Sexual Violence Data - Actions Needed to Improve Clarity and Address Differences Across Federal Data Collection Efforts, GAO, 2016

Sexual Violence Data - Actions Needed to Improve Clarity and Address Differences Across Federal Data Collection Efforts, GAO, 2016

• Sept. 1, 2016 • Locations: United States of America • Topics: Sexual Assault, Statistics/Trends, Prison Rape Elimination Act

Download original document:
Brief thumbnail

Document text

This text is machine-read, and may contain errors. Check the original document to verify accuracy.

United States Government Accountability Office

Report to the Ranking Member,
Permanent Subcommittee on
Investigations, Committee on Homeland
Security and Governmental Affairs,
U.S. Senate
July 2016

SEXUAL VIOLENCE
DATA
Actions Needed to
Improve Clarity and
Address Differences
Across Federal Data
Collection Efforts

GAO-16-546

July 2016

SEXUAL VIOLENCE DATA

Highlights of GAO-16-546, a report to the
Ranking Member, Permanent Subcommittee
on Investigations, Committee on Homeland
Security and Governmental Affairs, U.S.
Senate

Actions Needed to Improve Clarity and
Address Differences Across Federal Data
Collection Efforts

Why GAO Did This Study

What GAO Found

Concerns have grown about sexual
violence—in general, unwanted sexual
acts—in the United States, particularly
involving certain populations such as
college students, incarcerated
individuals, and military personnel.
Data on the occurrence of sexual
violence are critical to preventing,
addressing, and understanding the
consequences of these types of
crimes. GAO was asked to identify and
compare federal efforts to collect data
on sexual violence.

Four federal agencies—the Departments of Defense, Education, Health and
Human Services (HHS), and Justice (DOJ)—manage at least 10 efforts to collect
data on sexual violence, which differ in target population, terminology,
measurements, and methodology. Some of these data collection efforts focus on
a specific population that the agency serves—for example, the incarcerated
population—while others include information from the general population. These
data collection efforts use 23 different terms to describe sexual violence. Data
collection efforts also differ in how they categorize particular acts of sexual
violence. For example, the same act of sexual violence could be categorized by
one data collection effort as “rape,” whereas it could be categorized by other
efforts as “assault-sexual” or “nonconsensual sexual acts,” among other terms.
In addition, five data collection efforts—overseen by Education, HHS, and DOJ—
reflect inconsistencies between their measurements and definitions of sexual
violence. Further, these data collection efforts do not have publicly-available
descriptions of what is included in their respective measurements to allow
persons using the data to understand the differences, which may lead to
confusion for data users. Publicly-available measurement information could
enhance the clarity and transparency of sexual violence data. Data collection
efforts also differ in terms of the context in which data are collected, data
sources, units of measurement, and time frames.

This report addresses two questions:
(1) What are the federal efforts
underway to collect data on sexual
violence, and how, if at all, do these
efforts differ? (2) How do any
differences across the data collection
efforts affect the understanding of
sexual violence, and to what extent are
federal agencies addressing any
challenges posed by the differences?
GAO reviewed agency documentation
and academic literature, and
interviewed agency officials.

What GAO Recommends
GAO recommends that Education,
HHS, and DOJ make information that
is included in their measurements of
sexual violence publicly available.
GAO also recommends that OMB
establish a federal interagency forum
on sexual violence data. Education,
HHS, and DOJ agreed with the
recommendation. OMB stated that
convening a forum may not be the
most effective use of resources at this
time, in part because the data
collection efforts are not far enough
along in their research. However, OMB
said it will consider convening or
sharing information across agencies in
the future.
View GAO-16-546. For more information,
contact Gretta L. Goodwin at (202) 512-8777
or goodwing@gao.gov.

Differences in data collection efforts may hinder the understanding of the
occurrence of sexual violence, and agencies’ efforts to explain and lessen
differences have been fragmented and limited in scope. Differences across the
data collection efforts may address specific agency interests, but collectively, the
differences lead to varying estimates of sexual violence. For example, in 2011
(the most recent year of available data), estimates ranged from 244,190 rape or
sexual assault victimizations to 1,929,000 victims of rape or attempted rape.
These differences can lead to confusion for the public. Officials from federal
agencies and entities GAO spoke with who use federal data on sexual violence
emphasized that the differences across the data collection efforts are such that
the results are not comparable, and entities reported using data that best suited
their needs. Agencies have taken some steps to clarify the differences between
the data collection efforts. For example, two DOJ entities coauthored a statement
that describes the differences between their two efforts. In addition, agencies
have taken some steps to harmonize the data collection efforts—that is,
coordinate practices to achieve a shared goal. However, actions to increase
harmonization have been fragmented, generally only involving 2 of the 10 data
collection efforts at a time, and limited in scope. The Office of Management and
Budget (OMB) through its authority to coordinate federal statistics has previously
convened interagency working groups, such as the Interagency Working Group
for Research on Race and Ethnicity, to improve federal statistics. OMB has no
plans to convene a working group on sexual violence data. Additional
collaboration, facilitated by OMB, between agencies that manage data collection
efforts about which differences help or hinder the overall understanding of sexual
violence could help to clarify the scope of the problem of sexual violence in the
United States.
United States Government Accountability Office

Contents

Letter

1
Background
Four Federal Agencies Manage Ten Efforts to Collect Data on
Sexual Violence, which Differ in Target Population,
Terminology, Measurements, and Methodology
Differences in Data Collection Efforts May Hinder the
Understanding of the Occurrence of Sexual Violence; Agencies
Have Taken Steps to Explain and Lessen Differences, but
Efforts Have Been Fragmented and Limited in Scope
Conclusions
Recommendations for Executive Action
Agency Comments and Our Evaluation

25
33
33
34

Appendix I

Objectives, Scope and Methodology

Appendix II

Descriptive Tables of Federal Data Collection Efforts

Appendix III

List of Relevant Articles and Reports

Appendix IV

Comments from the Department of Education

Appendix V

Comments from the Department of Health and Human Services

Appendix VI

Comments from the Department of Justice

Appendix VII

GAO Contact and Staff Acknowledgments

Table 1: Federal Efforts That GAO Identified That Include Data on
Sexual Violence

Tables

Page i

GAO-16-546 Sexual Violence Data

Table 2: Key Terms Used by Federal Data Collection Efforts to
Describe Sexual Violence
Table 3: Units of Measurement Reported by Federal Data
Collection Efforts
Table 4: Estimates of the Number of Rapes or Rape Victims Per
Year, 2010-2014, across Selected Federal Data
Collection Efforts
Table 5: Terminology Federal Data Collection Efforts Use to
Characterize Specific Acts of Sexual Violence Where a
Victim Was Penetrated
Table 6: Terminology Federal Data Collection Efforts Use to
Characterize Specific Acts of Sexual Violence Where a
Victim Was Made to Penetrate
Table 7: Terminology Federal Data Collection Efforts Use to
Characterize Specific Acts of Nonpenetrative or
Noncontact Sexual Violence
Table 8: Terminology Federal Data Collection Efforts Use to
Characterize Sexual Violence When Contextual Factors of
Consent Are Involved
Table 9: Terminology Federal Data Collection Efforts Use to
Characterize Sexual Violence When Contextual Factors of
Force Are Involved
Table 10: Terminology Federal Data Collection Efforts Use to
Characterize Sexual Violence When Other Contextual
Factors Are Involved
Table 11: Criminal Justice and Public Health Contexts for Federal
Data Collection Efforts
Table 12: Data Sources for Federal Data Collection Efforts
Table 13: Time Frames Used in Federal Data Collection Efforts
Concerning Sexual Violence

14
21
25
41
43
45
47
49
51
53
54
55

Abbreviations
BJS
CDC
Clery Act data
CNSTAT
DOD
DOJ
DMDC

Page ii

Bureau of Justice Statistics
Centers for Disease Control and Prevention
Data reported to the Department of Education under the
Jeanne Clery Disclosure of Campus Security Policy and
Campus Crime Statistics Act
Committee on National Statistics
Department of Defense
Department of Justice
Defense Manpower Data Center

GAO-16-546 Sexual Violence Data

DSAID
Education
FBI
HHS
NCS-X
NCVS
NEISS-AIP
NIS
NISVS
OCR
OIRA
OMB
OPSE
OVC
OVW
PRA
PREA
RMWS
SAPRO
SSV
UCMJ
UCR-NIBRS
UCR-SRS
VAWA
WGRA
WGRR

Defense Sexual Assault Incident Database
Department of Education
Federal Bureau of Investigation
Department of Health and Human Services
National Crime Statistics Exchange
National Crime Victimization Survey
National Electronic Injury Surveillance System-All Injury
Program
National Inmate Survey
National Intimate Partner and Sexual Violence Survey
Office for Civil Rights
Office of Information and Regulatory Affairs
Office of Management and Budget
Office of Postsecondary Education
Office for Victims of Crime
Office on Violence Against Women
Paperwork Reduction Act
Prison Rape Elimination Act
RAND Military Workplace Study
Sexual Assault Prevention and Response Office
Survey of Sexual Victimization
Uniform Code of Military Justice
Uniform Crime Reporting Program-National IncidentBased Reporting System
Uniform Crime Reporting Program-Summary Reporting
System
Violence Against Women Act
Workplace and Gender Relations Survey of Active Duty
Members
Workplace and Gender Relations Survey of Reserve
Component Members

This is a work of the U.S. government and is not subject to copyright protection in the
United States. The published product may be reproduced and distributed in its entirety
without further permission from GAO. However, because this work may contain
copyrighted images or other material, permission from the copyright holder may be
necessary if you wish to reproduce this material separately.

Page iii

GAO-16-546 Sexual Violence Data

Letter

441 G St. N.W.
Washington, DC 20548

July 19, 2016
The Honorable Claire McCaskill
Ranking Member
Permanent Subcommittee on Investigations
Committee on Homeland Security and Governmental Affairs
United States Senate
Dear Senator McCaskill:
There is concern about sexual violence in the United States, particularly
involving college students, incarcerated individuals, and military
personnel. Research has shown that sexual violence has long-lasting
effects on victims, including sexually-transmitted diseases, eating
disorders, anxiety, depression, and post-traumatic stress disorder. 1
Further, the economic costs of rape, including medical and social
services, loss of productivity, decreased quality of life, and law
enforcement resources, are estimated to range from $41,247 to $150,000
per incident. 2
According to research, victims often do not report sexual violence to law
enforcement officials due to feelings of guilt, shame, and embarrassment;
fear of the perpetrator; or fear of not being believed, among other

For the purposes of this report, GAO uses the term “sexual violence” to refer to a range
of unwanted sexual acts (including, for example, contact and noncontact sexual acts) on
which federal agencies collect information. This report does not define sexual violence;
instead, this report describes how data collection efforts refer to, define, and measure
what we are referring to in general as “sexual violence.”

T. R. Miller, M. A. Cohen, and B. Wiersema, “Victim Costs and Consequences: A New
Look,” a final summary report presented to the National Institute of Justice, January 1996;
M. Delisi, A. Kosloski, M. Sween, E. Hachmeister, M. Moore, and A. Drury, “Murder by
numbers: Monetary costs imposed by a sample of homicide offenders,” The Journal of
Forensic Psychiatry & Psychology, vol.21, no. 4 (2010), 501‐513; M. A. Cohen and A. R.
Piquero, “New Evidence on the Monetary Value of Saving a High Risk Youth,” Journal of
Quantitative Criminology, Vol. 25 (2008); and H. Fang, M. T. French, and K. E.
McCollister, “The Cost of Crime to Society: New Crime‐Specific Estimates for Policy and
Program Evaluation.” Drug Alcohol Dependence, Vol. 108, no.1‐2 (2010).

Page 1

GAO-16-546 Sexual Violence Data

reasons. Therefore, the occurrence of sexual violence is considered to be
underestimated. 3
Congress, state and local governments, and the public depend on data
collected by the federal government to prioritize resources, design
policies and programs that prevent and address crimes of sexual
violence, and evaluate those policies and programs. Federal data on
sexual violence are also used in crafting initiatives to mitigate the effects
of these experiences on victims and in gauging the effectiveness of these
initiatives.
You asked us to identify the primary sources of information on sexual
violence collected by the federal government and to identify any
challenges that may result from the differences across the data collection
efforts. This report addresses the following questions: (1) What are the
federal efforts underway to collect data on sexual violence, and how, if at
all, do these efforts differ? (2) How do any differences across the data
collection efforts affect the understanding of sexual violence, and to what
extent are federal agencies addressing any challenges posed by the
differences?
For the purpose of our analysis, we selected data collection efforts which
focus on data that:
•

provided information on the extent to which acts of sexual violence
occur in the United States in a particular year (for example, the
number of times an act of sexual violence occurred or the number of
victims of sexual violence);

•

were collected since 2010;

•

were collected at least once every 2 years;

3
J. Du Mont, K.Miller, and T.L. Myhr, “The Role of “Real Rape” and “Real Victim”
Stereotypes in the Police Reporting Practices of Sexually Assaulted Women,” Violence
Against Women, Vol. 9, No. 4 (2003), 466-486; National Academy of Sciences, Estimating
the Incidence of Rape and Sexual Assault (Washington, D.C.: National Academies Press,
2014), 25; and J.H. Gardella, C.A. Nichols-Hadeed, J.M. Mastrocinque, J.T. Stone, C.A.
Coates, C.J. Sly, and C. Cerulli, “Beyond Clery Act Statistics: A Closer Look at College
Victimization Based on Self-Report Data,” Journal of Interpersonal Violence, vol.30, no.4
(2015), 641, 643.

Page 2

GAO-16-546 Sexual Violence Data

•

were reported publicly; 4 and

•

were not focused primarily on minors. 5

We set these criteria to focus our analysis on data collection efforts that
provide current and regular information that is available to the public on
sexual violence. To narrow our scope, we chose to not include data
collection efforts that primarily focused on sexual violence against minors.
To identify data collection efforts that met our criteria, we reviewed past
GAO reports and a Federal Bureau of Investigation (FBI) list of federal
agencies that may collect crime data. 6 We asked officials at those
agencies if they had any data collection efforts that met our criteria. 7 We
also asked academic experts and officials from entities that use federal
data on sexual violence about any additional federal data collection
efforts they were aware of that met our selection criteria. 8 We identified
experts and officials from entities that use federal data on sexual violence
by conducting background research and by asking those experts and
officials that we interviewed to recommend additional experts and entities
that use federal data on sexual violence, including victim advocacy
groups, other special interest groups, and other federal agencies.
To identify and describe differences across the data collection efforts, we
obtained information on the purpose, scope, and methodology of each
4

Selected data collection efforts include information on victims who are 12 and older.

GAO, Military Personnel: Actions Needed to Address Sexual Assaults of Male
Servicemembers, GAO-15-284 (Washington, D.C.: Mar.19, 2015); Higher Education:
Experts Cited a Range of Requirements as Burdensome, GAO-13-371 (Washington, D.C.:
Apr. 10, 2013); Federal Statistical System: Agencies Can Make Greater Use of Existing
Data, but Continued Progress is Needed on Access and Quality Issues, GAO-12-54
(Washington, D.C.: Feb. 24, 2012); and Domestic Violence, Sexual Assault, Dating
Violence, and Stalking: National Data Collection Efforts Underway to Address Some
Information Gaps, GAO-11-833T (Washington, D.C., Jul.13, 2011).

7
FBI compiled the list of agencies as part of a Director’s Initiative to identify agencies that
could report to the Uniform Crime Reporting Program.
8
For the purposes of this report, we use the term “data collection effort” to identify a
compilation of information on sexual violence that meet our selection criteria.

Page 3

GAO-16-546 Sexual Violence Data

data collection effort. We obtained this information through a review of
documents, such as user manuals and program descriptions. We also
conducted interviews with senior agency officials, senior officials at
entities that use federal data on sexual violence, and academic experts.
To identify how the differences affect understanding of sexual violence,
we obtained and reviewed federal reports and interviewed and reviewed
relevant documentation from agency officials, experts, and
representatives from entities that use federal data on sexual violence. We
also reviewed articles, conference papers, and government reports that
discuss differences across federal sexual violence data collection efforts.
To identify these articles, our research librarian conducted a search of
several bibliographic databases, such as ProQuest, Embase, and
Scopus.
To describe the extent to which federal agencies are addressing any
challenges posed by differences across the data collection efforts, we
interviewed senior agency officials and academic experts and obtained
relevant agency documentation.
We conducted this performance audit from March 2015 to July 2016, in
accordance with generally accepted government auditing standards.
Those standards require that we plan and perform the audit to obtain
sufficient, appropriate evidence to provide a reasonable basis for our
findings and conclusions based on our audit objectives. We believe that
the evidence obtained provides a reasonable basis for our findings and
conclusions based on our audit objectives.

Background

The federal government has implemented a number of initiatives to
address sexual violence or mitigate its effects. For example, the
Department of Justice’s (DOJ) Office on Violence Against Women
(OVW), created in 1995 in order to help implement the Violence Against
Women Act (VAWA), sponsors grant programs for local law enforcement
agencies, prosecutors and judges, health care providers, and other
organizations that assist victims of sexual violence by providing, for
example, forensic medical services in sexual violence cases in rural areas
and specialized counseling services for victims from underserved

Page 4

GAO-16-546 Sexual Violence Data

populations. 9 Another office within DOJ, the Office for Victims of Crime
(OVC), convened crime victim advocates and experts in 2013 as part of
the Vision 21 Initiative and recommended in their report, among other
things, that federal agencies collaborate and expand the collection and
analysis of data on all forms of criminal victimization. The Department of
Health and Human Services’ (HHS) Family Violence Prevention and
Services Program supports two national resource centers on domestic
violence and special-issue and culturally-specific resource centers. In
addition, HHS’s College Sexual Assault Policy and Prevention Initiative
was launched in 2016 and is intended to provide support for organizations
that are implementing policies and practices at postsecondary schools to
prevent sexual assault on their campuses. The Department of
Education’s Office for Civil Rights (OCR) issued guidance to colleges and
universities, one in 2011 and another in 2014, concerning the
responsibilities of those institutions under Title IX of the Education
Amendments of 1972 with regard to addressing sexual violence against
students. OCR’s guidance sets standards for the grievance procedures
institutions must adopt and publish to promptly and equitably resolve
complaints brought by students alleging sex discrimination (including acts
of sexual violence and sexual harassment), and recommends preventive
education and training programs designed to reduce the occurrence of
sexual violence on campus and improve institutions’ responses to sexual
violence on campus when it does occur. In 2004, Congress passed a law
that required the Secretary of Defense to develop, among other things, a
comprehensive policy for the Department of Defense (DOD) on the
prevention of sexual assaults involving servicemembers. 10 In response to
that statutory requirement, DOD established its sexual-assault prevention
and response program in 2005, and in 2008, DOD published its first
sexual assault prevention strategy.

The Violence Against Women Act (VAWA) of 1994 and its subsequent 2000, 2005, and
2013 reauthorizations sought to strengthen protections for women facing sexual and other
forms of violence, including domestic violence. For example, VAWA enhanced federal
domestic violence penalties, required states to honor protective orders from other states
and provided a range of services for victims, for example by establishing the National
Domestic Violence Hotline.
10
See Ronald W. Reagan National Defense Authorization Act for Fiscal Year 2005, Pub.
L. No. 108-375, § 577 (2004).

Page 5

GAO-16-546 Sexual Violence Data

Several of the federal government’s responses to sexual violence involve
data collection on the occurrence of sexual violence. For example, the
Prison Rape Elimination Act (PREA) of 2003 directed DOJ to carry out
studies of the incidence and effects of prison rape. The Jeanne Clery
Disclosure of Campus Security Policy and Campus Crime Statistics Act,
as amended by VAWA in 2013, requires that all institutions of higher
education that participate in federal student financial assistance programs
disclose statistics on certain crimes, including those related to sexual
violence, to the Department of Education (Education). Since 2005,
National Defense Authorization Acts have directed DOD to, among other
things, collect and report information on sexual assaults against service
members.
Under the Paperwork Reduction Act (PRA), the Office of Information and
Regulatory Affairs (OIRA) in the Office of Management and Budget
(OMB) is charged with improving the efficiency and effectiveness of
federal information resources, which includes functions relating to
statistical policy and coordination. 11 Specifically, with regard to statistics,
OMB’s responsibilities include:
•

Oversight and approval of data collection: OMB reviews statistical
information collections as part of its responsibility under the PRA to
approve all federal agency information collections that will be
administered to 10 or more people to ensure adherence with PRA
standards for minimizing information collection burdens and
maximizing the practical utility of information collected by federal
agencies, including eliminating unnecessary duplication.

•

Guidance and standards: OMB develops and oversees
governmentwide policies, principles, standards, and guidelines for
collecting and disseminating statistical information.

•

Coordination: OMB coordinates the activities of the federal statistical
system, including ensuring the integrity, objectivity, and utility of
federal statistics.

•

Oversight of budgets: OMB ensures that statistical agencies’ budget
proposals are consistent with systemwide priorities for maintaining
and improving the quality of federal statistics.

44 U.S.C. § 3504.

Page 6

GAO-16-546 Sexual Violence Data

Other entities also provide guidance to agencies that conduct statistical
work. For example, the National Academy of Sciences’ Committee on
National Statistics (CNSTAT) publishes Principles and Practices for a
Federal Statistical Agency for newly appointed cabinet secretaries at the
beginning of each presidential administration. Principles and Practices
outlines basic principles for statistical agencies to carry out their missions
effectively, as well as practices designed to help implement them. 12
Different entities use federal data on sexual violence, including, for
example victim advocacy groups, other special interest groups, and other
federal agencies. Officials at victim advocacy groups we spoke with
publish reports on topics related to sexual violence and lobby Congress
for laws and programs designed to address the needs of victims. Other
groups include law enforcement associations and campus safety groups
that provide training and educational materials for law enforcement and
campus safety personnel. Federal agencies also use data on sexual
violence, for example to inform grant making decisions regarding
research and program development.

Many of the standards included in Principles and Practices for a Federal Statistical
Agency have been included in Statistical Policy Directive 1: Fundamental Responsibilities
of Federal Statistical Agencies and Recognized Statistical Units. Committee on National
Statistics, Division on Behavioral and Social Sciences and Education, National Research
Council, Principles and Practices for a Federal Statistical Agency (Washington, D.C.:
National Academies Press, 2013). Federal Register, vol. 29, no. 231 (Dec. 2, 2014),
71614.

Page 7

GAO-16-546 Sexual Violence Data

Four Federal
Agencies Manage
Ten Efforts to Collect
Data on Sexual
Violence, which Differ
in Target Population,
Terminology,
Measurements, and
Methodology
Federal Agencies Collect
Sexual Violence Data on
the General Population
and on Segments of the
Population

Four federal agencies manage at least 10 data collection efforts that
include data on sexual violence, among other things. 13 Some of these
data collection efforts focus on a target population that the agency serves.
For example, Education’s Clery Act data collection effort obtains
information on the occurrence of sexual violence at institutions of higher
education. 14 DOD’s Defense Sexual Assault Incident Database (DSAID)
and the Workplace and Gender Relations Survey of Active Duty Members
(WGRA) include and collect data on sexual violence involving military

We identified 10 data collection efforts that met our criteria, for example that data are
collected recently (i.e., 2010 or after); are reported publicly; and do not focus primarily on
minors. For a full description of the criteria we used in selecting data collection efforts, see
appendix I. We asked agency officials to identify data collection efforts that met these
criteria. It is possible that agencies have other data collection efforts that met our criteria
and were not included in the scope of this review.
14

The Clery Act data collection effort includes information on sexual violence that occurred
on or near campuses owned or controlled by postsecondary schools that participate in
federal student aid programs under Title IV of the Higher Education Act of 1965. Also,
information on sexual violence at institutions of higher education has recently been
collected by the Bureau of Justice Statistics (BJS). In 2015, BJS, along with RTI
International, conducted the Campus Climate Survey Validation Study. The survey was
web-based, confidential, and completed by 23,000 undergraduates (15,000 women and
8,000 men) at 9 colleges and universities. Because the study was a one-time effort, it did
not meet the criteria for inclusion in our analysis.

Page 8

GAO-16-546 Sexual Violence Data

service members. 15 Others, such as the FBI’s Uniform Crime Reporting
Program (UCR) data collection efforts compile data from law enforcement
agencies on the general population. 16 Those data collection efforts that
include information from the general population differ in terms of the ages
of respondents or individuals from whom reports of sexual violence are
taken. For example, the National Intimate Partner and Sexual Violence
Survey (NISVS) collects data from individuals who are 18 and older while
the National Crime Victimization Survey (NCVS) collects data on
household members who are 12 and older; both the Uniform Crime
Reporting Program-Summary Reporting System (UCR-SRS) and the
Uniform Crime Reporting Program-National Incident-Based Reporting
System (UCR-NIBRS) include data from law enforcement agencies on
criminal incidents involving people of all ages. Table 1 includes

GAO is currently conducting a separate review of DSAID, examining, among other
things, the extent to which DSAID has met its mandated requirements and how
modernization efforts address known issues. Regarding the WGRA, in 2014 DOD
contracted with RAND to conduct an independent assessment of the survey and, if
necessary, to update its methodology and to administer the 2014 WGRA. That year RAND
created and administered two versions of the survey. One version of the survey employed
DOD’s prior measure of “unwanted sexual contact” to estimate the past-year prevalence
of sexual assault in DOD. The other survey version—called the RAND Military Workplace
Study (RMWS)—employed a new measure of sexual violence that was designed to align
with the terminology used and corresponding categories of crimes specified in Articles 80
and 120 of the Uniform Code of Military Justice. In its Annual Report on Sexual Assault in
the Military for Fiscal Year 2015, DOD stated that the next WGRA will utilize the same
methodology as developed by RAND. For the purposes of this report, we refer broadly to
this data collection effort as the “WGRA.” DOD also administers the Workplace and
Gender Relations Survey of Reserve Component Members (WGRR). Because WGRR
methods align with WGRA methods, this report focuses on the Workplace and Gender
Relations Survey of Active Duty Members.
16
The FBI’s Uniform Crime Reporting Program-Summary Reporting System (UCR-SRS)
and Uniform Crime Reporting Program-National Incident-Based Reporting System (UCRNIBRS) program are related, for example, in that NIBRS data can be summarized and
used to populate UCR-SRS data fields. However the data collection efforts differ, for
example in terms of measurements and definitions of sexual violence, as discussed later
in this report.

Page 9

GAO-16-546 Sexual Violence Data

information about each of the 10 data collections discussed in this report,
including their respective target populations. 17
Table 1: Federal Efforts That GAO Identified That Include Data on Sexual Violence

Year
Data Collection Effort
Effort
Began
Agency
Component
Department Sexual Assault Defense Sexual 2012
of Defense Prevention and Assault Incident
(DOD)
Response
Database
(SAPRO)
(DSAID)a
Defense
Manpower
Data Center
(DMDC)

Department
of
Education
(Education)

Office of Postsecondary
Education
(OPSE)

Department
of Health &
Human
Services
(HHS)

Centers for
Disease
Control and
Prevention
(CDC)

Summary
DSAID includes case information on
unrestricted and restricted reports of
allegations of sexual assault committed
against members of the U.S. Armed
Forces, during or prior to their service.b
Workplace and 1995;
WGRA is a survey that calculates
Gender
began
prevalence data on sexual assault, sexual
Relations
collecting harassment, and gender discrimination
Survey of Active sexual
among active duty military personnel.
Duty Members
violence
d
(WGRA)
data in
2006
Clery Act data
1990
Clery Act data are collected under the
Jeanne Clery Disclosure of Campus
Security Policy and Campus Crime
Statistics Act, which requires all
postsecondary institutions participating in
federal student financial assistance
programs to disclose statistics on certain
crimes, including sexual offenses that
occur on or near campuses that the
institution owns or controls.
National
2000
NEISS-AIP collects data from a nationally
Electronic Injury
representative sample of hospital
Surveillance
emergency departments in the United
System-All Injury
States regarding visits for injuries.
Program
(NEISS-AIP)e

Target
Population
Military
populationc

Military
population
(active duty)

Institutions of
higher
education
receiving
federal student
aid funds

General
population

This report identifies the various federal agency holdings of sexual violence data by
referring to their commonly used names, for example, DSAID, which is the name of a
DOD system that holds military sexual violence data, and Clery Act data, which describes
the sexual violence data reported by institutions of higher education to the Department of
Education under the Clery Act. The use of these names is for the purpose of identifying
the various data collection efforts, and is not intended to describe the full extent of the
information holdings or the use of such systems.

Page 10

GAO-16-546 Sexual Violence Data

Agency

Component

Year
Data Collection Effort
Effort
Began
National Intimate 2010
Partner and
Sexual Violence
Survey (NISVS)

Department Bureau of
National Crime
of Justice
Justice
Victimization
(DOJ)
Statistics (BJS) Survey (NCVS)

1973

National Inmate
Survey (NIS)

2007

Survey of
Sexual
Victimization
(SSV)
Federal Bureau Uniform Crime
of Investigation Reporting
(FBI)
Program –
Summary
Reporting
System (UCRSRS)
Uniform Crime
Reporting
Program –
National
Incident-Based
Reporting
System (UCRNIBRS)

2004

1929

1989

Target
Population
General
population
(noninstitutionalized
adults, age 18
and over)
NCVS is a household survey that collects General
data from about 160,000 individuals from a population
nationally representative sample of about (noninstitutionalized,
90,000 households on the frequency,
characteristics, and consequences of
age 12 and
criminal victimization in the United States. over)
Summary
NISVS is a telephone survey that collects
information on intimate partner violence,
sexual violence, and stalking of adults in
the United States.

NIS falls under the Prison Rape
Elimination Act of 2003 (PREA), which
requires BJS to obtain data on the
incidence and effects of prison rape from a
sample of federal, state, county, and
municipal correctional facilities. NIS
collects sexual assault incident data
directly from inmates at those facilities.
SSV also falls under PREA and collects
information from the correctional facilities
on reported allegations of sexual
victimization by inmates or staff.
UCR-SRS consists of reported data on
criminal offenses received from over
18,000 city, university/college, county,
state, tribal, and federal law enforcement
agencies nationwide.

Inmates of adult
correctional
facilities,
including
juvenilesf

UCR-NIBRS collects detailed data from
over 6,500 law enforcement agencies for
each single incident reported to police
where one or more criminal offenses were
committed.

General
population

Inmates of adult
and juvenile
correctional
facilitiesf
General
population

Source: GAO analysis of agency documentation. | GAO-16-546
a

DSAID does not include information on child sexual abuse or sexual assault between intimate
partners.
b

An unrestricted report of sexual assault in the military is a report that is provided to military criminal
investigation organizations and/or law enforcement for investigation. A restricted report of sexual
assault in the military is a report that allows victims to confidentially access medical care and
advocacy services without triggering an investigation. Victims may convert their restricted report to an
unrestricted report at any time and participate in the military justice process.

Page 11

GAO-16-546 Sexual Violence Data

DSAID includes data on some civilians (e.g., spouses of servicemembers) who are eligible to receive
services.
d

In 2014 DOD contracted with RAND to conduct an independent assessment of the WGRA and, if
necessary, to update the WGRA methodology and to administer the 2014 WGRA. That year RAND
created and administered two versions of the survey. One version of the survey employed DOD’s
prior measure of “unwanted sexual contact” to estimate the past-year prevalence of sexual assault in
DOD. The other survey version—called the RAND Military Workplace Study (RMWS)—employed a
new measure of sexual violence that was designed to align with the terminology used and
corresponding categories of crimes specified in Articles 80 and 120 of the Uniform Code of Military
Justice. In its Annual Report on Sexual Assault in the Military for Fiscal Year 2015, DOD stated that
the next WGRA will utilize the same methodology as developed by RAND. For the purposes of this
report, we refer broadly to this data collection effort as the “WGRA.” DOD also administers the
Workplace and Gender Relations Survey of Reserve Component Members (WGRR). Because
WGRR methods align with WGRA methods, this report focuses on the Workplace and Gender
Relations Survey of Active Duty Members.
e

Data for NEISS-AIP is collected by the Consumer Product Safety Commission under an interagency
agreement with HHS/CDC.
f

NIS and SSV data cover federal, state, and local prisons and jails, as well as correctional facilities
operated by the U.S. military and U.S. Immigration and Customs Enforcement, as well as privately
operated prisons and jails. Neither data collection effort includes court or police lockup facilities. SSV
reports on adult (including juveniles in adult facilities) and juvenile facilities separately; NIS only
reports on adult facilities, including juveniles in those facilities.

Data collection efforts that are focused on target populations—such as,
the military population, institutions of higher education, and the
incarcerated population—provide information on the problem of sexual
violence within those groups and thus may be helpful for informing policy
affecting those groups. For example, Education officials told us that
Education’s Office of Federal Student Aid administers inquiries to specific
campuses if Clery Act data show unusually high incidences of certain
crimes, including rape. The Bureau of Justice Statistics (BJS) reports that
the National Inmate Survey (NIS) and the Survey of Sexual Victimization
(SSV) data provide helpful information for understanding and addressing
the problem of sexual violence in prisons, jails, and juvenile correctional
facilities. 18 In March 2015, we reported on the importance of using military
data on sexual violence to inform program decisionmaking. 19
The 5 data collection efforts whose target population is a segment of the
national population are the result of specific congressional mandates,
while the 5 data collection efforts that focus on the general population are

BJS is a division of the DOJ’s Office of Justice Programs.

GAO-15-284.

Page 12

GAO-16-546 Sexual Violence Data

discretionary initiatives arising from broad agency missions. 20 For
example, BJS is mandated under the terms of the Prison Rape
Elimination Act (PREA) of 2003 to collect data on sexual violence in
prisons, jails, and other detention facilities. BJS conducts both the NIS
and SSV in response to that mandate. 21 In contrast, according to HHS’s
Centers for Disease Control and Prevention (CDC) officials, the agency
launched NISVS as part of its public health mission, and with support
from the National Institute of Justice and DOD as a result of requests
from organizations in the field of sexual violence prevention.
The extent to which the data collection efforts focus on sexual violence
also varies. Some of the data collection efforts collect information solely
or primarily on the occurrence of sexual violence, such as the SSV or
DSAID. Other data collection efforts have a larger focus. For example, the
UCR-SRS, UCR-NIBRS, and NCVS include information on a broad
spectrum of crimes, and the National Electronic Injury Surveillance
System–All Injury Program (NEISS-AIP) includes information on a wide
variety of types of injuries.

Data Collection Efforts
Differ in the Terms They
Use to Describe Sexual
Violence, and
Measurements and
Definitions for Some
Efforts are Inconsistent

Data collection efforts use a range of terms to describe sexual violence in
publicly-available agency documentation. Specifically, the 10 data
collection efforts use a total of 23 different terms to describe sexual
violence. Table 2 shows the terms that data collection efforts use to
describe sexual violence.

DSAID, WGRA, Clery Act data, NIS, and SSV have target populations that represent a
small subset of the overall population—the military, institutions of higher education
communities, or the incarcerated. The National Electronic Injury Surveillance System–All
Injury Program (NEISS-AIP), NISVS, NCVS, UCR-SRS and UCR-NIBRS focus on the
general population.
21

BJS’s PREA data collection activities also include two additional data collection efforts,
the National Former Inmate Survey and the National Survey of Youth in Custody, which
did not meet the selection criteria for inclusion in our review.

Page 13

GAO-16-546 Sexual Violence Data

Table 2: Key Terms Used by Federal Data Collection Efforts to Describe Sexual Violence
Clery Act NEISSdata
AIP

NISVS

NCVS

NIS

SSV

UCRSRS

UCRNIBRS

—

Being Made to Penetrate
Someone Else

—

Fondling

—

Forcible Sodomy

—

Nonconsensual Sexual Acts

—

Noncontact Unwanted Sexual
Experiences

—

Nonpenetrative Sexual Assault

—

Penetrative Sexual Assault

—

Sex Offenses

—

Rape

—

Sexual Assault

—

Sexual Assault with an Object

—

Sexual Coercion

—

Sodomy

—

DSAIDa

WGRAb

—

Aggravated Sexual Contact

—

Assault-sexual

—

Attempted Penetrative Sexual
Assault

—

Attempts to Commit Offenses

Abusive Sexual Contact

Staff Sexual Harassment

—

Staff Sexual Misconduct

—

Unwanted Sexual Contact

—

Unwilling Activity

—

Willing Activity

—

Table legend:
DSAID: Defense Sexual Assault Incident Database, Department of Defense (DOD)
WGRA: Workplace and Gender Relations Survey of Active Duty Members, Department of Defense
Clery Act data: Data reported to the Department of Education under the Jeanne Clery Disclosure of Campus Security Policy and Campus Crime
Statistics Act
NEISS-AIP: National Electronic Injury Surveillance System-All Injury Program, Department of Health and Human Services
NISVS: National Intimate Partner and Sexual Violence Survey, Department of Health and Human Services
NCVS: National Crime Victimization Survey, Department of Justice
NIS: National Inmate Survey, Department of Justice
SSV: Survey of Sexual Victimization, Department of Justice
UCR-SRS: Uniform Crime Reporting Program-Summary Reporting System, Department of Justice

Page 14

GAO-16-546 Sexual Violence Data

UCR-NIBRS: Uniform Crime Reporting Program-National Incident-Based Reporting System, Department of Justice
Source: GAO analysis of agency documentation. | GAO-16-546
a

In 2014 DOD contracted with RAND to conduct an independent assessment of the WGRA and, if
necessary, to update the WGRA methodology and to administer the 2014 WGRA. That year RAND
created and administered two versions of the survey. One version of the survey employed DOD’s
prior measure of “unwanted sexual contact” to estimate the past-year prevalence of sexual assault in
DOD. The other survey version—called the RAND Military Workplace Study (RMWS)—employed a
new measure of sexual violence that was designed to align with the terminology used and
corresponding categories of crimes specified in Articles 80 and 120 of the UCMJ. In its Annual Report
on Sexual Assault in the Military for Fiscal Year 2015, DOD stated that the next WGRA will utilize the
same methodology as developed by RAND. For the purposes of this report, we refer broadly to this
data collection effort as the “WGRA” and use the measures of sexual violence developed by RAND—
”penetrative sexual assault,” “nonpenetrative sexual assault,” and “attempted penetrative sexual
assault”—for our analysis.
c

The National Inmate Survey uses the term “abusive sexual contacts only.”

The Survey of Sexual Victimization uses the term “abusive sexual contacts.”

Agencies may collect data on certain sexual offenses, for example, sexual harassment, in other data
collection efforts not included in this review.

Given the variation in terminology, data collection efforts may
characterize the same sex act using different terms. For example,
regarding sexual violence involving vaginal penetration of a victim, 6 data
collection efforts include this act of sexual violence in their measurement
of “rape,” 2 include it in their measurements of “nonconsensual sexual
acts” or “staff sexual misconduct,” 2 include it in their measurements of
“sexual assault” or “assault-sexual,” 1 includes it in its measurement of
“sexual coercion,” 1 includes it in its measurement of “penetrative sexual
assault,” and 1 includes it in its measurement of “sexual assault with an
object.” 22 See tables 5 through 7 in app. II for additional information on
acts of sexual violence included in measurements of sexual violence by
data collection effort.
It is also the case that one data collection effort may use multiple terms to
characterize a particular act of sexual violence, depending on the

22
Some data collection efforts include certain acts of sexual violence in their measurement
of more than one term. For example, UCR-NIBRS includes vaginal penetration of a victim
in its measurement of “rape” or “sexual assault with an object,” depending on the
circumstances of the incident.

Page 15

GAO-16-546 Sexual Violence Data

contextual factors that may be involved, such as whether the perpetrator
used physical force. For example, if a victim is penetrated vaginally,
NISVS may characterize that particular act as either “rape” or “sexual
coercion,” and the decision as to which term is most appropriate is based
on the contextual factors surrounding the act. As such, NISVS
characterizes vaginal penetration of a victim as “rape” if the act involves
the use of physical force or threats to physically harm the victim. On the
other hand, NISVS characterizes this same act as “sexual coercion” if the
act occurs after the victim is verbally pressured in a nonphysical way, for
example if the perpetrator uses their influence or authority.
Based on our analysis, data collection efforts rarely use the same
terminology to describe sexual violence; however, when they do, there
are some differences in the particular acts of sexual violence and
contextual factors that they include in their measurements of those terms.
For example, 4 of the 6 data collection efforts that use the term “rape”
consider whether actual physical force was used and the other two do
not. Three of the 6 that use the term “rape” consider whether the threat of
physical force was used and the other 3 do not. See tables 8 through 10
in app. II for additional information on contextual factors included in
measurements of sexual violence by data collection effort.
In general, measurements of sexual violence closely relate to definitions
of sexual violence in federal data collection efforts. For 5 of the data
collection efforts we reviewed, the acts of sexual violence and contextual
factors that are included in the measurements generally align with the
acts of sexual violence and contextual factors that are included in the
definitions.
However, for the remaining 5 data collection efforts, some of the acts of
sexual violence or contextual factors are included in both their
measurements and definitions and others are not. Further, these data
collection efforts do not have publicly-available descriptions of what is
included in their respective measurements to allow persons using the
data to understand the differences. Specifically, Clery Act data include
attempted rapes in its measurement of rape, but do not include attempted
rapes in its definition of rape. NEISS-AIP includes acts of sexual violence
involving penetration of a victim with an object and acts of sexual violence
involving a victim being made to penetrate someone else with an object in
its measurement of assault-sexual, but does not explicitly include these
acts of sexual violence in its definition of assault-sexual or in the
description of assault-sexual in the NEISS-AIP coding manual.
Additionally, NCVS includes the contextual factors of “victim unable to

Page 16

GAO-16-546 Sexual Violence Data

consent or refuse” and “victim alcohol/drug facilitated” in its
measurements of rape and sexual assault, but does not include these
contextual factors in its definitions of rape and sexual assault. Similarly,
SSV includes attempted nonconsensual sexual acts in its measurement
of nonconsensual sexual acts, but does not include attempts in its
definition of nonconsensual acts. NIS includes the act of victim
penetration with an object in its measurements of nonconsensual sexual
acts and staff sexual misconduct, but does not include it in its definitions
of nonconsensual sexual acts and staff sexual misconduct.
The National Academy of Sciences’ Principles and Practices for a Federal
Statistical Agency states that data releases from a statistical program
should include the methods and assumptions used for data collection and
reporting. Similarly, OMB guidelines regarding the federal Information
Quality Act call for agencies that disseminate government information to
ensure its utility, objectivity, and integrity, which refers to information
reproducibility and transparency. Additionally, federal internal control
standards state that an agency’s information requirements should
consider the expectations of both internal and external users and that
reliable internal and external information sources should provide data that
faithfully represent what they purport to represent. 23 Education officials
told us that they are updating The Handbook for Campus Safety and
Security Reporting and expect to issue the updated handbook in summer
2016. This may provide an opportunity for Education to eliminate
discrepancies between the Clery Act’s sexual violence measurements
and definitions. 24 Regarding NEISS-AIP, CDC officials told us that the
definition of “assault-sexual” is intended to include the range of sexual
assault experiences that victims presenting to the emergency department
may report. BJS officials told us that it is not possible to enumerate every
act of sexual violence that is included under NCVS’s terms of sexual
violence and that data users can make their own determinations about
what acts of sexual violence are included in the measurements. BJS
officials also told us that the definitions included in the NIS and SSV

GAO, Standards for Internal Control in the Federal Government, GAO-14-704G
(Washington, D.C.: Sept. 10, 2014).

In responding to a draft of this report, Education informed us that it issued the updated
handbook in June 2016, which provides clarification on its measurements of sexual
violence; however, the differences between the measurements and definitions remain.

Page 17

GAO-16-546 Sexual Violence Data

summary reports are intended for the general public, and acknowledged
that the reports present ambiguity regarding which acts of sexual violence
and contextual factors are included in the measurements. CDC and BJS
officials told us that researchers who are interested in descriptions of
what is included in the measurements for NEISS-AIP, NIS, and SSV
could access coding information, which is available at the University of
Michigan’s National Archive of Criminal Justice Data of the Interuniversity
Consortium for Political and Social Research. However, a layperson who
is not a researcher may not know how to access this information.
If data users are seeking to understand what sexual acts and contextual
factors are included in a data collection effort’s measurement of sexual
violence, they may read the definitions of terms contained in reports of
those data collections. If the definitions of the terms are different from
what the data collection effort includes in its measurement of those terms,
data users may lack clarity about what acts of sexual violence and
contextual factors the efforts are including in their measurements of
sexual violence.

Federal Sexual Violence
Data Collection
Methodologies Differ by
Context, Data Source,
Units of Measurement,
and Time Frames
Context

Federal agencies generally collect data on sexual violence within one of
two contexts—criminal justice or public health. For the purposes of this
report, “criminal justice” describes data collection efforts that refer to acts
of sexual violence as “crimes” or “offenses.” “Public health” describes
data collection efforts that seek to understand the health implications of
acts of sexual violence. Of the 10 data collection efforts within our scope,
7 collect data primarily in a criminal justice context, and 2 collect data
primarily in a public health context, and 1 data collection effort combines
both contexts. Table 11 in app. II outlines which data collection efforts fall
into each category.
According to agency officials, context can determine how each data
collection effort is designed and how the data are collected—and context
may inform what is included in the measurements of sexual violence in
the data collection efforts. The data collection efforts of BJS and FBI

Page 18

GAO-16-546 Sexual Violence Data

included in this report have a criminal justice focus, intended to collect
information on crimes, victims, or trends. CDC officials stated that their
data collection efforts—NEISS-AIP and NISVS—which both have a public
health focus, are more concerned with assessing the health impacts of
victimization and informing violence prevention efforts and are less
concerned with categorizing incidents as crimes. In some public health
surveys, interviewers begin by asking basic health and lifestyle questions
to establish a rapport with the interviewee and introduce the concepts of
public health and experiences rather than crimes and criminal events.
According to BJS and CDC officials, some studies with a focus on
criminal justice may ask questions about sexual violence as crimes that
have occurred, whereas some studies with a public health focus may ask
questions about violent sexual experiences by describing specific acts of
sexual violence while avoiding criminal terminology. For instance, NISVS
does not use the word rape when questioning interviewees, whereas
NCVS asks respondents if anyone has been attacked for example by
rape, attempted rape, or other type of sexual attack. NISVS program
documentation states that the term rape may carry a stigma or have
different meanings to different people, so the survey poses multiple
questions about behaviorally specific sexual acts, without using the term
rape.

Data Sources

Federal sexual violence data primarily come from two sources:
information reported to authorities and information obtained from victim
surveys. Some data collection efforts compile information reported to
relevant authorities. Other federal data collection efforts obtain their data
from surveys where agencies attempt to identify victims from a larger
population and invite them to share information about their experiences
with sexual violence. Table 12 in app. II outlines which data collections
fall into each category.
Information reported to authorities may originate from situations in which
a victim or observer reports an alleged act of sexual violence to law
enforcement, campus, or prison authorities or to military officials. Data
collection efforts vary in how and to whom the information is submitted.
For DSAID, restricted and unrestricted reports of sexual violence are
reported by victims to sexual assault response coordinators or victim
advocates who input information on the incident into DSAID. Whereas for
Clery Act data, all institutions of higher education that receive federal
student financial aid are required to report campus security data
(including information on sexual violence) to the Department of
Education.

Page 19

GAO-16-546 Sexual Violence Data

Some data collection efforts obtain information on sexual violence
through surveys. Each survey uses different methods to collect data from
their subjects. For instance, NISVS employs a random digit dialing
telephone survey while the NCVS uses a mix of face-to-face and
telephone interviewing.
Both types of data sources involve tradeoffs. With respect to information
reported to authorities, according to agency documentation and a senior
official from a law enforcement special interest group, data collection
efforts that provide information on crimes reported to authorities are
useful for administrative and funding decisions related to law
enforcement. For example, the Bureau of Justice Assistance uses UCRSRS data, in part, to determine how much grant funding should be
awarded to state, local, and tribal governments through the Edward Byrne
Memorial Justice Assistance Grant program, which made $255.7 million
in funding available to states, territories, and localities in fiscal year 2015.
However, one limitation is that these data may underestimate the scope
of the problem, since acts of sexual violence are historically
underreported to authorities. 25
As such, obtaining information through a survey may identify more
instances of sexual violence than efforts that rely on information reported
to authorities. However, surveys also have their limitations. Surveys are
subject to variable response rates over time, and different surveys may
have different response rates, which may affect the resulting estimates
and the validity of the data. For example, response rates of the data
collection efforts included in our review range from 24 percent for WGRA
in 2012 to 33 percent for NISVS in 2011 to 84 percent for NCVS in
2014. 26 Survey results may also be subject to response biases, for
example the tendency for a respondent to provide untruthful but socially
acceptable responses or the tendency for individuals who either have or
have not experienced the action (e.g., sexual violence) to not participate

25
Du Mont, Miller, and Myhr, “The Role of “Real Rape,” 468; National Academies of
Sciences, Estimating the Incidence, 40; and Gardella, Nichols-Hadeed, Mastrocinque,
Stone, Coates, Sly, and Cerulli, “Beyond Clery Act Statistics,” 641, 643.
26

The NISVS 2011 weighted cooperation rate, which reflects the proportion of persons
contacted who agreed to participate in the survey and who were determined to be eligible,
was 83.5 percent.

Page 20

GAO-16-546 Sexual Violence Data

in the survey (which can lead to nonresponse bias). 27 Also, according to
BJS officials, obtaining information directly from victims creates a burden
on survey respondents and interview subjects, and research and officials
from an entity that uses federal data on sexual violence stated that such
surveys may face difficulties in getting people to discuss victimization
experiences with strangers during interviews. Furthermore, administrative
costs associated with surveys and interviews can affect the practical
frequency of data collection.

Units of Measurement

Different federal data collection efforts measure and report different
aspects of the occurrence of sexual violence. Some data collection efforts
report the number of incidents that involved an act of sexual violence,
some report the number of unique victims of sexual violence, and some
report information about the number of times an act of sexual violence
occurred. On the surface it may appear that the number of incidents that
involve an act of sexual violence and the number of times an act of sexual
violence occurred are synonymous, but that is not necessarily the case.
Multiple offenses could occur in the same incident, and one incident could
involve multiple victims. For example, a perpetrator could both rob and
sexually assault someone in the same incident, or a perpetrator could
carry out sexually violent acts against multiple victims in the same
incident.
Table 3 outlines the units of measurement that each data collection effort
reports.
Table 3: Units of Measurement Reported by Federal Data Collection Efforts
Data
Collection
Effort

Number of Incidents or
Reports that Involve
Sexual Violence

Number of
Victims of Sexual
Violence

Number of Times an
Act of Sexual
Violence Occurred

DSAID

WGRA

—

Clery Act

—

NEISS-AIP

—

NISVS

According to DOD officials, the Defense Manpower Data Center conducts non-response
bias studies on the WGRA to identify potential non-response bias.

Page 21

GAO-16-546 Sexual Violence Data

Data
Collection
Effort

Number of Incidents or
Reports that Involve
Sexual Violence

Number of
Victims of Sexual
Violence

Number of Times an
Act of Sexual
Violence Occurred

NCVS

NIS

—

SSV
UCR-SRS
UCR-NIBRS

—

Table legend:
DSAID: Defense Sexual Assault Incident Database, Department of Defense (DOD)
WGRA: Workplace and Gender Relations Survey of Active Duty Members, Department of Defense
Clery Act data: Data reported to the Department of Education under the Jeanne Clery Disclosure of
Campus Security Policy and Campus Crime Statistics Act
NEISS-AIP: National Electronic Injury Surveillance System-All Injury Program, Department of Health
and Human Services
NISVS: National Intimate Partner and Sexual Violence Survey, Department of Health and Human
Services
NCVS: National Crime Victimization Survey, Department of Justice
NIS: National Inmate Survey, Department of Justice
SSV: Survey of Sexual Victimization, Department of Justice
UCR-SRS: Uniform Crime Reporting Program-Summary Reporting System, Department of Justice
UCR-NIBRS: Uniform Crime Reporting Program-National Incident-Based Reporting System,
Department of Justice
Source: GAO analysis of agency documentation. | GAO-16-546
a

In its annual report on sexual assault in the military, DOD reports the number of reports of sexual
violence by category of offenses (for example, the total number of reports involving “rape”).
b

WGRA and NISVS collect some data about the number of times an act of sexual violence occurs
among their respective subject populations, but do not report totals or estimates of those data.
c

NCVS collects information on the number of incidents involving sexual violence and the number of
times an act of sexual violence occurred, but its reporting of incident counts is limited.
d

The 2013 and 2010 NIS reports presents information on the number of acts of sexual violence
experienced in ranges (e.g., 1, 2, 3 to 5, 6 to 10), and by percent of victims that experienced a
number of acts of sexual violence in a specific range. The 2007 NIS report also uses ranges to
present information on the number of acts of sexual violence experienced.
e

SSV reports data on the number of victims of substantiated incidents of sexual violence.

UCR-SRS defines an incident as an event in which one or more criminal offenses occur. In its annual
report, “Crime in the United States,” UCR-SRS reports information about offenses involving sexual
violence. In its annual report, “Hate Crime Statistics,” UCR-SRS reports information on both incidents
and offenses, including for hate crimes involving rape.
g

UCR-NIBRS reports the number of victims of sexual violence offenses but does not report a number
of unique victims due to possible duplication.

The 4 data collection efforts that report the number of times an act of
sexual violence occurred—NCVS, NIS, UCR-SRS, and UCR-NIBRS—
may not record the exact number of acts of sexual violence from each
individual incident in some cases. These data collection efforts count one

Page 22

GAO-16-546 Sexual Violence Data

act of sexual violence per victim per incident, meaning they would capture
acts against multiple victims in a single incident but not multiple acts
against each individual victim.
The data source each federal data collection effort uses is an important
determinant of the units of measurement that each study uses in reporting
its results. Data collection efforts that use information reported to
authorities generally measure the number of incidents or reports that
involve sexual violence. Those efforts may or may not publish a count of
the number of acts of sexual violence that occurred. Such efforts are
generally not set up to count the number of unique victims across
incidents; they may report a count of victims, but have no mechanism to
avoid the double-counting of victims who experience multiple incidents
and are thus not measuring the same quantity that surveys that seek to
report a number of unique victims measure. By contrast, data collection
efforts that use information obtained from victim surveys generally
measure and report the number of victims rather than the number of
incidents that involved sexual violence. These efforts also collect data on
the number of separate times each individual respondent has been a
victim of sexual violence. However, the agencies operating those studies
are cognizant of the challenges associated with asking respondents about
multiple incidents, particularly with respect to the respondent’s ability to
accurately recall multiple experiences of sexual violence. For example, a
DOD official stated that WGRA asks questions about how many times a
crime has been committed against the respondent, but DOD does not
report a total number of times that an act of sexual violence occurred
because respondents’ recall of multiple events may be subject to memory
biases and confirming details of each act may be overly burdensome in a
survey. 28

Time Frames

Federal agencies collect sexual violence data for different periods of time
and report the data at different frequencies. Table 13 in app. II outlines
the time frames for each federal data collection effort.

28
Telescoping is an example of memory bias, where a respondent displaces the
experiences in time, for example perceiving remote events as having occurred more
recently. Recall bias is another example of memory bias, where a respondent may not
remember a past event with accuracy and completeness.

Page 23

GAO-16-546 Sexual Violence Data

The data collection efforts cover different intervals of time for which an act
of sexual violence occurred. For instance, NISVS asks whether each
respondent has experienced sexual violence during the previous 12
months and during the respondent’s lifetime. 29 NCVS asks whether each
respondent has experienced sexual violence during the previous 6
months. By contrast, the data collection efforts that compile reports from
authorities may capture information at the point in time when the event
was reported to those authorities.
Some data collection efforts release their results annually, whereas
others do so less often. For example, NISVS collected data annually
except in 2014 and issues reports periodically. NISVS most recently
reported results to the public in 2014 using data collected in 2011. Most
data collection efforts that compile reports from authorities release their
results annually.
One expert we interviewed stated that the period of time for which data on
sexual violence are collected and how often data are publicly reported
affects each data collection effort’s results regarding the occurrence of
sexual violence, an observation also found in academic literature and a
nongovernmental report. 30 For example, data on lifetime experience of
sexual violence may yield larger numbers of rape and sexual assault than
data on experiences of sexual violence in the last 6 or 12 months.
According to officials from 2 entities that use federal data on sexual
violence, data that are reported annually may be more useful for trend
analysis than data reported less frequently.

NISVS also asks survey respondents about their age of first victimization.

30
R. Bachman, “Measuring Rape and Sexual Assault: Successive Approximations to
Consensus,” Paper presented to the National Academy of Sciences, Jun.6, 2012; and
National Academy of Sciences, Estimating the Incidence of Rape and Sexual Assault
(Washington, D.C.: National Academies Press, 2014).

Page 24

GAO-16-546 Sexual Violence Data

Differences in Data
Collection Efforts May
Hinder the
Understanding of the
Occurrence of Sexual
Violence; Agencies
Have Taken Steps to
Explain and Lessen
Differences, but
Efforts Have Been
Fragmented and
Limited in Scope

The differences across the data collection efforts may hinder
understanding of the extent of sexual violence, and agencies have taken
steps to clarify differences and harmonize the data collection efforts.
However, these efforts have been fragmented and more could be done to
increase understanding of the problem of sexual violence.

Differences in Data
Collection Efforts
Collectively May Hinder
Understanding of the
Extent of Sexual Violence

Collectively, the differences across federal data collection efforts lead to
differing estimates of sexual violence, for example rape, in the United
States, as shown in selected data collection efforts on the general
population in table 4.

Table 4: Estimates of the Number of Rapes or Rape Victims Per Year, 2010-2014, across Selected Federal Data Collection
Efforts
Estimated Number of Rapes or Victims per Yeara
What crime or
action the count
includes

Unit of Measure

National Intimate Partner
and Sexual Violence
Survey (NISVS)

Rapeb

Victims

National Crime
Victimization Survey
(NCVS)

Rape/sexual assaultd

Victimizations

Uniform Crime ReportingSummary Reporting
System (UCR-SRS)

Forcible rape/ Rapee

Offenses reported

Data collection effort

2010

2011

2012

1,270,000

1,929,000

not yet
available

2013

2014

268,570

244,190

346,830

300,170

284,350

85,593

84,175

85,141

113,695

116,645

not yet data not
availablec collected

Source: GAO analysis of agency documentation. | GAO-16-546
a

Information in the table reflects the most current data available from selected data collection efforts.

Page 25

GAO-16-546 Sexual Violence Data

The NISVS estimate includes completed and attempted rape of female victims. NISVS contains
estimates of lifetime prevalence for male victims of rape, but the case count for male victims of rape
does not allow the Centers for Disease Control and Prevention (CDC) to produce a reliable estimate
for a 12-month prevalence rate.
c

According to CDC officials, in 2013 pilot data on survey changes was collected.

In the Bureau of Justice Statistics’ annual report of NCVS data, “Criminal Victimization,” data on the
number of sexual violence victimizations are reported as “rape/sexual assault” which includes rape or
sexual assault.
e

The Federal Bureau of Investigation (FBI) broadened its definition of rape for UCR-SRS in 2013 and
removed the word “forcible” from the name of the offense. The data in this table for years prior to
2013 use FBI’s old definition of “forcible rape;” for 2013 and 2014, the data use the revised definition.

According to research and individuals we spoke with who are familiar with
the data, differences in federal data on sexual violence may confuse the
public. 31 A National Academy of Sciences’ Panel that studied the
incidence of rape in the United States reported in 2014 that the data
collection efforts’ different purposes and methodologies produce different
results, which creates confusion for the public, law enforcement,
policymakers, researchers, and victim advocacy groups. Additionally,
officials from four entities that use federal data told us that they believe
the public does not understand data from federal sources on sexual
violence. Officials from three entities that use federal data on sexual
violence stated that they—and the media—cite a range of sources and
may not always, or adequately, explain the details of the data collection
efforts. In addition, the public may not take the time to understand the
differences among the data collection efforts. For example, an official
from one entity told us that the entity frequently uses the results from one
particular data collection effort to educate the public on sexual violence,
but the official was not aware of certain methodological details and
limitations of the data. Also, as previously discussed, some data
collection efforts’ measurements and definitions do not align and
information on what is included in these measurements is not publicly
available, which may lead to confusion for data users. Further, officials
from the federal agencies and entities we spoke with that use federal data
on sexual violence emphasized that the differences across the data
collection efforts are such that the results are not comparable.

J. Lynch, “Clarifying Divergent Estimates of Rape from Two National Surveys,” The
Public Opinion Quarterly, vol. 60, no. 3 (1996); and National Academy of Sciences,
Estimating the Incidence, 32.

Page 26

GAO-16-546 Sexual Violence Data

Officials we spoke with who use the data stated that differences in
measurements, definitions, and methodology across the data collection
efforts can lead to confusion. Officials at one entity stated that they found
challenges in using federal data on sexual violence because varying
measurements and definitions across the data collection efforts make it
difficult to compare data. However, even in instances where the acts of
sexual violence and contextual factors that are included in measurements
and definitions are similar across data collection efforts, other differences
create challenges. For example, officials at the National Center for
Campus Public Safety stated that whereas the Clery Act data program
and UCR-SRS use the same definition for rape, the methodologies are
different—the Clery Act data program collects information on allegations
made in “good faith,” whereas the UCR-SRS includes only information on
incidents resulting in a police report—which results in different estimates
of rape and may lead to confusion for users who try to compare the data.
Because there is wide variation in the results, entities that use federal
data on sexual violence have a choice of which data to use, and entities
reported using data that best suited their needs. For example, officials
from one entity told us that they use NCVS data because it includes
information on incidents not reported to the police and is user-friendly,
and officials from another entity told us they use NCVS data because it
has a larger sample size than other data collection efforts. Officials from
another entity stated that they use NISVS because it includes the most
expansive set of acts of sexual violence and contextual factors in its
measurement of “rape” and estimates lifetime prevalence rates.

Page 27

GAO-16-546 Sexual Violence Data

Agencies Have Taken
Some Steps to Clarify
Differences across Sexual
Violence Data Collection
Efforts, but Efforts to
Harmonize the Data
Collection Efforts Have
Been Fragmented

Federal agencies have acknowledged that differences exist among data
collection efforts, for example in terms of methodology, context, and data
sources, which has led some agencies to take steps to identify and
explain differences across the data collection efforts. In addition, some
federal agencies have taken steps to lessen the differences among data
collection efforts by focusing on harmonization—that is, coordination of
practices to enhance data collection to achieve a shared goal. However,
such efforts have been fragmented—that is, they are limited in scope and
generally involve two data collection efforts at a time. 32
Two ongoing efforts are intended to clarify the differences across two
data collection efforts:
•

BJS and FBI coauthored a statement describing the differences
between UCR and NCVS. In 1995, BJS and FBI coauthored
statements entitled “The Nation’s Two Crime Measures,” which
describe similarities and differences between the UCR program and
NCVS. The statement was updated in September 2014. BJS and FBI
post similar but different versions of the statement on their websites.
BJS’s statement provides a side-by-side description of FBI’s UCR
program and BJS’s NCVS including, for example, information on
historical background, data sources, and time frames for data
collection and reporting. Both statements also include a section on
comparing UCR and NCVS data, which describes the data collection
efforts’ similarities (e.g., they both have somewhat similar subsets of
serious crimes, such as rape, robbery, aggravated assault, burglary,
theft, and motor vehicle theft ) and key differences (e.g., definitions of
certain crimes). The statements conclude with a description of the two
data collection efforts’ strengths (e.g., UCR provides data on the
number of crimes reported to law enforcement and NCVS provides
data on the number and types of crimes not reported to law
enforcement).

•

CDC and BJS have discussed publishing a statement that
compares sexual violence statistics in NISVS and NCVS. At a
meeting in November 2015, CDC and BJS discussed coauthoring a

According to our work, fragmentation refers to those circumstances in which more than
one federal agency (or more than one organization within an agency) is involved in the
same broad area of national need and opportunities exist to improve efficiency.

Page 28

GAO-16-546 Sexual Violence Data

statement about NISVS and NCVS that would describe the
differences and similarities of the two data collection efforts.
There are five efforts underway or recently implemented that are intended
to increase harmonization across the data collection efforts, including: 33
•

Education adopted FBI’s UCR-SRS definition of rape for use in
the Clery Act data. Education, in its 2014 rule implementing the
VAWA 2013 reauthorization, changed the definition of rape that is
used in the Clery Act data to match UCR-SRS’ definition. The
definition used by both UCR-SRS and the Clery Act data is
“Penetration, no matter how slight, of the vagina or anus with any
body part or object or oral penetration by a sex organ of another
person, without the consent of the victim.” Education began collecting
data using the new definition of rape for calendar year 2014.

•

BJS sponsored the National Academy of Sciences’ CNSTAT
Panel on Estimating the Incidence of Rape and Sexual Assault in
BJS Household Surveys. In March 2011, BJS charged the panel to
“assess the quality and relevance of statistics on rape and sexual
assault from NCVS and other surveys contracted for by other federal
agencies as well as surveys conducted by private organizations,”
examining issues such as the “legal definitions in use by the states for
these crimes, best methods for representing the definitions in survey
instruments so that their meaning is clear to respondents, and best
methods for obtaining as complete reporting as possible of these
crimes in surveys, including methods whereby respondents may
report anonymously.” The panel, which was comprised of experts in
research and sexual assault response, held five in-person meetings
and issued 15 recommendations in a final report published in 2014.
For example, the panel recommended that BJS’s definitions of sexual
violence be expanded to include victimizations when the victim does
not have the capacity to consent to the sexual actions of the offender
and that this research be conducted in a coordinated manner because
many of the issues to be investigated are interrelated. The panel also
recommended that the survey questionnaire should have a neutral

Federal agencies have a range of other efforts underway to improve individual data
collection efforts. For example, BJS, with support from FBI, has launched the National
Crime Statistics Exchange (NCS-X) to generate detailed national estimates of crimes
reported to law enforcement by actively working to create a sample of 400 law
enforcement agencies to initiate their reporting of crime data to UCR-NIBRS.

Page 29

GAO-16-546 Sexual Violence Data

context, such as a health survey. 34 BJS officials told us that some of
their current work addresses some of the panel’s recommendations.
For example, BJS is currently conducting a methodological
comparison of NCVS with a public health approach that includes a
five-city comparison study and consultation with CDC. The five-city
comparison study involves an expanded scope of sexual violence,
which includes questions regarding consent, and BJS is testing a
range of behavior-specific questions. BJS officials told us that they are
planning to issue a report on the progress of the project in spring
2016. 35
•

BJS and FBI commissioned a National Academy of Sciences’
CNSTAT Panel on Modernizing the Nation’s Crime Statistics.
Commissioned in 2013, the panel will assess and make
recommendations for the development of a modern set of crime
measures in the United States and the best means for obtaining them.
The review will focus, among other things, on full and accurate
measurement of criminal victimization events and their attributes,
considering types of crime (and their definitions), including the current
scope of crime types covered by existing FBI and BJS data
collections; gaps in knowledge of contemporary crime; development
of international crime classification frameworks that should be
considered in increasing international comparability; and the optimal
scope of crime statistics to serve the needs of the full array of data
users and stakeholders—-federal agencies, other law enforcement
agencies, Congress, other actors in the justice system (such as the
courts and corrections officials), researchers, and the general public.
Panel membership includes academics in the field of criminal justice
and statistics and stakeholders who use and provide the data that the
government collects. According to the chair of the panel, as part of the
first phase of work, the panel has developed an initial
conceptualization for classifying all types of crime, including rape and
sexual assault. The recommended classification and its justification
appear in the panel’s first report, which was released in May

34
The panel did not make recommendations on how surveys using a public health
approach might be improved. With limited time and resources, the panel made the
decision to focus its analysis on NCVS with the intent to make specific recommendations
to BJS for estimates of rape and sexual assault.
35
In responding to a draft of this GAO report, BJS did not comment on the progress of this
project.

Page 30

GAO-16-546 Sexual Violence Data

2016. The panel has also begun the second phase of its work, which
will suggest the means for gathering data for the comprehensive
crime classification, including information from non-BJS or FBI
sources, and recommend how crime data collection should proceed in
practice. The panel plans to consider possible coordination among
agencies to produce more comprehensive reports on data, instead of
one agency producing one report and another agency producing a
separate but related report. The panel intends to finish its second
phase of work in 2016 and issue a final report in early 2017.
•

CDC, partnering with BJS, plans to convene a Technical Expert
Panel to examine ways to improve NISVS. As part of OMB’s review
of a CDC information collection proposal, OMB requested that CDC
and BJS officials convene a panel of experts in survey methods to
improve NISVS’s methodology, including increasing the response rate
and minimizing non-response bias. The agencies identified and
recruited a panel of experts and plan to meet in spring 2016. 36

•

CDC provided DOD with an adapted dataset from NISVS. In 2010,
NISVS included two random samples of active duty women and wives
of active duty men in addition to a random sample of the general U.S.
population. Using identical survey methods, data were collected in the
first two quarters of 2010. According to DOD officials, CDC provided a
subset of NISVS data to DOD so that DOD received information on
crimes that fell under military law to enable DOD to do an “apples to
apples” comparison of CDC and DOD data. CDC officials informed us
that they are working on a follow-up military population study in 2016.

However, these various federal efforts to clarify and harmonize sexual
violence data have been fragmented. CNSTAT’s Principles and Practices
for a Federal Statistical Agency (2013) calls for federal agencies that
produce similar federal statistics, with different missions, to coordinate
and collaborate with each other to meet current information needs and
provide new or more useful data than a single system can provide. While
the guidance applies primarily to the 13 federal statistical agencies, OMB
officials stated that the report provides best practices for all federal data
collection activities. The guidance encourages collaborative interagency
efforts and highlights the importance of agencies developing standard
definitions as a way to maximize the value and comparability of data.

36
In responding to a draft of this GAO report, CDC did not comment on the progress of
this panel.

Page 31

GAO-16-546 Sexual Violence Data

However, the coordination that has occurred across the agencies that
collect data on sexual violence has been limited. Specifically, coordination
has been bilateral—generally involving only 2 of the 10 data collection
efforts at a time and limited in scope. Agency officials expressed
skepticism that broader harmonization efforts could benefit federal data
on sexual violence, stating that each data collection effort is designed—
through target population, measurements and definitions, and
methodology—to fulfill a certain purpose. Agency officials told us that
changes to any of these differences may undermine the specific purpose
of each data collection effort. However, harmonization does not
necessarily entail making the data collection efforts identical; instead, it
could entail agencies considering how they could make their efforts more
complementary and, as appropriate, more alike without compromising
their programmatic needs.
The Paperwork Reduction Act (PRA), among other things, established a
process for OMB to oversee agency information collection efforts in order
to improve the quality and use of federal information while reducing
collection burdens, including through the coordination of federal
statistics. 37 Per its authority under the PRA, OMB has convened
interagency working groups to assess differences across data collection
efforts and determine which of those differences are beneficial and which
are unnecessary. For example, OMB has convened the following
interagency groups:
•

The Interagency Working Group for Research on Race and Ethnicity
was formed in 2014 to exchange research findings, identify
implementation issues, and collaborate on a shared research agenda
to improve federal statistics on race and ethnicity.

•

The Interagency Working Group on Measuring Relationships in
Federal Household Surveys, which was established in 2010,
convenes representatives from a variety of federal agencies involved
in the collection, dissemination, or use of household relationship data
to address the challenges in measuring household relationships,
including same sex couples.

•

The Federal Interagency Forum on Child and Family Statistics was
formally established in 1997 to develop priorities for collecting

44 USC § 3504 (c)&(e).

Page 32

GAO-16-546 Sexual Violence Data

enhanced data on children and youth, improve the reporting and
dissemination of information on the status of children to the policy
community and the general public, and produce more complete data
on children at the state and local levels.
We asked OMB if there are plans to convene a similar group for
harmonizing data on sexual violence. OMB staff stated that they did not
have plans to form an interagency group on the topic, but instead they
plan to invest limited resources strategically by engaging with BJS on its
redesign of NCVS and with CDC on its information quality of NISVS.
OMB staff also stated that their plans are to ensure that both agencies
are taking advantage of the insights gained as each agency undergoes
redesign and technical consultations in the next couple of years.
However, other data collection efforts, in addition to NCVS and NISVS,
also influence policy decisions on sexual violence. Depending upon the
outcome of the work being conducted by BJS and CDC, OMB may
encourage other data collection efforts to adapt or adopt insights gained
as appropriate to their respective programmatic missions. In the absence
of broader harmonization efforts, agency sexual violence data continue to
be inconsistent and incomparable, leading to confusion about the data
and lack of clarity about the scope of the problem of sexual violence in
the United States.

Conclusions

Differences in data collection efforts—particularly in terms of what is
included in measurements and definitions of sexual violence and
methodologies—collectively can lead to confusion. Without publiclyavailable information on which acts of sexual violence and contextual
factors are included in the measurements of sexual violence, data users
may lack clarity about what each data collection effort’s results represent.
Additionally, entities that use federal data may misunderstand the data
and develop policies that may not be based on the full extent of the
problem. In the absence of collaboration among agencies that manage
data collection efforts, it is unclear which differences enhance and which
impair the overall understanding of sexual violence, and as a result, policy
makers and the public lack coordinated information by which to address
the problem.

Recommendations for
Executive Action

To enhance the clarity and transparency of sexual violence data that is
reported to the public, we recommend that the Secretary of Education
direct the Assistant Secretary for the Office of Postsecondary Education,
the Secretary of Health and Human Services direct the Director of CDC,

Page 33

GAO-16-546 Sexual Violence Data

and the Attorney General direct the Director of BJS to make information
on the acts of sexual violence and contextual factors that are included in
their measurements of sexual violence publicly available. This effort could
entail revising their definitions of key terms used to describe sexual
violence so that the definitions match the measurements of sexual
violence.
To help lessen confusion among the public and policy makers regarding
federal data on sexual violence, we recommend that the Director of OMB
establish a federal interagency forum on sexual violence statistics. The
forum should consider the broad range of differences across the data
collection efforts to assess which differences enhance or hinder the
overall understanding of sexual violence in the United States.

Agency Comments
and Our Evaluation

We provided a copy of our report to DOD, Education, HHS, DOJ and
OMB for their review and comment. The agencies provided technical
comments, which we incorporated as appropriate. Education, HHS, and
DOJ also provided written comments, which are reprinted in appendices
IV, V, and VI, respectively. The OMB liaison to GAO provided us with
comments via email, which are summarized below.
DOJ, Education, and HHS agreed with our recommendation that, in order
to enhance the clarity and transparency of sexual violence data, they
should make information on the acts of sexual violence and contextual
factors that are included in their measurements of sexual violence publicly
available. In their written comments, DOJ and Education described
actions they have recently taken or plan to take to implement the
recommendation. DOJ stated that beginning in calendar year 2017, BJS
will provide the exact computer code used to construct its measures of
sexual violence as well as additional information on how sexual violence
is defined and measured. Education stated that in June 2016, the
department released an updated version of The Handbook for Campus
Safety and Security Reporting, which provides additional details on what
acts of sexual violence and contextual factors are included in the data
collection effort’s sexual violence measurements. HHS stated that the
department is committed to improving the quality of the data and the
clarity of the descriptions and definitions of sexual violence.
In an email responding to our recommendation that OMB establish a
federal interagency forum on sexual violence statistics, OMB stated that it
did not believe convening a forum at this time was the most strategic use
of resources. OMB stated that other interagency groups it has convened

Page 34

GAO-16-546 Sexual Violence Data

were typically about statistical methods or measurement issues that
would affect a wide swath of government and for which OMB guidance or
a best practice working paper would be forthcoming. OMB noted that
there are only four agencies involved in collecting sexual violence data,
and regarded none to be conducting or far enough along in its research
for OMB to develop guidance or identify best practices at this time. OMB
does, however, plan to follow closely and participate in CDC’s and BJS’s
ongoing technical work, and will consider convening or sharing
information across agencies when that work is further along. We
understand the importance of allowing time for a data collection effort to
mature before providing guidance or best practices. However, considering
that 7 of the 10 data collection efforts have been in place for more than 10
years, and several have been in place for multiple decades, we disagree
with OMB’s assertion that none of the data collection efforts are far
enough along for OMB to provide guidance and best practices. DOJ and
Education also commented on the recommendation to OMB. DOJ stated
that BJS welcomes OMB efforts to coordinate data collection and
reporting on sexual violence and stands ready to participate in an
interagency forum. Education stated that efforts to “harmonize” definitions
should not be pursued solely to achieve symmetry for its own sake, but
that the focus should be on the needs of each individual program. We
agree that the data collection efforts should continue to meet the needs of
individual agencies; however, considering the number of federal data
collection efforts, the range of differences across them, and the potential
for causing confusion, it would be beneficial for agencies to discuss these
differences and determine whether they are, in fact, necessary.
As agreed with your office, unless you publicly announce the contents of
this report earlier, we plan no further distribution until 30 days from the
report date. At that time, we will send copies of this report to the Director
of OMB, the Attorney General, the Secretaries of Defense, Education,
and Health and Human Services, and other interested parties. The report
will also be available at no charge on the GAO web site at
http://www.gao.gov.
If you or your staff members have any questions about this report, please
contact me at (202) 512-8777 or goodwing@gao.gov. Contact points for
our Offices of Congressional Relations and Public Affairs may be found

Page 35

GAO-16-546 Sexual Violence Data

on the last page of this report. GAO staff who made key contributions to
this report are listed in appendix VII.
Sincerely yours,

Gretta L. Goodwin, Acting Director
Homeland Security and Justice Issues

Page 36

GAO-16-546 Sexual Violence Data

Appendix I: Objectives, Scope and
Methodology
Appendix I: Objectives, Scope and
Methodology

Our objectives for this report were to address the following questions:
(1) What are the federal efforts underway to collect data on sexual
violence, and how, if at all, do these efforts differ?
(2) How do any differences across the data collection efforts affect the
understanding of sexual violence, and to what extent are federal agencies
addressing any challenges posed by the differences?
To address the first question, we identified federal efforts to collect data
on sexual violence, for which the data:
•

provided information on the extent to which acts of sexual violence
occur in the United States in a particular year (for example, the
number of times a rape or sexual assault has occurred or the number
of victims of rape and sexual assault);

•

were collected recently (i.e., 2010 or after);

•

were collected periodically (i.e., at least once every 2 years);

•

were reported publicly; 1 and

•

were not focused primarily on minors.

To identify efforts that met these criteria, we reviewed past GAO reports
and a Federal Bureau of Investigation (FBI) list of federal agencies that
may collect crime data. 2 We asked officials at those agencies if they had

Federal law enforcement agencies have information on sexual violence (e.g.,
investigations and case processing) in their case management systems, but we chose not
to include those data in our scope. We chose to focus on data that are available to the
public, because those data are used to influence policy decisions and the public’s
understanding of these crimes.
2
FBI compiled the list of agencies as part of a Director’s Initiative to identify agencies that
could report to the Uniform Crime Reporting Program’s Summary Reporting System.

Page 37

GAO-16-546 Sexual Violence Data

Appendix I: Objectives, Scope and
Methodology

any data collection efforts that met our criteria. 3 Additionally, we asked
experts in the field (for example, academic researchers) and officials from
victim advocacy groups and other special interest groups about any
additional federal data collection efforts they were aware of that met our
selection criteria. 4 We initially identified experts and entities that use
federal data on sexual violence by conducting background research.
Then, we interviewed experts and officials from the entities identified in
our research, and from these contacts we identified additional entities that
use federal data on sexual violence. In all, we spoke with officials from
three victim advocacy groups, five other special interest groups (for
example, law enforcement associations and campus safety groups), three
other federal agencies, and two academic experts. 5 Based on this work,
we identified 10 data collection efforts that met our criteria across four
federal agencies, including the Departments of Defense, Education,
Health and Human Services, and Justice.

Several federal agencies have data collections that include information on sexual
violence but are not included in this study because they did not meet our criteria. For
example, the Department of Health and Human Services’ (HHS) Child Maltreatment
Reports contain child abuse and neglect data collected via the National Child Abuse and
Neglect Data System (which is administered by HHS’s Administration for Children and
Families’ Children’s Bureau); however, because these reports focus on children they did
not meet our criteria. Another example of a study not included in our review is the
“Campus Sexual Assault Study” [C. P. Krebs, C. H. Lindquist, T .D. Warner, B. S. Fisher,
S. L. Martin, The Campus Sexual Assault Study: Final Report, 2007, National Institute of
Justice (Washington, D.C.: December 2007)] which was prepared with Department of
Justice funds in 2007. This was a one-time study conducted prior to 2010 and thus did not
meet our criteria of being collected periodically and recently. It is possible that agencies
have other data collection efforts that met our criteria and were not included in the scope
of this review.
4

For the purposes of this report, we use the term “data collection effort” to identify a
compilation of information on sexual violence that meet our selection criteria.
5

The victim advocacy groups we spoke with included the National Alliance to End Sexual
Violence, the National Center for Victims of Crime, and the Rape, Abuse & Incest National
Network. The other special interest groups we spoke with included the International
Association of Chiefs of Police, the International Association of Campus Law Enforcement
Administrators, the National Center for Campus Public Safety, the National Sexual
Violence Resource Center, and the Police Foundation. The other federal agencies we
spoke with included the Department of Justice’s National Institute of Justice, the Office for
Victims of Crime, and the Office on Violence Against Women. The two academic experts
we spoke with were Professor James Lynch and Professor Janet Lauritsen.

Page 38

GAO-16-546 Sexual Violence Data

Appendix I: Objectives, Scope and
Methodology

To identify and describe differences across the data collection efforts, we
obtained information on the purpose, scope, and methodology of each
data collection effort. We obtained this information through a review of
documents, such as user manuals and program descriptions. We also
conducted interviews with senior agency officials, senior officials at
entities that use federal data on sexual violence, and academic experts.
Using documentary and testimonial information, we compared the
similarities and differences of the data collection efforts with respect to
target population; context in which data were collected; source of the
data; unit of measurement; time frames (for example, when data are
collected and how often data are reported by the federal agency); and
terminology and measurements of sexual violence. To compare
similarities and differences of terminology of sexual violence, we used
agency documents to identify for each data collection effort the terms
used to describe sexual violence. To identify for each data collection
effort what acts of sexual violence and contextual factors are included in
measurements of sexual violence, we reviewed agency documentation
and interviewed agency officials.
To identify how the differences affect understanding of sexual violence,
we obtained and reviewed federal reports and interviewed and reviewed
relevant documentation from agency officials, experts, and officials from
entities that use federal data on sexual violence. We asked these officials
and experts whether, in their experience, any of the differences made it
difficult for people who may use the data (e.g., Congress, policy makers,
academics, the general public) to understand the extent to which sexual
violence occurs in the United States. We also asked them to identify any
difficulties or challenges, of which they were aware, that have resulted
from the differences across federal efforts to collect data on sexual
violence. Because these officials and experts were not selected as a
representative sample, the information obtained from these interviews
applies solely to this set of officials and experts, and cannot be
generalized to others.
We also reviewed articles, conference papers, and government and
nongovernment reports that discuss differences across federal sexual
violence data collection efforts. To identify articles, a research librarian
conducted a search of several bibliographic databases, such as
ProQuest, Embase, and Scopus, using terms such as “rape data” or
“sexual assault statistics,” among others. The search looked for peerreviewed articles, books, and conference papers published during or after
2005. This search yielded 36 publications, 16 of which were relevant to
our research objective on the impact of the differences across the data

Page 39

GAO-16-546 Sexual Violence Data

Appendix I: Objectives, Scope and
Methodology

collection efforts and 20 which were not. In reviewing the identified
publications, we found and reviewed an additional 9 articles and reports
that were pertinent. See app. III for a list of articles and reports that we
reviewed and determined to be relevant for our analysis. The librarianassisted literature search was conducted in August 2015, and we
reviewed literature from that search and identified additional sources from
August 2015 to May 2016.
To describe the extent to which federal agencies are addressing any
challenges posed by differences across the data collection efforts, we
interviewed senior agency officials and academic experts and obtained
relevant documentation. We asked agency officials about what, if any,
steps their agency has taken, or planned to take, to address some of the
difficulties or challenges that may have resulted from the differences
across the data collection efforts. We also asked agency officials as well
as officials from entities that use federal data on sexual violence if they
were aware of any additional steps being taken by other federal agencies,
state agencies, or other national entities, etc., to address some of the
difficulties or challenges that may have resulted from the differences
across the data collection efforts.
We conducted this performance audit from March 2015 to July 2016, in
accordance with generally accepted government auditing standards.
Those standards require that we plan and perform the audit to obtain
sufficient, appropriate evidence to provide a reasonable basis for our
findings and conclusions based on our audit objectives. We believe that
the evidence obtained provides a reasonable basis for our findings and
conclusions based on our audit objectives.

Page 40

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal
Data Collection Efforts

Appendix II: Descriptive Tables of Federal Data Collection Efforts

Tables 5 through 10 provide information on the acts of sexual violence and the contextual
factors included in the measurements of sexual violence by federal data collection efforts
identified by GAO.
Table 5: Terminology Federal Data Collection Efforts Use to Characterize Specific Acts of Sexual Violence Where a Victim Was Penetrated
Acts of
Sexual
Violence

DSAIDa

WGRAb

Clery Act
data

NEISS-AIP

NISVS

NCVS

NIS

SSV

UCR-SRS

UCR-NIBRS

Victim was penetrated
Vaginally

Rape, Sexual
Assault

Penetrative
Sexual Assault

Rape

AssaultSexualc

Rape, Sexual
Coercion

Rape

Nonconsensual
Sexual Acts, Staff
Sexual Misconduct

Nonconsensual
Rape
Sexual Acts, Staff
Sexual
Misconductc

Rape, Sexual
Assault With
An Object

Anally

Rape, Sexual
Assault,
Forcible
Sodomy

Penetrative
Sexual Assault

Rape

AssaultSexualc

Rape, Sexual
Coercion

Rape

Nonconsensual
Sexual Acts, Staff
Sexual Misconduct

Nonconsensual
Rape
Sexual Acts, Staff
Sexual
c
Misconduct

Sodomy,
Sexual Assault
With An Object

Orally

Rape, Sexual
Assault,
Forcible
Sodomy

Penetrative
Sexual Assault

Rape

AssaultSexualc

Rape, Sexual
Coercion

Rape

Nonconsensual
Sexual Acts, Staff
Sexual Misconduct

Nonconsensual
Rape
Sexual Acts, Staff
Sexual
Misconductc

Sodomy

With a body Rape, Sexual
part
Assault

Penetrative
Sexual Assault

Rape

AssaultSexualc

Rape

Rapec

Nonconsensual
Sexual Acts, Staff
Sexual Misconduct

Nonconsensual
Rape
Sexual Acts, Staff
Sexual
Misconductc

Rape

With an
object

Rape, Sexual
Assault

Penetrative
Sexual Assault

Rape

Assaultd
Sexual

Rape

Nonconsensual
Nonconsensual
Rape
Sexual Acts, Staff
Sexual Acts, Staff
Sexual Misconductd Sexual
c
Misconduct

Table legend:
DSAID: Defense Sexual Assault Incident Database, Department of Defense (DOD)
WGRA: Workplace and Gender Relations Survey of Active Duty Members, Department of Defense
Page 41

GAO-16-546 Sexual Violence Data

Sexual Assault
With An Object

Appendix II: Descriptive Tables of Federal Data Collection Efforts

Clery Act data: Data reported to the Department of Education under the Jeanne Clery Disclosure of Campus Security Policy and Campus Crime Statistics Act
NEISS-AIP: National Electronic Injury Surveillance System-All Injury Program, Department of Health and Human Services
NISVS: National Intimate Partner and Sexual Violence Survey, Department of Health and Human Services
NCVS: National Crime Victimization Survey, Department of Justice
NIS: National Inmate Survey, Department of Justice
SSV: Survey of Sexual Victimization, Department of Justice
UCR-SRS: Uniform Crime Reporting Program-Summary Reporting System, Department of Justice
UCR-NIBRS: Uniform Crime Reporting Program-National Incident-Based Reporting System, Department of Justice
Source: GAO analysis of testimonial evidence and agency documentation. | GAO-16-546
a

DSAID uses the following key terms to categorize sexual violence incidents that occurred prior to 2012 because these
were the terms that were included in older versions of the Uniform Code of Military Justice (UCMJ): “aggravated sexual
assault,” “indecent assault,” and “wrongful sexual contact.” However, for the purposes of this review, we only included
DSAID terms that are derived from the current version of the UCMJ and are used for incidents occurring on or after June
28, 2012: “abusive sexual contact,” “aggravated sexual contact,” “attempts to commit offenses,” “forcible sodomy,” “rape,”
and “sexual assault.” According to a DOD official, DOD's Military Criminal Investigation Organizations (MCIOs) confirm that
the acts of sexual violence that are described in this table are included in the measurements of these terms per their
inclusion in the UCMJ definitions. MCIOs review reports for legal sufficiency and confirm and document the factors in each
case.
b

In 2014 DOD contracted with RAND to conduct an independent assessment of the WGRA and, if necessary, to update the
WGRA methodology and to administer the 2014 WGRA. That year RAND created and administered two versions of the
survey. One version of the survey employed DOD’s prior measure of “unwanted sexual contact” to estimate the past-year
prevalence of sexual assault in DOD. The other survey version—called the RAND Military Workplace Study (RMWS)—
employed a new measure of sexual violence that was designed to align with the terminology used and corresponding
categories of crimes specified in Articles 80 and 120 of the UCMJ. In its Annual Report on Sexual Assault in the Military for
Fiscal Year 2015, DOD stated that the next WGRA will utilize the same methodology as developed by RAND. For the
purposes of this report, we refer broadly to this data collection effort as the “WGRA” and use the measures of sexual
violence developed by RAND—“penetrative sexual assault,” “nonpenetrative sexual assault,” and “attempted penetrative
sexual assault”—for our analysis.
c

Agency officials reported that the measurement of this sexual violence term includes this specific act of sexual violence,
though reference to the act of sexual violence may be implicit rather than explicit.
d

Agency officials reported that the measurement of this sexual violence term includes this specific act of sexual violence,
though agency documentation neither implicitly nor explicitly supports this claim.

Page 42

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data Collection Efforts

Table 6: Terminology Federal Data Collection Efforts Use to Characterize Specific Acts of Sexual Violence Where a Victim Was Made to Penetrate
Acts of
Sexual
Violence

DSAIDa

WGRAb

Clery Act
data

NEISS-AIP

NISVS

NCVS

NIS

SSV

UCR-SRS

UCR-NIBRS

Victim was made to penetrate
Vaginally

—

Assault-Sexualc

Being Made
To Penetrate
Someone
Else

—

Nonconsensual
Sexual Actsc, Staff
Sexual Misconductc

Nonconsensual
Rapec
Sexual Actsc, Staff
Sexual Misconductc

Rapec, Sexual
Assault With
An Objectc

Anally

Forcible
Sodomy

—

Assault-Sexualc

Being Made
To Penetrate
Someone
Else

—

Nonconsensual
Sexual Actsc, Staff
Sexual Misconductc

Nonconsensual
Rapec
Sexual Actsc, Staff
Sexual Misconductc

Sodomyc,
Sexual Assault
With An
c
Object

Orally

Forcible
Sodomy

—

Assault-Sexualc

Being Made
To Penetrate
Someone
Else

—

Nonconsensual
Sexual Acts, Staff
c
Sexual Misconduct

Nonconsensual
Rapec
Sexual Actsc, Staff
c
Sexual Misconduct

Sodomyc

With a body
part

—

Assault-Sexualc

—

Nonconsensual
Sexual Actsc, Staff
Sexual Misconductc

Nonconsensual
Rapec
c
Sexual Acts , Staff
Sexual Misconductc

Rapec

With an
object

—

Assault-Sexuald

—

Nonconsensual
Sexual Actsd, Staff
Sexual Misconductd

Nonconsensual
Rapec
Sexual Actsc, Staff
Sexual Misconductc

Sexual Assault
With An
Objectc

Table legend:
DSAID: Defense Sexual Assault Incident Database, Department of Defense (DOD)
WGRA: Workplace and Gender Relations Survey of Active Duty Members, Department of Defense
Clery Act data: Data reported to the Department of Education under the Jeanne Clery Disclosure of Campus Security Policy and Campus Crime Statistics Act
NEISS-AIP: National Electronic Injury Surveillance System-All Injury Program, Department of Health and Human Services
NISVS: National Intimate Partner and Sexual Violence Survey, Department of Health and Human Services
NCVS: National Crime Victimization Survey, Department of Justice
NIS: National Inmate Survey, Department of Justice
SSV: Survey of Sexual Victimization, Department of Justice
Page 43

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data Collection Efforts

UCR-SRS: Uniform Crime Reporting Program-Summary Reporting System, Department of Justice
UCR-NIBRS: Uniform Crime Reporting Program-National Incident-Based Reporting System, Department of Justice
Source: GAO analysis of testimonial evidence and agency documentation. | GAO-16-546
a

In 2014 DOD contracted with RAND to conduct an independent assessment of the WGRA and, if necessary, to update the
WGRA methodology and to administer the 2014 WGRA. That year RAND created and administered two versions of the
survey. One version of the survey employed DOD’s prior measure of “unwanted sexual contact” to estimate the past-year
prevalence of sexual assault in DOD. The other survey version—called the RAND Military Workplace Study (RMWS)—
employed a new measure of sexual violence that was designed to align with the terminology used and corresponding
categories of crimes specified in Articles 80 and 120 of the UCMJ. In its Annual Report on Sexual Assault in the Military for
Fiscal Year 2015, DOD stated that the next WGRA will utilize the same methodology as developed by RAND. For the
purposes of this report, we refer broadly to this data collection effort as the “WGRA” and use the measures of sexual
violence developed by RAND—“penetrative sexual assault,” “nonpenetrative sexual assault,” and “attempted penetrative
sexual assault”—for our analysis.
c

Page 44

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data Collection Efforts

Table 7: Terminology Federal Data Collection Efforts Use to Characterize Specific Acts of Nonpenetrative or Noncontact Sexual Violence
Acts of Sexual
Violence

DSAIDa

WGRAb

Clery Act
data

NEISS-AIP

NISVS

NCVS

NIS

SSV

UCR-SRS

UCR-NIBRS

Nonpenetrative contact
Touching/
Fondling

Aggravated
Sexual
Contact,
Abusive
Sexual
Contact

Nonpenetrative
Sexual Assault

Fondling

AssaultSexual

Unwanted
Sexual Contact

Sexual
Assault

Abusive Sexual
Contacts Only

Abusive
Sex Offenses
Sexual
Contacts, Staff
Sexual
Misconduct

Kissing

Aggravated
Sexual
c
Contact ,
Abusive
Sexual
Contactc

Nonpenetrative
Sexual Assaultc

—

Unwanted
Sexual Contact

Sexual
Assaultc

—

Sex Offensesc

Fondlingc

Fondling

Noncompleted or noncontact sexual acts
Attempted
Acts

Attempts To
Commit
Offenses

Attempted
Raped
Penetrative
Sexual Assault

AssaultSexual

Rape, Being
Made to
Penetrate
Someone Else

Rape, Sexual
Assault

Abusive
Sexual
Contacts
Onlyd

Nonconsensu
al Sexual
Actsd, Staff
Sexual
Misconduct

Rape, Sex
Offenses

Rape,
Sodomy,
Sexual
Assault With
An Object,
Fondling

Indecent
Exposure,
Invasion of
Privacy,
Voyeurism

—

Noncontact
Unwanted
Sexual
Experiences

—

Staff Sexual
Misconduct

Sex Offenses

—

Non-Contact
Sexual
Violence

—

Noncontact
Unwanted
Sexual
Experiences

—

Staff Sexual
Harassment

—

Table legend:
DSAID: Defense Sexual Assault Incident Database, Department of Defense (DOD)
Page 45

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data Collection Efforts

WGRA: Workplace and Gender Relations Survey of Active Duty Members, Department of Defense
Clery Act data: Data reported to the Department of Education under the Jeanne Clery Disclosure of Campus Security Policy and Campus Crime Statistics Act
NEISS-AIP: National Electronic Injury Surveillance System-All Injury Program, Department of Health and Human Services
NISVS: National Intimate Partner and Sexual Violence Survey, Department of Health and Human Services
NCVS: National Crime Victimization Survey, Department of Justice
NIS: National Inmate Survey, Department of Justice
SSV: Survey of Sexual Victimization, Department of Justice
UCR-SRS: Uniform Crime Reporting Program-Summary Reporting System, Department of Justice
UCR-NIBRS: Uniform Crime Reporting Program-National Incident-Based Reporting System, Department of Justice
Source: GAO analysis of testimonial evidence and agency documentation. | GAO-16-546
a

In 2014 DOD contracted with RAND to conduct an independent assessment of the WGRA and, if necessary, to update the
WGRA methodology and to administer the 2014 WGRA. That year RAND created and administered two versions of the
survey. One version of the survey employed DOD’s prior measure of “unwanted sexual contact” to estimate the past-year
prevalence of sexual assault in DOD. The other survey version—called the RAND Military Workplace Study (RMWS)—
employed a new measure of sexual violence that was designed to align with the terminology used and corresponding
categories of crimes specified in Articles 80 and 120 of the UCMJ. In its Annual Report on Sexual Assault in the Military for
Fiscal Year 2015, DOD stated that the next WGRA will utilize the same methodology as developed by RAND. For the
purposes of this report, we refer broadly to this data collection effort as the “WGRA” and use the measures of sexual
violence developed by RAND—“penetrative sexual assault,” “nonpenetrative sexual assault,” and “attempted penetrative
sexual assault”—for our analysis.
c

Page 46

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data Collection Efforts

Table 8: Terminology Federal Data Collection Efforts Use to Characterize Sexual Violence When Contextual Factors of Consent Are Involved
Contextual
Factors

DSAIDa

WGRAb

Clery Act
data

NEISS-AIP

NISVS

NCVS

NIS

SSV

UCR-SRS

UCR-NIBRS

Nonconsensual
Sexual Acts,
Abusive Sexual
Contacts Only,
Unwilling Activity,
Staff Sexual
Misconduct

Nonconsensual
Sexual Acts,
Abusive Sexual
Contacts Only,
Staff Sexual
Misconduct

Rape

Rape,
Sodomy,
Sexual
Assault With
An Object,
Fondling

Consent
Lack of Consent/ Rape, Sexual
Against Victim's Assault,
Will
Aggravated
Sexual Contact,
Abusive Sexual
Contact,
Forcible
Sodomy

Penetrative
Rape,
Sexual
Fondling
Assault,
Nonpenetrative
Sexual Assault

AssaultSexual

Rape, Sexual
Rape,
Coercion, Being Sexual
Made to
Assault
Penetrate
Someone Else,
Unwanted Sexual
Contact, Noncontact
Unwanted Sexual
Experiences

Victim Unable to
Consent (or
Refuse)
(including
asleep/
unconscious or
illness/disability)

Rape, Sexual
Assault,
Aggravated
Sexual Contact,
Abusive Sexual
Contact

Penetrative
Raped,
Sexual
Fondling
Assault,
Nonpenetrative
Sexual Assault

AssaultSexual

Rape, Being
Made to
Penetrate
Someone Else

Raped,
Sexual
Assaultd

—

Nonconsensual
Sexual Acts,
Abusive Sexual
Contacts

Rape

Rape,
Sodomy,
Sexual
Assault With
An Object,
Fondling

Victim
Alcohol/Drug
Facilitated

Rape, Sexual
Assault,
Aggravated
Sexual Contact,
Abusive Sexual
Contact

Penetrative
Raped,
Sexual
Fondlingc
Assault,
Nonpenetrative
Sexual Assault

AssaultSexual

Rape, Being
Made to
Penetrate
Someone Else

Raped,
Sexual
Assaultd

Nonconsensual
Sexual Acts,
Staff Sexual
Misconduct

Nonconsensual
Sexual Acts,
Abusive Sexual
Contacts

Rape

Rapec,
Sodomyc,
Sexual
Assault With
An Objectc,
Fondlingc

Willing Sexual
Activity

—

Willing Activity

Staff Sexual
Misconduct

—

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data Collection Efforts

NISVS: National Intimate Partner and Sexual Violence Survey, Department of Health and Human Services
NCVS: National Crime Victimization Survey, Department of Justice
NIS: National Inmate Survey, Department of Justice
SSV: Survey of Sexual Victimization, Department of Justice
UCR-SRS: Uniform Crime Reporting Program-Summary Reporting System, Department of Justice
UCR-NIBRS: Uniform Crime Reporting Program-National Incident-Based Reporting System, Department of Justice
Source: GAO analysis of testimonial evidence and agency documentation. | GAO-16-546
a

DSAID uses the following key terms to categorize sexual violence incidents that occurred prior to 2012 because these
were the terms that were included in older versions of the Uniform Code of Military Justice (UCMJ): “aggravated sexual
assault,” “indecent assault,” and “wrongful sexual contact.” However, for the purposes of this review, we only included
DSAID terms that are derived from the current version of the UCMJ and are used for incidents occurring on or after June
28, 2012: “abusive sexual contact,” “aggravated sexual contact,” “attempts to commit offenses,” “forcible sodomy,” “rape,”
and “sexual assault.” According to a DOD official, DOD's Military Criminal Investigation Organizations (MCIOs) confirm that
the contextual factors that are described in this table are included in the measurements of these terms per their inclusion in
the UCMJ definitions. MCIOs review reports for legal sufficiency and confirm and document the factors in each case.
b

In 2014 DOD contracted with RAND to conduct an independent assessment of the WGRA and, if necessary, to update the
WGRA methodology and to administer the 2014 WGRA. That year RAND created and administered two versions of the
survey. One version of the survey employed DOD’s prior measure of “unwanted sexual contact” to estimate the past-year
prevalence of sexual assault in DOD. The other survey version—called the RAND Military Workplace Study (RMWS)—
employed a new measure of sexual violence that was designed to align with the terminology used and corresponding
categories of crimes specified in Articles 80 and 120 of the UCMJ. In its Annual Report on Sexual Assault in the Military for
Fiscal Year 2015, DOD stated that the next WGRA will utilize the same methodology as developed by RAND. For the
purposes of this report, we refer broadly to this data collection effort as the “WGRA” and use the measures of sexual
violence developed by RAND—“penetrative sexual assault,” “nonpenetrative sexual assault,” and “attempted penetrative
sexual assault”—for our analysis.
c

Agency officials reported that the measurement of this sexual violence term includes this specific contextual factor, though
reference to the contextual factor may be implicit rather than explicit.
d

Agency officials reported that the measurement of this sexual violence term includes this specific contextual factor, though
agency documentation neither implicitly nor explicitly supports this claim.

Page 48

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data Collection Efforts

Table 9: Terminology Federal Data Collection Efforts Use to Characterize Sexual Violence When Contextual Factors of Force Are Involved
Contextual
Factors

DSAIDa

WGRAb

Physical
Force

Rape, Sexual
Assault,
Aggravated Sexual
Contact, Abusive
Sexual Contact,
Forcible Sodomy

Penetrative
Sexual
Assault,
Nonpenetrative
Sexual Assault

Threat of
Physical
Force

Rape, Sexual
Assault,
Aggravated Sexual
Contact, Abusive
Sexual Contact

Coercion

Rape, Sexual
Assault,
Aggravated Sexual
Contact, Abusive
Sexual Contact

Clery Act
data

NEISS-AIP

NISVS

—

AssaultSexual

Penetrative
Sexual
Assault,
Nonpenetrative
Sexual Assault

—

Penetrative
Sexual
Assault,
Nonpenetrative
Sexual Assault

—

NCVS

NIS

SSV

UCR-SRS

UCR-NIBRS

Rape, Being
Rape, Sexual
Made to
Assault
Penetrate
Someone Else

Nonconsensual
Sexual Acts, Staff
Sexual
Misconduct

Nonconsensual
Sexual Acts,
Abusive Sexual
Contacts, Staff
Sexual
Misconduct

—

Rapec,
Sodomyc,
Sexual Assault
With An
Objectc,
c
Fondling

AssaultSexual

Rape, Being
Rape, Sexual
Made to
Assault
Penetrate
Someone Else

Nonconsensual
Sexual Acts, Staff
Sexual
Misconduct

Nonconsensual
Sexual Acts,
Abusive Sexual
Contacts

—

AssaultSexual

Sexual
Coercion

Nonconsensual
Sexual Acts, Staff
Sexual
Misconduct

Nonconsensual
Sexual Acts,
Abusive Sexual
Contacts

—

Force

Rape

Table legend:
DSAID: Defense Sexual Assault Incident Database, Department of Defense
WGRA: Workplace and Gender Relations Survey of Active Duty Members, Department of Defense
Clery Act data: Data reported to the Department of Education under the Jeanne Clery Disclosure of Campus Security Policy and Campus Crime Statistics Act
NEISS-AIP: National Electronic Injury Surveillance System-All Injury Program, Department of Health and Human Services
NISVS: National Intimate Partner and Sexual Violence Survey, Department of Health and Human Services
NCVS: National Crime Victimization Survey, Department of Justice
NIS: National Inmate Survey, Department of Justice
SSV: Survey of Sexual Victimization, Department of Justice
UCR-SRS: Uniform Crime Reporting Program-Summary Reporting System, Department of Justice
UCR-NIBRS: Uniform Crime Reporting Program-National Incident-Based Reporting System, Department of Justice
Source: GAO analysis of testimonial evidence and agency documentation. | GAO-16-546

Page 49

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data Collection Efforts

DSAID uses the following key terms to categorize sexual violence incidents that occurred prior to 2012 because these
were the terms that were included in older versions of the Uniform Code of Military Justice (UCMJ): “aggravated sexual
assault,” “indecent assault,” and “wrongful sexual contact.” However, for the purposes of this review, we only included
DSAID terms that are derived from the current version of the UCMJ and are used for incidents occurring on or after June
28, 2012: “abusive sexual contact,” “aggravated sexual contact,” “attempts to commit offenses,” “forcible sodomy,” “rape,”
and “sexual assault.” According to a DOD official, DOD's Military Criminal Investigation Organizations (MCIOs) confirm that
the contextual factors that are described in this table are included in the measurements of these terms per their inclusion in
the UCMJ definitions. MCIOs review reports for legal sufficiency and confirm and document the factors in each case.
b

In 2014 DOD contracted with RAND to conduct an independent assessment of the WGRA and, if necessary, to update the
WGRA methodology and to administer the 2014 WGRA. That year RAND created and administered two versions of the
survey. One version of the survey employed DOD’s prior measure of “unwanted sexual contact” to estimate the past-year
prevalence of sexual assault in DOD. The other survey version—called the RAND Military Workplace Study (RMWS)—
employed a new measure of sexual violence that was designed to align with the terminology used and corresponding
categories of crimes specified in Articles 80 and 120 of the UCMJ. In its Annual Report on Sexual Assault in the Military for
Fiscal Year 2015, DOD stated that the next WGRA will utilize the same methodology as developed by RAND. For the
purposes of this report, we refer broadly to this data collection effort as the “WGRA” and use the measures of sexual
violence developed by RAND—“penetrative sexual assault,” “nonpenetrative sexual assault,” and “attempted penetrative
sexual assault”—for our analysis.
c

Agency officials reported that the measurement of this sexual violence term includes this specific contextual factor, though
reference to the contextual factor may be implicit rather than explicit.

Page 50

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data Collection Efforts

Table 10: Terminology Federal Data Collection Efforts Use to Characterize Sexual Violence When Other Contextual Factors Are Involved
Contextual
Factors

DSAIDa

WGRAb

Clery Act
data

NEISSAIP

NISVS

NCVS

NIS

SSV

UCRSRS

UCR-NIBRS

Other contextual factors
Influence/
Authority

—

AssaultSexual

Sexual
Coercion

—

Nonconsensual Sexual Staff Sexual
Acts, Abusive Sexual
Misconduct, Staff
Contact Only, Staff
Sexual Harassment
Sexual Misconduct

—

Fraudulent
Representation/
Professional
Purpose

Sexual Assault,
Abusive Sexual
Contact

Penetrative
Sexual
Assault,
Nonpenetrative
Sexual Assault

—

Rape,
Sodomy,
Sexual
Assault With
An Object,
Fondling

Fear

Rape, Sexual
Assault,
Aggravated Sexual
Contact, Abusive
Sexual Contact

Penetrative
Sexual
Assault,
Nonpenetrative
Sexual Assault

—

Abusive/
Humiliating/
Exploitative

Rape, Sexual
Assault,
Aggravated Sexual
Contact, Abusive
Sexual Contact

Penetrative
Sexual
Assault,
Nonpenetrative
Sexual Assault

—

AssaultSexual

Sexual
Coercion

—

Abusive Sexual
Contacts, Staff Sexual
Misconduct

—

For Sexual
Gratification

Rape, Sexual
Assault,
Aggravated Sexual
Contact, Abusive
Sexual Contact

Penetrative
Fondling
Sexual
Assault,
Nonpenetrative
Sexual Assault

—

Staff Sexual
Misconduct

—

Fondling

Non-Threat
Pressure

—

AssaultSexual

Sexual
Coercion

—

Nonconsensual Sexual Nonconsensual
Acts, Staff Sexual
Sexual Acts, Abusive
Misconduct
Sexual Contacts

—

Table legend:
DSAID: Defense Sexual Assault Incident Database, Department of Defense (DOD)
Page 51

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data Collection Efforts

DSAID uses the following key terms to categorize sexual violence incidents that occurred prior to 2012 because these
were the terms that were included in older versions of the Uniform Code of Military Justice (UCMJ): “aggravated sexual
assault,” “indecent assault,” and “wrongful sexual contact.” However, for the purposes of this review, we only included
DSAID terms that are derived from the current version of the UCMJ and are used for incidents occurring on or after June
28, 2012: “abusive sexual contact,” “aggravated sexual contact,” “attempts to commit offenses,” “forcible sodomy,” “rape,”
and “sexual assault.” According to a DOD official, DOD's Military Criminal Investigation Organizations (MCIOs) confirm that
the contextual factors that are described in this table are included in the measurements of these terms per their inclusion in
the UCMJ definitions. MCIOs review reports for legal sufficiency and confirm and document the factors in each case.
b

In 2014 DOD contracted with RAND to conduct an independent assessment of the WGRA and, if necessary, to update the
WGRA methodology and to administer the 2014 WGRA. That year RAND created and administered two versions of the
survey. One version of the survey employed DOD’s prior measure of “unwanted sexual contact” to estimate the past-year
prevalence of sexual assault in DOD. The other survey version—called the RAND Military Workplace Study (RMWS)—
employed a new measure of sexual violence that was designed to align with the terminology used and corresponding
categories of crimes specified in Articles 80 and 120 of the UCMJ. In its Annual Report on Sexual Assault in the Military for
Fiscal Year 2015, DOD stated that the next WGRA will utilize the same methodology as developed by RAND. For the
purposes of this report, we refer broadly to this data collection effort as the “WGRA” and use the measures of sexual
violence developed by RAND—“penetrative sexual assault,” “nonpenetrative sexual assault,” and “attempted penetrative
sexual assault”—for our analysis.
c

According to DOD officials, “influence/authority” is not currently included in DSAID’s measurements of sexual violence, but
DOD has a legislative proposal to include this contextual factor in the Uniform Code of Military Justice’s definitions of
sexual violence. If that proposal is passed, the contextual factor of “influence/authority” would be implied in the
measurement of sexual violence in DSAID.

Page 52

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data
Collection Efforts

Tables 11 through 13 provide information on the contexts, data sources,
and time frames used by federal data collection efforts identified by GAO.
Table 11: Criminal Justice and Public Health Contexts for Federal Data Collection
Efforts
Context
Data Collection Effort

Criminal Justice

Public Health

DSAID

—

WGRA

—

NEISS-AIPa

Clery Act data

—

NISVS

—

NCVS

—

NIS

—

SSV

—

UCR-SRS

—

UCR-NIBRS

—

Table legend:
DSAID: Defense Sexual Assault Incident Database, Department of Defense
WGRA: Workplace and Gender Relations Survey of Active Duty Members, Department of Defense
Clery Act data: Data reported to the Department of Education under the Jeanne Clery Disclosure of
Campus Security Policy and Campus Crime Statistics Act
NEISS-AIP: National Electronic Injury Surveillance System-All Injury Program, Department of Health
and Human Services (HHS)
NISVS: National Intimate Partner and Sexual Violence Survey, Department of Health and Human
Services
NCVS: National Crime Victimization Survey, Department of Justice
NIS: National Inmate Survey, Department of Justice
SSV: Survey of Sexual Victimization, Department of Justice
UCR-SRS: Uniform Crime Reporting Program-Summary Reporting System, Department of Justice
UCR-NIBRS: Uniform Crime Reporting Program-National Incident-Based Reporting System,
Department of Justice
Source: GAO analysis of agency documentation. | GAO-16-546
a

The Consumer Product Safety Commission collects NEISS-AIP data under an interagency
agreement with HHS’s Centers for Disease Control and Prevention .

Page 53

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data
Collection Efforts

Table 12: Data Sources for Federal Data Collection Efforts
How Federal Agency Receives Data
Information Reported
to Authorities

Information Obtained
from Victim Surveys

DSAID

—

WGRA

—

NEISS-AIP

—

NISVS

—

NCVS

—

NIS

—

SSV

—

UCR-SRS

—

UCR-NIBRS

—

Data Collection Effort

Clery Act data
a

Table legend:
DSAID: Defense Sexual Assault Incident Database, Department of Defense
WGRA: Workplace and Gender Relations Survey of Active Duty Members, Department of Defense
Clery Act data: Data reported to the Department of Education under the Jeanne Clery Disclosure of
Campus Security Policy and Campus Crime Statistics Act
NEISS-AIP: National Electronic Injury Surveillance System-All Injury Program, Department of Health
and Human Services (HHS)
NISVS: National Intimate Partner and Sexual Violence Survey, Department of Health and Human
Services
NCVS: National Crime Victimization Survey, Department of Justice
NIS: National Inmate Survey, Department of Justice
SSV: Survey of Sexual Victimization, Department of Justice
UCR-SRS: Uniform Crime Reporting Program-Summary Reporting System, Department of Justice
UCR-NIBRS: Uniform Crime Reporting Program-National-Incident Based Reporting System,
Department of Justice
Source: GAO analysis of agency documentation. | GAO-16-546
a

The Consumer Product Safety Commission collects NEISS-AIP data under an interagency
agreement with HHS’s Centers for Disease Control and Prevention .

Page 54

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data
Collection Efforts

Table 13: Time Frames Used in Federal Data Collection Efforts Concerning Sexual Violence
Period of time for which
sexual violence took
Data Collection Effort
place

How often data are
collected by data
collector

How often data are
reported by federal
agency

Period of time covered
in the report

DSAID

point in timea

ongoing

annually

fiscal yearb

WGRA

previous 12 months, since
joining the military, prior to
joining the military, lifetime

biennially

covers previous year

Clery Act data

point in time

annual submissions by
institutions of higher
education to Education

annually

Includes data for 3 most
recent calendar years
(e.g., 2011 includes FY
08,09, and 10)

NEISS-AIP

point in time

daily

annually

covers the previous
calendar year

previous 12 months &
lifetime

annually (except 2014)c

periodically (last report
was in 2014, with 2011
data)

covers the previous
calendar year

NCVS

previous 6 months

every 6 months from
participating households

annually

covers victimizations
reported during the same
calendar year

NIS

previous 12 months, or
admission to current
facility

periodically

SSV

annually

covers the previous
calendar year

UCR-SRS

point in time

ongoing by police; monthly
submissions by authorities
to FBI

annually

covers the previous
calendar year

UCR-NIBRS

point in time

ongoing by police; monthly
submissions by authorities
to FBI

annually

covers the previous
calendar year

NISVS

Table legend:
DSAID: Defense Sexual Assault Incident Database, Department of Defense (DOD)
WGRA: Workplace and Gender Relations Survey of Active Duty Members, Department of Defense
Clery Act data: Data reported to the Department of Education under the Jeanne Clery Disclosure of Campus Security Policy and Campus Crime
Statistics Act
NEISS-AIP: National Electronic Injury Surveillance System-All Injury Program, Department of Health and Human Services (HHS)
NISVS: National Intimate Partner and Sexual Violence Survey, Department of Health and Human Services
NCVS: National Crime Victimization Survey, Department of Justice
NIS: National Inmate Survey, Department of Justice
SSV: Survey of Sexual Victimization, Department of Justice
UCR-SRS: Uniform Crime Reporting Program-Summary Reporting System, Department of Justice
UCR-NIBRS: Uniform Crime Reporting Program-National Incident-Based Reporting System, Department of Justice
Source: GAO analysis of agency documentation. | GAO-16-546
a

DSAID contains reports of sexual violence that occurred prior to the victim’s military service.

Page 55

GAO-16-546 Sexual Violence Data

Appendix II: Descriptive Tables of Federal Data
Collection Efforts

Data from DSAID is included in DOD’s annual report on sexual assault in the military.

According to HHS’s Centers for Disease Control and Prevention officials, they plan to collect NISVS
data biennially starting in 2016.

Page 56

GAO-16-546 Sexual Violence Data

Appendix III: List of Relevant Articles and
Reports
Appendix III: List of Relevant Articles and
Reports

National Academy of Sciences, Committee on National Statistics,
Estimating the Incidence of Rape and Sexual Assault. Washington, DC,
US: National Academies Press, Washington, DC, 2014.
[Unknown]. “Overcoming Challenges Related to Data Collection and
Measurement.” Forced Migration Review, vol. 27, no. Jan (2007): 28-29,.
Addington, L. and C. Rennison. “Rape Co-Occurrence: Do Additional
Crimes Affect Victim Reporting and Police Clearance of Rape?” Journal
of Quantitative Criminology, vol. 24, no. 2 (2008): 205-226.
Bachman, R. “Measuring Rape and Sexual Assault: Successive
Approximations to Consensus.” A paper commissioned for the National
Academy of Sciences, June 6, 2012.
Campbell, R, A. E. Adams, and D. Patterson. “Methodological Challenges
of Collecting Evaluation Data from Traumatized Clients/Consumers - A
Comparison of Three Methods.” American Journal of Evaluation, vol. 29,
no. 3 (2008): 369-381.
Chen, Y. and S. E. Ullman. “Women’s Reporting of Sexual and Physical
Assaults to Police in the National Violence Against Women Survey.”
Violence Against Women, vol. 16, no. 3 (2010): 262-79.
Clay-Warner, J. and J. McMahon-Howard. “Rape Reporting: ‘Classic
Rape’ and the Behavior of Law.” Violence and Victims, vol. 24, no. 6
(2009): 723-743.
Cohen, M.A. and A. R. Piquero. “New Evidence on the Monetary Value of
Saving a High Risk Youth,” Journal of Quantitative Criminology, vol. 25
(2008).
Cohn, A. M., H. M. Zinzow, H. S. Resnick, and D. G. Kilpatrick.
“Correlates of Reasons for Not Reporting Rape to Police: Results from a
National Telephone Household Probability Sample of Women with
Forcible Or Drug-Or-Alcohol Facilitated/Incapacitated Rape.” Journal of
Interpersonal Violence, vol. 28, no. 3 (2013): 455-473.
Cook, S. L., C. A. Gidycz, M. P. Koss, and M. Murphy. “Emerging Issues
in the Measurement of Rape Victimization.” Violence Against Women, vol.
17, no. 2 (2011): 201-18.

Page 57

GAO-16-546 Sexual Violence Data

Appendix III: List of Relevant Articles and
Reports

Delisi, M., A. Koloski, M. Sweeny, E. Hachmeister, M. Moore, A. Drury.
“Murder by Numbers: Monetary Costs Imposed by a Sample of Homicide
Offenders,” The Journal of Forensic Psychiatry & Psychology, vol.21, no.
4 (2010).
Fang, H., M. T. French, and K. E. McCollister. “The Cost of Crime to
Society: New Crime-Specific Estimates for Policy and Program
Evaluation.” Drug Alcohol Dependence, vol. 108, no.1-2 (2010).
Du Mont, J., K. Miller, and T.L. Myhr, “The Role of “Real Rape” and “Real
Victim” Stereotypes in the Police Reporting Practices of Sexually
Assaulted Women.” Violence Against Women, vol. 9, no. 466 (2003).
Fisher, B. S. “The Effects of Survey Question Wording on Rape
Estimates: Evidence from a Quasi-Experimental Design.” Violence
Against Women, vol. 15, no. 2 (2009): 133-47.
Gardella, J. H., C. A. Nichols-Hadeed, J. M. Mastrocinque, J. T. Stone, C.
A. Coates, C. J. Sly, and C. Cerulli. “Beyond Clery Act Statistics: A Closer
Look at College Victimization Based on Self-Report Data.” Journal of
Interpersonal Violence, vol. 30, no. 4 (2015).
Lynch, J. P. “Clarifying Divergent Estimates of Rape from Two National
Surveys.” The Public Opinion Quarterly, vol. 60, no. 3 (1996) 410-430.
Lynch, J. P. and J. L. Lauritsen. “Modernizing the Nation’s Crime
Statistics.” The Criminologist, vol. 40, no. 2 (2015).
Miller, T.R., M. A. Cohen, and B. Wiersema, “Victim Costs and
Consequences: A New Look.” A final summary report presented to the
National Institute of Justice, January 1996.
Murphy, S. B., K. M. Edwards, S. Bennett, S. J. Bibeau, and J. Sichelstiel.
“Police Reporting Practices for Sexual Assault Cases in which ‘the Victim
does Not Wish to Pursue Charges’.” Journal of Interpersonal Violence,
vol. 29, no. 1 (2014): 144-156.
Palermo, T. and A. Peterman. “Undercounting, Overcounting and the
Longevity of Flawed Estimates: Statistics on Sexual Violence in Conflict.”
Bulletin of the World Health Organization, vol. 89, no. 12 (2011): 924-925.
Saltzman, L. E., K. C. Basile, R. R. Mahendra, M. Steenkamp, E. Ingram,
and R. Ikeda. “National Estimates of Sexual Violence Treated in

Page 58

GAO-16-546 Sexual Violence Data

Appendix III: List of Relevant Articles and
Reports

Emergency Departments.” Annals of Emergency Medicine, vol. 49, no. 2
(2007): 210-7.
Simon, T. R., M. Kresnow, and R. M. Bossarte. “Self-Reports of Violent
Victimization among U.S. Adults.” Violence and Victims, vol. 23, no. 6
(2008): 711-726.
Violence Against Women: A Statistical Overview, Challenges and Gaps in
Data Collection and Methodology and Approaches for Overcoming Them.
A report of the expert group meeting organized by the UN Division for the
Advancement of Women in collaboration with the Economic Commission
for Europe and World Health Organization, April 11-14, 2005.
Weiss, K. G. “‘You Just Don’t Report that Kind of Stuff’: Investigating
Teens’ Ambivalence Toward Peer-Perpetrated, Unwanted Sexual
Incidents.” Violence and Victims, vol. 28, no. 2 (2013): 288-302.
Wolitzky-Taylor, K. B., H. S. Resnick, J. L. McCauley, A. B. Amstadter, D.
G. Kilpatrick, and K. J. Ruggiero. “Is Reporting of Rape on the Rise? A
Comparison of Women with Reported Versus Unreported Rape
Experiences in the National Women’s Study-Replication.” Journal of
Interpersonal Violence, vol. 26, no. 4 (2011): 807-832.

Page 59

GAO-16-546 Sexual Violence Data

Appendix IV: Comments from the
Department of Education
Appendix IV: Comments from the Department
of Education

Page 60

GAO-16-546 Sexual Violence Data

Appendix IV: Comments from the Department
of Education

Page 61

GAO-16-546 Sexual Violence Data

Appendix IV: Comments from the Department
of Education

Page 62

GAO-16-546 Sexual Violence Data

Appendix V: Comments from the Department
of Health and Human Services
Appendix V: Comments from the Department
of Health and Human Services

Page 63

GAO-16-546 Sexual Violence Data

Appendix V: Comments from the Department
of Health and Human Services

Page 64

GAO-16-546 Sexual Violence Data

Appendix VI: Comments from the
Department of Justice
Appendix VI: Comments from the Department
of Justice

Page 65

GAO-16-546 Sexual Violence Data

Appendix VI: Comments from the Department
of Justice

Page 66

GAO-16-546 Sexual Violence Data

Appendix VI: Comments from the Department
of Justice

Page 67

GAO-16-546 Sexual Violence Data

Appendix VII: GAO Contact and Staff
Acknowledgments
Appendix VII: GAO Contact and Staff
Acknowledgments

GAO Contact

Gretta L. Goodwin, (202) 512-8777, goodwing@gao.gov

Staff
Acknowledgments

In addition to the contact named above, individuals making key
contributions to this report were Kristy Love, Assistant Director; Meghan
Squires, Analyst-in-Charge; Tim Young; Kirsten Leikem; Janelle House;
David Alexander; and David Plocher. Diana Maurer, Tom Jessor, Janet
Temko-Blinder, Tovah Rom and Eric Hauswirth also provided valuable
assistance.

(441279)

Page 68

GAO-16-546 Sexual Violence Data

GAO’s Mission

The Government Accountability Office, the audit, evaluation, and
investigative arm of Congress, exists to support Congress in meeting its
constitutional responsibilities and to help improve the performance and
accountability of the federal government for the American people. GAO
examines the use of public funds; evaluates federal programs and
policies; and provides analyses, recommendations, and other assistance
to help Congress make informed oversight, policy, and funding decisions.
GAO’s commitment to good government is reflected in its core values of
accountability, integrity, and reliability.

Obtaining Copies of
GAO Reports and
Testimony

The fastest and easiest way to obtain copies of GAO documents at no
cost is through GAO’s website (http://www.gao.gov). Each weekday
afternoon, GAO posts on its website newly released reports, testimony,
and correspondence. To have GAO e-mail you a list of newly posted
products, go to http://www.gao.gov and select “E-mail Updates.”

Order by Phone

The price of each GAO publication reflects GAO’s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO’s website,
http://www.gao.gov/ordering.htm.
Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537.
Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional information.

Connect with GAO

Connect with GAO on Facebook, Flickr, Twitter, and YouTube.
Subscribe to our RSS Feeds or E-mail Updates.
Listen to our Podcasts and read The Watchblog.
Visit GAO on the web at www.gao.gov.

To Report Fraud,
Waste, and Abuse in
Federal Programs

Contact:
Website: http://www.gao.gov/fraudnet/fraudnet.htm
E-mail: fraudnet@gao.gov
Automated answering system: (800) 424-5454 or (202) 512-7470

Congressional
Relations

Katherine Siggerud, Managing Director, siggerudk@gao.gov, (202) 5124400, U.S. Government Accountability Office, 441 G Street NW, Room
7125, Washington, DC 20548

Public Affairs

Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800
U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, DC 20548

Please Print on Recycled Paper.

The Habeas Citebook Ineffective Counsel Side