim is promiscuous

Victim is not credible

Coding Scheme
0 = Records did not note the victim is drunk/high
when interacting with law enforcement personnel or
is a regular drug user
1= Records noted that the victim is drunk/high
when interacting with law enforcement personnel or
is a regular drug user
0 = Records did not note that the victim is a sex
worker (e.g., a prostitute, a “deal gone bad,” “on the
street”, etc.)
1= Records noted that the victim is a sex worker
(e.g., a prostitute, a “deal gone bad,” “on the street”,
etc.)
0 = Records did not note that the victim has
previously “done this before.” “This” refers to
reporting a rape (and in some cases not participating
in the ensuing investigation), being raped, and/or
having a rape kit done
1= Records noted that the victim has “done this
before.” “This” refers to reporting a rape (and in
some cases not participating in the ensuing
investigation), being raped, and/or having a rape kit
done
0 = Records did not note that the victim is “mental”
or has a mental illness
1 = Records noted that the victim is “mental” or has
a mental illness
0 = Records did not note that victim is promiscuous
1 = Records noted that the victim is promiscuous
0 = Records did not note that victim is not credible
or has a history of lying (separate from lying about
the specific assault reported)
1 = Records noted that the victim is not credible or
has a history of lying (separate from lying about the
specific assault reported)

78

Victim is a regular drug user. The most frequently occurring characterological
legitimizing myth was “victim is a regular drug user” and referred to notes in the police files that
the victim was a regular drug user, or was high/drunk at the time he/she talked to police. It is
important to note that a victim being drunk or under the influence of drugs at the time of the
assault did not qualify for this code. The notes had to specify that the victim’s alcohol/drug use
extended beyond the timing of the assault, so as to indicate regular alcohol/drug use. Fourteen of
the 248 cases included in the study (5.7%) had this legitimizing myth endorsed in the police file.
Case 167 provides an example of how this myth was endorsed in police records. Based
on the police files, case 167 involved a 33 year-old Black female who was raped by a 33 year-old
Black male. The victim met the perpetrator, her friend, at a liquor store. He then attacked her and
pulled her into an abandoned building. The perpetrator strangled the victim, threatened to kill her
with a gun, and then raped her. After the assault, the victim fled to a fire department and reported
the assault. The victim was then transported to a hospital by a medic unit. The victim was 4
months pregnant at the time of the assault and was experiencing vaginal pain and bleeding while
being questioned by the police. As showcased in Figure 18, law enforcement personnel noted
that the room “reeked of alcohol.”

79

Figure 18. Case file notes documenting “victim is a regular drug user”

“Writer made [name redacted] hospital to take statement from compl. Compl
stated that she was in pain. The doctor came into the room, stated that the compl
may be having complications from the pregnancy…the compl is 4mths pregnant,
but admitted to being intoxicated. The room reeked of alcohol…”

Additionally, law enforcement noted that they were concerned as to if the victim was
telling the truth and instructed her that “she should be worried about telling the police the truth
or she would be charged with making a false felony report” (See Figure 19).

80

Figure 19. Additional case file notes for case 167

“The compl was advised that she must be truthful regarding the events that led up
to the rape…she stated that she didn’t feel comfortable talking abouts the events
due to the fact that her boyfriend was at the hospital, and he didn’t believe her
story. I explained to her that she shouldn’t be worried about her boyfriend, she
should be worried about telling the police the truth or she would be charged
w/making a false felony report.”

Case 167 was classified as unfounded by law enforcement personnel; the police file did
not have any record of an arrest or a referral to the prosecutor. The SAK associated with case
167 was tested upon discovery of the unsubmitted rape kits in 2009. Male DNA was isolated in
the sample and a DNA profile was created, though it has not been linked to an offender.
Victim is a sex worker. The second most frequently occurring characterological
legitimizing myth was “victim is a sex worker” and referred to statements that the victim was a
sex worker (e.g., a prostitute, a “deal gone bad,” “on the street,” etc.). Thirteen of the 248 cases
included in the study (5.2%) endorsed this legitimizing myth.
81

Case 139 provides an example of how this myth was endorsed in police records. Based
on the police files, case 139 involved a 41 year-old Black female who was raped by a 45-50
year-old Black male. The victim was walking down the street when the perpetrator, someone she
knew from the neighborhood, approached her, produced a knife, and forced her to a vacant
dwelling. The perpetrator raped the victim, beat her with a brick, and stabbed her in the shoulder.
After the assault, the perpetrator urinated on the victim and then told the victim to wait before
leaving the location or he would beat her again. After the perpetrator left, the victim fled the
location carrying her pants and shoes, and ran into a police station screaming for help. A medic
unit was called and the victim was conveyed to the hospital. Law enforcement personnel noted in
the report that upon entering the police station, the victim was only wearing a shirt covered in
blood, had blood on her head, face, and arms, and was bleeding from her head and shoulder. Law
enforcement personnel noted that they believed the victim to be a sex worker in their
investigator’s scene sheet as showcased in Figure 20 below.

Figure 20. Investigator’s scene sheet documenting “victim is a sex worker”

“COMMENTS/REMARKS: I took this complainant’s statement. She is familiar
with perp. He has asked to use her crackpipe in the past. She is a prostitute.”
Law enforcement personnel noted that they tried to contact the victim by phone
once and in person once. No further action was taken on the case; there is no record of an
82

arrest or a referral to the prosecutor in the case files. The SAK corresponding to Case 251
was analyzed upon discovery of the kits in 2009; there was no male DNA in the SAK to
produce a DNA profile.
Victim has “done this before.” The third most frequently occurring
characterological legitimizing myth was “victim has “done this before.”” “This” referred
to reporting a rape (and in some cases not participating in the ensuing investigation),
being raped, or having a rape kit previously done. Nine of the 248 cases included in the
study (3.6%) included an endorsement of this characterological legitimizing myth.
Case 149 provides an example of how this myth was endorsed in police records.
Based on the police files, case 149 involved a 25 year-old Black female who was raped
by a 36 year-old Black male. The victim was standing on the street when the perpetrator
pulled up in a vehicle, asked if she had a crack stem, and if she wanted to get high. The
victim got into the vehicle with the perpetrator. The perpetrator pulled behind a restaurant
building and produced a handgun. He threatened to kill the victim if she did not have sex
with him. After the assault, the perpetrator then forced the complainant out of the car and
drove away. The victim went to the front of the restaurant building and saw police, who
were onsite for an unrelated call. Police noted that the victim was seen in front of the
restaurant, crying. She was wearing a torn shirt, a jacket tied around her waist, and no
pants. The police also note that the victim “appeared intox/high and stated she had been
smoking crack earlier.” A medic unit arrived on the scene and conveyed the victim to the
hospital. After attempting to contact the victim, as noted in Figure 21, the police officers
noted that they were unable to establish the elements of a crime and that the victim had
“made other CSC reports and had not followed through with the cases.”

83

Figure 21. Progress notes documenting “victim has “done this before””

“Rec case, compl was not at [name of hospital] when sex crimes went to talk to
her. She is homeless. Called [name of shelter], compl not at that shelter. Also,
compl has made other CSC reports and has not followed through with the cases.
Compl is a crack addick and was high when she talked to the officers that went to
the scene.”

There are no records of an arrest or a referral to the prosecutor. When the SAK was tested
upon the discovery of the unsubmitted kits in Detroit, a DNA profile resulted and provided an
identity for the offender.
Victim is “mental.” The fourth most frequently occuring characterological legitimizing
myth was “victim is “mental.”” This legitmiizing myth referes to statements in which the records
noted that the victim was “mental” (i.e., this was how law enforcement personnel described the
victim) or had a mental illness. Eight of the 248 cases included in the study (3.2%) included an
endorsement of “victim is “mental.””
Case 120 provides an example of how this myth was endorsed in police records. Based
on the police files, case 120 involved a 36 year-old White female who was raped by a 42 yearold White male. The victim was in her room at the hospital when the perpetrator entered the
84

room, grabbed the victim by the shoulders, and pushed her down onto the bed. After the assault,
the perpetrator went into the bathroom and masturbated, then left the room. The victim went to
the cafeteria and told an employee, then returned to the floor of the hospital where the incident
happened and told another employee. Law enforcement personnel noted that the victim was
“mental,” had “done this before,” and was not injured, as shown in Figure 22.

Figure 22. Initial police report documenting “victim is “mental””

“C [circumstance]: Wtr’s made loc, compl abv (mental patient) stated that abv
perp who works at the hospital as a matenence worker came into her room
pretending to be cleaening, walked over to her bed...
“Compl is a mental patient and that according to her family, she’s done this
before. Compl was seen by Dr. [name redacted] who ran test on her and found no
sign of rape.”
In addition to noting that the victim was “mental,” had “done this before,” and was not
injured (see Figure 85), law enforcement personnel also noted that the victim’s “story changed
several times” (i.e., “victim is lying” circumstantial legitimizing myth), that the victim was “in a

85

calm state” (i.e., “victim is not upset” circumstantial legitimizing myth), and that the victim may
have consented to sex (i.e., “victim consented” circumstantial legitimizing myth). There was no
record of an arrest or a referral to the prosecutor in the case files. The SAK associated with Case
120 was tested after discovery of the unsubmitted SAKs in 2009. Male DNA was isolated from
the SAK and a DNA profile was created, though there was no information on the identity of the
offender.
Victim is promiscuous. The fifth most frequently occurring characterological
legitimizing myth was “victim is promiscuous” and referred to statements regarding the
victim’s sexual history that described her as promiscuous. Six of the 248 cases included
in the study (2.4%) had this legitimizing myth endorsed.
Case 198 provides an example of how this myth was endorsed in police records.
Based on the police files, case 198 involved a 14 year-old Black female who was raped
by a 40 year-old Black male. The victim was at a party when the perpetrator, who the
victim knew but who she would not describe as a friend, grabbed her by the neck and
forced her into the basement. The perpetrator pushed the victim down onto the bed. The
victim screamed for help, but could not be heard over the loud music from the party
upstairs. After the assault, the perpetrator fled and the victim’s aunt found the victim in
the basement crying. The victim was then transported to the hospital. Law enforcement
personnel suggested that the victim was promiscuous by noting her prior sexual acts in
the initial police report (see Figure 23 below).

86

Figure 23. Initial police report documenting “victim is promiscuous”

“Victim stated that she was not a virgin before this incident…writers then spoke
to victim’s mother who stated that the victim has had sex multiple times and has
had three abortions.”

Law enforcement personnel also noted in the police report that the “victim was calm”
(i.e., “victim is not upset” circumstantial legitimizing myth), that parts of the victim’s story
changed as she told it (i.e., “victim is lying” circumstantial legitimizing myth), and that the
victim “has had a rape kit done on her more than once” (i.e., “victim has “done this before””
characterological legitimizing myth). The police files documented that this case was referred to
the prosecutor and a warrant was issued. The perpetrator was subsequently arrested and pled
guilty to criminal sexual conduct.
Victim is not credible. The last characterological legitimizing myth was “victim is not
credible.” While all of the characterological legitimizing myths suggest that the victim is not
credible, this legitimizing myth only referred to statements in the police file that explicitly stated
that the victim was not credible or had a history of lying (i.e., separate from lying about the
specific assault reported). This myth was just as common as stating the victim was promiscuous
with six out of the 248 cases included in the study (2.4%) having an endorsement of this
legitimizing myth.

87

Case 202 provides an example of how this myth was endorsed in police records. Based
on the police files, Case 202 involved a 13 year-old Black female who was raped by a 25 yearold Black male. The victim ran away from home and was staying at a friend’s house. She started
to walk back home after dark and the perpetrator, someone familiar to the victim, approached
her. The victim kept walking and the perpetrator came up behind and forced her behind a vacant
house. The perpetrator forced her to undress and showed her a kitchen knife, threatening to stab
her if she said anything. The perpetrator then raped her. After the assault, the victim went home,
told her mother, and was transported to the hospital. In Figure 24, the law enforcement officer
made notes about the victim that presented her as unreliable and a liar. At the end of the excerpt,
the law enforcement officer explicitly stated that the victim “cannot be deemed credible.”

88

Figure 24. Progress notes documenting “victim is not credible”

“[date redacted] rec case, compls mother called and stated the compl she took
her vehicle and almost hit her. Compl is 13 years old, advised motehr to call 911
and she did. Mother also said the compl is a run away and has lied before. She is
hanging around with the wrong crowd and she can’t control her. The compl told
her mother that the house would be broken into and it was, twice, compls things
were not touched, only the mothers things were missing…
“[name redacted] will be contacted [date redacted]. Compl gave a totaly
different description and age of the perp and different account of what had
happen. This compl cannot be deemed credible and this case is closed
unfounded.”

In addition, the law enforcement officer noted that the victim was lying about this
specific assault; accordingly, the case was also coded for “victim is lying” circumstantial
legitimizing myth. The case was deemed unfounded; there was no record of an arrest or a referral
to the prosecutor in the case files. The SAK associated with Case 202 was tested following the
discovery of the unsubmitted SAKs in 2009; no male DNA samples were identified in the SAK.
89

Investigatory Blame Legitimizing Myths
Investigatory blame legitimizing myths blamed the victim for the fact that the police
conducted a less-than-thorough investigation. To be clear, these legitimizing myths did not
blame the victim for the rape, or suggest the rape did not happen; they blamed the victim for the
investigation not advancing as far as it might have been able to otherwise because the victim was
considered not willing or not able to participate in the process. These legitimizing myths
suggested that the case proceeded as it did because the victim was not cooperating, did not
provide enough information, was not able to be contacted, or was a “weak” victim that would not
hold up well during trial.
Investigatory blame legitimizing myths were far more common than circumstantial and
characterological legitimizing myths. Of the 248 cases included in the study, 102 (41.1%) had at
least one investigatory blame legitimizing myth endorsed. Of the 102 cases with at least one
investigatory blame legitimizing myth endorsed, 88 (35.5%) cases has only one investigatory
blame legitimizing myth endorsed, thirteen (5.2%) cases had two investigatory blame
legitimizing myths endorsed, and one case (0.4%) had three investigatory blame legitimizing
myths endorsed (see Table 3 on page 64 for count data). There were four different sub-categories
of investigatory blame legitimizing myths. These categories are listed in Table 6 along with how
they were coded.

90

Table 6. Investigatory blame legitimizing myths and coding scheme
Investigatory Blame Legitimizing Myth
Victim is uncooperative

Victim doesn’t have enough information

Victim has no phone/address for contact

Victim or case is weak

Coding Scheme
0 = Records did not note that the victim was
uncooperative, hostile, or intentionally withholding
information
1= Recorded noted that the victim was
uncooperative, hostile, or intentionally withholding
information
0 = Records did not note that victim did not have or
could not remember enough information
1 = Records noted that the victim did not have or
could not remember enough information (e.g., did
not know the name of her rapist)
0 = Records did not note a problem in contacting
the victim
1 = Records noted that law enforcement personnel
did not have a working phone number or address
for the victim, or were otherwise unable to contact
the victim
0 = Records did not record that the victim or case
was weak or incompetent
1 = Recorded noted that the victim or case was
weak or incompetent

Victim is uncooperative. The most frequently occurring investigatory blame
legitimizing myth, and most frequently occurring legitimizing myth overall, was “victim is
uncooperative.” This legitimizing myth referred to statements that the victim was uncooperative
or hostile during the investigation, didn’t care about the investigation, or was intentionally
withholding information. A total of 72 cases out of the 248 cases included in the study (29%) had
endorsement of this legitimizing myth. It was many times difficult to tell why or how the victim
was uncooperative; police records would only note that the victim was uncooperative.

91

Case 179 provides an example of how this myth was endorsed in police records, and even
provides additional detail as to what led to the victim being deemed uncooperative. Based on the
police files, case 179 involved a 16 year-old female who was raped by 5 Black males, ranging in
age from 14-20 years old. The victim went to visit her 15 year-old male friend at home. The
victim and her friend then walked over to a second house together. The victim and the 15 yearold male friend went into a room in the basement to have sex. After they finished having sex, the
victim heard other voices in the room arguing over whose turn was next; the victim was then
raped vaginally, orally, and anally by at least three different perpetrators, though more
perpetrators were believed to be in the room. The 15 year-old male friend of the victim who
initially had consensual sex with the victim identified the other males in the room to be four
males of various ages: one 14 year-old Black male, two seventeen year-old Black males, and a
20 year-old Black male. Near the time of the assault, a witness saw ten Black males running out
of the house. Additionally, police officers at the scene of the crime found freshly opened
condoms lying beside the bed, along with a folded, freshly-soiled sheet and a freshly-soiled
mattress pad. Figure 25 provides an excerpt from the case progress notes and provides detail on
what led up to the victim being deemed uncooperative by law enforcement personnel.

92

Figure 25. Progress notes documenting “victim is uncooperative”

“P/C [phone call] to complt’s home to get clarification of additional info. Complt
states she does not know who was in room but could hear voices not faces. Writer
advised complt that she mentioned in her statement that she went over to [name of
15 year-old male redacted] house then to house on [street name redacted] with
intent to have sex with him (consentual) and that [name of 15 year-old male
redacted]’s only fifteen under the age of consent and that she could be charged as
well because she is 16 (age of consent). Writer then t/t [talked to] complt’s
mother [name of mother redacted] advised of above, who then asked complt what
did she want to do. Complt then replied to mother that she just wanted to drop
case, “to just forget it.” Case CRTP [complainant refused to prosecute], advised
[name of supervising officer redacted] on same.”
93

The excerpt from Case 179, showcased in Exhibit 17, documented that the victim
elected not to participate in the investigation (i.e., “victim is uncooperative”) after being
threatened by law enforcement that charges may be brought against her for having sex
with a minor. At the time the victim decided “to just forget it,” the case had already
been referred to the prosecutor. Several of the perpetrators had also been arrested. The
police officer informed the prosecutor that the victim no longer wanted to proceed with
the case and the prosecutor explained that it was best to “wait to see if the compl would
change her mind again and go through with prosecuting the case.” The prosecutor went
on to say that “the decision to drop the charges was now up to the courts.” The police
files do not indicate what happened next in the case. The SAK associated with Case 179
was tested after the unsubmitted kits were discovered in Detroit in 2009. One of the DNA
profiles from the SAK testing was identified as one of the 17 year-old perpetrators
arrested for the assault.
Victim doesn’t have enough information. The second most frequently occurring
investigatory blame legitimizing myth was “victim doesn’t have enough information.”
The previous investigatory blame legitimizing myth, “victim is uncooperative” includesd
cases in which the victim was intentionally withholding information; this legitimizing
myth referred to cases in which the victim did not remember or did not have enough
information on the perpetrator or assault, but was not considered to be withholding
information intentionally. Twenty-four of the 248 cases included in the study (9.7%) had
endorsement of this legitimizing myth.
Case 235 provides an example of how this myth was endorsed in police records.
Based on the police files, Case 235 involved a 22 year-old Black female who was raped

94

by a Black male in his twenties. The perpetrator was giving the victim a ride home from a
night club when he choked her and forced her to perform oral sex on him. After the
assault, the victim got out of the vehicle and the perpetrator kicked her in the chest before
driving off. Law enforcement personnel noted that they saw bruising on the victim’s
chest at the time the report was taken. Progress notes in the file noted that the victim was
interested in pursuing the case, but only knew the perpetrator by a nickname and the
name of the production company where he worked. Law enforcement personnel were not
able to find the production company listed in the yellow pages and asked the victim to go
to the location of the production company to get the address. The victim provided an
address to law enforcement personnel; law enforcement personnel noted that they were
still unable to find the location. Figure 26 below is an excerpt from the case progress
notes in which law enforcement personnel noted that they were waiting on the victim to
contact them with additional information on the perpetrator.

95

Figure 26. Progress notes documenting “victim doesn’t have enough information”

“Writer advised compl that I couldn’t find the comp [company] she said that they
probably moved. Advised her of the case. Until I find the production company I
can’t do anything. She said ok, she will try to find out his name and call writer
back.”
This was the last entry in the case file; there was no record of an arrest or a referral.
Discovery of the unsubmitted rape kits in 2009 caused the SAK associated with Case 235 to be
tested; no male DNA was isolated from the SAK.
Victim has no phone/address for contact. The third most frequently occurring
investigatory blame legitimizing myth was “victim has no phone/address for contact” and
referred to statements in the police files that law enforcement personnel did not have a working
phone number or address for the victim, or were otherwise unable to contact the victim. Sixteen
of the 248 cases included in the study (6.5%) had this legitimizing myth endorsed.
Case 335 provides an example of how this myth was endorsed in police records. Based
on the police files, case 335 involved a 32 year-old Black female who was raped by a 25 year-old
Black male. The victim had just exited a store when the perpetrator approached her from behind
and put an unknown object, believed to be a gun, behind her head. The perpetrator threatened to
96

kill the victim if she turned around. The perpetrator forced the victim into his vehicle and then
raped her. After the assault, the victim got out of the vehicle and ran to a nearby address where
the police arrived to take the report. Progress notes in the case file noted that law enforcement
personnel attempted to phone the complainant three times and were unsuccessful in their
attempts. They then sent a letter to the victim regarding the case, as documented in Figure 27,
and the letter was returned to sender.

Figure 27. Progress notes documenting “victim has no phone/address for contact”

“Received letter sent. Compl cannot be located and has not attempted to call me.
Shows no interest in case. To locate.”

After not being able to reach the victim, law enforcement personnel noted that the
victim must not be interested in the case and took no further action. There is no record of
an arrest or a referral to the prosecutor in the case files. The SAK associated with Case
335 was tested upon discovery of the SAKs in Detroit in 2009; no male DNA was
isolated from the SAK.
Victim or case is weak. The least frequently occurring investigational blame
legitimizing myth was “victim or case is weak” and referred to statements that state
explicitly that the victim or case is weak or incompetent and that the victim would not be
97

able to hold up at trial. Like all other investigatory blame legitimizing myths, this myth
did not necessarily minimize or discount the rape (in contrast to circumstantial and
characterological legitimizing myths). Instead, this myth was used to justify law
enforcement personnel’s inaction on the case. This legitimizing myth was endorsed in
five of the 248 cases included in the sample (2.0%).
Case 106 provides an example of how this myth was endorsed in police records.
Based on the police files, case 106 involved a 4 year White female who was raped by a
22 year-old White male; the perpetrator was the victim’s half-brother. The victim
reported to her mother that it burned to go to the bathroom. The mother inquired as to if
anyone had touched her; the victim said yes and described the assault. The mother kicked
the perpetrator out of the house; he was currently residing with the victim’s family. The
mother then took the victim to the hospital. Law enforcement personnel noted that while
at the hospital, the mother “was advised that the victims private area was red unknown
from what but it appeared irregular.” As displayed in Figure 28, an interview was
conducted with the victim; the interview tape was then sent to prosecutor’s office and law
enforcement personnel typed up a warrant for the prosecutor to approve (i.e., referred to
the prosecutor). The police records noted that the prosecutor viewed the interview tape
and then informed law enforcement personnel that “we did not have a case.” Figure 28 is
an excerpt from the case progress notes in which law enforcement personnel blamed the
victim for their incomplete investigation (i.e., no arrest on the case).

98

Figure 28. Progress notes documenting “victim or case is weak”

“[date redacted] interview. The compl was very weak and all over the place not
very competent.”

There was no documentation of an arrest in the police file. Upon discovery of the SAKs
in 2009, the kit associated with Case 106 was tested; no male DNA was isolated from the kit.
Summary. Directed and conventional content analysis revealed fifteen different
legitimizing myths that were conceptually grouped into three categories: circumstantial,
characterological, and investigatory blame legitimizing myths. Sixty percent of the cases
included in the study had at least one legitimizing myth endorsed, with as many as six
legitimizing myths on a single case. Investigatory blame legitimizing myths were the most
common (40.1% of all cases), with nearly 30% of victims being called “uncooperative” and 10%
being blamed for not being able to provide enough information to law enforcement personnel,
such as the full name of the perpetrator, where the perpetrator lived, or where the perpetrator
worked. While prior literature has found that victims tend to be blamed for the assault (e.g., see
Kettrey, 2013), blaming the victim for a less-than-thorough investigation has not been previously
documented in the rape myth literature (i.e., all four codes in the investigatory blame
legitimizing myth category were identified during conventional content analysis coding).
Circumstantial and characterological legitimizing myths more closely aligned with
traditional rape myths, as they pertained to what qualifies as ‘real’ rape and who can be raped,
99

respectively. Four-of-the-five circumstantial legitimizing myths and two-of-the-six
characterological legitimizing myths documented in Study 1 were included in the a priori coding
scheme used during directed content analysis as they had been operationalized in prior literature
(Kettrey, 2013). The only new circumstantial legitimizing myth to emerge in Study 1 during
conventional content analysis was “victim didn’t act like a victim afterwards.” The four new
characterological legitimizing myths to emerge in Study 1 during conventional content analysis
were “victim has “done this before,”” “victim is “mental,”” victim is promiscuous, and victim is
not credible.

100

STUDY 2: EXAMINING RELATIONSHIPS BETWEEN LEGITIMIZING MYTHS AND
OUTCOMES
Method
Sample
Study 2 used the same dataset as Study 1 to examine how the different types of
legitimizing myths identified in Study 1 related to the investigatory effort, case outcomes,
specific social identity factors of the victim and perpetrator and the number of perpetrators across
sexual assault cases (see Sample on page 47 for additional detail). The total sample size for
Study 2 matched that of Study 1 with N=248 case files.
Preparing Case Files for Coding
Consistent with Study 1, this project utilized archival data and required no direct contact
with participants. However, the files did contain identifying and sensitive participant
information. As described in greater detail in Study 1, all data were maintained electronically on
password protected computers and only the investigators had access to the data. Identifying
information was redacted as soon as possible after cases had been coded in accordance with
procedures described in Study 1 and consistent with approval from the Michigan State
University IRB.
Coder Training
Case files included in Study 2 were coded by four coders (i.e., the project director and
three undergraduate research assistants; the same set of coders used in Study 1) for the
investigative steps taken by law enforcement personnel, case outcomes, and the age, race, and
sex of the victim and perpetrator(s) (see Measures below). The two-stage approach used to train

101

coders as described in the Coder Training section of Study 1 was also used for Study 2, and will
not be presented in its entirety here.
Coding Procedures
Per Study 1, coders had already completed IRB training and signed a confidentiality
agreement prior to viewing (e.g., for training purposes) or coding any data. The project director
provided coders with coding instructions and the codebook defining all codes for Study 2 (see
Appendix A for the coding instructions; Appendix C for the codebook defining investigative
steps, case outcomes, and specific social identity factors of the victim and perpetrator[s] and
Appendix D for the complete list of codes used in Study 2). As in Study 1, the codebook was
reviewed with all coders to ensure they understood each operationalization and how to code the
data accurately and consistently. The coders used the codebook to code N = 6 randomly selected
cases as a group. The same set of twelve cases double coded during coder training in Study 1
were double coded during coder training in Study 2. Kappa was calculated to examine intercoder reliability in coding investigative steps and case outcomes. The overall training kappa
score across the twelve cases was 0.78. As in Study 1, thirty percent of the remaining uncoded
cases were selected to be double coded (N = 69). A new kappa score (i.e., did not include the
kappa calculated during training) was calculated after ten percent of the double coded files had
been coded (N = 7). This process continued throughout the coding process with a final total
kappa score for investigative steps and case outcomes of 0.95.
Measures
Number of Circumstantial Legitimizing Myths (endogenous variable). Each case file
was read for evidence of each of the circumstantial legitimizing myths listed in Table 4 (page 65)
during Study 1 and coded as present (1) or not (0). The legitimizing myths were then summed to

102

produce the total number of circumstantial legitimizing myths endorsed in each case file (see
Table 3 on page 64 for count distributions). A case could have up to five circumstantial
legitimizing myths endorsed, though the number of circumstantial legitimizing myths endorsed
on a single case in this sample ranged from 0-4 with 63 cases (25.4%) having at least one
circumstantial legitimizing myth. On average, a case had 0.36 circumstantial legitimizing myths
endorsed (SD = 0.717). Number of circumstantial legitimizing myths was a count variable. There
were no missing data on this variable.
Number of Characterological Legitimizing Myths (endogenous variable). Each case
file was read for evidence of each of the characterological legitimizing myths listed in Table 5
(page 78) during Study 1 and coded as present (1) or not (0). The legitimizing myths were then
summed to produce the total number of characterological legitimizing myths endorsed in each
case file (see Table 3 on page 64 for count distribution). A case could have up to six
characterological legitimizing myths endorsed, though the number of characterological
legitimizing myths endorsed on a single case in this sample ranged from 0-3 with 42 cases
(16.9%) having at least one characterological legitimizing myth. On average, a case had 0.23
characterological legitimizing myths endorsed (SD = 0.560). Number of characterological
legitimizing myths was a count variable. There were no missing data on this variable.
Number of Investigatory Blame Legitimizing Myths (endogenous variable). Each
case file was read for evidence of each of the investigatory blame legitimizing myths listed in
Table 6 (page 91) during Study 1 and coded as present (1) or not (0). The legitimizing myths
were then summed to produce the total number of investigatory blame legitimizing myths
endorsed in each case file (see Table 3 on page 64 for count distribution). A case could have up
to four investigatory blame legitimizing myths endorsed, though the number of investigatory

103

blame legitimizing myths endorsed on a single case in this sample ranged from 0-3 with 102
cases (41.13%) having at least one investigatory blame legitimizing myth. On average, a case
had 0.47 investigatory blame legitimizing myths endorsed (SD = 0.616). Number of
investigatory blame legitimizing myths was a count variable. There were no missing data on this
variable.
Case Outcome (endogenous variable). Case outcomes were coded into one of four
categories: (1) a suspect was arrested and the case was referred to the prosecutor (i.e., arrest and
referral; N = 68 cases; 27.4%); (2) no arrest was made, but the case was referred to the
prosecutor (i.e., no arrest and referral; N = 19 cases; 7.7%); (3) a suspect was arrested, but the
case was not referred to the prosecutor (i.e., arrest and no referral; N = 12 cases; 4.8%); or (4) no
arrest was made and the case was not referred to the prosecutor (i.e., no arrest and no referral; N
= 149 cases; 60.1%). Case outcome was a nominal variable with the fourth category—no arrest
and no referral—used as the reference category for analysis. There were no missing data on this
variable.
Number of Investigative steps (endogenous variable). Each of the investigative steps
listed in Table 7 was coded as present (1) or not (0). The steps were then summed to produce the
total number of investigative steps taken on each case. A case could have up to ten investigative
steps, though the number of investigative steps taken on a single case in this sample ranged from
0-9 with an average of 3.38 steps per case (SD = 1.930). Number of investigative steps was a
count variable. There were no missing data on this variable.

104

Table 7. Investigative steps summed to produce “Number of Investigative steps” variable
Investigatory Step
Coding Scheme
Evidence technicians 0 = No documentation of evidence technicians arriving at the crime scene
at scene
1 = Documentation of evidence technicians arriving at the crime scene as
indicated by an “evidence tech report,” in “scene investigation” notes, or
as indicated in the initial report
Photographs at scene 0 = No photographs from the scene of the crime

Canvassed

1 = Photographs from the scene of the crime
0 = No completed canvass sheets or other documentation in the initial
report that investigators/police canvassed the area

Progress Notes

1 = Completed canvass sheets or other documentation in the initial report
that investigators/police canvassed the area
0 = No progress notes appear

Victim statement

1 = At least one progress note was recorded in the file as recorded on the
“progress notes” document
0 = No victim statement (in a transcript format)

Witness statement

1 = Statement from the victim (in a transcript format)
0 = No witness statement (other than the victim)

SAK to lab

1 = Statement from the witness (other than the victim)
0 = No lab request form or lab report for processing of the SAK

1 = Lab request form or lab report for processing the SAK
Medical release form 0 = No completed medical release form (i.e., signed by victim or
guardian)

Suspect lineup

Suspect interview

1 = Completed medical release form (i.e., signed by victim or guardian)
0 = No documentation of a suspect lineup
1 = Documentation of a suspect lineup, as indicated by the “showup
and/or lineup” file or written in the case notes (i.e., initial report of
progress notes)
0 = No documentation of a suspect interview (including if the suspect
refused to provide an interview)
1 = Documentation of a suspect interview (or the suspect refusing to
provide an interview), as indicated by the “interrogation record” or in the
case notes (i.e., initial report or progress notes)

105

Victim Sex (exogenous). The sex of the victim was recorded as female (0) or male (1).
Victim sex was a binary variable with female victims as the reference category. The majority of
sexual assault victims associated with the cases in the study was female (95.6% female; 4.4%
male). There were no missing data on this variable.
Perpetrator Sex (exogenous variables). The sex of the perpetrator was recorded as
female (0) or male (1). In the case of multiple perpetrators, the sex of the first perpetrator listed
was recorded. Perpetrator sex was a binary variable with female perpetrators as the reference
category.15 All perpetrators in the study were male (100%). There were no missing data on this
variable.
Victim Race (exogenous variables). The race of the victim was recorded as non-White
(0) or White (1). Victim race was a binary variable with non-White victims as the reference
category. The majority of victims were non-White (N = 215; 86.7% non-White; N = 33; 13.3%
White), and more specifically, Black/African-American (N = 214; 86.3%). There were no
missing data on this variable.
Perpetrator Race (exogenous variable). The race of the perpetrator was recorded as
non-White (0) or White (1). In the case of multiple perpetrators, the race of the first perpetrator
listed was recorded. Perpetrator race was a binary variable with non-White perpetrators as the
reference category. The majority of perpetrators were non-White (N = 229; 92.3% non-White; N
= 14; 5.6% White; N = 5; 2.0% missing), and more specifically, Black/African-American (N =
228; 91.9%). Missing data on this variable were recorded as 999 and set to missing in analysis.
Victim Age (exogenous variable). Victim age ranged from 2-81 years old (M = 23.36;
SD = 11.46) and each case was coded into one of three categories: (1) victim is 15 years old or
15

Female perpetrators were used as the reference category for purposes of coding consistency, not because female
perpetrators were expected to be the norm (i.e., female victims were used as the reference category for victim sex).
As discussed, the vast majority of sexual assault perpetrators are male (Black et al., 2011).

106

younger (N = 66; 26.6%); (2) victim is 16-25 years old (N = 96; 38.7%); or (3) victim is 26 years
old or older (N = 86; 34.7%). These categories were created based on the age of consent in the
state of Michigan and prior literature. The first category (i.e., victims 15 years old or younger)
included cases corresponding to victims that could not legally consent to sex in the state of
Michigan (i.e., the age of consent in the state of Michigan is 16 years old; Section 750.520b-g of
the Michigan Penal code). The second category (i.e., victims 16-25 years old) included cases
corresponding to victims that have been identified in prior literature to be deemed less credible
by law enforcement personnel as compared to younger and older victims (Heenan & Murray,
2006; Kelly et al., 2005; LaFree, 1981; Rose & Randall, 1982; Spohn & Spears, 1996; Triggs et
al., 2009) The third category (i.e., victims 26 years old or older) included all other cases in the
sample. Victim age was a categorical variable and was dummy coded in the analysis with cases
involving victims 15 years old or younger as the reference category as these cases involved
victims that could not legally consent to sex. There were no missing data on this variable.
Victim and Perpetrator Age Difference (exogenous variable). On average,
perpetrators were 5.7 years older than the victim (SD = 11.87), though this ranged from the
perpetrator being 47 years younger than to 53 years older than the victim. The age difference
between the victim and the perpetrator was coded into one of three categories: (1) the
perpetrator’s age is within 5 years of the victim’s age (N = 95; 38.3%) (2) the perpetrator is 5
years or more older than the victim (N = 102; 41.1%); or (3) the perpetrator is 5 years or more
younger than the victim (N = 25; 10.1%). In the case of multiple perpetrators, the age of the first
perpetrator listed was used to calculate the age difference between the victim and the perpetrator.
Victim and perpetrator age difference was a categorical variable and was dummy coded in the

107

analysis with perpetrators within 5 years of the victim’s age as the reference category. Missing
data (N = 26; 10.5%) on this variable were recorded as 999 and set to missing in analysis.
Multiple Perpetrators (exogenous variable). Each case was coded for if it had multiple
perpetrators (1) or not (0). Multiple perpetrators was a binary variable with a single perpetrator
as the reference category. The majority of cases involved a single perpetrator (N = 190; 76.6%).
There were no missing data on this variable.
Data Analysis
Rationale for path analysis and an exploratory analytic approach. Path analysis16
was used to examine the number of characterological, circumstantial, and investigatory blame
legitimizing myths as predictors of the number of investigative steps taken in each case and the
final case outcome; the relationship between the different types of legitimizing myths; the
number of investigative steps and the final case outcome; and the impact of victim sex,
perpetrator sex, victim race, perpetrator race, victim age, victim and perpetrator age difference,
and multiple perpetrators on these variables (see Figure 29). Path analysis is a flexible approach
that allows for the empirical examination of theorized models that explain how variables are
related to one another. Via path analysis, the analyst is able to test relationships between
different variables by constraining and relaxing parameters within the model, assessing for
improved model fit along the way (see Barrett, 2007; Bollen, 1989; Byrne, 2012). Model fit
refers to the extent to which the expected values in the variance covariance matrix generated by
the model estimates is discrepant from the observed variance covariance matrix (Bollen, 1989;

16

Full structural equation models include a structural model that examines the relationships between observed
variables or between latent constructs and a measurement model that examines how observed items load onto the
latent constructs. Accordingly, some would call a model that includes only the structural component a “path
analysis” while others may refer to this more generally as structural equation modeling (SEM). Path analysis will be
used here to refer to a structural equation model that examines the relationships between observed variables. See
Bollen, 1989.

108

Tabachnick & Fidell, 2007). By constraining and/or relaxing parameters within the model (i.e.,
relationships between different variables), the analyst is able to compare different models to one
another, and then select a final model that most adequately represents the observed data. An
additional key advantage of path analysis over multiple regression is its ability to include
multiple endogenous variables (Tabachnick & Fidell, 2007). Accordingly, path analysis was a
natural fit for this analysis as it is flexible, able to handle complex models that test varying
relationships, and can include multiple endogenous variables.

109

Figure 29. Current study’s statistical model
Victim
Sex

Perpetrator
Sex

Victim
Race

Perpetrator
Race

Victim
Age

Characterological
Legitimizing Myths
*

Victim and Perpetrator
Age Difference

Multiple
Perpetrators

Number of
Investigative Steps

Circumstantial
Legitimizing Myths
Case Outcome
Investigatory Blame
Legitimizing myths

*The single-headed arrow indicates that characterological legitimizing myths were regressed onto circumstantial legitimizing myths.
However, neither the research design nor existing literature provides justification for this to be interpreted as a causal relationship.
Therefore this was interpreted as an association with no assumption as to which myth preceded the other. Circumstantial and
characterological legitimizing myths, however, were interpreted to precede investigatory blame legitimizing myths as they frequently
followed this temporal sequence in the police reports. For example, the report would state that the victim was a sex worker (i.e.,
characterological legitimizing myth) before stating that the victim was uncooperative (i.e., investigatory blame legitimizing myth).

110

The analytic approach used to identify the statistical model that best represented the data
can be described as exploratory. The statistical model (see Figure 29) specified that the
characterological and circumstantial legitimizing myths were to predict investigatory blame
legitimizing myths; that the characterological and circumstantial legitimizing myths were to
predict one another; that all legitimizing myths were to predict investigative steps and final case
outcomes; and that the investigative steps were to predict final case outcomes. The victim and
perpetrator variables were then allowed to predict all other variables in the model. Allowing so
many relationships to be tested in a single model embodies an exploratory analytic approach as
all possible relationships between the exogenous and endogenous variables were examined
empirically.
Identifying the final model. To identify the statistical model that most adequately
represented the observed data, all relationships as indicated by single-headed arrows in Figure 6
were free to be estimated using MPlus Version 7.11 software (Muthén & Muthén, 1998-2012).
The initial model, as depicted in Figure 29, included variables with limited variance that
prevented convergence; these variables were trimmed from the model. This resulted in a revised
model that did not include victim sex (96% of the victims were female) or perpetrator sex (100%
of the perpetrators were male). The revised model is pictured in Figure 30. Assessment for model
fit, as described below, began with this model. Cases with missing data on any of the included
variables were excluded from analysis (i.e., listwise deletion). There were missing data on
perpetrator race (N = 5 cases) and on perpetrator age (N = 26 cases). Models that included either
of these two variables, therefore, had smaller sample sizes; the sample sizes for each model
tested are included in the results section.

111

Figure 30. Revised statistical model for analysis
Perpetrator Race
Victim Age

Victim Race

Characterological
Legitimizing Myths
*

Victim and
Perpetrator Age
Difference

Multiple Perpetrators

Number of Investigative
Steps

Circumstantial
Legitimizing Myths
Case Outcome
Investigatory Blame
Legitimizing myths

*Indicates an association, not a causal relationship.

112

To produce a more parsimonious solution and conserve statistical power, non-significant
variables (i.e., variables that did not have a significant relationship with any other variables in
the model) were trimmed from the model. The constrained model (i.e., with fewer variables) was
then compared to the less constrained model to assess for changes in model fit via the Bayesian
information criterion (BIC). The BIC is an index for comparing fit across models (Barrett, 2007;
Bollen, 1989; Bollen, Harden, Ray, & Zavisca, 2014; Byrne, 2012; Krueger, Hicks, Patrick,
Iacono, & McGue, 2002; Raftery, 1995; Schreiber, Nora, Stage, Barlow, & King, 2006). The
BIC provides a quantitative index of the extent to which the predicted values from the model
variance-covariance matrix correspond to the observed values, in consideration of the number of
parameters in the model (Bollen et al., 2014; Krueger et al., 2002). In other words, when fitting a
model, the analyst could choose to add parameters in order to improve fit. However, the BIC
penalizes models with a greater, rather than fewer, number of free parameters (Bollen, 1989;
Bollen et al., 2014; Krueger et al., 2002). In doing so, the BIC helps to identify the best-fitting,
most parsimonious model. When comparing models to one another, the recommendation is to
choose the model with the smaller BIC (Barrett, 2007; Bollen, 1989; Bollen et al., 2014); a BIC
change of 10 is considered very strong evidence in favor of selecting the model with the smaller
BIC (Krueger et al., 2002; Raftery, 1995). Accordingly, the BIC for the more constrained model
(i.e., with fewer free parameters) was compared to the less constrained model (i.e., with more
free parameters) and the model with the smaller BIC was selected. Each time a variable was
removed from the model (i.e., the variable was constrained to not have relationships with any
other variable in the model), change in model fit was assessed using the BIC. This process
continued to produce a model in which each variable in the model significantly predicted at least
one other variable in the model and no included variables prevented model convergence.

113

After non-significant variables were trimmed from the model, non-significant
relationships were trimmed from the model. A single variable in the model could have a
significant relationship to some other variables in the model as well as non-significant
relationships with other variables in the model. In other words, some arrows connecting variables
were significant while other arrows were not. As before, BIC was assessed to determine change
in fit with each new iteration of the model. This process was to result in a final model in which
only significant relationships were included.

Results
The overall goal of Study 2 was to examine empirically the relationships between
legitimizing myths, the investigative steps taken in each case, and the case outcome, all within
the context of specific factors of the victim and perpetrator. These relationships were analyzed
via SEM using maximum-likelihood estimation with robust standard errors (MLR) in MPlus
Version 7.11 software (Muthén & Muthén, 1998-2012).17 First, the full model (i.e., Model 1),
with all 5 exogenous variables predicting all 5 endogenous variables, as well as the endogenous
variables predicting one another was tested (See Figure 31a). Twenty-seven cases were excluded
from analysis as data were missing on the race of the perpetrator or on the victim and perpetrator
age difference, bringing the sample size for analysis to N = 221 and producing a model with BIC
= 2516.961. Missing data on the victim and perpetrator age difference was responsible for 22
cases being excluded and was also non-significant in the model. Therefore, Model 2 constrained
the relationship between victim and perpetrator age and all other variables in the Model to zero
(see Figure 31b). The sample analyzed in Model 2 was restricted to the same N = 221 cases used

17

The entire model-building process is provided in the Results section. However, only the results for the final model
will be explained.

114

in Model 1 so that the model BICs could be compared to one another. Model 2 produced a BIC =
2454.963. Given a BIC change of over 10, there was very strong evidence that Model 2 provided
a better fit than Model 1 supporting the exclusion of victim and perpetrator age difference from
the model and allowing cases that had previously been excluded due to missingness on this
variable alone (N = 22) to be included in the analysis (i.e., not subject to listwise deletion).

115

Figure 31. Models 1 and 2 tested via path analysis
Figure 31a. Model 1

Victim Race

Perpetrator Race

Victim Age

Characterological
Legitimizing Myths
*

Victim and
Perpetrator Age
Difference

Multiple
Perpetrators

Number of
Investigative Steps

Circumstantial
Legitimizing Myths
Case Outcome
Investigatory Blame
Legitimizing myths
N = 221; BIC = 2516.961

*Indicates an association, not a causal relationship.

116

Figure 31 (cont’d)
Figure 31b. Model 2

Perpetrator Race
Victim Race

Victim Age

Characterological
Legitimizing Myths
*

Multiple
Perpetrators

Number of
Investigative Steps

Circumstantial
Legitimizing Myths
Case Outcome
Investigatory Blame
Legitimizing myths

N = 221; BIC = 2454.963

*Indicates an association, not a causal relationship.

117

Model 3 (see Figure 32a) imposed the same constraints as Model 2 and allowed the full
sample to be included in analysis. Five cases were excluded from analysis as data were missing
on the race of the perpetrator only bringing the sample size for analysis to N = 243 and
producing a model with BIC = 2679.010. All of the variables in Model 3 had at least one
significant relationship with other variable(s) in the model. Model 4 (see Figure 32b) then
constrained all non-significant relationships in Model 3 to be zero and resulted in a BIC =
2527.270 providing very strong evidence that Model 4 is a better fit than Model 3. Not all
relationships entered into Model 4, however, remained significant. Specifically, perpetrator race
no longer had a significant relationship with the case outcome, as indicated by the dashed line in
Figure 32b. Model 5 (see Figure 32c) then constrained all non-significant relationships in Model
4 to be zero and resulted in a BIC = 2516.384, again providing very strong evidence that Model 5
is a better fit than Model 4. In addition, the improvement in fit and parsimony supported the
exclusion of perpetrator race from the model and allowed cases that had previously been
excluded to due to missingness on perpetrator race (N = 5) to be included in the analysis (i.e., not
subject to listwise deletion)

118

Figure 32. Models 3-5 tested via path analysis
Figure 32a. Model 3

Victim Race

Perpetrator Race

Victim Age

Characterological
Legitimizing Myths
*

Multiple
Perpetrators

Number of
Investigative Steps

Circumstantial
Legitimizing Myths
Case Outcome
Investigatory Blame
Legitimizing myths

N = 243; BIC = 2679.010
*Indicates an association, not a causal relationship.

119

Figure 32 (cont’d)
Figure 32b. Model 4

Victim Age

Multiple
Perpetrators

Victim Race

Characterological
Legitimizing Myths
*

Perpetrator Race

Number of
Investigative Steps

Circumstantial
Legitimizing Myths
Case Outcome
Investigatory Blame
Legitimizing myths

N = 243; BIC = 2527.270
*Indicates an association, not a causal relationship.

120

Figure 32 (cont’d)
Figure 32c. Model 5

Victim Age

Multiple
Perpetrators

Victim Race

Characterological
Legitimizing Myths
*

Number of
Investigative Steps

Circumstantial
Legitimizing Myths
Case Outcome
Investigatory Blame
Legitimizing myths

N = 243; BIC = 2516.384
*Indicates an association, not a causal relationship.

121

Model 6 (see Figure 33) imposed the same constraints as Model 5 and allowed the full
sample to be included in analysis. There were no missing data for any variables included in this
model, providing the full sample size of N = 248 and producing a model with BIC = 2569.235.
Model 6 was selected as the final model. Figure 34 depicts this model with corresponding
regression coefficients.18 All of the regression coefficients presented in Figure 34 were
significant at p < 005. Table 8 provides the regression coefficients for predictive relationships
between the exogenous and endogenous variables in the final, trimmed model.19

18

Note that “victim race” has been moved to the bottom of this model so as to prevent arrows from overlapping with
one another. While Model 6 is displayed slightly differently in figure 12 as compared to figure 11, it is the same
model.
19
Standardized regression coefficients are not provided because they are not available for models with count or
nominal predictor variables (Muthén & Muthén, 1998-2012).

122

Figure 33. Model 6 tested via path analysis

Victim Age

Multiple
Perpetrators

Victim Race

Characterological
Legitimizing Myths
*

Number of
Investigative Steps

Circumstantial
Legitimizing Myths
Case Outcome
Investigatory Blame
Legitimizing myths

N = 248; BIC = 2569.235
*Indicates an association, not a causal relationship.

123

Figure 34. Final model with regression coefficients

Multiple
Perpetrators

Victim Age

0.76

-1.12, -0.71**

*
0.52

Circumstantial
Legitimizing Myths

Characterological
Legitimizing Myths

0.33
0.053, 0.66***

Victim Race

-0.20

-0.27
Investigatory Blame
Legitimizing myths

0.07+

Number of
Investigative Steps
2.02, 1.59, 1.55++

Case Outcome

-0.80

N = 248; BIC = 2569.235
* Indicates an association, not a causal relationship; **Cases with victims aged 0-15 years old were used as the reference category.
Cases with victims aged 16-25 years old and 26 years or older were significantly less likely (b = -1.12, p = 0.000; b = -0.71, p = 0.016,
respectively) to have circumstantial legitimizing myths endorsed as compared to cases with victims aged 0-15 years old; *** Cases
with victims aged 0-15 years old were used as the reference category. Cases with victims aged 16-25 years old and 26 years or older
were significantly more likely (b = 0.53, p = 0.021; b = 0.66, p = 0.009, respectively) to have investigatory blame legitimizing myths
endorsed as compared to cases with victims aged 0-15 years old. +This value is an odds ratio for predicting that a case resulted in an
arrest and a referral, as compared to cases that had no arrest and no referral. These myths did not significantly predict just an arrest
(and no referral) or just a referral (and no arrest) as compared to cases that had no arrest and no referral to the prosecutor; ++These
values are odds ratios for predicting that a case resulted in (1) an arrest and a referral, (2) no arrest and a referral, and (3) an arrest and
no referral, respectively, as compared to cases that had no arrest and no referral. The number of investigate steps significantly
predicted all three case outcomes.
124

Table 8. Regression coefficients from the final model
Circum.
Legitimizing
Myths
b
S.E.

Char.
Legitimizing
Myths
b
S.E.

Invest. Blame
LMs

-1.117 0.297

04

--

0.534

-0.707 0.294

04

--

Victim
Race

04

--

04

Multiple
Perps

04

--

Circum.
LMs

--

Victim
Age: 16-25
years old1
Victim
Age: 26+
years old1

Char. LMs
Invest.
Blame LMs
Number
Invest
Steps

Number of
Investigative
Steps
b
S.E.

Case Outcome:
arrest and
referral2, 3
b
S.E.

Case Outcome:
no arrest and
referral2, 3
b
S.E.

Case Outcome:
arrest and no
referral2, 3
b
S.E.

0.232

04

--

04

--

04

--

04

--

0.656

0.250

04

--

04

--

04

--

04

--

--

-0.802

0.311

04

--

04

--

04

--

04

--

0.755

0.319

04

--

04

--

04

--

04

--

04

--

--

0.5246

0.1136

0.327

0.096

04

--

04

--

04

--

04

--

0.5246

0.1136

--

--

04

--

-0.200

0.061

04

--

04

--

04

--

04

--

04

--

04

--

-0.268

0.059

0.070

0.862

0.5005

0.515

0.9375

0.539

04

--

04

--

04

--

--

--

2.016

0.119

1.587

0.130

1.547

0.192

b

S.E.

1

Reference category = cases with victims aged 0-15 years old; 2Reference category = cases that had no arrest and no referral to the
prosecutor; 3All estimates for predictors of the case outcome are in odds ratios as case outcome is a nominal endogenous variable;
4
Parameter was non-significant in previous models and constrained to zero in the final model; 5These values were not statistically
significant and were included in the final model as they were necessary for dummy coding. All other values in the table were
significant at p < 0.05 and are bold; 6 Characterological legitimizing myths were regressed onto circumstantial legitimizing myths.
However, neither the research design nor existing literature provides justification for this to be interpreted as a causal relationship. As
such, this was interpreted as an association and therefore appears in two places in the table to reiterate this point.
125

To assess the relationships between the exogenous and endogenous variables, parameter
estimates from the final model (i.e., Model 6; Figure 12) were examined. Aim 1 of Study 2 was
to examine the relationships between the different types of legitimizing myths, as identified in
Study 1, and investigational effort and case outcomes. With each additional circumstantial
legitimizing myth documented in the police report (e.g., victim is not upset enough), a case was
significantly more likely to have characterological legitimizing myths (e.g., victim is a sex
worker) and investigatory blame legitimizing myths endorsed (e.g., victim is uncooperative) (b =
0.524, p = 0.000; b = 0.327, p = 0.001, respectively). In other words, cases that had more myths
related to specific circumstances of the assault (e.g., victim wasn’t upset enough, wasn’t injured,
or actually consented) were likely to have more myths related to the character of the victim (e.g.,
victim is a regular drug user, sex worker, or not credible) and to have more myths that blame the
victim for a less-than-thorough investigation (e.g., victim is uncooperative, doesn’t have enough
information, or is weak).
With each additional characterological (e.g., victim is a sex worker) or investigatory
blame legitimizing myth (e.g., victim is uncooperative) documented in the police report, a case
was likely to have fewer investigative steps completed (b = -0.200, p = 0.001; b = -0.268, p =
0.000, respectively). This meant that cases that had more myths related to the character of the
victim (e.g., that the victim is a regular drug user, sex worker, or not credible) or myths that
blame the victim for a less-than-thorough investigation (e.g., that the victim is uncooperative,
doesn’t have enough information, or is weak) were likely to have fewer investigative steps
completed (e.g., a suspect interrogation, canvassing of the area, or photographs taken at the scene
of the crime). However, only investigatory blame legitimizing myths directly predicted the case
outcome, such that cases with more investigatory blame legitimizing myths endorsed were

126

significantly less likely to result in an arrest and a referral (OR = 0.070, p = 0.002). Investigatory
blame legitimizing myth endorsement, however, did not predict the other two categories of
possible case outcomes: no arrest with a referral to the prosecutor (OR = .500 p = 0.178) or an
arrest with no referral to the prosecutor (OR = 0.937, p = 0.904). That is, cases that had more
investigatory blame legitimizing myths documented in the police files (e.g., victim is
uncooperative, doesn’t have enough information, or is weak) were only significantly less likely
to have both an arrest and a referral on the case—the single outcome that means the case may
move forward in the criminal justice system.
Aim 2 of Study 2 was to examine the relationship between investigational effort across
sexual assault cases and the final case outcomes. The number of investigative steps predicted all
possible case outcomes with each additional step increasing the likelihood of an arrest and a
referral (OR = 2.016, p = 0.000), no arrest with a referral (OR = 1.587, p = 0.000), or an arrest
with no referral (OR = 1.547, p = 0.023). That is, if law enforcement personnel conducted a more
thorough investigation with more investigational steps completed, the case was much more likely
to have an opportunity of proceeding in the criminal justice system (i.e., an arrest and a referral).
Finally, Aim 3 of Study 2 was to examine the impact of specific social identity factors of
the victim and perpetrator (i.e., sex, race, and age) and the number of perpetrators on the types of
legitimizing myths endorsed, investigational effort, and case outcomes. Cases in which the
victims were 16-25 years-old and cases in which the victims were over the age of 25 had
significantly fewer circumstantial legitimizing myths endorsed (e.g., victim is not upset enough)
as compared to victims under 16 years old (b = -1.117, p = 0.000; b = -0.707, p = 0.016,
respectively). Additionally, cases in which the victims were 16-25 years-old and cases in which
the victims were over the age of 25 had significant more investigatory blame legitimizing myths

127

endorsed (e.g., victim is uncooperative) as compared to victims under 16 years old (b = 0.534, p
= 0.021; b = 0.656, p = 0.019, respectively). In other words, cases with victims under the age of
16 were significantly more likely to have myths related to specific circumstances of the assault
(e.g., that the victim wasn’t upset enough, wasn’t injured, or actually consented) and
significantly less likely to have myths that blamed them for the less-than-thorough investigation
(e.g., victim is uncooperative, doesn’t have enough information, or is weak). Cases with White
victims had significantly fewer investigatory blame legitimizing myths endorsed (e.g., victim is
uncooperative) as compared to cases with victims of Color (b= -0.802, p = 0.010). This means
that victims of Color were more likely to be blamed for a less-than-thorough investigation (e.g.,
victim is uncooperative, doesn’t have enough information, or is weak). Finally, cases with
multiple perpetrators had significantly more characterological legitimizing myths endorsed (e.g.,
victim is uncooperative) as compared to cases with a single perpetrator (b = 0.755, p = 0.018).
That is, gang rapes (i.e., more than one perpetrator) had significantly more myths related to the
character of the victim (e.g., victim is a regular drug user, sex worker, or not credible) as
compared to cases with a single perpetrator.

128

DISCUSSION
Prior social science literature has documented the criminal justice system’s negligent
response to sexual assault via the high rates of sexual assault case attrition through the system
(e.g., see Lonsway & Archaumbault, 2012). This body of research has also provided tangible
proof of the lack of effort put into the investigation stage of the criminal justice system process
via the large quantity of unsubmitted SAKs in jurisdictions across the country (e.g., see
Campbell et al., 2014 forthcoming; Shaw & Campbell, 2013; Strom & Hickman, 2010).
However, not all cases slip through the cracks. Rather, how a case progresses through the
criminal justice system varies depending on specific factors of the victim, suspect, and assault,
including but not limited to the age of the victim and the number of perpetrators involved (e.g.,
see Campbell et al., 2009; Shaw & Campbell, 2013). Taken together, these studies cleary show
“what” is happening when victims report to law enforcement and the “who,” “when,” and
“where” factors associated with case progression.
The purpose of the current project was to illuminate the “why” and “how”—why does the
criminal justice system respond to sexual assault in this way? And, more specifically, how do
police explain their response, given that the majority of sexual assault cases fall out of the system
while under their purview (e.g., see Campbell et al., 2014)? If we can identify how law
enforcement personnel explain their response to sexual assault, we can design and implement
interventions to change it, potentially improving the overall criminal justice system response.
To do this, the current project drew upon social dominance theory, as it provides a
comprehensive theoretical model for understanding and examining mechanisms that support and
sustain group-based social hierarchies (Sidanius & Pratto, 2001, 2011). Specifically, social
dominance theory was used to conceptualize the negligent criminal justice system response to

129

sexual assault as a form of institutional discrimination. Within the social dominance theory
framework, legitimizing myths are used to provide justification for institutional discrimination,
effectively supporting and reinforcing it. Legitimizing myths, then, are the “how” within this
theory. Therefore, the current project sought to identify the legitimizing myths used by police to
justify the current criminal justice system response to sexual assault.
Study 1 documented the types and extent of legitimizing myths in police records of
sexual assault cases. Law enforcement personnel’s endorsements of legitimizing myths in
relation to sexual assault (e.g., rape myths) had been previously measured via questionnaires and
surveys (see Edwards et al., 2011), so it was unknown if police attitudes towards rape would be
observable in their investigations. Study 2 examined the empirical relationships between the
different types of legitimizing myths documented in Study 1 with the investigatory effort and
case outcomes across cases in order to assess whether legitimizing myths are used to justify law
enforcement personnel’s response to sexual assault. Study 2 also examined how specific social
identity factors of the victim and perpetrator (i.e., the context of the assault) and the number of
perpetrators affected the use of legitimizing myths, investigatory effort, and case outcomes. In
building a model that incorporated all significant relationships between these variables, the
current project sought to identify where change efforts should be targeted to improve the
criminal justice system response to sexual assault. The findings for each study will be
summarized below, comparing and contrasting the current findings to the existing literature. The
limitations of the current project will then be presented, followed by a discussion of implications
for policy and practice, as well as directions for future research.

130

Study 1: Documenting Legitimizing Myths
Rape myths have been identified as one type of legitimizing myths in the United States
that operate to justify or excuse related discrimination, prejudice, and oppression (Edwards et al.,
2011; Pratto et al., 1994; Sidanius & Pratto, 2001, pp. 85-87). Study 1 examined whether
legitimizing myths (e.g., rape myths) were observable in police records of sexual assault case
investigations and documented the extent and types of legitimizing myths endorsed. Three
different types of legitimizing myths emerged from the data: circumstantial, characterological,
and investigatory blame legitimizing myths.
Circumstantial and characterological legitimizing myths aligned very closely with prior
literature that has assessed attitudes towards rape via rape myth acceptance scales. Table 9
presents the different sub-categories of circumstantial and characterological legitimizing myths
endorsed in the current project, alongside statements that are frequently included in rape myth
acceptance scales (Lonsway & Fitzgerald, 1995; McMahon & Farmer, 2011; Page, 2010; Payne
et al., 1999). This table highlights how most of the observations of circumstantial and
characterological legitimizing myths documented in the current project closely aligned with
statements on rape myth acceptance scales in the literature measuring attitudes towards rape.

131

Table 9. Alignment between circumstantial and characterological legitimizing myths in current
project with the prior literature
Circumstantial Legitimizing Myths
Suggest rape didn’t happen based on specific circumstances of the sexual assault
Sub-categories in Current Project
Victim is lying

Rape Myth Acceptance Scale Items20
“Women falsely report rape to call attention to
themselves.”

Victim is not injured

“A rape probably didn’t happen if the woman has no
bruises or marks.”

Victim consented

“Many so-called rape victims are actually women
who had sex and “changed their mind” afterwards.”

Victim is not upset

NONE

Victim didn’t act like a victim afterwards

NONE

Characterological Legitimizing Myths
Suggest rape didn’t happen based on specific characteristics of the victim
Sub-categories in Current Project
Victim is a regular drug-user

Rape Myth Acceptance Scale Items
“If a girl is raped while she is drunk, she is at least
somewhat responsible for letting things get out of
hand.”

Victim is a sex worker

“In any rape case one would have to question
whether the victim is promiscuous or has a bad
reputation.”

Victim has “done this before”

“In any rape case one would have to question
whether the victim is promiscuous or has a bad
reputation.”

Victim is “mental”

“A lot of times, girls who claim they were raped
have emotional problems.”

Victim is promiscuous

“If a girl acts like a slut, eventually she is going to
get into trouble.”

Victim is not credible

“In any rape case one would have to question
whether the victim is promiscuous or has a bad
reputation.”

20

(Lonsway & Fitzgerald, 1995; McMahon & Farmer, 2011; Page, 2010; Payne et al., 1999)

132

As can be seen in Table 9, all but two sub-categories of circumstantial legitimizing myths
directly align with items on rape myth acceptance scales; victim is not upset and victim didn’t act
like a victim afterwards did not have exact corresponding items. Table 9 also highlights that
characterological legitimizing myths as documented in the current project were frequently
represented by one common item questioning the victim’s reputation in rape myth acceptance
scales (i.e., “in any rape case one would have to question whether the victim is promiscuous or
has a bad reputation”). Regardless, there is ample overlap between the circumstantial and
characterological legitimizing myths delineated in the current project with the prior literature on
rape myth acceptance. Therefore, the key contribution of the current project in relation to the
identification of circumstantial and characterological legitimizing myths in police records is that
there is now evidence that the negative attitudes about victims among police documented via
survey research are also observed in their investigational practices.
The third category of legitimizing myths identified in Study 1 was investigatory blame
legitimizing myths. This type of myth blames the victim for the less-than-thorough investigation
carried out by police, as the victim was considered to be unwilling or unable to assist in the
process. Specifically, investigatory blame legitimizing myths included statements that the victim
was uncooperative; didn’t have enough information; didn’t have a phone or address for contact;
or was weak and would not make a good witness or victim at trial. Rape myth acceptance scales
include statements that assign blame to the victim, but the statements on these surveys tap beliefs
about victims’ culpability for the assault itself, not for the subsequent investigation (e.g., see
Kettrey, 2013; Lonsway & Fitzgerald, 1995; McMahon & Farmer, 2011; Payne et al., 1999). In
other words, prior rape myth acceptance studies have found that victims are blamed for the
assault; when victim-blaming is observed in police records corresponding to sexual assault case

133

investigations, the victim is instead blamed for the less-than-thorough police response. Indeed,
legitimizing myths that blamed the victim for police inaction were the most common with over
forty percent of the cases in the current project having this type of legitimizing myth endorsed.
These new findings can be combined with the prior literature documenting victimblaming attitudes for the assault itself to suggest that victims who choose to report the assault to
police are subject to being blamed for (1) allowing the rape to happen in the first place, and (2)
for a stalled investigation by law enforcement personnel. The experiences of sexual assault
survivors who choose to report their assault to police are frequently referred to as “secondary
victimization” or the “second rape” (Campbell, 2005; Campbell & Raja, 2005; Campbell et al.,
2001; P. Y. Martin & Powell, 1994). These interactions are defined by the cold or impersonal
reception from legal personnel who lack empathy, express disbelief, blame the victims for the
assault, and even deny them services (Campbell, 2005, 2008; Campbell & Raja, 2005; Logan et
al., 2005; Madigan & Gamble, 1991; P. Y. Martin, 2005; P. Y. Martin & Powell, 1994). If
blaming the victim for the assault itself (e.g., her poor decisions led to the rape) is part of what
defines “secondary victimization,” then blaming the victim yet again for the poor investigation
carried out by police could be conceptualized as “tertiary victimization.” The survivor was first
victimized during the assault, a second time by legal personnel as the survivor was blamed for
being assaulted, and a third time by law enforcement personnel as the survivor was blamed for
police not conducting a thorough investigation. If conceptualized in this way, over forty percent
of the victims associated with cases in the current project were subject to tertiary victimization.

134

Study 2: Examining Relationships Between Legitimizing Myths and Outcomes
Study 2 investigated the relationships between the different types of legitimizing myths,
investigatory effort, and case outcomes, within the context of specific social identity factors of
the victim and perpetrator (i.e., sex, race, and age) and the number of perpetrators. While the
final model (see Figure 34, page 124) includes many different variables with varying
relationships to one another, the design for Study 2 was predicated upon a rather simple
conceptual model. Figure 35 is a reprint of Figure 5; Figure 5 was originally presented on page
44 to provide a representation of the relationships to be examined. The results of the final model
(see Figure 34, page 124) will be discussed below in terms of what they tell us about the
relationships depicted in Figure 35.

Figure 35. A reprint of Figure 5
Victim and Perpetrator
Social Identity Factors
(Unequal Intergroup Context)
Box
4

Investigatory Effort
(Institutional Discrimination)
Rape Myths and Other
Justifications
(Legitimizing Myths)

Box
2

Box
1

Case Outcomes
(Institutional
Discrimination)
Box
3

135

The role of legitimizing myths in predicting investigatory effort and case outcomes.
Of particular interest was investigating the relationship between legitimizing myths, and
investigatory effort and case outcomes (i.e., the arrows between Boxes 1 and 2 and between
Boxes 1 and 3 in Figure 35). Social dominance theory suggests that legitimizing myths are the
mechanism (i.e., the “why” and “how”) by which institutional discrimination is supported and
reinforced (Sidanius & Pratto, 2001, 2011). In short, the legitimizing myths identified in Study 1
fulfilled this role as they predicted the number of investigational steps completed on cases, as
well as the case outcomes. However, not all legitimizing myths were equally influential; the
different types of legitimizing myths documented in Study 1 had different relationships with the
number of investigational steps completed and case outcomes.
Investigatory blame legitimizing myths (i.e., that blame the victim for a less-thanthorough investigation) were the only legitimizing myths that directly predicted the case
outcome; this type of legitimizing myth also had indirect influence on the case outcome, via the
number of investigative steps completed. This means that cases with more investigatory blame
had significantly fewer investigative steps completed and were then significantly less likely to
proceed forward to the prosecution stage of the criminal justice system process. However,
because investigatory blame also directly predicted the case outcome, it didn’t always matter
how many investigative steps were completed on a case. That is, even if a sexual assault case had
all possible investigative steps completed, if police stated that the victim was uncooperative, did
not have enough information, did not have a phone or address for contact, and/or was described
as weak or incompetent, the case was significantly less likely to be referred to the prosecutor’s
office and have an associated arrest. Study 1 conceptualized investigatory blame legitimizing
myths as tertiary victimization. Study 2 revealed that tertiary victimization was particularly

136

damaging in that once an investigatory blame legitimizing myth was endorsed on a case, the
likelihood that the case would move forward to prosecution dropped to just seven percent (see
OR = 0.07 in Table 8).
Circumstantial and characterological legitimizing myths also impacted the case outcome,
though their influence was always indirect via the number of investigative steps completed
and/or the number of investigatory blame legitimizing myths endorsed. Therefore, the
endorsement of circumstantial and/or characterological legitimizing myths on a given case hurt
the case’s likelihood of moving forward in the criminal justice system by influencing the number
of investigative steps completed and/or the number of investigatory blame legitimizing myths
endorsed. However, it did not have the same direct impact on case progression as the
investigatory blame legitimizing myths. For example, if police noted that the victim was not at
all upset after the assault (i.e., they endorsed a circumstantial legitimizing myth), they were more
likely to also state that the victim was uncooperative (i.e., they endorsed an investigatory blame
legitimizing myth). If they stated that the victim was uncooperative, the number of investigative
steps completed on the case decreased, as did the likelihood that the case would proceed in the
criminal justice system. On the other hand, if police stated that the victim was not at all upset
after the assault (i.e., they endorsed a circumstantial legitimizing myth) but did not go on to also
state the victim was uncooperative (i.e., they endorsed an investigatory blame legitimizing
myth), there was no impact on the number of investigational steps completed or the case
outcome.21 In this way, the circumstantial and characterological legitimizing myths were not as
damaging to case progression as the investigatory blame legitimizing myths.

21

The language in the example provided states that each type of myth was endorsed or not (i.e., suggests a binary
code). This was done to aid in understanding and as an attempt to simplify very complex relationships. As has been
noted throughout this document, the legitimizing myth variables included in Study 2 were all count variables.
Additionally, the example suggests that the endorsement of one specific (sub-category of) legitimizing myth affected

137

Overall, while the specific relationships between the different types of legitimizing myths
and the case outcomes were nuanced, they all in some way (i.e., directly or indirectly) predicted
the likelihood that a case would move onto the prosecution stage of the criminal justice system;
as the number of legitimizing myths increased, the likelihood that the case would be referred to
the prosecutor and have an associated arrest decreased. These legitimizing myths operated as
predicted by social dominance theory to justify the criminal justice system response to sexual
assault (Sidanius & Pratto, 2001, 2011), confirming the arrow drawn between Boxes 1 and 2 and
between Boxes 1 and 3 in Figure 35. Police provided explanations for their negligent response to
sexual assault by stating that the reported offense wasn’t really rape given the specific
circumstances of the assault, the characteristics of the rape victim, or by blaming the victim for
law enforcement personnel’s less-than-thorough investigation. Based on the current project
findings, these legitimizing myths explain “why” law enforcement personnel behave in the way
that they do in responding to sexual assault cases and “how” they explain their response.
The influence of victim and perpetrator social identity factors. Social dominance
theory argues that it is important to consider how the identities of individuals interacting with
one another in a given context (i.e., unequal intergroup context) influence the mechanistic
relationship between legitimizing myths (i.e., the circumstantial, characterological, and
investigatory blame legitimizing myths) and institutional discrimination (i.e., sexual assault case
investigatory effort and outcomes) (Sidanius & Pratto, 2001, 2011) (i.e., the arrows between Box
4 and Boxes 1-3 in Figure 35). Of the specific variables examined in Study 2, the age of the
victim, race of the victim, and whether the case involved multiple perpetrators significantly

the likelihood of another specific (sub-category) legitimizing myth being endorsed. This too is done to aid in
understanding and as an attempt to simplify very complex relationships. In reality, it could have been any specific
(sub-category) legitimizing myth in the relevant category.

138

predicted legitimizing myth endorsement; none of these variables significantly predicted
investigatory effort or case outcomes directly.
The age of the victim predicted what types of legitimizing myths were likely to be
endorsed. Specifically, cases with victims over the age of consent (i.e., 16 years old or older) had
significantly more investigatory blame legitimizing myths endorsed (e.g., the victim is
uncooperative) whereas victims under the age of consent had significantly more circumstantial
legitimizing myths endorsed (e.g., the victim is not upset). In other words, police were more
likely to say that a victim over the legal age of consent was uncooperative, didn’t have enough
information, was without a phone/address for contact, or was weak and unable to hold up at trial.
Police were more likely to say that a minor victim was lying, not injured, consented, not upset, or
didn’t act like a victim afterwards as compared to victims over the legal age of consent.
Study 2 suggests that something changes in how police explain their response to sexual
assault once the victim is old enough to legally consent to sex. Law enforcement personnel seem
to shift their emphasis from justifying inaction by denying a rape happened (e.g., because he/she
wasn’t injured) to blaming the victim for being unwilling and/or unable to participate in the
subsequent investigation (e.g., because he/she was uncooperative). This shift in police
explanations for case outcomes is important as the investigatory blame legitimizing myths are
more damaging to the case overall, as they directly predict the case outcome. Because law
enforcement personnel are more likely to endorse investigatory blame legitimizing myths for
victims over the legal age of consent, these case are then likely to have significantly fewer
investigative steps completed and less likely to be referred to the prosecutor.
These findings are consistent with prior literature documenting that cases with adolescent
victims over the age of consent were less likely to have their SAK submitted to a crime lab for

139

analysis, less likely to be referred to the prosecutor, and more likely to be classified as a false
report, as compared to cases with victims under the age of consent (Campbell et al., 2010;
Heenan & Murray, 2006; Kelly et al., 2005; Rose & Randall, 1982; Shaw & Campbell, 2013;
Triggs et al., 2009). In fact, the significant effect of age documented in these prior studies may
have been due to investigatory blame legitimizing myths, as suggested by the results of the
current project, though this association is speculative given that prior studies have not examined
investigatory blame.
It is important to note that while social dominance theory would have predicted an effect
of age on the use of legitimizing myths and/or case outcomes, the predicted effect was not
necessarily in the observed direction. Social dominance theory would categorize younger
individuals as members of the subordinate group and older individuals as members of the
dominant group and that as someone gets older, they have more social value (Sidanius & Pratto,
2001, 2011). Therefore, it would be expected that younger individuals would be less likely to be
referred to the prosecutor’s office. The findings of the current study do not support this
expectation and instead suggest a more complex relationship between victim age and the
criminal justice system response that takes into consideration the social and legal implications of
particular ages (e.g., the age of consent).
The race of the victim was also associated with legitimizing myth endorsement. This
finding was surprising in the current project given that there was limited variance in victim race;
only thirteen percent of the cases in the sample involved a White victim. The rate of
investigatory blame legitimizing myths endorsed on cases involving White victims was different
enough from the rate on cases involving victims of Color for a significant association to be
detected, even with the limited variance. Specifically, cases with White victims had significantly

140

fewer investigatory blame legitimizing myths endorsed as compared to cases with victims of
Color.
Prior research examining the influence of victim race on the criminal justice system
response to sexual assault has been mixed. Some studies have found no effect of race
(Kerstetter, 1990); some have found cases with victims of Color are not taken as seriously, are
more likely to be unfounded by police, and are less likely to be prosecuted (D. Black, 1978;
Frohmann, 1997; LaFree, 1981; Reiss, 1971; Rose & Randall, 1982; Smith & Klein, 1983;
Wriggins, 1983); still others have found that cases with victims of Color are more likely to have
a suspect identified (though not arrested), less likely to be unfounded, and more likely to have
their rape kit submitted to the crime lab for analysis (Bryden & Lengnick, 1997; Horney &
Spohn, 1996; Shaw & Campbell, 2013). Study 2 did not find a direct impact of victim race on
the number of investigational steps completed or the case outcome, aligning with prior literature
that has found no effect of race. However, there was an indirect effect of race on the number of
investigational steps and case outcomes via the number of investigatory blame legitimizing
myths endorsed. Law enforcement personnel were more likely to state that victims of Color were
uncooperative, didn’t have enough information, had no phone/address for contact, and/or were
weak and unable to hold up as solid victims at trial. The significant effect of race in prior
literature may have had less to do with the specific race of the victim, and more to do with the
frequency of legitimizing myths that blame the victim for the less-than-through investigative
response. Again, because prior literature has not documented the observations of legitimizing
myth endorsements in police investigations, this cannot be known with certainty.
Regardless, Study 2 highlighted that victims of Color were more likely than White
victims to be blamed for a less-than-thorough investigation, which then predicted fewer

141

investigative steps being completed and a lower likelihood that the case would move forward to
prosecution. Social dominance theory would categorize victims of Color as belonging to the
subordinate group, with their White counterparts as belonging to the dominant group (Sidanius &
Pratto, 2001, 2011). Within this framework, it would be expected that cases involving victims of
Color would not proceed into the prosecution stage of the criminal justice system response to
sexual assault. That was documented here, and legitimizing myths that blame the victim for the
less-than-thorough investigation were used to justify this differential response.
Finally, in regards to characteristics of the assault, cases with multiple perpetrators (i.e.,
gang rapes) were likely to have more characterological legitimizing myths as compared to cases
with a single perpetrator. Shaw and Campbell (2013) found that cases involving multiple
perpetrators were less likely to have their SAK submitted to the crime lab for analysis as
compared to cases with a single perpetrator. Study 2 is consistent with these prior findings and
furthermore suggests the means by which law enforcement personnel explain this differential
response: victims involved in gang rapes were drug users, sex workers, had “done this before,”
were “mental,” were promiscuous, or were just not credible. As such, it wasn’t really rape, so
fewer investigative steps were completed on the case (e.g., SAKs are less likely to be submitted,
as documented by Shaw & Campbell, 2013), and the case was less likely to move onto the
prosecution stage of the criminal justice system response.
As predicted by social dominance theory, specific factors of the assault such as the age
and race of the victim, as well as the number of perpetrators involved, predicted the use of
legitimizing myths in explaining the criminal justice system response to sexual assault,
confirming the arrow drawn between Box 4 and Box 1 in Figure 35. However, the impact of
specific case factors on investigatory effort and case outcomes was indirect. Therefore, the

142

arrows between Box 4 and Boxes 2-3 in Figure 35 are not supported by the current study. That
is, when specific case factors, including the victim age, victim race, and number of perpetrators,
influenced the investigatory effort and case outcomes, the relationship was fully mediated by the
endorsement of legitimizing myths. This strengthens the role that the legitimizing myths play in
providing the “how” and “why” as their endorsement was necessary for the identified case
factors to influence the case outcome.
The relationship between investigatory effort and case outcomes. In the current
project, investigatory effort and case outcomes were both considered to be forms of institutional
discrimination, as suggested by social dominance theory. However, investigatory effort and case
outcomes were conceptualized and examined in the current study as two separate constructs in
order to assess their relationship with one another. If a thorough investigation were conducted on
every sexual assault case prior to deciding the case outcome, there should have been no
relationship between these two variables.
Study 2 found a significant relationship between these two variables; with each additional
investigative step completed on a case, the case was 1.55 times more likely to have only an
arrest, 1.59 times more likely to have only a referral and twice as likely to have both an arrest
and a referral. This suggests that law enforcement personnel decide on a case outcome (i.e., if
they will refer the case to the prosecutor and arrest a suspect or not) prior to conducting a
thorough investigation and completing all possible investigative steps. Prior research has found
that cases are more likely to be referred to the prosecutor’s office when additional evidence has
been collected (Campbell et al., 2009). However, the current study suggests that causation may
flow in the opposite direction; law enforcement personnel decide which cases should continue

143

onto the prosecution stage and then complete additional investigative steps to prepare it for
referral. This finding confirms the arrow drawn between Box 2 and Box 3 in Figure 35.
Figure 36, below, provides an updated version of the conceptual model, based on the
findings of the current study and which specific relationships were confirmed. The criminal
justice system has long been defined by social dominance theory as one of the most important
hierarchy-enhancing institution (Sidanius et al., 1994). The specific variables and relationships
depicted in Figure 36 provide a visual display of how this system mechanistically operates so as
to support and reinforce the status quo. That is, the criminal justice system helps to maintain the
disproportionate allocation of social value between dominant and subordinate groups by
providing (i.e., via their investigatory effort and case outcomes), and then justifying (i.e., via
legitimizing myths), a systematically different response to marginalized individuals (e.g., victims
of color). In short, these processes are maintained so as to not shift the current power structure
and to keep marginalized individuals in the margins.

144

Figure 36. The conceptual model illustrating confirmed relationships only

Victim and Perpetrator
Social Identity Factors
(Unequal Intergroup Context)
Box
4

Investigatory Effort
(Institutional
Discrimination)

Rape Myths and Other
Justifications
(Legitimizing Myths)

Box
2

Box
1

Case Outcomes
(Institutional
Discrimination)
Box
3

Limitations
The current project does have several key limitations. First, this project relied on paper
records corresponding to sexual assault cases that dated back nearly thirty years (i.e., 19802009). The condition of the paper records presented limitations in coding for information.
Specifically, it was not possible to code for co-occurring crimes as this information was not
systematically recorded by law enforcement personnel. Many times, the specific charges related
to each case would not be delineated until the case was referred to the prosecutor’s office. Prior
literature has found that law enforcement personnel are more likely to “solve” (i.e., clear by
arrest or exception) a rape if it co-occurs with another crime (i.e., robbery, burglary, or property
theft) as compared to rapes that do not have co-occurring crimes (Addington & Rennison, 2008).
It is possible, and perhaps likely, that cases with co-occurring crimes in the current project were
145

handled in a systematically different way than rapes that did not have co-occurring crimes.
However, this could not be examined empirically as this information could not be gleaned
consistently from the paper records.
Additionally, over the thirty years that these cases accrued, the associated police
department moved six times. Police records were likely lost along the way, explaining why only
248 sexual assault cases out of the 400 cases in the original random sample were able to be
included in the current project. While there is nothing to suggest that the excluded cases are
systematically different from the included cases, it is unknown as no police records were
available. Related to this point, included case files could have also had missing information.
Paperwork may have been misplaced, or police may not have recorded all investigative steps
taken on a specific case, for example. Because this project relied on archival records, if
something was not recorded, it was not able to be included in analyses.
Second, the current project suggests that the endorsement of a specific legitimizing myth
predicted the number of investigative steps taken on a case and the case outcome. However,
there was no way to determine causal inference. It is possible that law enforcement personnel did
not conduct a thorough investigation or decided on a case outcome due to extenuating
circumstances, and then manufactured a justification for their action and/or inaction. For
example, an investigating officer may have had to close a case prematurely due to time
constraints. The officer may have then noted that the victim didn’t have a phone and so was
unable to reach the victim. The current study cannot state with certainty that legitimizing myths
came first, followed by a decision on how to proceed with the case as opposed to a decision
being made on how to proceed, and then justified via legitimizing myths.

146

Third, the current study was conducted using sexual assault case records from the city of
Detroit, which is a racially homogenous city (US Census Bureau, n.d.). Accordingly, the sample
was also racially homogenous, with 87% of the victims and 92% of the perpetrators being
individuals of Color. While victim race did significantly predict the number of investigatory
blame legitimizing myths in the current project, perpetrator race was not significantly related to
any variables in the path analysis model. It is possible that perpetrator race did significantly
predict the types of legitimizing myths endorsed, the number of investigational steps completed,
and/or case outcomes, but that these relationships were not able to be detected in the current
project due to limitations in variance. Similarly, the sex of the victims and/or perpetrators may
have had significant relationships with the other variables examined in the current project.
However, sexual assault is a gendered crime primarily perpetrated by men against women (M. C.
Black et al., 2011). Because 96% of the victims in the sample were female and 100% of the
perpetrators were male, there was not ample variance to detect these relationships.
Finally, it is important to acknowledge that some important considerations in examining
the criminal justice system response to sexual assault were simply beyond the scope of the
current study. The criminal justice system relies on resources and infrastructure for success.
However, recent research has found that stakeholder groups involved in the response to sexual
assault in Detroit, including the police department, crime lab, prosecutor’s office, medical
system, and victim service agencies, were operating in a state of chronic resource depletion
during the time period associated with these sexual assault case investigations (Campbell et al.,
2014 forthcoming). The current study did not examine how limited resources impacted case
progress. Furthermore, because no other cities have responded to unsubmitted rape kits in police
property in the same fashion as Detroit (i.e., their systematic, empirically-informed approach of

147

first drawing a random sample of all unsubmitted kits to better understand the scope of the
problem), it is unknown if what has been documented in the current project is comparable to
other cities.
Implications for Policy and Practice
Despite these limitations, the findings of this project can be used to inform policy and
practice change within the criminal justice system. Changing the criminal justice system
response to sexual assault can seem daunting for two key reasons: (1) the high rates of sexual
assault case attrition seem to be somewhat inherent in the system (e.g., see Lonsway &
Archambault, 2012; Spohn & Tellis, 2012) and (2) systems change is complex and challenging
(Foster-Fishman, Nowell, & Yang, 2007). Fortunately, within any system, there are “leverage
points” at which small changes in one thing can produce big changes in everything else
throughout the system (Foster-Fishman et al., 2007; Meadows, 1999).
The system change literature defines leverage points (or levers for change) as parts of a
system that maintain and constrain system patterns and come in two general types: apparent
structures and deep structures (see Foster-Fishman et al., 2007). Apparent structures are the
visible elements of a system that might explain why and how a system operates as it does. This
includes policies, practices, procedures, roles, responsibilities, resources, and power structures.
In the criminal justice system, this would include policies that mandate how evidence is to be
handled (e.g., strict policies on maintaining the change of custody) and the organizational chart
indicating the chain of command (e.g., patrol officers answer to sergeants, sergeants answer to
lieutenants, etc.). Deep structures are below the surface and refer to the normative elements of
the system. This includes attitudes, values, beliefs, expectations, and tacit assumptions that drive
behavior for members of the system. Within the criminal justice system, this would include

148

values placed on different case assignments (e.g., some cases are considered more high-profile as
compared to others and thus may be more valued) and assumptions about how victims should
interact with law enforcement personnel (e.g., victims should readily accessible if they want their
case investigated). The literature on systems and organizational change suggests that for change
to be sustained over time, it is necessary that attention is given to both apparent and deep
structures (Foster-Fishman et al., 2007; Howard, Logue, Quimby, & Schoeneberg, 2009). The
current project has helped in identifying several leverage points for change operating at the
apparent and deep levels.
Changing apparent structures via policy change. In order to reduce rates of sexual
assault case attrition within the criminal justice system, a greater proportion of sexual assault
cases would need to transition from the investigation stage, overseen by law enforcement, to the
prosecution stage, overseen by the prosecutor’s office. This means that more sexual assault cases
would need to be referred to the prosecutor’s office and have an arrestee. Of all the variables
tested, only two directly predicted case outcomes: the number of investigative steps completed
on a case and the number of investigatory blame legitimizing myths endorsed. The effects of all
other variables in the model had to “go through” one of these two pathways—through the
number of investigative steps or through the number of investigatory blame legitimizing myths—
in order to impact the outcome of the case. For example, the number of characterological
legitimizing myths endorsed in a sexual assault case did not directly predict if the case moved
onto the prosecution stage; instead, the number of characterological legitimizing myths predicted
the number of investigative steps, which then predicted the case outcome. As such, these two
variables represent leverage points for change. If the number of investigative steps completed on
a case were to increase or the number of investigatory blame legitimizing myths endorsed on a

149

case was to decrease, it would increase the likelihood of a more desirable case outcome, such as
a referral and an arrest, meaning the case would have the opportunity to progress forward in the
criminal justice system. Therefore, changes to current system policies regulating sexual assault
case investigations (i.e., small changes in one thing) could leverage change in the rates of case
attrition (i.e., big change everywhere else).
Fortunately, policy changes could be implemented that target these two leverage points
for change (i.e., increasing the number of investigative steps completed in a case and decreasing
the number of investigatory blame legitimizing myths endorsed on a case). In most police
agencies, a supervising officer is responsible for monitoring cases to ensure that progress is
being made and that the investigation is thorough and accurate (Zoller, Normore, & McDonald,
2014). In order to leverage the number of investigative steps completed on a sexual assault case,
a policy could be implemented that a supervising officer will not sign off on a case until all
possible investigative steps have been completed and/or documented rationale is provided for
any steps that have not been completed. For example, a suspect line-up is typically unnecessary
in a rape case with a known offender. In such cases, the investigating officer or detective should
note that the suspect line-up was not completed for this reason, satisfying the policy
requirements. In the current study, an average of 3.38 steps was taken on each case, out of ten
possible steps. If supervising officers require this type of documentation on all cases prior to
their sign-off, it would increase the number of investigative steps taken, and potentially improve
the case outcome.
Some of these changes have already been implemented in jurisdictions across the
country, such as requiring all SAKs to be submitted to a crime lab for analysis (See National
Center for Victims of Crime, 2014). For example, in 2010, Illinois was the first state to pass into

150

law (§ 725 ILCS 202/5) a policy requiring all law enforcement agencies to submit all DNA
evidence gathered from reported sex crimes to the crime lab, and for evidence to undergo testing
(National Center for Victims of Crime, 2014; Twohey, 2010). Other jurisdictions have followed
their example, including the city of Detroit in which passed Public Act 227 in 2014 requiring all
sexual assault kits to be submitted by law enforcement personnel to a crime lab for analysis
within 14 days of receipt, and tested within 90 days thereafter (Egan, 2014).
In order to leverage the number of investigatory blame legitimizing myths endorsed on a
case, policies could be implemented that supervising officers will not sign off on a case that
blames the victim for law enforcement personnel’s less-than-thorough investigation or low case
outcome. For example, if an investigating officer recorded that he/she tried to call the victim
twice and could not locate him/her, and then took no further action on the case, the supervising
officer would not sign off. The supervising officer would require that all investigative steps are
completed on the case and that the case is developed to its full potential to prepare it for referral
to the prosecutor’s office. In the current study, over 40% of cases had at least one investigatory
blame legitimizing myth endorsed. If supervising officers would not sign off on cases providing
these types of justification for no further action, it may increase the number of investigative steps
completed on a case (especially when paired with the prior policy recommendation), and
improve the case outcome.
It is important to note, however, that the number of investigatory blame legitimizing
myths impacted the case outcome through two different pathways. The number of investigatory
blame legitimizing myths endorsed predicted the number of investigative steps completed on a
case, which then predicted the case outcome. The number of investigatory blame legitimizing
myths also predicted the case outcome directly, and circumvented the number of investigative

151

steps completed on the case. In other words, for some cases, the number of investigative steps
completed on a case did not matter—as the law enforcement officer endorsed more investigatory
blame legitimizing myths, the likelihood that the case would have both an arrest and a referral
and continue on in the criminal justice system went down, regardless of how many investigative
steps were completed on the case. This means that a policy that requires all possible investigative
steps to be completed and/or documented rationale provided for any uncompleted steps may not
be sufficient to bring about the desired behavioral change. There also needs to be explicit
attention given to the investigatory blame legitimizing myth, deeming it as an unacceptable
reason to not refer the case onto the prosecutor or to not arrest an associated suspect. It is
unknown if jurisdictions have implemented these types of policies.
In order to implement the proposed policy changes describe herein regarding how cases
are to be investigated and supervised in any given jurisdiction (i.e., requiring all possible
investigative steps to be carried out, documentation of rationale for any steps not carried out, and
stipulations on necessary conditions for supervisor case review sign-off), it is necessary first to
determine the ‘model investigation.’ In other words, jurisdictions will need to delineate what
steps need to be completed in order to claim an investigation is comprehensive and thorough.
This will likely include, but is not limited to, canvassing the area surrounding the scene of the
crime; taking photographs at the scene of the crime; collecting statements from the victim(s),
witnesses, and suspects; and maintaining progress notes. Law enforcement personnel, including
patrol officers and supervising officers, could serve as subject matter experts (SMEs) in
delineating the full range of investigative steps that should be a part of the ‘model investigation.’
Involving police in a collaborative process may increase their sense of ownership and relevance

152

in this organizational change, making it more likely that they will embrace it and facilitate
implementation (Harnar & Preskill, 2007; Patton, 2008).
The purpose of the ‘model investigation’ is to serve as a standard of comparison for all
future investigations of sexual assault cases. Investigating detectives could use the ‘model
investigation’ as a checklist to ensure they have carried out all possible and necessary steps
during the course of their evaluation, and supervising officers could use it to check off all
possible investigative steps during case review. In order for the ‘model investigation’ to impact
future investigations, it must become a transparent model that is shared among all law
enforcement personnel in an agency and is tied to the agency’s vision (Buchanan et al., 2005).
Supervising officers and leaders within the organization play a unique role in that they have the
responsibility to model how the ‘model investigation’ should be used and ensure its translation to
reporting officers (e.g., see Zaccaro & Banks, 2001 for a discussion of the importance of
leadership in organizational change efforts.).
Several states across the country have already created ‘model investigations’ to guide
criminal justice system practitioners in their response to sexual assault. Massachusetts, New
Hampshire, New Mexico, and North Dakota are a few of the states that have developed what are
frequently referred to as ‘model policies’ (e.g., see North Dakota Council on Abused Women's
Services/Coalition Against Sexual Assault in North Dakota, 2011; Patrick, Murray, & Burke,
2009) The Michigan Domestic and Sexual Violence Prevention and Treatment Board
(MDSVPTB) has been leading similar efforts in Michigan. As recommended herein, MDSVPTB
has involved representatives from several law enforcement agencies and organizations across the
state, as well as advocates and prosecutors, to help support future implementation efforts (G.

153

Krieger, personal communication, November 10, 2014). MDSVPTB has also created one-page
checklists that patrol and investigating officers can use on the job.
Changing deep structures via training. In addition to leveraging change at the apparent
level with new policies, it is also essential to identify leverage points for change below the
surface, by targeting normative elements within the system (Foster-Fishman et al., 2007; Howard
et al., 2009). The current project provided insight into the underlying culture and values that
support the current criminal justice system response to sexual assault and, in doing so, identified
additional levers for change.
Circumstantial and characterological legitimizing myths align well with traditional rape
myths about what qualifies as ‘real’ rape and who can be raped, respectively (Edwards et al.
2011; Kettrey, 2013). The documentation of these myths in sexual assault case investigations
reflect police officer’s attitudes and beliefs about sexual assault and are therefore normative
elements within the criminal justice system. While circumstantial and characterological
legitimizing myths did not directly predict case outcomes, they did predict the number of
investigatory blame legitimizing myths endorsed and the number of investigative steps
completed on a case, which then affected the case outcome. Therefore, initiatives that target and
attempt to change the endorsement of circumstantial and characterological legitimizing myths in
sexual assault case investigations will challenge the normative elements of the criminal justice
system and indirectly affect case outcomes, thus providing an additional leverage point for
change. Indeed, many argue that the normative elements of a system must be targeted for any
change to be sustainable, as these elements are the root causes of system problems (FosterFishman et al., 2007).

154

Ongoing training that challenges traditional rape myths should be implemented. It is
important to reiterate that training is not the change, but rather plays a supporting role under the
assumption that changes in underlying values and beliefs will help to support the success and
sustainability of policy-level change (see Foster-Fishman et al., 2007; Howard et al., 2009 for a
discussion of the importance of attending to deep structure levers for change in addition to
apparent structures.). Implementing policy-level changes in how cases are to be investigated and
the process of case supervision may catalyze change in underlying values about what qualifies as
‘real’ rape and who can be raped. Training can provide a venue to support and facilitate these
value change processes, which will then help to make the policy-level changes more effective.
Training should attend to learning and unlearning processes. Whereas learning is the
acquisition of new knowledge, unlearning is the discarding of old knowledge or routines (Tsang
& Zahra, 2008). Law enforcement personnel need to learn the new ‘model investigation’ for their
jurisdiction, as well as up-to-date information on how to conduct victim-centered, offenderfocused investigations.22 For example, many jurisdictions across the country have sought
Rebecca Campbell’s training in the Neurobiology of Sexual Assault so as to equip their law
enforcement personnel with a greater understanding of victims’ responses to trauma (Campbell,
2012). This new information can help challenge underlying attitudes and beliefs regarding what
qualifies as ‘real’ rape and who can be raped. It is also essential that police unlearn processes and
practices that have proved to be misleading and unsuccessful so as to make way for their new
knowledge and training (Tsang & Zahra, 2008). For example, the current project found that law
enforcement personnel conduct less-than-thorough investigations and then blame victims for it.

22

Victim-centered responses prioritize the needs of the victims, support their efforts for justice and healing, and seek
to preserve their safety and dignity. Offender-focused responses draw attention to the actions, behaviors,
characteristics, and prior criminality of the perpetrator in order to hold him responsible for his crime (Wisconsin
Coalition Against Sexual Assault, 2009).

155

This response (i.e., conducting a less-than-thorough investigation and blaming victims for it) is
misleading, as it suggests that police are not invested in criminal justice, and unsuccessful, as
perpetrators are not held accountable for their actions. An approach that incorporates learning
and unlearning will help law enforcement personnel identify what behaviors and beliefs need to
cease (via unlearning) and what new strategies should fill the void (via learning).
Implications for Future Research
The current project highlights several interrelated lines of inquiry for future research so
as to advance our current understanding of the role of legitimizing myths in explaining the
criminal justice system response to sexual assault. First, it is essential to expand the sample
examined in the current project. The current project relied on police records corresponding to a
random sample of unsubmitted SAKs in Detroit. This sample was selected intentionally as it
represented cases that had somehow been subjected to the less-than-adequate criminal justice
system response. It is necessary to expand the sample to include cases corresponding to
unsubmitted SAKs in other jurisdictions, as well as to other samples that include cases that have
progressed through the criminal justice system.
It is essential to expand the sample to include cases corresponding to unsubmitted SAKs
in other jurisdictions. This will help to determine if the documentation of legitimizing myths in
sexual assault case investigations corresponding to unsubmitted SAKs are unique to Detroit, or if
legitimizing myths also are observable in sexual assault case investigations corresponding to
unsubmitted SAKs in other jurisdictions. In short, this will help to answer if this is just
happening to cases with unsubmitted SAKs in Detroit, or if it is happening to cases with
unsubmitted SAKs across the country. Given the documented struggles Detroit has had with
resources and infrastructure, it is possible that they blamed victims for investigational progress,

156

rather than their own frustrating organizational culture of scarcity (See Campbell, 2014
forthcoming for an in-depth discussion of the context in Detroit). It is also essential to expand the
sample to include cases that progressed through the criminal justice system. While some of the
cases in the current study did transition to prosecution (27% of cases had a referral and arrest),
the vast majority of cases did not. This will help to determine if the documentation of
legitimizing myths in sexual assault case investigations are limited to cases that have been
subjected to the less-than-thorough criminal justice system response (e.g., a SAK not submitted),
or if they also appear in records for cases that have progressed to prosecution, and perhaps even
resulted in a conviction of the offender. This will help to answer if this is just happening to cases
with unsubmitted SAKs, or if this is happening to most cases within the criminal justice system.
Expanding the sample to include cases corresponding to unsubmitted SAKs in other jurisdictions
and cases that have progressed further through the criminal justice system will illuminate the
pervasiveness of the problem and can inform future change strategies.
Second, it is important to expand the methodological approach utilized in the current
study to include more than just archival methods. The current project relied on sexual assault
police records to document observations of legitimizing myth endorsement in the police response
to sexual assault. The written police files provide only one perspective—that of the law
enforcement officer/investigator, and that which they were willing to record in writing. This
approach does not allow us to understand what police were thinking in responding to these
sexual assault cases. Furthermore, this approach does not allow us to understand perspectives of
other individuals involved in the criminal justice system response to sexual assault that may have
influenced or been affected by law enforcement personnel’s endorsement of different
legitimizing myths; for example, the perspective of the prosecutor, advocate, medical provider,

157

and the victim. Future research, therefore, should complement the current project by relying on
alternative research strategies, such as interviews, to illuminate this process from additional
perspectives (e.g., the victim perspective).
Third, it is important to expand the domain of investigation beyond the criminal justice
system. The current project examined the endorsement of legitimizing myths by law enforcement
personnel in the criminal justice system. A portion of the legitimizing myths endorsed related to
what qualifies as ‘real’ rape (i.e., circumstantial legitimizing myths) and who can be raped (i.e.,
characterological legitimizing myths). These legitimizing myths align well with more traditional
rape myths and have been found to be endorsed by law enforcement personnel, as well as the
general population (e.g., see Edwards et al., 2011; Lonsway & Fitzgerald, 1995; Page, 2010).
However, the current project also identified a third type of legitimizing myth that blamed the
victim for police conducting a less-than-thorough investigation. This type of legitimizing myth
had not been previously documented in the literature and therefore it is unknown to what extent
it is endorsed by the general population.
It is possible that this legitimizing myth is only used in the context of sexual assault
investigations so as to provide a reason for the case not going forward, as was argued in the
current project. Alternatively, this type of myth could also be endorsed in the general population
to serve a different purpose: as another means to minimize the assault by defining what qualifies
as ‘real’ rape. That is, if the victim is not willing to participate in the investigation, it wasn’t
really rape. Examining the extent to which these types of myths appear in the general population
would help to understand how this myth is being used; if it is used only by legal personnel in the
domain of sexual assault case investigations to explain why a case does not progress or, if it is
also used by the general population as another means to minimize and discount the rape.

158

Finally, future research should also focus on evaluating if policy changes intended to
improve the criminal justice system response to sexual assault are having their intended effect.
For example, policies that require all SAKs to be submitted to the crime lab for analysis within a
set period of time may result in more SAKs being tested and more cases proceeding forward in
the criminal justice system process. Alternatively, it is also possible that these types of policies
will only relocate the problem. Instead of having a pile of unsubmitted kits, ‘submit all SAKs’
policies may result in a pile of submitted, but yet to be tested kits, if these policies are
implemented without necessary resources for crime labs to receive and process the new influx of
SAKs. If the crime lab is able to keep up with the increase in submitted SAKs, the problem may
then become a pile of submitted, tested, and yet to be prosecuted SAKs, if these policies are
implemented without necessary resources for police to conduct thorough investigations and for
prosecution to ensue. If adequate resources, infrastructure, and training are not provided to
stakeholders affected by these new policy changes, they may be ineffective. Therefore, particular
emphasis should be given to examining what contextual factors are necessary for policy and
practice change success across jurisdictions.
Conclusion
The criminal justice system response to sexual assault needs our ongoing attention if we
are to improve it, especially in light of the current findings that victims are subjected to tertiary
victimization in this system. Fortunately, the results of the current project suggest specific
strategies that may be implemented to change how these cases move through the criminal justice
system, and potentially improve a victim’s opportunity for criminal justice. Recent national
attention and significant legislative changes to the criminal justice system response (see National
Center for Victims of Crime, 2014) are promising. However, it is essential that researchers

159

partner with communities in order to evaluate these new policies and practices to see if they are
having their intended effects and what supports are necessary for their success. Through
researcher-practitioner partnerships, we may be able to have our greatest impact by using science
to inform practice and by investing in practice-informed research.

160

APPENDICES

161

APPENDIX A:

CODING INSTRUCTIONS

162

DESCRIPTION OF THIS DOCUMENT
This document provides instructions for coding the 400 Project Evaluation police reports. It
includes instructions for accessing the raw data, coding the raw data, and saving the coded
data. Keep this with while you are coding so you can quickly reference it as needed. This
document has also been stored on dropbox.

163

ACCESSING THE RAW DATA
To access the raw data, the project director will need to log you into a password protected
research team computer. You will only be able (and permitted) to access the raw data when
you are in research team space on a password protected research team computer. If you need
to step away from your computer at any time, you will need to log off of the computer and
retrieve the project director to log you back in. THE LOGGED IN COMPUTER SHOULD NOT BE
LEFT UNATTENDED AT ANY TIME.
Below are steps for accessing the raw data:
1.
2.
3.
4.
5.

The project director will log you into a password protected computer.
Open the “dropbox.”
Within “dropbox,” open “400 PROJECT EVALUATION DATA
Within “400 PROJECT EVALUATION DATA,” open “Raw Police Records.”
In this folder, you will find one file for each case included in this project. File names
consist of the victim names and case numbers.
6. In each file, you will find all of the raw police reports and records—these are what you
will code for this project.

Remember:






The logged in computer should not be left unattended at any time.
If you have to leave your computer for any reason (e.g., to use the bathroom, to step
out to make a call, to leave for the day), you must log off of the computer. You will then
need to get the project director to log you back into the computer.
Do not delete, move, or otherwise remove any files from the “raw police records”
folder.
Do not save any changes you may make (accidentally or intentionally) to any files in the
“raw police records” folder.

164

ACCESSING THE DATA SPREADSHEETS
You will record the information you code from the raw data into the data spreadsheets. Each
member of the research team will have their own data spreadsheet that they will use to record
the coded data. You will only record your codes into your data spreadsheet. The project
director will be responsible for combining the information recorded across data spreadsheets
into a master spreadsheet.
Below are steps for accessing the data spreadsheets:
1.
2.
3.
4.
5.

The project director will log you into a password protected computer.
Open the “dropbox.”
Within “dropbox,” open “400 PROJECT EVALUATION DATA.”
Within “400 PROJECT EVALUATION DATA,” open “Data Spreadsheets.”
Within “Data Spreadsheets,” open the folder that has your name in it
(“Spreadsheets_YOURNAME”). You should only regularly access your folder.
6. Within “Spreadsheets_YOURNAME,” open the file, “Spreadsheet_YOURNAME_original.”
7. Your spreadsheet is password protected. When you open the document, it will ask for a
password. The password is “DISSERTATION”
8. The cases that have been assigned to you for coding will be listed in your spreadsheet.
9. The spreadsheet will include columns for each of the variables you are to code, in
addition to identifying information for each case. The variable categories include:
a. Victim demographics
b. Suspect demographics
c. Investigative steps
d. Legitimizing myths
e. Date coded
10. When you code (detailed in next section), you will enter information into every column
in the spreadsheet.
11. After you have made changes/updates to your spreadsheet, always “save as” with the
date. For example, “Spreadsheet_JESS_2-13-2014.”

165

Remember:








The logged in computer should not be left unattended at any time.
If you have to leave your computer for any reason (e.g., to use the bathroom, to step
out to make a call, to leave for the day), you must log off of the computer. You will then
need to get the project director to log you back into the computer.
Only work in your assigned spreadsheet.
When you make updates/changes to your spreadsheet, “save as” with the most recent
date. For example, “Spreadsheet_JESS_2-13-2014.” The project director will regularly
archive previous files.
You should not make any changes to the “Spreadsheet_YOURNAME_original.” Only the
project director will make changes to this file.

166

CODING THE RAW DATA
To code the raw data, you will need to access the raw data and data spreadsheets at the same
time; you will be coding information out of the raw data and inputting it into the data
spreadsheets. This is where you will use what you know about how to access the raw data and
the data spreadsheets (previous sections).This is what you were brought onto the team to do.
Take your time. Here you will find the overall instructions of how to do this. For the specifics on
the items you will be coding, go to the “Codebook.”
Below are steps for coding the raw data:
1. Open your data spreadsheet (see directions above). The first time you do this, open the
original spreadsheet: “Spreadsheet_YOURNAME_original.” For subsequent coding
sessions, open the most recent spreadsheet: “Spreadsheet_YOURNAME_MOST RECENT
DATE.”
2. Save a new copy of the data spreadsheet to work in. Click on “file,” and select, “save
as.” Rename the data spreadsheet with today’s date. For example,
“Spreadsheet_YOURNAME_2-13-2014.” This file should be saved in your spreadsheet
folder, “Spreadsheet_YOURNAME” (Your spreadsheet folder is in the “Data
Spreadsheets” folder).
3. Identify the next case to code. The first time you do this, it will be the first case in your
data spreadsheet. For subsequent coding sessions, pick up where you left off. For
example, if the last case you coded was on line 3, the next case to code will be on line 4.
4. Locate the raw data for the case you are to code using the case’s identifying
information. The raw data will be in the “raw police records” (see directions above).
5. Verify that the information in the police file matches the identifying information in the
data spreadsheet. If this information does not match, notify the project director.
6. Close the raw data file.
7. In the “Raw Police Records” folder, right click the raw data file and click, “copy.”
8. Go into the “Coded Police Records” folder, right click, and paste a copy of the raw data
file into this folder.
9. Rename the raw data file with the Case ID number (ranging from 101-352), the word
“CODED,” and your initials. For example, if the project director coded case 101, it would
be named “101_CODED_JS”
10. Using this copy of the file, code the police record for the information detailed in the
“codebook.” You should view every record in the case file to ensure that you did not
miss any relevant information. You should start with the “Crisnet Report” as it has the
initial report in it.
11. Highlight any information you record from the raw police record using the highlight tool
in adobe pdf (if, for some reason, the highlight tool is not working, insert comments to
167

mark the information you recorded). For example, you will record the age of the victim
for every case file you view. You should highlight on the police record where you
retrieved this information.
12. Record all information in the appropriate column in the data spreadsheet.
13. Record the date that you coded the file in the data spreadsheet.
14. After you have filled in all columns for a case in the data spreadsheet, click save in the
data spreadsheet. Save at least after each case is completely coded; you may save in the
data spreadsheet more frequently if you like.
15. You need to also save the highlights/comments you made in the raw police record (the
second copy of it you made with steps 7-9 above). In the pdf, click “file” and select
“save.”
16. Continue working and coding files until your coding time comes to an end.
17. Be aware of the time. Do not start coding a new case unless you have time to finish it.
18. At the end of your coding session, be sure that you have saved all of your work in the
appropriate locations with the appropriate labels

Remember:












The logged in computer should not be left unattended at any time.
If you have to leave your computer for any reason (e.g., to use the bathroom, to step
out to make a call, to leave for the day), you must log off of the computer. You will then
need to get the project director to log you back into the computer.
Before you start coding a raw police record, “save as” per the labeling scheme and
location noted above. The “raw police records” folder should remain unchanged for the
entirety of project coding. Be sure to save your highlights/codes in the new copy of the
file you created.
Only work in your assigned spreadsheet.
When you make updates/changes to your spreadsheet, “save as” with the most recent
date. For example, “Spreadsheet_JESS_2-13-2014.” The project director will regularly
archive previous files.
Be sure to save your work in the data spreadsheet after each coded case (and more
frequently if you like).
Do not start coding a case file if you don’t have time to finish it.
Keep a list of questions that arise while you are coding and ask the project director for
clarifications. The goal is to code the police files as accurately and consistently as
possible. Do not be afraid to ask questions!

168





The project director will be monitoring the coding process. There will be several cases
that are coded by more than one coder and the inter-coder reliability (kappa) will be
assessed. We need to maintain 80% agreement, so take your time.
Information coded from the raw police files will only be recorded in the data
spreadsheets. You are not permitted (and have signed confidentiality agreements in
accordance with this) to record and/or remove any information from the raw police
files. Failure to comply with this agreement will result in a failing grade for the semester.

169

OTHER IMPORTANT NOTES
It is important to note that these case files were reopened by a collaborative team in 2009.
When coding if specific investigative steps or the endorsement of legitimizing myths occurred,
be sure that you only code information from the original file (i.e., documented dates post-2009
should not be included in the coding or subsequent analyses).
It is also important to note that the “Checklist” included in each file may indicate that a certain
investigatory step was completed (i.e., the checkbox is checked). This is not sufficient
documentation that the investigatory step was completed. You must find additional
documentation of the investigatory step to code it as a “1.” Additional detail is provided with
each variable in the table below.
Also note, anything in the file dated after 8/17/2009 should NOT be coded (the kits were
“discovered” on 8/17/2009).

170

APPENDIX B:

CODING SCHEME USED FOR DIRECTED CONTENT ANALYSIS OF LEGITIMIZING
MYTHS (STUDY 1)

171

Table 10. Directed content analysis codes
Legitimizing Myths
Victim didn’t fight
0=No mention of if the victim did or did not fight back, screamed, or
back
tried to run away
1=Records note that the victim did not fight back, scream, or try to run
away
2=Records note that the victim did fight back, scream, or try to run away
NOTE: Code the “Crisnet Report” and “Progress Notes” for legitimizing
myths. Additionally, review all other documents in the file (e.g.,
“Checklist,” “Witness Statement”) for notes taken by law enforcement
personnel. Anything written in script like format, first person language
from the perspective of the victim (or other witness) (i.e., in the
“Witness Statement” document) should NOT be coded as a legitimizing
myth.

Victim exaggerates
or lies

Document the source in the “source” column.
0=No mention of if the victim is exaggerating, lying, telling the truth, or
questioning of the victim’s story (e.g., if it “lines up” or seems plausible)
1=Records note that the victim is exaggerating or lying about the rape
itself or the impact/effects of the rape, or questions the victim’s story
(e.g., if it “lines up” or seems plausible)
2=Records note that the victim is telling the truth/not lying about the
rape itself of the impact/effects of the rape, or questioning the victim’s
story (i.e., if it “lines up” or seems plausible)
NOTE: Code the “Crisnet Report” and “Progress Notes” for legitimizing
myths. Additionally, review all other documents in the file (e.g.,
“Checklist,” “Witness Statement”) for notes taken by law enforcement
personnel. Anything written in script like format, first person language
from the perspective of the victim (or other witness) (i.e., in the
“Witness Statement” document) should NOT be coded as a legitimizing
myth.
Document the source in the “source” column.

172

Table 10 (cont’d)
Victim consented

Legitimizing Myths (cont’d)
0=No mention of the victim’s consent or non-consent
1=Records note that the victim consented to engage in consensual
sexual activity with perpetrator(s) at some point during the assault (and
perhaps changed mind) or on previous occasions
2=Records note that the victim expressed non-consent to all sexual acts
NOTE: Code the “Crisnet Report” and “Progress Notes” for legitimizing
myths. Additionally, review all other documents in the file (e.g.,
“Checklist,” “Witness Statement”) for notes taken by law enforcement
personnel. Anything written in script like format, first person language
from the perspective of the victim (or other witness) (i.e., in the
“Witness Statement” document) should NOT be coded as a legitimizing
myth.

No bruises/marks

Document the source in the “source” column.
0=No mention of if the victim did or did not have bruises, marks,
injuries, or appeared disheveled
1=Records note that the victim did not have any bruises, marks, injuries,
or appear disheveled
2=Records note that the victim had bruises, marks, injuries, or appeared
disheveled
NOTE: Code the “Crisnet Report” and “Progress Notes” for legitimizing
myths. Additionally, review all other documents in the file (e.g.,
“Checklist,” “Witness Statement”) for notes taken by law enforcement
personnel. Anything written in script like format, first person language
from the perspective of the victim (or other witness) (i.e., in the
“Witness Statement” document) should NOT be coded as a legitimizing
myth.
Document the source in the “source” column.

173

Table 10 (cont’d)
Not upset enough

Legitimizing Myths (cont’d)
0=No mention of the victim’s emotional demeanor
1=Records note that the victim did not appear upset or distraught,
seemed distracted, or exhibited emotions that would be unexpected
given situation/circumstance
2=Records note that the victim did appear upset or distraught, or
exhibited emotions that would be expected given
situation/circumstance
NOTE: Code the “Crisnet Report” and “Progress Notes” for legitimizing
myths. Additionally, review all other documents in the file (e.g.,
“Checklist,” “Witness Statement”) for notes taken by law enforcement
personnel. Anything written in script like format, first person language
from the perspective of the victim (or other witness) (i.e., in the
“Witness Statement” document) should NOT be coded as a legitimizing
myth.

Victim is a sex
worker

Document the source in the source column.
0=No mention of if the victim is a sex worker
1=Records note that the victim is a sex worker
2=Records note that the victim is not a sex worker (uncommon)
NOTE: Code the “Crisnet Report” and “Progress Notes” for legitimizing
myths. Additionally, review all other documents in the file (e.g.,
“Checklist,” “Witness Statement”) for notes taken by law enforcement
personnel. Anything written in script like format, first person language
from the perspective of the victim (or other witness) (i.e., in the
“Witness Statement” document) should NOT be coded as a legitimizing
myth.
Document the source in the “source” column.

174

Table 10 (cont’d)
Victim is drunk/high

Legitimizing Myths (cont’d)
0=No mention of if the victim is drunk/high when interacting with law
enforcement personnel or is a regular drug user
1=Records note that the victim is drunk/high when interacting with law
enforcement or is a regular drug user
2=Records note that the victim is not drunk/high when interacting with
law enforcement or is not a regular drug user (uncommon)
NOTE: Code the “Crisnet Report” and “Progress Notes” for legitimizing
myths. Additionally, review all other documents in the file (e.g.,
“Checklist,” “Witness Statement”) for notes taken by law enforcement
personnel. Anything written in script like format, first person language
from the perspective of the victim (or other witness) (i.e., in the
“Witness Statement” document) should NOT be coded as a legitimizing
myth.

Document the source in the “source” column.

175

APPENDIX C:

CODEBOOK FOR INVESTIGATIVE STEPS, CASE OUTCOMES, AND SOCIAL IDENTITY
FACTORS (STUDY 2)

176

Table 11. Investigative steps, case outcomes, and social identity factors codebook
Victim Demographics (All info can usually be found in the Crisnet/initial report)
Victim’s first name
Enter victim’s first name
Victim’s last name
Enter victim’s last name
Victim Sex
0=Female
1=Male
999=Unknown
Victim Age
Enter victim’s age
999=Unknown
Victim Race
0=African-American
1=Arab American/Chaldean
2=Asian American/Pacific Islander
3=Caucasian
4=Hispanic/Latino
5=Multi-Racial
999=Unknown
Suspect Demographics (All info can usually be found in Cristnet/initial report)
Perp Sex
0=Female
1=Male
2=Male and Female
999=Unknown
Perp Age
Enter perpetrator’s age
999=Unknown

Perp Race

Perp Notes

NOTE: If there are multiple perpetrators, enter the age of the first
perpetrator in this column and the age of additional perpetrators in the
“Perp Notes” column.
0=African-American
1=Arab American/Chaldean
2=Asian American/Pacific Islander
3=Caucasian
4=Hispanic/Latino
5=Multi-Racial
999=Unknown
NOTE: If there are multiple perpetrators, enter the race of the first
perpetrator in this column and the race of additional perpetrators in the
“Perp Notes” column.
If the assault involved multiple perpetrators, enter the age and race of
the additional perpetrators in this column.
0=no additional perpetrators

177

Table 11 (cont’d)
Final Case Outcome

Progress Notes

Photos at Scene

Victim Statement

Investigative steps
Enter the final case outcome. Potential case outcomes include (but are
not limited to): MI-UTEEC, unfounded, to locate, referred to the
prosecutor, warrant denied, warrant issued, closed on information and
belief, and CRTP
999=Missing
NOTE: Read all documents in the file as this information may be
recorded in different places. It frequently (though not always) appears
in the “Crisnet Report,” “Progress note(s),” or “Checklist” folder(s). It
may be labeled with “case outcome,” “case disposition,” “dictated,”
“disp,” “dispo,” or “disposition,” among other things. If different case
outcomes appear on different documents, enter the most recent case
outcome. The “checklist” outcome is most frequently the most recent
outcome, though it is not dated. If the outcome on the “checklist” does
not match the most recent outcome in the other files, note this.
0=No progress notes in the file (or the “progress notes” document
appears, but there are no notes written on it)
1=At least one progress note in the file
NOTE: Progress notes appear in the “Progress Note(s)” folder. Read all
documents in the file in case they were misfiled.
0=No crime scene photos in the file
1=Crime scene photos in the file
NOTE: Photos at scene appear in the “Crime Scene Photo(s)” folder.
Read all documents in the file in case they were misfiled.
0=No statement from victim in the file
1=Statement from victim in the file
NOTE: The statement from the victim may appear in the “Witness
Statement(s)” folder or written in a transcript format in the “Crisnet
Report” (i.e., it MUST be written in transcript format). Read all
documents in the file in case the victim statement was recorded
elsewhere or these documents were misfiled.

178

Table 11 (cont’d)
Witness Statement

Canvass Sheets

Lab Request Form

Lab Report

Investigative steps (cont’d)
0=No witness statement(s) in the file
1=Witness statement(s) in the file
NOTE: This refers to a statement from a witness OTHER than the victim.
The statement may appear in the “Witness Statement(s)” folder, be
written in a transcript format in the “Crisnet Report,” or there may just
be notes in the “Crisnet Report” or “Progress Notes” that a witness
provided information (i.e., it does not need to be written in transcript
format). Read all documents in the file in case the witness statement(s)
was recorded elsewhere or these documents were misfiled. The
“Checklist” may indicate that a witness statement was in the file, but
this may be referring to the victim statement.
0=No canvass sheets in the file
1=Canvass sheets in the file
NOTE: Canvass sheets appear in the “Canvass” folder. Read all
documents in the file in case they were misfiled.
0=No lab request form for the rape kit to be tested in the file
1=Lab request form for the rape kit to be tested in the file
NOTE: The “Request for Laboratory Service” appears in the “Lab
Report(s)” folder. It may also be called the “Detroit Police Forensic
Services Division Serology/Trace Evidence Unit.” Read all documents in
the file in case it was misfiled.
0=No lab report for the rape kit in the file
1=Lab report for the rape kit in the file (the lab report may just say that
they did not do any testing on the kit, but this still counts as a report
NOTE: With this variable, we are trying to document if the kit did indeed
go to the lab. It is possible that the police officer did not complete a lab
request form, but still submitted the kit to the crime lab. Documents
produced from the crime lab prior to August 17, 2009 qualify here and,
if they appear, the variable should be coded as a 1. Read all documents
in the file in case it was misfiled.

179

Table 11 (cont’d)
Medical Release
Form

Suspect Lineup

Evidence techs on
scene

Suspect Brought in
for Interview

Arrest Suspect

Investigative steps (cont’d)
0=No completed medical release form in the file (i.e., there is no form at
all, or it not signed off by the patient).
1=Completed medical release (i.e., signed by patient) form in the file
NOTE: Completed medical release form appears in the “Medical
Documents” folder. The release form may be a stand-alone document,
or it may be a check box/initial space on a document for medical care
consent. Read all documents in the file in case it was misfiled.
0=No documentation of a suspect lineup in the file
1=Documentation of a suspect lineup in the file
NOTE: This usually appears in the “Showup and or lineups” folder. Read
all documents in the file in case it was misfiled.
0=No documentation that evidence technician(s) arrived on scene in the
file
1=Documentation that evidence technician(s) arrived on scene in the file
NOTE: This is usually documented in the “Evidence Technician Report” in
the “Evidence Tech Report(s)” folder or in the “Scene Investigation”
folder. Read all documents in the file in case they were misfiled.
0=No documentation of a suspect being brought in for an interview in
the file
1=Documentation of a suspect brought in for an interview in the file
NOTE: This may appear in the “Crisnet Report” or “Progress Note(s).”
This is most frequently documented with the “Suspect Interrogation”
form. Sometimes, the a suspect is brought in for an interview, but
refuses to provide any statement. This should still be coded as a “1.”
Read all documents in the file in case it appears somewhere else.
0=No documentation of a suspect arrest in the file
1=Documentation of a suspect arrest in the file
NOTE: This may appear in the “Crisnet Report” or “Progress Note(s).”
Read all documents in the file in case it appears somewhere else.

180

Table 11 (cont’d)
Referral to
Prosecutor

Other

Investigative steps (cont’d)
0=No documentation that the case was referred to the prosecutor in file
1=Documentation that the case was referred to the prosecutor in file
NOTE: This is most always documented with the “investigator’s report.”
This may also appear in the “Crisnet Report,” “Progress Note(s),” or
“Checklist.” Read all documents in the file in case it appears somewhere
else. Referral to prosecutor may be listed as the final case
outcome/disposition. Also, if the case disposition is warrant denied,
warrant issued, or some other outcome emanating from the court
system, the case was indeed referred to the prosecutor.
Enter any additional investigative steps taken by law enforcement
personnel that were not recorded/represented in a previous variable.
(e.g., car impounded for evidence; blood sample collected from suspect;
search warrant issued, etc.)
0=No additional investigative steps taken by law enforcement

181

APPENDIX D:

FINAL CODES (STUDY 2)

182

Table 12. Final codes
Victim is lying

Victim is not injured

Victim consented

Victim is not upset

Victim didn’t act
like a victim
afterwards

Total Number of
Circumstantial
Legitimizing Myths

Circumstantial Legitimizing Myths
0 = Records did not note that the victim is exaggerating, lying, and does
not call into question the victim’s story (e.g., if it “lines up” or seems
plausible)
1 = Records noted that the victim is exaggerating, lying, or calls into
question the victim’s story (e.g., if it “lines up” or seems plausible)
0 = Records did not note that the victim did not have bruises, marks,
injuries, or appeared disheveled
1 = Records noted that the victim did not have bruises, marks, injuries, or
appeared disheveled
0 = Records did not make any mention of consent or noted that the victim
did not consent to the sexual activity
1 = Records noted that the victim consented to part or all of the sexual
activity with the perpetrator on this occasion, or on previous occasions
0 = Records did not make any mention of the victim’s emotional
demeanor, or noted that the victim was upset, distraught, or exhibited
emotions that would be expected given the circumstance
1 = Records noted that the victim did not appear upset or distraught,
seemed distracted, or exhibited emotions that would be unexpected given
the circumstance
0 = Records did not make mention of how the victim’s actions following
the assault were unexpected given the circumstance
1 = Records noted that the victim’s actions following the assault were
unexpected given the circumstance
Enter the total number of circumstantial legitimizing myths

Victim is a regular
drug user

Characterological Legitimizing Myths
0 = Records did not note the victim is drunk/high when interacting with
law enforcement personnel or is a regular drug user

Victim is a sex
worker

1= Records noted that the victim is drunk/high when interacting with law
enforcement personnel or is a regular drug user
0 = Records did not note that the victim is a sex worker (e.g., a prostitute,
a “deal gone bad,” “on the street”, etc.)
1= Records noted that the victim is a sex worker (e.g., a prostitute, a
“deal gone bad,” “on the street”, etc.)

183

Table 12 (cont’d)
Victim has “done
this before”

Victim is “mental”

Victim is
promiscuous
Victim is not
credible

Total Number of
Characterological
Legitimizing Myths

Characterological Legitimizing Myths (cont’d)
0 = Records did not note that the victim has previously “done this
before.” “This” refers to reporting a rape (and in some cases not
participating in the ensuing investigation), being raped, and/or having a
rape kit done
1= Records noted that the victim has “done this before.” “This” refers to
reporting a rape (and in some cases not participating in the ensuing
investigation), being raped, and/or having a rape kit done
0 = Records did not note that the victim is “mental” or has a mental
illness
1 = Records noted that the victim is “mental” or has a mental illness
0 = Records did not note that victim is promiscuous
1 = Records noted that the victim is promiscuous
0 = Records did not note that victim is not credible or has a history of
lying (separate from lying about the specific assault reported)
1 = Records noted that the victim is not credible or has a history of lying
(separate from lying about the specific assault reported)
Enter the total number of charcterological legitimizing myths

Victim is
uncooperative

Investigatory Blame Legitimizing Myths
0 = Records did not note that the victim was uncooperative, hostile, or
intentionally withholding information

Victim doesn’t have
enough information

1= Recorded noted that the victim was uncooperative, hostile, or
intentionally withholding information
0 = Records did not note that victim did not have or could not remember
enough information

Victim has no
phone/address for
contact

Victim or case is
weak

1 = Records noted that the victim did not have or could not remember
enough information (e.g., did not know the name of her rapist)
0 = Records did not note a problem in contacting the victim
1 = Records noted that law enforcement personnel did not have a
working phone number, address, or were otherwise unable to contact the
victim
0 = Records did not record that the victim or case was weak or
incompetent
1 = Recorded noted that the victim or case was weak or incompetent

184

Table 12 (cont’d)
Victim is
uncooperative

Total Number of
Investigatory Blame
Legitimizing Myths
Case Outcome

Investigatory Blame Legitimizing Myths (cont’d)
0 = Records did not note that the victim was uncooperative, hostile, or
intentionally withholding information
1= Recorded noted that the victim was uncooperative, hostile, or
intentionally withholding information
Enter the total number of investigatory blame legitimizing myths

Case outcome
1 = arrest and referral
2 = no arrest and referral
3 = arrest and no referral
4* = no arrest and no referral

*Reference Category
Investigative steps
Evidence technicians 0 = No documentation of evidence technicians arriving at the crime scene
at scene
1 = Documentation of evidence technicians arriving at the crime scene as
indicated by an “evidence tech report,” in “scene investigation” notes, or
as indicated in the initial report
Photographs at scene 0 = No photographs from the scene of the crime

Canvassed

1 = Photographs from the scene of the crime
0 = No completed canvass sheets or other documentation in the initial
report that investigators/police canvassed the area

Progress Notes

1 = Completed canvass sheets or other documentation in the initial report
that investigators/police canvassed the area
0 = No progress notes appear

Victim statement

1 = At least one progress note was recorded in the file as recorded on the
“progress notes” document
0 = No victim statement (in a transcript format)

Witness statement

1 = Statement from the victim (in a transcript format)
0 = No witness statement (other than the victim)

SAK to lab

1 = Statement from the witness (other than the victim)
0 = No lab request form or lab report for processing of the SAK
1 = Lab request form or lab report for processing the SAK

185

Table 12 (cont’d)
Investigative steps
Medical release form 0 = No completed medical release form (i.e., signed by victim or
guardian)
1 = Completed medical release form (i.e., signed by victim or guardian)
Suspect lineup
0 = No documentation of a suspect lineup

Suspect interview

Total Number of
Investigative Steps

1 = Documentation of a suspect lineup, as indicated by the “showup
and/or lineup” file or written in the case notes (i.e., initial report of
progress notes)
0 = No documentation of a suspect interview (including if the suspect
refused to provide an interview)
1 = Documentation of a suspect interview (or the suspect refusing to
provide an interview), as indicated by the “interrogation record” or in the
case notes (i.e., initial report or progress notes)
Enter the total number investigative steps completed
Social Identity Variables

Victim Sex
Perpetrator Sex

Victim Race
Perpetrator Race

Victim Age

Victim and
Perpetrator Age
Difference
Multiple Perpetrators

0 = Female
1 = Male
0 = Male
1 = Female
This is recorded for the first perpetrator listed in the file.
0 = Non-White
1 = White
0 = Non-White
1 = White
This is recorded for the first perpetrator in the file.
1 = 0-16 years old or younger
2* = 17-25 years old
3 = 26+ years old
*Reference Category
1* = Perpetrator is within 5 years of the victim’s age
2 = Perpetrator is 5 years or more older than the victim
3 = Perpetrator is 5 years of more younger than the victim
0 = One perpetrator
1 = Two or more perpetrators

186

REFERENCES

187

REFERENCES

Adams, M., Bell, L. A., & Griffin, P. (2007). Teaching for diversity and social justice. New
York: Routledge.
Addington, L. A., & Rennison, C. M. (2008). Rape co-occurrence: Do additional crimes affect
victim reporting and police clearance of rape? Journal of Quantitative Criminology,
24(2), 205-226.
Amstadter, A. B., McCauley, J., Ruggiero, K. J., Resnick, H. S., & Kilpatrick, D. G. (2010).
Self-rated health in relation to rape and mental health disorders in a national sample of
women. American Orthopsychiatric Association, 81, 202-210.
Archambault, J., & Lonsway, K. A. (May 2012). Clearance methods for sexual assault cases End
Violence Against Women International (EVAWI).
Bachman, R. (1998). The factors related to rape reporting behavior and arrest: New evidence
from the National Crime Victimization Survey. Criminal Justice & Behavior, 25, 8-29.
Bakeman, R., & Gottman, J. M. (1986). Observing interaction: An introduction to sequential
analysis. New York: Cambridge University Press.
Barrett, P. (2007). Structural equation modelling: Adjudging model fit. Personality and
Individual differences, 42(5), 815-824.
Basile, K. C., Chen, J., Black, M. C., & Saltzman, L. E. (2007). Prevalence and Characteristics
of Sexual Violence Victimization Among US Adults, 20012003. Violence and victims,
22(4), 437-448.
Basile, K. C., & Smith, S. G. (2011). Sexual violence victimization of women: Prevalence,
characteristics, and the role of public health and prevention. American journal of Lifestyle
Medicine, 5(407-417).
Black, D. (1978). Dispute settlement by the police. New Haven, CT: Yale University Press.

188

Black, M. C., Basile, K. C., Breiding, M. J., Smith, S. G., Walters, M. L., Merrick, M. T., . . .
Stevens, M. R. (2011). The National Intimate Partner and Sexual Violence Survey
(NISVS): 2010 summary report. Atlanta, GA: National Center for Injury Prevention and
Control, Centers for Disease Control and Prevention.
Bollen, K. A. (1989). Structural equations with latent variables. New York: John Wiley & Sons,
Inc.
Bollen, K. A., Harden, J. J., Ray, S., & Zavisca, J. (2014). BIC and Alternative Bayesian
Information Criteria in the Selection of Structural Equation Models. Structural Equation
Modeling: A Multidisciplinary Journal, 21(1), 1-19.
Bonomi, A. E., Anderson, M. L., Rivara, F. P., & Thompson, R. S. (2007). Health outcomes in
women with physical and sexual intimate partner violence exposure. Journal of Women's
Health, 16(987-997).
Bosick, S. J., Rennison, C. M., Gover, A. R., & Dodge, M. (2012). Reporting violence to the
police: Predictors through the life course. Journal of Criminal Justice, 40(6), 441-451.
Bouffard, J. (2000). Predicting type of sexual assault case closure from victim, suspect, and case
characteristics. Journal of Criminal Justice, 28, 527-542.
Breitenbecher, K. H. (2006). The relationships among self-blame, psychological distresst, and
sexual victimization. Journal of Interpersonal Violence, 21, 597-611.
Brown, J., & King, J. (1998). Gender differences in police officers attitudes towards rape:
Results of an exploratory study. Psychology, Crime, & Law, 4, 265-279.
Bryant-Davis, T., Chung, H., & Tillman, S. (2009). From the margins to the center: Ethnic
minority women and the mental health effects of sexual assault. Trauma, Violence, &
Abuse, 10, 330-357.
Bryden, D. P., & Lengnick, S. (1997). Rape in the criminal justice system. Journal of Criminal
Law and Criminology, 1194-1384.
Buchanan, D., Fitzgerald, L. F., Ketley, D., Gollop, R., Jones, J. L., Lamont, S. S., . . . Whitby,
E. (2005). No going back: A review of the literature on sustaining organizational change.
International Journal of Management Reviews, 7(3), 189-205.

189

Burt, M. R. (1980). Cultural myths and supports for rape. Journal of personality and social
psychology, 38, 217-230.
Byrne, B. M. (2012). Structural equation modeling with Mplus: Basic concepts, applications,
and programming: Routledge.
Campbell, R. (2002). Emotionally involved: The impact of researching rape. New York, NY:
Routledge.
Campbell, R. (2005). What really happened? A validation study of rape survivors' help-seeking
experiences with the legal and medical systems. Violence and victims, 20, 55-68.
Campbell, R. (2008). The psychological impact of rape victims' experiences with the legal,
medical, and mental health systems. American Psychologist, 63(8), 702-717.
Campbell, R. (Producer). (2012, November 11, 2014). The Neurobiology of Sexual Assault.
Retrieved from http://nij.gov/multimedia/presenter/presentercampbell/Pages/welcome.aspx
Campbell, R., Adams, A. E., Wasco, S. M., Ahrens, C. E., & Sefl, T. (2009). Training
Interviewers for Research on Sexual Violence A Qualitative Study of Rape Survivors'
Recommendations for Interview Practice. Violence against women, 15(5), 595-617.
Campbell, R., Bybee, D., Ford, J. K., & Patterson, D. (2009). A systems change analysis of
SANE programs: Identifying mediating mechanisms of criminal justice system impact.
Washington, DC: National Institute of Justice.
Campbell, R., Bybee, D., Patterson, D., & Dworkin, E. (2009). Predicting sexual assault
prosecution outcomes: The role of medical forensic evidence collected by sexual assault
nurse examiners. Criminal Justice & Behavior, 36(7), 712-727.
Campbell, R., Dworkin, E., & Cabral, G. (2009). An ecological model of the impact of sexual
assault on women's mental health. Trauma, Violence, & Abuse, 10, 2250246.
Campbell, R., Greeson, M., Bybee, D., & Fehler-Cabral, G. (2012). Adolescent sexual assault
victims and the legal system: Building community relationships to improve prosecution
rates. American Journal of Community Psychology, 45(1/2), 141-154.

190

Campbell, R., Greeson, M., Bybee, D., Kennedy, A., & Patterson, D. (2011). Adolescent sexual
assault victims' experiences with SANE-SARTs and the criminal justice system.
Washington, DC: National Institute of Justice.
Campbell, R., & Johnson, C. R. (1997). Police officers' perceptions of rape: Is there consistency
between state law and individual beliefs? Journal of Interpersonal Violence, 12, 255-274.
Campbell, R., Lichty, L., Sturza, M., & Raja, S. (2006). Gynecological health impact of sexual
assault. Research in Nursing and Health, 29(5), 399-413.
Campbell, R., & Raja, S. (2005). The sexual assault and secondary victimization of rape victims:
Insights from mental health professionals who treat survivors of violence. Violence and
victims, 14, 261-275.
Campbell, R., Townsend, S., Bybee, D., Shaw, J., & Markowitz, J. (2012). Implementation of a
sexual assault nurse examiner (SANE) practitioner evaluation toolkit: Washington, DC:
US Department of Justice, Office of Justice Programs, National Institute of Justice.
Campbell, R., & Townsend, S. M. (2011). Defining the scope of sexual violence against women.
In C. Renzetti, J. Edleson & R. Bergen (Eds.), Sourcebook on violence against women (2
ed., pp. 95-109). Thousand Oaks, CA: SAGE.
Campbell, R., Townsend, S. M., Bybee, D., Shaw, J., Karim, N., & Markowitz, J. (2014). The
impact of sexual assault nurse examiner (SANE) programs on criminal justice case
outcomes: A multi-site replication study. Violence against women, 20(5), 607-625.
Campbell, R., Wasco, S. M., Ahrens, C. E., Sefl, T., & Barnes, H. E. (2001). Preventing the
"second rape:" Rape survivors' experiences with community service providers. 16,
12(1239-1259).
Cloutier, S., Martin, S. L., & Poole, C. (2002). Sexual assault among North Carolina women:
prevalence and risk factors. Journal of Epidemiology & Community Health, 56, 265-271.
Clum, G. A., Nishith, P., & Resick, P. A. (2001). Trauma-related sleep disturbance and selfreported physical health symptoms in treatment-seeking female rape victims. Journal of
Nervous and Mental Disease, 189, 618-622.
Daly, K., & Bouhours, B. (2010). Rape and attrition in the legal process: A comparative analysis
of five countries. Crime and justice, 39(1), 565-650.
191

DeLisi, M., Kosloski, A., Sween, M., Hachmeister, E., Moore, M., & Drury, A. (2010). Murder
by numbers: Monetary costs imposed by a sample of homicide offenders. The Journal of
Forensic Psychiatry & Psychology, 21(4), 501-513.
Department of Justice. (2013). A national protocol for sexual assault medical forensic
examinations adults/adolescents, second edition. Washington, DC: Author.
Du Mont, J., White, D., & McGregor, M. (2009). Investigating the medical forensic examination
from the perspectives of sexually assaulted women. Social Science & Medicine, 68(4),
774-780.
Edwards, K. M., Turchik, J. A., Dardis, C. M., Reynolds, N., & Gidycz, C. A. (2011). Rape
myths: History, individual and institutional-level presence, and implications for change.
Sex Roles, 65(11-12), 761-773.
Egan, P. (2014). Gov. Rick Snyder signs bill to speek rape kit testing. Detroit Free Press.
Retrieved from http://archive.freep.com/article/20140626/NEWS06/306260109/rape-kittesting-backlog
El-Mouehly, M. (2004). Violence against women: A public health problem. The Journal of
Primary Prevention, 25, 289-303.
Elliott, I., Thomas, S. D., & Ogloff, J. R. (2012). Procedural justice in contacts with the police:
the perspective of victims of crime. Police Practice and Research, 13(5), 437-449.
Feldman-Summers, S., & Palmer, G. C. (1980). Rape as viewed by judges, prosecutors, and
police officers. Criminal Justice & Behavior, 7, 19-40.
Field, H. S. (1978). Attitudes towards rape: A comparative analysis of police, rapists, crisis
counselors, and citizens. Journal of personality and social psychology, 36, 157-179.
Filipas, H. H., & Ullman, S. E. (2001). Social reactions to sexual assault victims from various
support systems. Violence and victims, 16, 673-692.
Fisher, S. Z. (1993). Just the facts, ma'am: Lying and the omission of exculpatory evidence in
police reports. New Eng. L. Rev., 28, 1-57.

192

Fleiss, J. L. (1971). Measuring nominal scale agreement among many raters. Psychological
Bulletin, 76, 378-382.
Foster-Fishman, P. G., Nowell, B., & Yang, H. (2007). Putting the system back into systems
change: A framework for understanding and changing organizational and community
systems. American Journal of Community Psychology, 39, 197-215.
Frazier, P., & Haney, B. (1996). Sexual assault cases in the legal system: Police, Prosecutor and
victim perspectives. Law and Human Behavior, 20, 607-628.
Frohmann, L. (1997). Discrediting victims'allegations of sexual assault: Prosecutorial accounts
of case rejections. Social Problems, 38(2), 213-226.
Galton, E. R. (1975). Police processing of rape complaints: A case study. American Journal of
Criminal Law, 4, 15-30.
Gidycz, C. A., Orchowski, L. M., King, C. R., & Rick, C. L. (2008). Sexual victimization and
health-risk behaviors. Journal of Interpersonal Violence, 23, 744-763.
Golding, J. M. (1999). Sexual assault history and long-term physical health problems: Evidence
from clinical and population epidemiology. Current Directions in Psychological Science,
8, 191-194.
Gylys, J. A., & McNamara, R. (1996). Acceptance of rape myths among prosecuting attorneys.
Psychological Reports, 79, 15-18.
Hanson, R. F., Self-Brown, S., Borntrager, C., Kilpatrick, D. G., Saunders, B. E., Resnick, H. S.,
& Amstadter, A. B. (2008). Relations among gender, violence exposure, and mental
health: The National Survey of Adolescents. American Journal of Orthopsychiatry,
78(3), 313-321.
Harnar, M. A., & Preskill, H. (2007). Evaluatiors' descriptions of process use: An exploratory
study. New Directions in Evaluation, 116, 27-44.
Hart, T. C., & Rennison, C. M. (2003). Reporting crime to the police, 1992-2000. Washington,
DC: US Department of Justice, Office of Justice Programs.

193

Heenan, M., & Murray, S. (2006). Study of reported rapes in Victoria 2000-2003: Summary
resaerch report. Melbourne: Office of Women's Policy, Department of Victorian
Communities.
Herman, J. L. (1992). Complex PTSD: A syndrome in survivors of prolonged and repeated
trauama. Journal of Trauma and Stress, 5(3), 377-391.
Horney, J., & Spohn, C. (1996). The influence of blame and believability factors on the
processing of simple versus aggravated rape cases. Criminology, 34, 135-162.
Howard, C., Logue, K., Quimby, M., & Schoeneberg, J. (2009). Framing change: A new
approach to change management analysis. OD Practitioner, 41(1), 25-31.
Hsieh, H.-F., & Shannon, S. E. (2005). Three approaches to qualitative content analysis.
Qualitative health research, 15(9), 1277-1288.
Human Rights Watch. (2009). Testing justice: The rape kit backlog in Los Angeles City and
County. New York, NY: Human Rights Watch.
Human Rights Watch. (2010). "I used to think the law would protect me:" Illinois's failure to test
rape kits. New York, NY: Human Rights Watch.
Jordan, C. E., Campbell, R., & Follingstad, D. (2010). Violence and women's mental health: The
impact of physical, sexual, and psychological aggression. Annual Review of Clinical
Psychology, 6, 607-628.
Kelly, L., Lovett, J., & Regan, L. (2005). A gap or a chasm? Attrition in reported rape cases
(Home Office Research Study 293). London: Home Office Research, Development and
Statistics Directorate.
Kerstetter, W. A. (1990). Gateway to justice: Police and prosecutorial responses to sexual
assaults against women. Journal of Criminal Law and Criminology, 81, 267-313.
Kettrey, H. H. (2013). Reading Playboy for the Articles The Graying of Rape Myths in Black
and White Text, 1953 to 2003. Violence against women, 19(8), 968-994.

194

Kilpatrick, D. G., Resnick, H. S., Ruggiero, K. J., Conoscenti, L. M., & McCauley, J. (2007).
Drug-facilitated, incapacitated, and forcible rape: A national study. Washington, DC: US
Department of Justice.
Kimerling, R., & Calhoun, K. S. (1994). Somatic symptoms, social support, and treatment
seeking among sexual assault victims. Journal of Consulting and Clinical Psychology,
62, 333-340.
Krueger, R. F., Hicks, B. M., Patrick, C. J., Iacono, W. G., & McGue, M. (2002). Etiologic
connections among substance dependence, antisocial behavior, and personality: Modeling
the externalizing spectrum. Journal of Abnormal Psychology, 111(3), 411-424.
LaFree, G. D. (1981). Official reactions to social problems: Police decisions in sexual assault
cases. Social Problems, 28, 582-594.
LaFree, G. D. (1989). Rape and criminal justice: The social construction of sexual assault.
Belmont, CA: Wadsworth.
Laxminarayan, M. (2012). Procedural justice and psychological effects of criminal proceedings:
The moderating effect of offense type. Social Justice Research, 25(4), 390-405.
Lisak, D. (2008). Understanding the predatory nature of sexual violence. from
http://www.wcsap.org/sites/www.wcsap.org/files/uploads/webinars/SV%20on%20Camp
us/UnderstandingthePredatoryNatureofSexualViolence.pdf
Logan, T., Evans, L., Stevenson, E., & Jordan, C. (2005). Barriers to sevices for rural and urban
survivors of rape. Journal of Interpersonal Violence, 20(5), 591-616.
Lonsway, K. A. (2010). Trying to move the elephant in the living room: Responding to the
challenge of false rape reports. Violence against women, 16(12), 1356-1371.
Lonsway, K. A., & Archambault, J. (2012). The "justice gap" for sexual assault cases: Future
directions for research and reform. Violence against women, 18(2), 145-168.
Lonsway, K. A., & Fitzgerald, L. F. (1995). Attitudinal antecedents of rape myth acceptance: A
theoretical and empirical reexamination. Journal of personality and social psychology,
68(4), 704.

195

Maddox, L., Lee, D. A., & Barker, C. (2012). The impact of psychological consequences of rape
on rape case attrition: The police perspective. Journal of Police and Criminal
Psychology, 27, 33-44.
Madigan, L., & Gamble, N. (1991). The second rape: Society's continued betrayal of the victim.
New York: MacMillan.
Martin, P. Y. (2005). Rape work: Victims, gender, and emotions in organizataion and community
context. New York: Routledge.
Martin, P. Y., & Powell, M. (1994). Accounting for the second assault: legal organizations'
framing of rape victims. Law and Social Inquiry, 20, 853-890.
Martin, S., & Macy, R. (2009). Sexual violence against women: Impact on high-risk health
behaviors and reproductive health. http://www.vawnet/org
Martin, S., Macy, R., & Young, S. (2011). Health and economic consequences of sexual
violence. In J. White, M. Koss & A. Kazdin (Eds.), Violence against women and
children: Mapping the terrain (pp. 173-195). Washington, DC: American Psychological
Association.
McMahon, S., & Farmer, G. L. (2011). An updated measure for assessing subtle rape myths.
Social Work Research, 35(2), 71-81.
Meadows, D. (1999). Leverage points: Places to intervene in a system: The Sustainability
Institute.
Muthén, L. K., & Muthén, B. O. (1998-2012). Mplus user's guide, 7th edition. Los Angeles, CA:
Muthén & Muthén.
National Association of Crime Victim Compensation Boards. (n.d.). Compensation for Crime
Victims. In N. A. o. C. V. C. Boards (Ed.). Washington, DC: Office for Victims of
Crime, US Department of Justice.
National Center for Victims of Crime. (2014). Laws About the Sexual Assault Kit Backlog. from
http://www.victimsofcrime.org/our-programs/dna-resource-center/untested-sexualassault-kits/sexual-assault-kit-backlog-laws

196

National Institute of Justice. (2010, October 26, 2010). Rape and Sexual Violence. from
http://www.nij.gov/topics/crime/rape-sexual-violence/pages/welcome.aspx
North Dakota Council on Abused Women's Services/Coalition Against Sexual Assault in North
Dakota. (2011). North Dakota Model Law Enforcementn Sexual Assault Policy
September 2011 (First Edition ed.). Bismarck, ND.
Office for Victims of Crime. (n.d.). Community-Based Advocates. from
ovc.ncjrs.gov/sartkit/develop/team-advocate-c.html
Page, A. D. (2008a). Gateway to reform? Police implications of police officers' attitudes towards
rape. American Journal of Criminal Justice, 33, 44-58.
Page, A. D. (2008b). Judging women and defining crime: Police officers' attitudes toward
women and rape. Sociological Spectrum, 28, 389-411.
Page, A. D. (2010). True colors: Police officers and rape myth acceptance. Feminist
Criminology, 5, 315-334.
Palm, K. M., & Follette, V. M. (2008). Sexual victimization and physical health: An examination
of exploratory mechanisms. Journal of Child Sexual Abuse, 17(2), 117-132.
Patrick, D. L., Murray, T. P., & Burke, K. M. (2009). Adult Sexual Assault Law Enforcement
Guidelines. Massachusetts: The Commonwealth of Massachusetts, Executive Office of
Public Safety and Security.
Patterson, D., & Campbell, R. (2010). Why rape survivors participate in the criminal justice
system. Journal of Community Psychology, 38(2), 191-205.
Patterson, D., & Campbell, R. (2012). The problem of untested sexual assault kits: Why are
some kits never submitted to a crime laboratory? Journal of Interpersonal Violence, 27,
2259-2275.
Patton, M. Q. (2008). Utilization-focused evaluation (4th edition). Thousand Oaks, CA: Sage.
Payne, D. L., Lonsway, K. A., & Fitzgerald, L. F. (1999). Rape myth acceptance: Exploration of
its structure and its measurement using the Illinois Rape Myth Acceptance Scale. Journal
of Research in Personality, 33(1), 27-68.
197

Perilloux, C., Duntley, J. D., & Buss, D. M. (2012). The costs of rape. Archives of sexual
behavior, 41(5), 1099-1106.
Pett, M. A. (1997). Nonparametric statistics in health care research: Statistics for small samples
and unusual distributions: Sage.
Pharr, S. (1997). Common Elements of Oppression Homophobia: A weapon of sexism (pp. 5364). Berkeley, CA: Chardon Press.
Pierce, S. J., & Zhang, Z. (2011). Projected numbers of evidence kits expected to reach different
stages of forensic analysis: A summary based on data from the 400 Project (pp. 1-16).
East Lansing, Michigan: Center for Statistical Training and Consulting (CSTAT),
Michigan State University.
Pratto, F., Sidanius, J., & Levin, S. (2006). Social dominance theory and the dynamics of
intergroup relations: Taking stock and looking forward. European Review of Social
Psychology, 17, 271-320.
Pratto, F., Sidanius, J., Stallworth, L. M., & Malle, B. F. (1994). Social dominance orientation: A
personality variable predicting social and political attitudes. Journal of personality and
social psychology, 67(4), 741.
Raftery, A. E. (1995). Bayesian model selection in social research. Sociological Methodology,
25, 111-163.
RAINN. (2009). Reporting Rape. from https://rainn.org/get-information/legalinformation/reporting-rape
Raphael, J., & Logan, T. (2011). The use (and misuse) of data on rape: Restoring sexual assault
to the national agenda. Paper presented at the CounterQuo Conference.
Reiss, A. J. (1971). The police and the public. New Haven, CT: Yale University Press.
Resnick, H. S., Acierno, R., Holmes, M., Dammeyer, M., & Kilpatrick, D. G. (2000). Emergency
evaluation and intervention with female victims of rape and other violence. Journal of
Clinical Psychology, 56, 1317-1333.

198

Resnick, H. S., Holmes, M. M., Kilpatrick, D. G., Clum, G. A., Acierno, R., Best, C., &
Saunders, B. E. (2000). Predictors of post-rape medical care in a national sample of
women. American Journal of Prevention Medicine, 19(4), 214-219.
Rose, V. M., & Randall, S. C. (1982). The impact of investigator perceptions on victim
legitimacy on the processing of rape/sexual assault cases. Symboli Interaction, 5, 22-36.
Rubin, J. (2009). DNA going untested in cases throughout L.A. County. Los Angeles Times.
Retrieved from latimes.com/news/local/la-me-dna31-2009mar31,0,4150400.story
website:
Schreiber, J. B., Nora, A., Stage, F. K., Barlow, E. A., & King, J. (2006). Reporting structural
equation modeling and confirmatory factor analysis results: A review. The Journal of
Educational Research, 99(6), 323-338.
Schuller, R., & Stewart, A. (2000). Police responses to sexual assault complaints: The role of
perpetrator/compainant intoxication. Law and Human Behavior, 24(5), 535-551.
Shaw, J., & Campbell, R. (2013). Predicting sexual assault kit submission among adolescent rape
cases treated in forensic nurse examiner programs. Journal of Interpersonal Violence,
28(18), 3400-3417.
Sidanius, J., Liu, J. H., Shaw, J. S., & Pratto, F. (1994). Social dominance orientation, hierarchy
attenuators and hierarchy enhancers: Social dominance theory and the criminal justice
system. Journal of Applied Social Psychology, 24(4), 338-366.
Sidanius, J., & Pratto, F. (2001). Social dominance: An intergroup theory of social hierarchy and
oppression. New York: Cambridge University Press.
Sidanius, J., & Pratto, F. (2011). Social dominance theory. Handbook of Theories of Social
Psychology: Volume Two, 2, 418.
Sidanius, J., Pratto, F., Sinclair, S., & Van Laar, C. (1996). Mother Teresa meets Genghis Khan:
The dialectics of hierarchy-enhancing and hierarchy-attenuating career choices. Social
Justice Research, 9(2), 145-170.
Skinner, K. M., Kressin, N., Frayne, S., Tripp, T. J., Hankin, C. S., Miller, D. R., & Sullivan, L.
M. (2000). The prevalence of military sexual assault among female veterans'
administration outpatients. Journal of Interpersonal Violence, 15(291-310).
199

Smith, D. A., & Klein, J. R. (1983). Police agency characteristics and arrest decisions. In G. P.
Whitaker & C. D. Phillips (Eds.), Evaluation performance of criminal justice agencies.
Beverly Hills: Sage.
Spohn, C., & Spears, J. (1996). The effect of offender and victim characteristics on sexual
assault case processing decisions. Justice Quarterly, 13, 647-679.
Strom, K. J., & Hickman, M. J. (2010). Unanalyzed evidence in law-enforcement agencies: A
national examination of forensic processing in police departments. Criminology & Public
Policy, 9(2), 381-404.
Svensson, P. (2013). Telephone companies to abandon land lines. Associated Press. Retrieved
from
http://www.salon.com/2013/07/09/telephone_companies_abandon_copper_phone_lines_
ap/
Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (2nd ed). Boston, MA:
Pearson.
Tasca, M., Rodriguez, N., Spohn, C., & Koss, M. P. (2013). Police Decision Making in Sexual
Assault Cases Predictors of Suspect Identification and Arrest. Journal of Interpersonal
Violence, 28(6), 1157-1177.
Tjaden, P., & Thoennes, N. (2000). Full report of the prevalence, incidence, and consequences of
violence against women: Findings from the National Violence Against Women Survey.
Washington: US Department of Justice, Office of Justice Programs.
Tjaden, P., & Thoennes, N. (2006). Extent, nature and consequences of rape victimization:
Findings from the National Violence Against Women Survey. Special report.
Washington, DC: National Institute of Justice and Centers for Disease Control and
Prevention.
Tofte, S. (2009). Detroit's rape kits must be checked. Detroit Free Press. Retrieved from
http://www.hrw.org/en/news/2009/10/02/detroit-s-rape-kits-must-be-checked website:
Triggs, S., Mossman, E., Jordan, J., & Kingi, V. (2009). Responding to sexual violence: Attrition
in the New Zealand criminal justice system. Wellington, NZ: Ministry of Women's
Affairs.

200

Tsang, E. W. K., & Zahra, S. A. (2008). Organizational unlearning. Human Relations, 61(10),
1435-1462.
Twohey, M. (2010). Illinois to test every rape kit: New law requiring DNA analysis of all
submitted samples comes after Tribue report found police often shelf the evidence. The
Chicago Tribune. Retrieved from http://articles.chicagotribune.com/2010-07-06/news/ctmet-rape-kit-law-20100706_1_untested-kits-crime-lab-dna-backlog
Ullman, S. E. (2010). Talking about sexual assault: Socieyt's response to survivors. Washington,
DC: American Psychological Association.
Ullman, S. E., Filipas, H. H., Townsend, S. M., & Starzynski, L. L. (2006). The role of victimoffender relationship in women's sexual assault experiences. Journal of Interpersonal
Violence, 21(798-819).
Ullman, S. E., Filipas, H. H., Townsend, S. M., & Starzynski, L. L. (2007). Psychosocial
correlates of PTSD symptom severity in sexual assault survivors. Journal of Traumatic
Stress, 20, 821-831.
US Census Bureau. (n.d.). American Fact Finder. from
http://factfinder2.census/gov/faces/nav/jsf/pages/searchresults.xhtml?refresh=t
US Department of Justice Federal Bureau of Investigation. (2010). Estimated number of arrests:
United States 2010. from http://www.fbi.gov/about-us/cjis/ucr/crime-in-theu.s/2010/crime-in-the-u.s.-2010/tables/10tbl29.xls
US Department of Justice Federal Bureau of Investigation. (January 2009). UCR Offense
Definitions. from http://www.ucrdatatool.gov/offenses.cfm
US Department of Justice Federal Bureau of Investigation. (March 2010). Estimated crime in
2010. from
http://www.ucrdatatool.gov/Search/Crime/State/RunCrimeOneYearofData.cfm
Wemmers, J.-A. (2013). Victims’ experiences in the criminal justice system and their recovery
from crime. International review of victimology, 19(3), 221-233.
Wisconsin Coalition Against Sexual Assault (Ed.). (2009). Prosecutor's Sexual Assault
Reference Book. Wisconsin: Wisconsin Office of Justice Assistance.

201

Wriggins, J. (1983). Rape, racism, and law. Harvard Women's Law Journal, 6, 140-141.
Yuan, N. P., Koss, M., & Stone, M. (2006). The psychological consequences of sexual trauma.
http://www.vawnet.org
Zaccaro, S. J., & Banks, D. J. (2001). Leadership, vision, and organizational effectiveness. In S.
J. Zaccaro & R. J. Kilmoski (Eds.), The nature of organizational leadership:
Understanding the performance imperatives confronting today's leaders (pp. 181-218).
San Francisco, CA: Jossey-Bass.
Zinzow, H. M., Resnick, H. S., Amstadter, A. B., McCauley, J., Ruggiero, K. J., & Kilpatrick, D.
G. (2010). Drug- or alcohol-facilitated, incapacitated, and forcible rape in relationship to
mental health among a national sample of women. Journal of Interpersonal Violence, 25,
2217-2236.
Zoller, K., Normore, A. H., & McDonald, S. (2014). A new way of leading for the law
enforcement supervisor: An adaptive leadership case study. The Police Chief, 81, 44-47.

202