ArticlePDF Available

Abstract

Using data from the National Violence Against Women Survey, the authors examine whether rapes committed after reforms were more likely to be reported to police than those committed before reforms. The authors also consider whether the gap between the reporting of simple versus aggravated rape has narrowed. They find that rapes committed after 1990 were more likely to be reported than rapes occurring before 1974. Aggravated rape continues to be more likely to be reported than simple rape, however, and this effect is stable over time. The authors conclude by discussing the implications of these findings for evaluating the success of rape reform statutes.
10.1177/1077801204271566VIOLENCE AGAINST WOMEN / February 2005Clay-Warner, Burt / REPORTING RAPE
Rape Reporting After Reforms
Have Times Really Changed?
JODY CLAY-WARNER
CALLIE HARBIN BURT
University of Georgia
Using data from the National Violence Against Women Survey, the authors examine
whether rapes committed after reforms were more likely to be reported to police than those
committed before reforms. The authors also consider whether the gap between the report
-
ing of simple versus aggravated rape has narrowed. They find that rapes committed after
1990 were more likely to be reported than rapes occurring before 1974. Aggravated rape
continues to be more likely to be reported than simple rape, however, and this effect is sta-
ble over time. The authors conclude by discussing the implications of these findings for
evaluating the success of rape reform statutes.
Keywords:
rape reform; rape reporting; sexual assault
Since the mid-1970s, sweeping changes have been made to laws
governing forcible rape. These changes include the alteration of
evidentiary requirements, establishment of rape shield statutes,
modification of resistance requirements, and, in some cases, even
redefining the crime of rape (e.g., Spohn & Horney, 1992). Prior to
the institution of these reforms, many victims were reluctant to
report rapes, fearing mistreatment from the criminal justice system
and retaliation from the perpetrator (e.g., Bryden & Lengnick,
1997; Legrand, 1973; Robin, 1977). In fact, under-reporting was so
widespread that some estimated that even if every rape reported
resulted in a trial, more than 90% of rapists would never see the
150
AUTHORS NOTE: Data used in this article were obtained through the Inter-
University Consortium for Political and Social Research housed at the University of
Michigan.
VIOLENCE AGAINST WOMEN, Vol. 11 No. 2, February 2005 150-176
DOI: 10.1177/1077801204271566
© 2005 Sage Publications
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
inside of a courtroom (McCahill, Meyer, & Fischman, 1979). Rec
-
ognizing that justice system reforms are only successful to the
extent that victims access the systems, one key goal of the reforms
was to increase reporting of rape (Bachman & Paternoster, 1993;
Bryden & Lengnick, 1997; Horney & Spohn, 1991). Despite the
significance of reporting to the overall success of the reforms,
however, only limited empirical research has examined changes
in rape reporting across time (Bachman & Paternoster, 1993).
Even less attention has been given to whether extralegal factors
that inhibited reporting, such as prior acquaintance with the
assailant, continue to exert the same influence today as they did in
the prereform era. The influence of extralegal factors is signifi
-
cant, as Estrich (1987) has argued that there has always been a dis
-
tinction in the legal system between aggravated and simple rape.
She defined aggravated rape, following Kalvin and Zeisel (1966;
see also Horney & Spohn, 1996), as an assault in which the victim
and assailant are unknown to each other, there are multiple assail-
ants, or violence is explicit, as evidenced by use of weapons and
victim injury. In simple rapes, none of these circumstances exists.
Although modern law has not drawn a distinction between these
two forms of sexual violence, considerable evidence suggests
that, indeed, aggravated rape and simple rape have been viewed
as different crimes by the public and have been treated differently
by the courts (Alder, 1987; Bryden & Lengnick, 1997; Estrich, 1987;
Kalvin & Zeisel, 1966; LaFree, 1981; McCahill et al., 1979). Estrich
also states that certain components of rape reform were written
with the explicit intent of reducing the distinction between aggra
-
vated and simple rape (e.g., rape shield laws). The question
remains, however, whether these reforms have been successful in
reducing the reporting gap between simple and aggravated rape.
Here we address the issue of changes in rape reporting by
focusing upon two central questions: (a) Has the overall likeli
-
hood of rape reporting changed since the advent of reforms? and
(b) Has the gap between the reporting of simple and aggravated
rape narrowed since the passage of statutory reforms? We begin
with a discussion of rape reform legislation and then review prior
research on rape reporting. Next, we examine changes in rape
reporting behavior through an analysis of data from the National
Violence Against Women (NVAW) Survey.
Clay-Warner, Burt / REPORTING RAPE 151
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
RAPE REFORM LEGISLATION
Feminists, frustrated by the institutionalized sexism in the
criminal justice system, and “law-and-order groups,” seeking to
increase rates of rape reporting and conviction, began pushing for
rape reform in the late 1960s (Spohn & Horney, 1992). Feminists
contended that rape victims were hesitant to report their victim
-
ization to the police because they feared both derogatory treat
-
ment by criminal justice officials and the invasion of privacy asso
-
ciated with rape trials (Robin, 1977). Legal scholars pointed out
that the definitional requirements and stringent standards of
proof associated with rape made rape statutes unlike any other
area of criminal law (Caringella-McDonald, 1985; Robin, 1977).
As a result, reporting rates were very low, convictions were diffi
-
cult to procure, and victims were often revictimized through the
criminal justice process (e.g., Caringella-McDonald, 1985; Largen,
1988; Sebba, 1974; Shapo, 1975).
In 1974, Michigan became the first state to respond to these crit-
icisms by enacting rape reform legislation (Galvin, 1985). Many
states followed suit; in the years subsequent to the commence-
ment of rape reforms, substantial changes have been made to stat-
utes in every state (Field & Bienen, 1980, p. 153; see also Berger,
Neuman, & Searles, 1994; Berger, Searles, & Neuman, 1988).
Although the reforms and laws enacted vary between states, they
all seek to shift the focus away from the victim and toward the
behavior of the defendant (e.g., Caringella-McDonald, 1985; Fut
-
ter & Mebane, 2001; Searles & Berger, 1987). This goal is reflected
in four general themes of rape reform legislation: (a) a redefinition
of rape, (b) the elimination or modification of the resistance
requirement, (c) the elimination or modification of the corrobora
-
tion requirement, and (d) the establishment of rape-shield laws
that limited or prohibited the admissibility of a victim’s sexual
history on cross-examination (Horney & Spohn, 1991).
Traditional rape laws have generally defined forcible rape as
“an act of sexual intercourse undertaken by a man with a woman,
not his wife, by force and against her will” (Bohmer, 1991, p. 318).
Most states’ definitions of rape now encompass acts other than
the “traditional” rape such as marital rape and forced oral or anal
sex (Marsh, Geist, & Caplan, 1982; Searles & Berger, 1987), and
they have replaced the single crime of rape with a series of graded
152 VIOLENCE AGAINST WOMEN / February 2005
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
offenses. These graded offenses reflect the continuum of violence
in sexual assault crimes, allow punishments to be prescribed that
are commensurate with the crime, and reduce the amount of dis
-
cretion given to criminal justice officials (Marsh et al., 1982). An
additional modification was the replacement of the term rape with
a more general term, such as sexual assault,
1
with the intentions of
emphasizing both that rape is a crime of violence, rather than one
resulting from uncontrollable sexual urges, and equating rape
with other violent assaults (Searles & Berger, 1987).
The resistance requirement also came under fire. In traditional
American legal theory, rape consisted in the concurrence of a
criminal act (the nonconsensual sexual intercourse) with criminal
intent (the intention or knowledge of having the intercourse with
-
out the victim’s consent; Estrich, 1987). Thus, there was no rape
unless the victim explicitly did not consent to sexual intercourse
and the perpetrator both recognized and disregarded this fact.
2
The most facile way of proving nonconsent was through victim
resistance. As a result, many jurisdictions required victims to
demonstrate resistance for the crime to qualify as rape. Many crit-
icized this requirement, because it is unique to the crime of rape.
As a consequence of the reform movement, most states elimi-
nated this requirement and/or reduced the state’s burden of
proof by stipulating the circumstances that constitute force
(Horney & Spohn, 1991).
Another significant evidentiary reform was the removal of the
corroboration requirement, which had prohibited conviction for
rape on the uncorroborated testimony of the victim. Legal corrob
-
oration was confirmation that “(a) the event occurred and is thus
not a complete fabrication, (b) the defendant has been identified
as the rapist, (c) there was actual ‘penetration’ of the woman, and
(d) force was present and ‘consent’ was absent” (Robin, 1977,
p. 138). As with the resistance requirement, reformers criticized
this condition by pointing to the fact that rape was the only crime
with such a requisite (Horney & Spohn, 1991). Every state, with
the exception of Nebraska, eventually eliminated the condition
that a rape complainant’s testimony be corroborated in the case of
forcible rape (Bryden & Lengnick, 1997).
3
Perhaps both the most controversial and noteworthy reform
was the creation of rape shield laws. In theory, these laws prevent
the introduction of the victim’s past sexual conduct and evidence
Clay-Warner, Burt / REPORTING RAPE 153
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
of character and reputation into cross-examination (Caringella-
McDonald, 1985; Hibey, 1973). Under traditional rape laws, a pro
-
pensity to consent to sexual intercourse, established by prior acts
as well as the victim’s character and reputation, were considered
probative evidence for the jury to utilize to determine whether the
victim was the type of person who would have willingly engaged
in intercourse with the defendant (Caringella-McDonald, 1985).
Thus, the defense used evidence of prior sex acts as being tanta
-
mount to acquiescence by contending that if a woman had had
sex with various men on many different occasions, it is likely that
she consented during the incident in question. As a result of this
focus on the victim, many women described the trial as a “second
rape” (e.g., Bryden & Lengnick, 1997; Legrand, 1973; Madigan &
Gamble, 1991; J. E. Williams & Holmes, 1981).
In response, rape shield laws were passed to limit or prevent
the defense from focusing on the behavior of the defendant rather
than on the behavior of the victim (Caringella-McDonald, 1985).
When legislatures passed rape shield laws, their reasons cited
included encouraging the reporting of rapes, protecting the pri-
vacy of rape victims, and thwarting the humiliation of victims
during trial, particularly in cases of acquaintance rape (Berger
et al., 1994; Kessler, 1992).
As Bachman and Paternoster (1993) noted, reformers believed
that changes in rape laws, especially rape shield statutes, would
make victims more likely to report sexual assaults to the authori-
ties. Women would no longer fear the humiliation at trial that
often results from focusing on the behavior and characteristics of
the victims. Women would also be more willing to come forward
as a result of changes in evidentiary requirements that made rape
convictions easier to obtain, because they would no longer see
their reporting as a futile effort (Bryden & Lengnick, 1997; Fairstein,
1993; Horney & Spohn, 1991).
CHANGES IN REPORTING
Research investigating the success of rape-reform legislation in
increasing reporting has been inconclusive. Early state-level stud
-
ies found rape reforms to be ineffective. In Michigan, where the
first and often considered the most comprehensive reforms were
enacted, researchers found no change in the number of rapes
154 VIOLENCE AGAINST WOMEN / February 2005
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
reported to the police (Marsh et al., 1982). Additionally, in a time-
series analysis of data pre- and postreforms and in interviews
with criminal justice officials, Marsh et al. (1982) found no evi
-
dence that the new laws affected reporting rates.
Spohn and Horney (1992) examined the effects of the reforms
in six cities: Chicago; Detroit; Philadelphia; Washington, D.C.;
Atlanta; and Houston. The first three cities were in states with
strong reforms, whereas the last three cities were located in juris
-
dictions with weak reforms. Examining monthly data, they found
that in the years from 1970 to 1984, legal reforms had no positive
effect in Chicago, Philadelphia, and Atlanta. In Washington, D.C.,
most likely by coincidence, there was a decrease in reporting after
the removal of the corroboration requirement. Spohn and Horney
observed in Detroit, the state with the strongest reforms, and in
Houston, the jurisdiction with the weakest reforms, an increase in
reporting rates that appeared in some way correlated with the
reforms. Further inquiry revealed, however, that the changes in
reporting rates were not related to the “substantive content” of
the reforms but, rather, suggested that “the increases resulted
from publicity surrounding the reforms rather than gradually
acquired knowledge of improved treatment of victims under the
new laws” (Spohn & Horney, 1992, pp. 101-102). This explanation
is plausible, because in both cities, the rate of reporting increased
immediately after the passage of the reforms and then stabilized
(Bryden & Lengnick, 1997).
Berger et al. (1994) examined the relationship between the
strength of rape reform legislation and reporting rates using the
1985 rape and sexual assault statutes in 48 states. Seven variables
were selected and arranged on an ordinal continuum ranging
from traditional to progressive statutes with regard to feminist
objectives for the reforms. They then tested whether the strength
of a state’s reform statute was related to either rape rates or to a
percentage increase in rape reporting in the postreform period.
Finding neither relationship to be statistically significant, Berger
et al. concluded that reforms had not appreciably affected official
rape rates. They did find, however, that the rape law reforms
that extended the range of persons protected by law were posi
-
tively associated with official rape rates. As a result, they con
-
tended that although the effects of rape reform were limited, it
Clay-Warner, Burt / REPORTING RAPE 155
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
would be premature to dismiss the reforms as ineffective (see also
Caringella-MacDonald, 1985).
Other research on changes in reporting has utilized national-
level data. Bryden and Lengnick (1997) considered crime statis
-
tics from both the Uniform Crime Reports (UCR) and the National
Crime Victimization Survey (NCVS) from the years 1973 to 1994.
When examining the NCVS, which interviews individuals 12
years of age and older from a stratified sample of U.S. households,
Bryden and Lengnick found that the number of reported rapes
declined piecemeal during that period. In the UCR, however,
which tallies the number of reported crimes, the number of rapes
reported increased from 24.5 in 1973 to 39.2 per 100,000 inhabit
-
ants in 1994.
4
Bryden and Lengnick’s explanation for the apparent
discrepancy is that reports of sexual assault have increased.
Bachman and Paternoster (1993) analyzed the NCVS, the UCR,
the National Prisoners Statistics program (NPS), and the National
Corrections Reporting Program (NCRP) to examine the extent to
which rape reporting changed in the years from 1973 to 1990 com-
pared to the change in reporting for robbery and nonsexual
assaults.
5
The NCVS data demonstrated that rape victims who
reported to police increased by 10% compared to a 4% increase for
assault and a 12% decrease for robbery. The UCR data corrobo-
rated this slight increase by revealing a 13% increase in rape
reporting during the years from 1973 to 1990, whereas assault
reports increased 46% and robbery 6% over the same time period.
Based on these findings, Bachman and Paternoster concluded
that “rape victims were slightly more likely to report their victim
-
izations after statutory reforms were in place” (1993, p. 566).
In sum, studies of rape law reform have been unable to docu
-
ment significant increases in reports that could be directly attrib
-
uted to the legislation. Most notably, Spohn and Horney’s (1992)
comparative analysis of six urban jurisdictions did not find differ
-
ences in reporting rates between strong-reform and weak-reform
jurisdictions (1992).
FACTORS AFFECTING RAPE REPORTING
It is not just rates of reporting that deserve scrutiny, however; it
is also important to examine whether extralegal factors associated
156 VIOLENCE AGAINST WOMEN / February 2005
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
with rape reporting have changed. An ancillary goal of rape
reform was to remove the stigma attached to certain types of rape
such as those committed by acquaintances or those in which no
additional injury occurred. As Estrich (1987) noted, these types of
rapes were not considered “real rapes” and were not treated seri
-
ously by either the public or the criminal justice system. In devel
-
oping her argument, Estrich cited the fact that rapes committed
by strangers (one type of aggravated rape) are more likely to be
reported than are rapes committed by acquaintances, most of
which would be classified as simple rape. Indeed, studies using
crime data gathered from the late 1970s to the mid-1980s have
concluded that women raped by strangers were significantly
more likely to report compared to women raped by acquaintances
(Feldman-Summers & Ashworth, 1981; Greenberg & Ruback,
1992; Lizotte, 1985; Smith & Nelson, 1976; L. S. Williams, 1984).
Consistent with Estrich’s (1987) assertion, situational danger
has also been found to affect reporting. Smith and Nelson (1976)
found that victim reporting varied directly with the amount of
danger experienced by the victim during her rape and that report-
ing was more likely if the rapist committed additional crimes at
the time of the attack. Similarly, Greenberg and Ruback (1992)
found that women were more likely to report their rapes if
degrading acts were inflicted. Other indicators of rape serious-
ness, such as weapon use (Amir, 1971; Bachman, 1998; Lafree,
1980; Lizotte & Wolfson, 1981) and the level of physical injury
(Bachman, 1993, 1998; Holmstrom & Burgess, 1978; LaFree, 1980;
Lizotte, 1985; Lizotte & Wolfson, 1981), have also been found to
increase the likelihood of reporting. In fact, Lizotte (1985) found
that although the seriousness of the incident was predictive of
crime reporting in general, serious injury had an even greater
effect on the likelihood of reporting in rape cases than in cases of
nonsexual assault.
6
Other factors not directly related to the simple-aggravated rape
dichotomy have also been found to be associated with rape
reporting. Lizotte (1985) found that both married rape victims
and highly educated victims were more likely to report. Greenberg
and Ruback (1992) reported that women were also more likely to
report their rape if the attack occurred outdoors and if the rapist
Clay-Warner, Burt / REPORTING RAPE 157
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
was African American. Social distance factors have also been
found to be important. Smith and Nelson (1976) found that
nonreporting was more likely if the victim and her rapist were of
equal age and social class. Similarly, Lizotte (1985) reported that
the more familiar the victim was with the offender, the less likely
she was to report the sexual assault to the police.
Most of these studies, however, have not examined whether
there have been changes in the impact of these extralegal factors
on reporting since the institution of rape reform, and none has
examined whether simple rapes are more likely to be reported
now than prior to reforms. The only relevant variable that has
been examined is acquaintance between the victim and offender.
Bachman (1993), using rape data from the NCVS from 1987 to
1990, found that the victim-offender relationship did not affect
the likelihood of reporting. Pollard (1995) and Ruback (1993)
questioned this finding on both statistical and conceptual grounds.
Ruback (1993) argued that this research simply demonstrated
that acquaintance was a less important factor than it once was and
that it was “premature to say that the victim-offender relationship
does not matter” (p. 278). Because Bachman did not compare rape
reporting pre- and postreform, however, we cannot conclude that
there has actually been a significant change across time in the
effect of the victim-offender relationship on rape reporting.
Thus, it is still unknown whether rapes were more likely to be
reported in the period after reforms than before or whether the
gap between the reporting of simple and aggravated rapes has
been attenuated. In addressing these questions, this study attempts
to avoid some of the problems associated with prior studies of
rape reporting. First, most of the studies were conducted in the
1970s and 1980s (e.g., Berger et al., 1994; Bridges, 1991; Caringella-
McDonald, 1985; Lizotte, 1985; Marsh et al., 1982). It is possible,
however, that the effects of rape reform on reporting would not be
seen so quickly after their enactment, as changes in reporting are
contingent upon changes in societal perceptions—either about
rape itself or about the way in which the crime of rape is treated in
the criminal justice system. In this way, the effects of reforms on
rape reporting may be either direct or indirect. Reforms may
directly affect reporting only when victims become aware of the
reforms and make decisions about reporting based upon this
158 VIOLENCE AGAINST WOMEN / February 2005
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
knowledge. A more likely scenario is that reforms have an indi
-
rect effect on reporting by contributing to a change in the way the
public views rape and rape victims, as was a goal of many femi
-
nist reformers (e.g., Bachman & Paternoster, 1993; Bryden &
Lengnick, 1997; Largen, 1988). Whether victims are more likely to
report as a result of direct knowledge of the reforms or because of
changes in the way society views rape victims, we should not
expect increases in reporting to be instantaneous with reforms.
Bachman and Paternoster (1993) suggested that this might be the
case, noting that attitudes about rape and victimization may have
changed in the past 15 years thus translating into significant oper
-
ational changes in the criminal justice system as well as in the way
victims respond to the crime.
Second, many studies have examined only a single state or
jurisdiction, which precludes any general conclusions about the
nationwide impact of the reforms. Finally, much of the previous
research has been limited by the available data, which has been
drawn primarily from the NCVS and official reports. Studies
employing the NCVS have been criticized for the way rape vic-
tims are identified. In critiquing Bachman (1993), Ruback (1993)
and Pollard (1995) pointed out that individuals who reported
their rapes to the police would be more likely to inform the NCVS
interviewers that they had been raped than women who had not
reported to the police. This bias could serve to inflate the percent-
age of acquaintance rapes that were reported to police. Reflecting
upon societal stigma and decisions concerning what actually con
-
stitutes rape, they argued that victims of acquaintance rapes who
do not report their rape are more likely to believe or convince
themselves that they were not actually victims of rape thus lead
-
ing to unreliable data (Koss, 1992; Pollard, 1995; Ruback, 1993).
On the other hand, studies that use official rape reports (e.g.,
UCR) are suspect because they can only measure whether the
total number of rape reports has increased; they cannot determine
whether the rate of reporting has increased, because official records
contain no measure of the number of rapes that are not reported.
In this way, it is quite possible that reports to the police could
increase because of an actual increase in crime and not because of
any increase in the rate at which crimes are reported. In fact, some
researchers have used changes in UCR rates as evidence of actual
Clay-Warner, Burt / REPORTING RAPE 159
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
changes in rates of rape commission (e.g., Baron & Straus, 1989)
thereby suggesting that an increase in official reports of rape is not
a reliable indicator of increases in the rate of rape reporting. This is
a limitation faced by all studies that use official reports of crime to
examine changes in reporting (e.g., Berger et al., 1994; Marsh
et al., 1982; Spohn & Horney, 1992).
In an attempt to overcome these data limitations, we address
the effects of rape reform on reporting using data from the NVAW.
Because the NVAW Survey employs a representative sample of
women and asks about both reported and unreported rape, it is a
more accurate measure of changes in reporting than data drawn
from police records. We also avoid some of the problems with
underreporting associated with the NCVS. Although the NCVS
was redesigned in 1992 to ask more directly about rape victimiza
-
tion, research indicates that significant underreporting remains a
problem. Fisher, Cullen, and Turner (2000) found that women
participating in the National College Women Sexual Victimization
Survey who were asked behaviorally specific questions reported a
rape prevalence rate 11 times higher than another group of partic-
ipants whose rape screening questions were drawn from the rede-
signed NCVS. Because the NVAW Survey uses behaviorally spe-
cific questions virtually identical to those used in the National
College Women Sexual Victimization Survey, the problems of
underreporting associated with the NCVS are less likely to occur
in the NVAW Survey. Also, the NVAW Survey is not described to
participants as a crime survey, as is the NCVS. Thus, a woman
could acknowledge having experienced sexual violence without
identifying as a crime victim. This would increase the likelihood
that women who did not report to police would reveal sexual vio
-
lence to the interviewer.
Our work also goes further than previous research by expand
-
ing the central question. We assert that any measure of the success
of rape reform in changing reporting behavior must examine not
only changes in the likelihood of rape being reported pre- and
postreform, but also must address changes in the effects of perti
-
nent extralegal factors on rape reporting (simple vs. aggravated
rape). Here we investigate both of these issues in multivariate
models.
160 VIOLENCE AGAINST WOMEN / February 2005
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
METHOD
DATA
Data fromthe NVAW Survey were analyzed (Tjaden & Thoennes,
1999). The NVAW Survey is a national telephone survey jointly
supported by the Centers for Disease Control and Prevention and
the National Institute of Justice. Participants were chosen through
random-digit dialing of residential telephone numbers. To ensure
that a nationally representative sample was obtained, the sample
was stratified by region, as defined by the U.S. Census Bureau
(Tjaden & Thoennes, 1999). Once contact was made with a house
-
hold resident, eligible adults were identified (those at least 18
years of age). In households in which more than one adult was eli
-
gible, the adult with the most recent birthday (day and month)
was asked to participate. Thus, there was only one respondent per
household. Both men and women were eligible for participation,
although the current study analyzes data only from the female
participants. The participation rate for women was 72.1% (n =
8,000).
7
Because this was a telephone-based survey, individuals
without telephones, those residing in group facilities or institu-
tions, and homeless persons are not represented.
A goal of the NVAW Survey was to determine lifetime preva-
lence rates of violent victimizations including sexual assault,
physical assault, stalking, and intimate partner violence. Respon-
dents were asked behaviorally specific questions to determine
whether they had experienced specific forms of violence. If vic
-
timization was reported, the interviewer asked respondents
detailed questions about the incident and the perpetrator, and a
separate report was filed for each incident. In the case of multiple
victimizations by the same perpetrator, women were asked these
same questions about the most recent victimization by this
perpetrator.
For purposes of this research, we are interested in incidents of
completed and attempted rape perpetrated by men against
women. Incidents of both forced vaginal, anal, or oral penetration
(rape) and attempts at such forced penetration (attempted rape)
are included in our analyses. We also include only rapes and
Clay-Warner, Burt / REPORTING RAPE 161
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
attempted rapes occurring to women who were at least 16 years of
age at the time of victimization. We exclude childhood sexual
assaults, because issues facing minors in reporting violence are
likely to be different from the issues facing adults; thus, it would
be inappropriate to include youths and adults in the same analy
-
sis. Also, rape law reform focused primarily on forcible rape stat
-
utes, which are often separate from child molestation statutes.
Therefore, an inclusion of child sexual assault cases could obscure
potential differences pre- and postreform.
Including only one reported victimization per study partici
-
pant further reduced our sample. As previously noted, the
NVAW Survey allows victims of sexual violence to identify multi
-
ple incidents of victimization, and a separate file is created for
each incident. It is inappropriate in most regression procedures to
include multiple cases from a single individual, however, as this
violates the assumption of uncorrelated errors. To address this
problem, we are including only the first incident of rape or
attempted rape reported by a participant (occurring after the age
of 16). Therefore, the unit of analysis is the rape incident, and
there is only one rape incident included per participant. Our final
sample is composed of 824 incidents of completed or attempted
rape.
MEASURES
The dependant variable was a dichotomous measure indicat
-
ing whether the assault was reported to police. Following previ
-
ous research (e.g., Bachman, 1998; Bryden & Lengnick, 1997) we
coded the reported variable as 1 if the victim indicated that the
incident had been reported to law enforcement either by the vic
-
tim or any other individual. Of the assaults that were reported,
the victim reported to police in approximately three quarters of
the cases. Friends or relatives of the victim reported most of the
other assaults. Atotal of 15% of the victimizations were reported.
To examine whether the likelihood of a rape being reported has
increased since the passage of rape reform legislation, we con
-
structed two dummy variables for the year in which the assault
occurred. Variables for the early reform period (1975-1989) and
the modern reform period (1990-1996) were included in the analy
-
sis with the prereform period (before 1975) as the left-out,
162 VIOLENCE AGAINST WOMEN / February 2005
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
comparison category. We chose 1974 as the cutoff for the
prereform period, because it was not until 1974 that the first rape
reform statute was passed (in Michigan). The period from 1975
until 1989 was characterized by ongoing changes in both statu
-
tory reform and reform by case law with virtually every state
enacting some form of statutory reform by the mid-1980s (Berger
et al., 1994). Thus, we labeled this period as early reform.
8
One
hundred and eighty-nine incidents occurred from 1990 until 1996.
Four hundred and twenty-four incidents occurred between 1975
and 1989. The remaining 211 assaults took place before 1975.
The variable aggravated distinguished between simple and
aggravated rapes using the criteria established by Estrich (1987)
and followed by Horney and Spohn (1996). A rape incident was
classified as aggravated if a stranger perpetrated the assault,
there were multiple assailants, if a weapon was used, or if an
injury occurred (e.g., burns, internal injuries, broken bones, sprains,
chipped teeth, and other wounds). Forty-two percent of the assaults
were aggravated (n = 342).
A number of control variables were included that have been
found in previous research to affect rape reporting. The variable
completed was created to distinguish between completed and
attempted rapes. Participants were asked a number of behavio-
rally specific questions to ascertain whether they had been sexu-
ally victimized. For example, one question asked,
Regardless of how long ago it happened, has a man or boy ever
made you have sex by using force or threatening to hurt you or
someone close to you? Just so there is no mistake, by sex we mean
putting a penis in your vagina.
Other questions address forced penetration of the vagina by
objects other than the penis as well as the forced penetration of the
mouth and anus by either a penis or other objects including fin
-
gers. The acts described in these questions meet legal definitions
of rape or forced sodomy in most states, and participants who
indicated being victimized in such a way were coded 1 for com
-
pleted rape. Participants were also asked whether an unsuccess
-
ful attempt at forced vaginal, anal, or oral penetration was made.
Incidents in which women indicated an unsuccessful attempt
were coded 0 for attempted rape. In our sample, there were 331
rape attempts (40%) and 493 completed rapes (60%).
Clay-Warner, Burt / REPORTING RAPE 163
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
Two alcohol/drug use variables were included. Victim drug/
alcohol use was coded 1 if the participant indicated that she was
under the influence of drugs and/or alcohol at the time of the
assault. Twenty percent of victims reported substance use at the
time of the rape. Victims were also asked whether the perpetrator
was using drugs or alcohol at the time of the incident. When vic
-
tims indicated that the perpetrator was using drugs, alcohol, or
both, the perpetrator drug/alcohol variable was coded 1. In 55%
of included cases, the victim reported that the assailant was using
alcohol and/or drugs.
Because previous research has found that attacks occurring out
of doors are more likely to be reported (Greenberg & Ruback,
1992), a variable for location of assault was included. Outside was
coded 1 if the assault took place at any out-of-doors location
including a street, alley, parking lot, car, park, rural area, lake,
dock, beach, or pool. If the assault occurred inside any public or
private building, the outside variable was coded 0.
9
Two demographic variables were included that have been
found in previous studies to be linked to rape reporting and/or
reporting of other serious violent crimes. Education has been pos-
itively linked to reporting of rape (Lizotte, 1985). Therefore, a
measure for college was included and coded 1 if the victim
reported having attended college (63%). Hindelang and Davis
(1977) found that minority victims of rape were more likely to
report than were White victims. Other studies have found, how-
ever, that African American women were less likely to report rape
than were White women (e.g., Greenberg & Ruback, 1992). To
account for racial differences in reporting, we included a dichoto
-
mous measure for race (African American = 1). Finally, we also
included a control variable for victim’s age in years at the time of
the assault.
RESULTS
CHANGES IN THE LIKELIHOOD OF REPORTING
The first question asks whether rapes committed after the pas
-
sage of reform statutes were more likely to be reported than those
committed before the reforms. To address this question, a series of
multivariate logistic regression models were constructed. The
164 VIOLENCE AGAINST WOMEN / February 2005
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
first model contains only the control variable, whereas the second
model adds the aggravated rape variable. The third model also
contains the dummy-coded variables for time (see Table 1).
As seen in the first model, the control variables contributed rel
-
atively little to the prediction of reporting. The model was statisti
-
cally significant, χ
2
= 14.74, p < .05, although the Nagelkerke R
2
value of .03 indicates that this set of variables is only weakly asso
-
ciated with reporting. The only significant variable was for victim
drug use, which indicates that, controlling for all other variables
in the equation, victims who were using drugs and/or alcohol at
the time of the assault were less likely to report to police than were
victims who were not using drugs or alcohol. When the variable
for aggravated rape was added (see Table 1, Model 2), the explan
-
atory power of the model increases significantly, –2 log likelihood
χ
2
(1, n = 824) = 93.21, p < .001. The odds ratio associated with the
aggravated rape variable (Exp β = 7.38) indicates that women
who were victims of aggravated rape were more than seven times
more likely to report than were victims of simple rape.
The addition of the two dummy-coded variables for year of the
assault also resulted in a significantly better fitting model, –2 log
likelihood χ
2
(2, n = 824) = 9.74, p < .01.Μost importantly, the vari-
able for the modern reform period was statistically significant,
indicating that rapes occurring during the modern reform period
(1990-1996) were more likely to be reported than rapes occurring
prior to reforms (see Table 1, Model 3). A rape occurring in the
modern reform period was 88% more likely to be reported than a
rape occurring prior to 1975 (Exp β= 1.88, p < .05). A sexual assault
that occurred in the middle reform period (1975-1989), however,
was no more likely to be reported than one occurring before the
enactment of reforms (Exp β = 0.91, p = .71).
In the full model, the variable for aggravated rape remains sig
-
nificant (p < .001). This indicates that aggravated rapes were sig
-
nificantly more likely to be reported than were simple rapes even
when taking into account the effects exerted by the time variables.
The additive model cannot tell us, however, whether the variable
for aggravated rape was more predictive of reporting prior to
reforms than in the periods following reforms. In other words,
has the gap between the reporting of simple and aggravated rape
narrowed over time? Answering this question requires the inclu
-
sion of an interaction term, as detailed below.
Clay-Warner, Burt / REPORTING RAPE 165
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
TABLE 1
Main Effects Models Predicting Rape Reporting (N = 824)
Model 1 Model 2 Model 3
β SE Exp ββ SE Exp ββ SE Exp β
Victim African American 0.17 0.29 1.19 0.39 0.32 1.47 0.29 0.33 1.33
Age at time of assault 0.02 0.01 1.02 0.02 0.01* 1.02 0.02 0.01 1.02
Attended college –0.09 0.19 0.92 –0.19 0.21 0.83 –0.18 0.21 0.83
Assaulted outside 0.21 0.25 1.23 0.11 0.27 1.12 0.19 0.27 1.21
Perpetrator drugs/alcohol 0.29 0.20 1.33 0.16 0.21 1.17 0.19 0.22 1.21
Victim drugs/alcohol –0.75 0.29* 0.47 –0.66 0.30* 0.52 –0.69 0.31 0.50
Completed rape 0.16 0.20 1.17 –0.25 0.21 0.78 –0.22 0.22 0.80
Aggravated rape 2.00 0.23*** 7.38 1.99 0.23*** 7.28
Early reform (1975-1989) –0.10 0.26 0.91
Modern reform (1990-present) 0.63 0.29* 1.88
Constant –2.21 0.35*** 0.11 –3.11 0.41*** 0.04 –3.13 0.43*** 0.04
–2 log-likelihood 736.31 643.10 633.36
χ
2
14.74* 107.95*** 117.68***
*p < .05. **p < .01. ***p < .001.
166
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
CHANGES IN THE LIKELIHOOD OF REPORTING
SIMPLE VERSUS AGGRAVATED RAPE
To examine whether rape type (simple vs. aggravated) had a
greater effect on reporting after the institution of reforms than
before, it was necessary to interact rape type with both time
period measures. These two interaction terms are included in the
analysis presented in Table 2. Neither term was significant thus
indicating that the effects of rape type on the likelihood of report
-
ing in the prereform period (before 1975) were not significantly
different from the effects of rape type on the likelihood of report
-
ing for either the early reform (1975-1989) or the modern reform
(1990-1996) eras.
DISCUSSION
These results indicate that although some changes have
occurred in the reporting of rape across time, not all goals related
to reporting have been realized. Importantly, there have been
changes in the likelihood of rape reporting over time. According
to the regression analyses, a rape that occurred in the modern
reform era (1990-1996) was significantly more likely to be
reported than one that occurred in the prereform period (before
1975). The size of the odds ratio indicates that this effect is sub-
stantively significant.
Reforms have been unsuccessful, however, in altering the
effects of rape type on reporting. Overall, aggravated rapes were
significantly more likely to be reported than simple rapes. These
effects were relatively stable across the three time periods; the
interaction terms revealed no significant differences in the effects
of rape type on reporting between the prereform era and either
period after reform. Thus, there is no statistical evidence to sug
-
gest that the gap in reporting between simple and aggravated
rapes has narrowed.
RECONCILING FINDINGS
WITH PREVIOUS RESEARCH
These findings are consistent with previous research suggest
-
ing that reforms have been only partially successful in changing
Clay-Warner, Burt / REPORTING RAPE 167
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
reporting behavior. Because this study asks somewhat different
questions and uses a different data source, however, the findings
reported here complement previous research by helping to draw
a more complete picture of the effects of reforms on reporting.
Earlier studies using official data reported that rape reforms
had little effect on the likelihood of reporting (Berger et al., 1994;
Spohn & Horney, 1992). As previously discussed, research using
only official data is limited, because it is impossible to determine
if the number of reported rapes has increased because of an
increase in reporting rates or an increase in the actual number of
rapes being committed. As a result, a decrease in the number of
rapes committed would obscure any changes in reporting across
time. Also, Spohn and Horney (1992) examined rape reporting
only through 1984. Our results indicate that significant changes in
reporting rates could not have been detected until after this
period.
Studies that utilized victimization data, however, found slight
increases in reporting postreform (Bachman & Paternoster, 1993;
Bryden & Lengnick, 1997). Our findings are consistent with this
research, although the effect sizes found in our data appear larger
than those in other studies. These other studies, however, used
data from the NCVS, which has been shown to dramatically
168 VIOLENCE AGAINST WOMEN / February 2005
TABLE 2
Interaction Model Predicting Rape Reporting (N = 824)
SE Exp β
Victim African American 0.28 0.33 1.32
Age at time of assault 0.02 0.01 1.02
Attended college –0.18 0.21 0.83
Assaulted outside 0.19 0.27 1.21
Perpetrator drugs/alcohol 0.19 0.22 1.21
Victim drugs/alcohol –0.70 0.31* 0.50
Completed rape –0.22 0.22 0.81
Aggravated rape 2.08 0.47*** 7.97
Early reform (1975-1989) 0.09 0.48 1.09
Modern reform (1990-present) 0.57 0.52 1.77
Early Reform × Aggravated Rape –0.27 0.57 0.77
Modern Reform × Aggravated Rape 0.09 0.61 1.09
Constant –3.21 0.53*** 0.04
–2 log-likelihood 632.85
χ
2
118.20***
*p < .05. ***p < .001.
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
underestimate rape victimizations (Fisher et al., 2000; see also
Koss, 1992), with some suggesting that those assaults not
reported to police are even more susceptible to underestimation
because women who do not report the crime to police often do not
identify themselves as victims thereby making them less likely to
report to an interviewer (Ruback, 1993). Because the NVAW Sur
-
vey uses behaviorally specific questions and does not make spe
-
cific reference to crime or victimization, a woman may report an
incident of sexual assault to the interviewer without actually
identifying herself as a crime victim. As a result, rape data in the
NVAW Survey are likely to be more accurate than rape data in the
NCVS. The fact that the NVAW Survey classifies women as rape
victims without the women themselves acknowledging their vic
-
timization also widens the scope of the analysis of reporting,
because it allows women to be included who may not be report
-
ing because they do not identify their victimization as a crime.
Because increasing awareness of rape as a crime, however, was
one goal of rape reform, it is important to include the universe of
rape victims and not only those who have acknowledged their
victimization. Including these unacknowledged rape victims also
lends further credibility to our significant findings for differences
in reporting across time.
Our general findings concerning simple rape and reporting are
also consistent with previous research. Although no other study
to our knowledge has specifically examined the effects of simple
versus aggravated rape on reporting, others have considered sep
-
arately whether certain aspects of rape that would distinguish an
assault as either simple or aggravated affect reporting. Following
Estrich (1987), we coded all stranger rapes as aggravated, as well
as any in which weapons were used or injury occurred. A great
deal of research finds that weapon use (Amir, 1971; Bachman,
1998; Lafree, 1980; Lizotte & Wolfson, 1981) and the degree of
physical injury (Bachman, 1993, 1998; Holmstrom & Burgess,
1978; LaFree, 1980; Lizotte, 1985; Lizotte & Wolfson, 1981) are pre
-
dictive of reporting. Assaults in which the offender has no prior
acquaintance with the victim are also considered aggravated, and
consistent with the results reported here, other studies find that
victims are more likely to report sexual violence when assaulted
by strangers (Feldman-Summers & Ashworth, 1981; Greenberg &
Clay-Warner, Burt / REPORTING RAPE 169
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
Ruback, 1992; Lizotte, 1985; Smith & Nelson, 1976; L. S. Williams,
1984).
What has not been previously examined, however, is whether
simple rapes were more likely to be reported in recent times than
before reforms or even immediately after their passage. Although
Bachman (1993) did not examine data from the prereform era, she
did find that in the period from 1987 to 1990, the prior relationship
between the victim and offender was not significantly related to
reporting. The prior relationship between the victim and
offender, of course, does not perfectly distinguish between simple
and aggravated rapes, because acquaintance rapes involving
weapons, injury, or multiple assailants would be considered
aggravated. Nonetheless, Bachman’s finding that acquaintance
rapes were no less likely to be reported in the late 1980s than were
stranger rapes suggests that the distinction between aggravated
and simple rape may also have little predictive power postreform
given that most acquaintance rapes take place without weapons
and additional injury (Burt, 1991). Our findings, however, indi-
cate not only that the distinction between aggravated and simple
rape remains an important predictor of reporting but also that the
impact of this variable did not differ significantly across time.
LIMITATIONS AND IMPLICATIONS
FOR RAPE LAW REFORM
Before considering the meaning of these findings for evaluat
-
ing the success of rape law reform, it is important to consider limi
-
tations associated with the analysis. One of the most significant
issues is possible memory decay. Because the NVAW Survey does
not employ a bounded survey design but instead seeks lifetime
prevalence data, survey participants are asked to provide details
about rapes that may have happened many years ago. As a result,
there may be inaccuracies in respondent recollections, and these
inaccuracies may increase over time. Underreporting may also
occur in the NVAW Survey, although as previously discussed, the
behaviorally specific screening questions reduce this risk. At the
same time, underreporting may be exacerbated by cohort effects
in that younger women may more readily identify forced sex than
older women who were likely socialized to hold more conserva
-
tive attitudes about women’s sexuality.
170 VIOLENCE AGAINST WOMEN / February 2005
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
Although we cannot know the extent to which the issues dis
-
cussed above actually affect the results reported here, we do
know that most of the problems associated with the NVAW Sur
-
vey are similar to problems of other victimization surveys.
Because the NVAW Survey uses behaviorally specific screening
questions that have been found to reduce underreporting, how
-
ever, the NVAW Survey is likely to produce a more representa
-
tive sample than many other surveys (e.g., NCVS). Furthermore,
the NVAW Survey remains the best source of data currently
available for analyzing changes in rape reporting, as it is the only
nationally representative survey that includes detailed informa
-
tion about both reported and unreported rapes occurring pre- and
postreform.
A more general consideration, however, is that it is impossible
to determine whether the changes that we find in rape reporting
are a direct result of the reforms or because of other societal
changes occurring at the same time. For example, it is quite possi-
ble that the feminist movement, which pushed for rape reform,
also brought about changes in the way women think about vio-
lence and their sexuality. The feminist movement stressed
women’s right to control their bodies and their sexuality. This
emphasis on ownership of one’s body has likely influenced the
way the public views forced sexual contact with more women
now identifying forced sexual contact as a criminal act. In this
way, the effects of the rape reform movement may be largely
symbolic.
Feminists involved in the reform movement, however, were
strongly motivated by the symbolic goals of rape reform. Such
groups were explicitly focused not only upon the measurable,
instrumental outcomes but also upon the ability of the law to
inform society about the gravity of sexual assaults (Bachman &
Paternoster, 1993; Bryden & Lengnick, 1997). Symbolic goals
focused on altering societal attitudes and providing visibility and
legitimacy to oppressed groups’ goals and values (Berger et al.,
1988). Although varying in intent, the effects of instrumental and
symbolic goals of rape law reform were designed to be comple
-
mentary, with alterations in society’s beliefs about “what rape ‘re
-
ally is’ and whom rape ‘really victimizes’” expected to increase
rape reporting (Bachman & Paternoster, 1993, p. 555). As a result,
increases in reporting can be seen as an indication of the success of
Clay-Warner, Burt / REPORTING RAPE 171
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
rape reform, whether the success is measured instrumentally or
symbolically.
One fact, however, still remains: The existing gap between
aggravated rape and simple rape is a clear indication that reforms
have not been completely successful. Not only did we find that
aggravated rapes were consistently more likely to be reported
than were simple rapes, but the importance of this classification in
predicting reporting was no less important in postreform periods
than in the period prior to reforms.
Reforms may have failed in improving the likelihood of report
-
ing simple rape for a number of reasons. Rational choice perspec
-
tives argue that victims report when they perceive that the bene
-
fits associated with reporting are greater than the costs. Reform
statutes were designed to reduce these costs by improving treat
-
ment of rape victims and increasing the benefits by improving the
chances that reporting would result in a conviction. Reforms may
have failed simply because they have not evoked enough changes
in the criminal justice system to make the benefits greater than the
costs. Indeed, rape victims continue to report dissatisfaction with
the criminal justice system (e.g., Frazier & Haney, 1996; Konradi,
1997, 1999). Evidence presented by Horney and Spohn (1996)
indicates, however, that there are few differences in criminal jus-
tice processing of simple versus aggravated rapes postreform
thus suggesting that the potential benefits associated with
reporting different types of rape are now roughly equivalent.
Rape victims also often incur great personal costs when report
-
ing simple rape, not the least of which is the risk of being ostra
-
cized by friends and family. It may be that the benefits obtained by
reporting do not currently outweigh these personal costs. If so,
then rape reform legislation will not be wholly effective until soci
-
etal attitudes also change. This issue brings the focus back to the
intersection of the instrumental and symbolic goals of reform. If
victims of simple rape are less likely to report than victims of
aggravated rape because of the ineffectiveness of reforms in alter
-
ing the criminal justice process, then instrumental goals have not
been met. If victims of simple rape are less likely to report than
victims of aggravated rape because of the ineffectiveness of
reforms in bringing about changes in societal perceptions of rape,
then symbolic goals have not been met. In either case, the fact that
simple rapes continue to be underreported relative to aggravated
172 VIOLENCE AGAINST WOMEN / February 2005
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
rapes is evidence that not all goals of rape reform have been
attained.
CONCLUSION AND DIRECTIONS
FOR FUTURE RESEARCH
The study reported here used a nationally representative data
set to examine changes in rape reporting since the institution of
rape reform. Because we employed a national data set, however,
we are addressing the effects of rape law reform in the aggregate.
Thus, although this research addresses questions concerning the
nationwide success of rape law reform, it also brings into view
other questions that need to be addressed concerning rape law
reform. For instance, are the findings reported here equally appli
-
cable across jurisdictions, or have some states been more success
-
ful than others in closing the reporting gap between simple and
aggravated rape? If some states have seen a significant increase in
the likelihood of reporting simple rape relative to aggravated
rape, then what specific statutes might explain the increase? Have
police and prosecutor policies that exist apart from statutory
reforms affected the reporting gap? Although these questions lie
outside our current frame, we hope that future research is able to
address these issues in state- and/or jurisdictional-level analyses.
Conducting such analyses would complement the data presented
here as well as guide future reform efforts.
Ultimately, our examination of data pre- and postreform indi
-
cate that although there has been an increase in the overall likeli
-
hood of a rape being reported, aggravated rapes continue to be
more likely to be reported than simple rapes. The importance of
rape type in predicting reporting remained unchanged across
pre- and postreform periods. Although we cannot determine
whether changes in reporting are directly attributable to reform,
the continuing gap between the reporting of aggravated and sim
-
ple rape suggests an unfinished agenda for rape law reform.
NOTES
1. Some states redefined the charge previously termed rape as criminal sexual conduct,
criminal sexual penetration, sexual battery, gross sexual imposition, or sexual abuse.
Clay-Warner, Burt / REPORTING RAPE 173
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
2. The Wisconsin Supreme Court in Brown v. State (1906) reversed a rape conviction on
the grounds that the woman had not resisted enough, although she both struggled and
screamed. The court’s reasoning was that “there must be the most vehement exercise of
every physical means or faculty within the woman’s power to resist the penetration of her
person, and this must be shown to persist until the offense is consummated” (p. 538).
3. Some states did retain a corroboration requirement for statutory rape (Kadish &
Schulhofer, 1995).
4. Controversy surrounds these statistics, as the number of reported rapes per 100,000
declined somewhat from 42.8 in 1992 to 39.2 in 1994.
5. The comparison was made to ascertain the impact of the law reforms rather than fac
-
tors affecting rape reporting trends, such as the increasing efficacy or punitiveness of the
criminal justice system as a whole (Bachman & Paternoster, 1993).
6. Interestingly, Lizotte (1985) found that weapon use and the presence of multiple vic
-
tims, however, increased the probability of reporting in nonsexual assaults, whereas no
significant effects were found for the reporting of rape.
7. The participation rate was “the number of completed interviews, including those
that were screened out as ineligible, divided by the total number of completed interviews,
screened-out interviews, refusals, and terminated interviews” (Tjaden & Thoennes, 1999,
p. 6).
8. These dummy-coded categories are also preferable to a continuous measure of time
(such as the year in which the rape occurred), because it may be many years after the
reforms that evidence of their effects on reporting can be seen. Rape reform can only affect
reporting to the extent that the general public is aware of the reforms and trusts in their
effectiveness. Thus, it is unlikely that rape reporting will increase in a linear fashion. As a
result, the use of a continuous measure of time would be inappropriate.
9. The National Violence Against Women Survey does not distinguish between assaults
occurring inside of a home and those occurring in the yard adjacent to the home. Therefore,
assaults occurring in a residential yard are necessarily coded as inside.
REFERENCES
Adler, Z. (1987). Rape on trial. London: Routledge & Keegan Paul.
Amir, M. (1971). Patterns in forcible rape. Chicago: University of Chicago Press.
Bachman, R. (1993). Predicting the reporting of rape victimization: Have rape reforms
made a difference? Criminal Justice and Behavior, 20, 254-270.
Bachman, R. (1998). The factors related to rape reporting behavior and arrest: New
evidence from the National Crime Victimization Survey. Criminal Justice and Behavior,
25, 8-29.
Bachman, R., & Paternoster, R. (1993). A contemporary look at rape law reform: How far
have we really come? Journal of Criminal Law and Criminology, 84, 554-574.
Baron, L., & Straus, M. A. (1989). Four theories of rape in American society. New Haven, CT:
Yale University Press.
Berger, R., Neuman, W. L., & Searles, P. (1994). The impact of rape law reform: An aggregate
analysis of police reports and arrests. Criminal Justice Review, 19, 1-23.
Berger, R., Searles, P., & Neuman, W. L. (1988). The dimensions of rape reform legislation.
Law and Society Review, 22, 331-357.
Bohmer, C. (1991). Rape and the law. In A. Parrot & L. Bechhofer (Eds.), Acquaintance rape:
The hidden crime (pp. 317-27). New York: John Wiley.
Bridges, J. S. (1991). Perceptions of date and stranger rape: A difference in sex role expecta
-
tions and rape supported beliefs. Sex Roles, 24, 291-308.
174 VIOLENCE AGAINST WOMEN / February 2005
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
Brown v. State, 106 N.W. 536 (127 Wis. 193, 1906).
Bryden, D., & Lengnick, S. (1997). Rape in the criminal justice system. Journal of Criminal
Law and Criminology, 87, 1194-1384.
Burt, M. R. (1991). Rape myths and acquaintance rape. In A. Parrot & L. Bechhofer (Eds.),
Acquaintance rape: The hidden crime (pp. 26-40). New York: John Wiley.
Caringella-McDonald, S. (1985). The comparability in sexual and nonsexual assault case
treatment: Did statute change meet objective? Crime and Delinquency, 31, 206-222.
Estrich, S. (1987). Real rape. Cambridge, MA: Harvard University Press.
Fairstein, L. (1993). Sexual violence: Our war against rape. New York: William Morrow.
Feldman-Summers, S., & Ashworth, C. D. (1981). Factors related to intentions to report a
rape. Journal of Social Issues, 37, 53-70.
Field, H., & Bienen, L. (1980). Jurors and rape: A study of psychology and the law. Lexington,
MA: Lexington Books.
Fisher, B. S., Cullen, F. T., & Turner, M. G. (2000). The sexual victimization of college women.
Washington, DC: U.S. Department of Justice, National Institute of Justice.
Frazier, P. A., & Haney, B. (1996). Sexual assault cases in the legal system: Police, prosecutor
and victim perspectives. Law and Human Behavior, 20, 607-628.
Futter, S., & Mebane, W. R. (2001). The effects of rape law reform on rape case processing.
Berkeley Women’s Law Journal, 16, 72-111.
Galvin, J. (1985). Rape: A decade of reform. Crime and Delinquency, 31, 163-168.
Greenberg, M., & Ruback, B. (1992). After the crime: Victim decision making. New York:
Plenum.
Hibey, R. A. (1973). The trial of a rape case. American Criminal Law Review, 10, 309-334.
Hindelang, M. J., & Davis, B. J. (1977). Forcible rape in the United States: A statistical pro-
file. In D. Chappell, R. Geis, & G. Geis (Eds.), Forcible rape: The crime, the victim, and the
offender (pp. 87-114). New York: Columbia University Press.
Holmstrom, L. L., & Burgess, A. W. (1978). The victim of rape: Institutional reactions. New
York: Wiley-Interscience.
Horney, J., & Spohn, C. (1991). Rape law reform and instrumental change in six urban juris-
dictions. Law and Society Review, 25, 117-153.
Horney, J., & Spohn, C. (1996). The influence of blame and believability factors on the pro-
cessing of simple versus aggravated rape cases. Criminology, 34, 135-162.
Kadish, S. H., & Schulhofer, S. (1995). Criminal law and its processes (5th ed.). New York:
Norton.
Kalvin, H., & Zeisel, H. (1966). The American jury. Boston: Little, Brown.
Kessler, E. (1992). Pattern of sexual conduct evidence and present consent: Limiting the
admissibility of sexual history evidence in rape prosecutions. Women’s Rights Law
Reporter, 14, 79-96.
Konradi, A. (1997). Too little too late: Prosecutors’ pre-court preparation of rape survivors.
Law & Social Inquiry, 22, 1101-1154.
Konradi, A. (1999). “I don’t have to be afraid of you”: Rape survivors’ emotion manage
-
ment in court. Symbolic Interaction, 22, 45-77.
Koss, M. P. (1992). The underdetection of rape: Methodological choices influence incidence
estimates. Journal of Social Issues, 48, 61-75.
LaFree, G. D. (1980). The effect of sexual stratification by race on official reactions to rape.
American Sociological Review, 45, 842-854.
LaFree, G. D. (1981). Official reactions to social problems: Police decisions in sexual assault
cases. Social Problems, 28, 582-594.
Largen, M. A. (1988). Rape-law reform: An analysis. In A. W. Burgess (Ed.), Rape and sexual
assault II (pp. 271-192). New York: Garland.
Legrand, C. E. (1973). Rape and rape laws: Sexism in society and law. California Law Review,
61, 919-941.
Clay-Warner, Burt / REPORTING RAPE 175
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
Lizotte, A. (1985). Uniqueness of rape: Reporting assaultive violence to the police. Crime
and Delinquency, 31, 169-190.
Lizotte, A., & Wolfson, F. (1981, August). The social location of rape and reporting: A victimiza
-
tion study. Paper presented at the annual meeting of the American Sociological Associa
-
tion, Montreal, Canada.
Madigan, L., & Gamble, N. C. (1991). The second rape: Society’s continued betrayal of the victim.
London: MacMillan.
Marsh, J. C., Geist, A., & Caplan, N. (1982). Rape and the limits of reform. Boston: Auburn
House.
McCahill, T., Meyer, L., & Fischman, A. (1979). The aftermath of rape. Lexington, KY:
Lexington Books.
Pollard, P. (1995). Rape reporting as a function of victim-offender relationship: Acritique of
the lack of effect reported by Bachman (1993). Criminal Justice and Behavior, 22, 74-80.
Robin, G. D. (1977). Forcible rape: Institutional sexism in the criminal justice system. Crime
and Delinquency, 23, 136-153.
Ruback, G. (1993). Comment on Bachman: The victim offender relationship does affect vic
-
tims’ decisions to report sexual assaults. Criminal Justice and Behavior, 20, 271-279.
Searles, P., & Berger, R. J. (1987). The current status of rape reform legislation: An examina
-
tion of state statutes. Women’s Rights Law Reporter, 10, 25-43.
Sebba, L. (1974). The requirement of corroboration in sex offenses. In I. Drapin & E. Viano
(Eds.), Victimology: A new focus (pp. 29-46). Lexington, MA: D. C. Health.
Shapo, H. S. (1975). Recent statutory developments in the definition of forcible rape. Vir-
ginia Law Review, 67, 1500-1543.
Smith, L., & Nelson, L. D. (1976, August). Predictors of rape victimization reportage.Paperpre-
sented at the annual meeting of the American Sociological Association, New York.
Spohn, C., & Horney, J. (1992). Rape law reform: A grassroots revolution and its impact. New
York: Plenum.
Tjaden, P., & Thoennes, N. (1999). Violence and threats of violence against women and men in the
United States, 1994-1996. Denver, CO: Center for Policy Research.
Williams, J. E., & Holmes, K. A. (1981). The second assault. Westport, CT: Greenwood.
Williams, L. S. (1984).The classic rape: When do victims report? Social Problems, 31, 459-477.
Jody Clay-Warner is an assistant professor of sociology at the University of
Georgia where she conducts research on women’s responses to sexual violence.
She also studies procedural justice in the workplace and in legal settings.Most
recently, her work has appeared in Violence and Victims and Social Psychol
-
ogy Quarterly.
Callie Harbin Burt is a doctoral student in the Department of Sociology at the
University of Georgia. In her M.A. thesis, she examined the mitigating effects of
gender identity on adolescents’ reactions to strain.
176 VIOLENCE AGAINST WOMEN / February 2005
at ARIZONA STATE UNIV on February 26, 2014vaw.sagepub.comDownloaded from
... Findings related to victim alcohol consumption and police reporting have been replicated by other researchers (Krebs et al., 2009;Lindquist et al., 2013) and among general population samples. Seriousness of physical injury or demonstrable harm (Bachman, 1993;Chon, 2014;Pino & Meier, 1999;Williams, 1984); presence of a weapon or use of force (Clay-Warner & Burt, 2005;Du Mont et al., 2003;Felson & Paré, 2005;Lizotte, 1985;Orcutt & Faison, 1988); victim-offender relationship (Chon, 2014;Du Mont et al., 2003;Felson & Paré, 2005); and victim characteristics such as age, education, race, and socioeconomic status (Clay-Warner & Burt, 2005) have influenced willingness to report. Survivors have been more likely to report to police if their experiences fit the classic rape narrative (Du Mont et al., 2003). ...
... Findings related to victim alcohol consumption and police reporting have been replicated by other researchers (Krebs et al., 2009;Lindquist et al., 2013) and among general population samples. Seriousness of physical injury or demonstrable harm (Bachman, 1993;Chon, 2014;Pino & Meier, 1999;Williams, 1984); presence of a weapon or use of force (Clay-Warner & Burt, 2005;Du Mont et al., 2003;Felson & Paré, 2005;Lizotte, 1985;Orcutt & Faison, 1988); victim-offender relationship (Chon, 2014;Du Mont et al., 2003;Felson & Paré, 2005); and victim characteristics such as age, education, race, and socioeconomic status (Clay-Warner & Burt, 2005) have influenced willingness to report. Survivors have been more likely to report to police if their experiences fit the classic rape narrative (Du Mont et al., 2003). ...
Article
Full-text available
Using a randomly assigned 2 (victim race) × 2 (alcohol consumption) between-subjects factorial design, this study used surveys from a sample of 571 undergraduate students at a mid-sized, public university in the United States to determine the effect of procedural justice on police referral after reading a sexual assault disclosure vignette. Multivariate binary logistic regression models demonstrated that positive perceptions of procedural justice increased police referral following sexual assault disclosure. Victim alcohol consumption and rape myth acceptance decreased police referral. Victim race, victim alcohol consumption, and participant sex did not moderate the effect of procedural justice on police referral. Implications are discussed.
... According to the findings of these studies, educational level has influence on individual's reporting rate. That is, people with higher level of education are more likely to report crime incidents to the police than those with less educational level (Clay-Warner & Burt, 2005). However, there have been contentions to this finding. ...
... Goudriaanb et al. (2006) found that victims with a job were less likely to report crime compared with unemployed victims. Finally, regarding the effect of victims' education, some researchers have noted that highly educated victims are more likely to report incidents to the police than less educated victims (Boateng, 2016;Clay-Warner & Burt, 2005). ...
Article
Relatively limited attention has been paid by scholars to explore the crime reporting behavior of white-collar crime victims, especially in developing countries, such as Azerbaijan, where some forms of white-collar crimes are widespread. Using the dataset of the first nationwide victimization survey (n = 1,214) in Azerbaijan, the current paper attempts to explore the determinants of crime reporting among 4 specific white-collar offense (fraud, request for bribery, sale of unsafe good and sale of unsafe food for consumption) among randomly selected, yet unrepresentative subsample of victims (n = 186). Offenses were categorized in two groups for analysis—financial/economic offenses and non-financial/economic offenses, hence two models in a binary logistic regression analysis. The results indicate that the extent of financial loss predicts the crime reporting behavior for victims of financial offenses. Those with a higher level of the perceived probability of being victimized by a violent crime were more likely to have notified the authorities of their victimization. Applicable for non-financial crimes only, the level of income has an inverse association with a decision to invoke the law. The relationship of a victim to an offender predicted a decision to contact the authorities—cases in which the offender was identified as a stranger were less likely to be reported to the law enforcement authorities than cases in which the offender had not been identified. No socio-demographic variable has a predictive capacity for either crime category. In addition, as the main motives for not mobilizing the law, almost half of the cases have been resolved in between the offender and victims, such as through compensation. The findings have several theoretical implications for white-collar crime literature. Suggestions for further research, as well as the limitations, are discussed toward the end of the paper.
... -Interest Group Activity Feminist mobilization, such as efforts by the Women's Equity Action League and Take Back the Night, provides another avenue through which women can collectively advocate policy preferences related to campus sexual assault. Historically, feminist mobilization has influenced various aspects of rape law reform, such as legal classifications (McMahon-Howard et al., 2009), police responses (Clay-Warner & Burt, 2005), and university reporting of rape incidents (Boyle et al., 2017). In an analysis of 413 universities from 2010 to 2013, Boyle et al. (2017) found that rape reporting and compliance with the Clery Act is higher among universities in states with more local American Association of University Women (AAUW) chapters and higher rates of feminist mobilization. ...
Article
Full-text available
The persistent problem of sexual assault on college campuses is receiving attention in both the public sphere and state legislatures. Although a considerable body of research examines various aspects of campus sexual assault, such as rates and reporting, scholars have not examined how state characteristics and interstate dynamics influence the policy process related to campus sexual assault. This gap is compounded by an underemphasis on gender in theories of state policy adoption, even as a record number of women serve in state legislatures. Drawing on a data set that captures the introduction and enactment of campus sexual assault legislation between 2007 and 2017, David R. Johnson and Liang Zhang examine in this article how the state policy adoption and diffusion framework explains the introduction and enactment of campus sexual assault policy. The results of their study show that the number of forcible sex offenses at public colleges, the number of female Democrats in state senates, contributions from women’s interest groups, gubernatorial power, Republican influence, and bipartisan sponsorship influence the campus sexual assault policy process, with varying influence across legislative stages. The authors discuss the implications of their findings for researchers interested in policy adoption and gender issues as well as for advocates working on campus sexual assault policy reform.
Article
Sexual assault (SA) decision-making literature primarily focuses on criminal-legal actors and often overlooks victim decision making. This relative dearth in research is problematic, as victims are principal gatekeepers of the criminal-legal process who influence whether perpetrators are arrested and prosecuted. Subsequent victim support is also contingent on the reporting decision. Overall, this body of research would benefit from a better understanding of how victims activate and participate with the criminal-legal system and the potential impact of these decisions on criminal-legal processes. Moreover, victim decision making is often situated in a theoretical analyses. Victim decision making is complex and should be studied within a criminological decision-making framework. Therefore, the current study relies on National Crime Victimization Survey (NCVS) data and applies a focal concerns perspective (FCP), informed by rape culture concepts, to examine why victims of sexual violence may or may not choose to report to legal authorities. The current study offers initial support for the application of FCP to victim reporting decisions. We found that victims consider each of the focal concerns (FC). Victims were more likely to report when offenders threatened them with harm (i.e., suspect blameworthiness), when the offense occurred in a private location (i.e., protection of the community), and when they sought help from victim support agencies or medical treatment (i.e., practical considerations). Additionally, we found that Black victims were more likely to report than other racial-ethnic groups (i.e., perceptual shorthand). These findings highlight a nexus between reporting to police and help-seeking via support agencies. Importantly, the results emphasize the importance for police to implement cultural competence and antiracist training to better support Black victims.
Article
A rape victim possesses a scarce resource: information about the crime. Thus, a victim's decision to report the crime to police, to allocate that resource, becomes an economic choice. A victim cannot receive social support or legal justice without revealing such information, but doing so creates real costs—social recrimination and lost privacy—with no guarantee of offender apprehension. This article explores the economics of the reporting and chronic nonreporting of rape in the context of this information‐allocation problem. The empirical analysis addresses the extent to which social‐support availability and evidentiary factors influence the reporting decision. Dichotomous and multinomial logit results, obtained using National Crime Survey data on a sample of rape victims, reveal how various demographic and crime‐specific factors explain the decision to report and the selection of specific reasons for not reporting. Some of these factors reflect circumstances addressable as matters of procedure or policy.
Article
Most sexual assault victims make the conscious decision to not report to the police. Concerns about the police appear to be prominent in victims' decision-making, but we do not know specifically what these concerns entail, and therefore cannot address them. A qualitative inquiry is needed to understand perceptions of the police in reporting decisions, from the voices of victims themselves. Participants ( N = 175) responded to an open-ended survey question about their decision to not report to the police following sexual victimization. Analyses revealed that police perceptions shaped the decision to not report for 36% ( n = 63) of participants. Nonreporting decisions related to the police fell into two themes: concerns of police treatment and previous/vicarious negative experiences with the police. We include recommendations for police and a call for alternative reporting options.
Article
This study is the first cross-national work to examine the impact of gender equality on both female homicide and rape victimization. The data set from the United Nations Office on Drugs and Crime (UNODC) was analyzed for 70 countries. Several gender equality measures were significantly and positively related to rape victimization, but were mostly not significant to female homicide. Findings for rape victimization were consistent with the backlash hypothesis, but such findings may be related to the limitations of police rape rates, such as different legal definitions and police-reporting behaviors across countries.
Article
Full-text available
Relying on the official rape rates, previous cross-national researchers found a significant and positive relationship between gender equality and the rates of reported rapes, and they used such findings as supporting evidence of the backlash hypothesis that increased gender equality leads to augmented violence against females. However, such a conclusion may be faulty because of the contested validity and reliability of the official rape rate. This study examined how the variation in a comprehensive legal definition of rape is associated with the rates of reported rapes across 74 countries by using the rates of reported rapes from the United Nations Office of Drugs and Crime. The regression analysis suggests that the comprehensive legal definition of rape, more than a country’s gender equality and developmental level, is related to the official rape rate. A country records a high official rape rate when the country includes a broad range of sexual assaults in the category of rape crime, suggesting a reevaluation of the existing studies on this topic.
Conference Paper
This thesis focuses on the relationship between feminism, post-traumatic symptoms, and acknowledgment of sexual assault in women. Part one is a literature review, which forms a conceptual introduction to the research area. The review highlights a paucity of research on the three concepts of feminism, PTSD and acknowledgment together, thus addresses the relationship between these concepts in pairs. The review also gives some context to the history of rape acknowledgment research, bringing attention to conflicting findings thus far. Part two, the empirical paper, studied women who had experienced an unwanted sexual encounter. The study sought to understand the relationship between feminist attitudes and self-identification, acknowledgment of sexual assault, and post-traumatic symptoms. In line with the literature review, the results suggested a complex relationship between acknowledgment and post-traumatic symptoms. Regression analysis indicate that characteristics of the assault statistically predict acknowledgment and post-traumatic symptoms, and the possible role of feminist attitudes is discussed. Part three contains a critical appraisal of the thesis. Reflections on the process of researching this topic are discussed, and methodological choices are considered and explained. It concludes with a discussion about what was learnt from the research, and how this has influenced thinking about the subject matter.
Chapter
Having presented the findings from 20 studies involving a variety of methodologies, we now introduce a theoretical framework that gives conceptual coherence to these findings. The proposed model, which emerged from our own empirical work and the work of others, offers several advantages. It provides an explanatory tool for comparing the decisions of various types of victims. In addition, it allows for the integration of the present findings with the existing literature on victim decision making. Further, the model suggests directions for future research, while at the same time laying the foundation for public policy decisions (see Chapter 10).