U.S. Department of Justice
Office of Justice Programs

Bureau of Justice Statistics

Status Report
June 30, 2004

Data Collections for the Prison
Rape Elimination Act of 2003
On September 4, 2003, President
George W. Bush signed into law the
Prison Rape Elimination Act of 2003
(P.L. 108-79). The legislation requires
the Bureau of Justice Statistics (BJS)
to develop a new national data collection on the incidence and prevalence of
sexual assault within correctional facilities. This report updates Implementing
the Prison Rape Elimination Act of
2003, dated February 5, 2004.
There have been only a few studies on
the prevalence of sexual assault within
correctional facilities. These studies
are typically small in scale, covering
only a few facilities, and generalizations to the national correctional population are not appropriate. The magnitude of sexual assault among prisoners
is not currently well understood. BJS is
tasked with developing reliable methods to measure the problem so that it
can be addressed and eliminated.
Victimization, particularly sexual
assault by a same-sex perpetrator, is a
sensitive event that introduces complexities for collecting self-report data.
There are ethical concerns, a need to
insure respondent confidentiality to
ease fears of reprisal, and a general
reluctance to fully report past incidents.
Data collection from juveniles also
requires consent from parents or legal
guardians. Corrections administrators
have concerns about legal liability,
reliability of self-reports, and potential
disruption of facility operations.

Highlights of Public Law 108-79
• Sec. 4 (a)(1). The Bureau of Justice Statistics ... shall carry out, for each
calendar year, a comprehensive statistical review and analysis of the incidence
and effects of prison rape. The statistical review and analysis shall include, but
not be limited to the identification of the common characteristics of —

(A) both victims and perpetrators of prison rape; and
(B) prisons and prison systems with a high incidence of prison rape.
• Sec. 4 (a)(4). The review and analysis ... shall be based on a random sample,
or other scientifically appropriate sample, of not less than 10 percent of all
Federal, State, and county prisons, and a representative sample of municipal
prisons.
• Sec. 4 (a)(6). Federal, State, or local officials or facility administrators that
receive a request from the Bureau ... will be required to participate in the
national survey and provide access to any inmates under their legal custody.
• Sec. 4 (b)(3)(A). ... there is established, within the Department of Justice, the
Review Panel on Prison Rape .... The duty of the Panel shall be to carry out,
for each calendar year, public hearings concerning the operation of the three
prisons with the highest incidence of prison rape and the two prisons with the
lowest incidence of prison rape within each category of facilities ....
• Sec. 4 (c)(1). Not later than June 30 of each year, the Attorney General shall
submit a report on the activities of the Bureau and the Review Panel, with
respect to prison rape, for the preceding calendar year ....
• Sec. 4 (c)(2). The report required under paragraph (1) shall include ...

(B)(ii) a listing of those institutions in the representative sample, separated
into each category ... and ranked according to the incidence of prison rape
in each institution; and
(C) a listing of any prisons in the representative sample that did not cooperate with the survey ...
• Sec. 8 (c)(4). For each fiscal year, any amount that a State receives for that
fiscal year under a grant program covered by this subsection shall not be used
for prison purposes ... unless the chief executive of the State submits to the
Attorney General a certification that neither the State, nor any political subdivision or unit of local government within the State, is listed in a report ... pursuant
to section 4(c)(2)(C).

BJS is developing new methods to
address current deficiencies

of control over who completes the
forms and under what settings.

At present there is no reliable collection
methodology for measuring prison
rape. Most prior attempts to measure
the prevalence of prison rape relied on
either personal interviews of inmates or
use of mail-in, self-administered inmate
questionnaires. Personal interviews of
inmates have generally yielded low
positive response rates (below 1%).
These low rates of reporting do not
permit further analyses of victim,
perpetrator, and facility characteristics.
More recent studies with self-administered questionnaires have yielded
higher prevalence rates (around 20%
with a broad definition of sexual
assault). However, the credibility of
such studies remains in question due
to low questionnaire completion rates
(e.g., 25% response rate) and loss

To address existing deficiencies, BJS
is developing and testing the use of
Audio Computer-Assisted SelfInterviews (known as “audio-CASI”).
The audio-CASI methodology involves
inmates responding to a computer
questionnaire using a touch-screen,
following audio instructions delivered
via headphones. It is anticipated that
removing the presence of a personal
interviewer will increase an inmate’s
willingness to report sensitive information. Audio-CASI also permits inmates
with limited literacy to report victimization. At the same time, the conditions
under which inmates complete the
survey can be controlled, unlike past
surveys in which inmates completed a
written questionnaire in various
environments.

How will rape be measured?
Section 10 of the Prison Rape Elimination Act defined the term “rape” as:
... the carnal knowledge, oral
sodomy, sexual assault with an
object, or sexual fondling of a
person, forcibly or against that
person’s will; or not forcibly or
against the person’s will, where
the victim is incapable of giving
consent because of his or her
youth or his or her temporary or
permanent mental or physical
incapacity; or
...the carnal knowledge, oral
sodomy, sexual assault with an
object, or sexual fondling of a
person achieved through the exploitation of the fear or threat of physical violence or bodily injury.
After consulting with experts in sexual
victimization, prison rape researchers,
and corrections administrators and
practitioners, BJS intends to operationalize this definition by disaggregating sexual assault into three
categories of inmate-on-inmate
sexual violence and all incidents
of staff sexual misconduct.

2

Categories of sexual violence will
reflect uniform definitions formulated
by the National Center for Injury
Prevention and Control, in “Sexual
Violence Surveillance: Uniform
Definitions and Recommended Data
Elements,” Center for Disease Control
and Prevention (CDC).

To identify the best measurement strategy, BJS will test audio-CASI in two
interview settings: (1) inmates in
prisons and jails and residents in
juvenile facilities; and (2) inmates and
residents who are soon-to-be released.
BJS will also test supplemental survey
methodologies, including computerassisted interview methods (CAI) and
paper and pencil interviews (PAPI), to
collect information from soon-to-bereleased jail inmates and recently
released prisoners in parole offices.
Despite the expected improvements
from audio-CASI, inmates may still
perceive that their responses will not
be kept confidential and may continue
to fear retribution as long as they
remain in prison or jail. BJS will test
various methods of surveying released
inmates to determine if rates of sexual
assault differ by incarceration status.
Inmate-on-inmate abusive sexual
contacts include:
y

y

contact of any inmate without his
or her consent, or of an inmate
who is unable to consent or
refuse;
intentional touching, either directly
or through the clothing, of the
genitalia, anus, groin, breast, inner
thigh, or buttocks of an inmate.

These categories are: 1) completed
non-consensual sexual acts; 2)
attempted non-consensual sexual
acts; and 3) abusive sexual contacts.
Inmate-on-inmate non-consensual
sexual acts include:

All sexual acts involving staff are
considered misconduct and are
covered under the Act, including:

y

y

y

y
y

contact of any inmate without his
or her consent, or of an inmate
who is unable to consent or
refuse; and
contact between the penis and the
vulva or the penis and the anus
involving penetration, however
slight; or
contact between the mouth and
the penis, vulva, or anus; or
penetration of the anal or genital
opening of another person by a
hand, finger, or other object.

Data Collections for the Prison Rape Elimination Act of 2003

y

y

y

any behavior of a sexual nature
directed toward an inmate by an
employee, volunteer, official
visitor, or agency representative;
all completed, attempted, threatened, or requested sexual acts
between staff and inmates;
any incident of intentional
touching of the genitalia, anus,
groin, breast, inner thigh, or
buttocks with the intent to abuse,
arouse, or gratify sexual desire;
incidents of indecent exposure,
invasion of privacy for sexual
gratification, or staff voyeurism.

Multiple measures of sexual assault
will be used to assess reliability and
to rank facilities
Due to the complexities associated
with collecting sensitive data within
correctional facilities serving a variety
of functions and supervising different
populations, BJS will obtain multiple
measures of sexual assault victimization. BJS will disaggregate the number
of incidents by type of facility and
distinguish non-consensual acts from
abusive sexual contacts. (See box on
page 2.) This approach will permit BJS
to analyze the validity of different data
collection methodologies as well as
their impact on facility-level estimates.
The need to provide anonymity to
victims and to give assurances of
confidentiality of their responses
prevents validating reports of sexual
assault through follow-ups and official
records. Instead, by incorporating
multiple measures within the audioCASI surveys and comparing results
across multiple modes of data collection, BJS will attempt to validate victim
self-reports through independent
measures.
The audio-CASI surveys will include
questions about victimizations other
than sexual assault. These questions
will be used to assess the reliability of
the victim self-reports. Also to assess
reliability, victims will be asked a series
of questions about the circumstances
surrounding each incident.
BJS will conduct an annual
administrative records collection
Concurrent with developing and testing
self-report survey methodologies, BJS
will conduct an annual administrative
records collection. During 2004 BJS
will survey all Federal and State prison
systems, State juvenile systems, and a
representative sample (not less than
10%) of local jails and privately or
locally operated juvenile facilities. The
administrative records collection is
expected to include 3,269 of the 8,727
facilities covered by the Act.

Number of facilities covered under
the Prison Rape Elimination
Act of 2003

Facility type
Total
Prisons
Public - Federal
Public - State
Private
Local jails
Public
Private
Juvenile facilities
Public
Private
Other facilities
Indian country jails
Military-operated
ICE-operatedb

Number Sampled for
of
collection
facilities in 2004
8,727
3,269
84
1,320
264

Alla
Alla
30

3,318
47

390
10

1,211
2,323

a

All
194

70
59
31

10
10
10

a
The administrative records collection
will cover all 50 State prison and juvenile
systems and the Federal Bureau of
Prisons.
b
Includes facilities operated by or exclusively for the Bureau of Immigration and
Customs Enforcement, formerly the
U.S. Immigration and Naturalization
Service.

The main objectives of the administrative collection are to determine:
y the number of reported incidents of
inmate-on-inmate sexual violence
and staff-on-inmate sexual misconduct, by gender;
y how prison systems and facilities
record these incidents (e.g., in
disciplinary, grievance, investigative, or medical files);
y what information is recorded (e.g.,
allegations, confirmed incidents,
only incidents involving serious
bodily harm, or threats);
y where the incidents occur (e.g., in
the victim’s cell/room, in a common
area, or outside of the facility);
y what additional data are available
(for purposes of administrative
collections in future years).
The 2004 administrative collection will
provide an understanding of what
corrections officials know, what information is recorded, how allegations
and confirmed incidents are handled,
what disciplinary or administrative and
legal sanctions are imposed on perpetrators, and what treatment is provided
to victims of sexual assault.

The results from this collection will
provide BJS with system- and facilitylevel estimates of the incidents of
sexual assault for calendar year 2004.
As required under the Act, BJS will
report the findings at a facility level
when possible and at a system level
when facility-level estimates are not
appropriate. Facilities and systems with
the highest and lowest rates of sexual
assault will be identified.
Findings from the 2004 administrative
collection will also be used in future
surveys. In 2005 the administrative
survey will be expanded to collect information on individual incidents of sexual
assault occurring in State and Federal
prisons and State juvenile systems. In
2006 the survey will collect data on
incidents occurring in local jails and
privately or locally operated juvenile
facilities. The incident-level data collections will provide detailed data on
circumstances surrounding each
reported incident. These data will
enable BJS to analyze in depth the
incidence and effects of sexual victimization within correctional facilities.
BJS has identified agents to
assist in the data collections
The Governments Division of the U.S.
Census Bureau will act as BJS’ collection agent for the 2004, 2005, and
2006 surveys of administrative records.
BJS has developed separate questionnaires for the Federal Bureau of
Prisons, State prison systems, local jail
jurisdictions, State juvenile systems,
and local and private facilities. BJS and
Census staff are currently testing these
questionnaires and developing optimal
sampling designs. Data collection will
begin in early 2005, following approval
from the Office of Management and
Budget (OMB).
In April 2004 after a competitive award
process, BJS selected the Research
Triangle Institute (RTI) to design,
develop, and test the audio-CASI
surveys for use in adult prisons and
jails. The cooperative agreement
award with RTI covers a 24-month
period.

Data Collections for the Prison Rape Elimination Act of 2003

3

BJS and RTI staff are currently developing a draft questionnaire to include
detailed items that measure the
incidence of physical and sexual
victimization during a 12-month reference period. These items are based on
BJS’ National Crime Victimization
Survey and have been adapted for use
in correctional facilities. The draft
questionnaire will undergo cognitive
testing in six facilities in September
2004. The cognitive tests will evaluate
the question wording and comprehension of the survey concepts by inmates.
Results of the cognitive tests will be
used to revise the draft audio-CASI
questionnaire. The draft questionnaire
and supporting survey methods will be
submitted to OMB for review and
approval.
Upon approval by OMB, the survey will
be field-tested in spring 2005. The field
test will include approximately 2,500
respondents from up to 25 State and
Federal prisons and local jails.
A separate data collection agent will
measure juvenile victimization
To address the complexities of collecting data from youth under the supervision of juvenile justice authorities, BJS
has solicited a separate collection
agent to develop and test audio-CASI

methods in juvenile facilities. A formal
announcement of an 18-month
cooperative agreement is expected in
July 2004.
Efforts to collect data from youth have
the additional burden of obtaining
consent from parents or legal guardians prior to the participation in the
survey. While administrators in some
juvenile facilities are able to consent on
behalf of the parent or guardian, other
administrators are not, depending on
the laws of the State or locality. A
further complication is that most State
child abuse reporting laws require
disclosure of an incident of sexual
assault to the appropriate authorities.
These legal and ethical requirements
pose special challenges in designing
survey methods that will satisfy internal
review boards (IRBs) and guarantee
confidentiality to the youth.
The cooperative agreement will include
developing and testing supplemental
PAPI or suitable CAI surveys. These
surveys are needed because of the
high turnover rates of juveniles in facilities. BJS staff and the collection agent
will work collaboratively to develop
procedures for interviewing youth who
flow through sampled facilities over a
12-month period to produce accurate
facility-level prevalence rates.

Cognitive tests are expected to occur
in up to 10 juvenile facilities in late
2004, followed by a national pretest in
up to 30 facilities in mid-2005. Based
on the pretest results, the collection
agent will formulate a national implementation plan that describes strategies for overcoming the legal, ethical,
and logistical complexities related to
the juvenile data collection.
Census Bureau to assist in data
collection from former inmates
Finally, BJS is finalizing an interagency
agreement with the Demographic
Surveys Division (DSD) of the U.S.
Census Bureau for the development
and testing methodologies for collecting data on sexual assault from former
prisoners under active parole supervision and soon-to-be-released jail
inmates. A final agreement is expected
in July 2004.
DSD will assist BJS staff in designing
and testing PAPI and other CAI survey
methods. Although the surveys will be
self-administered, Census field staff
will oversee survey procedures to
ensure respondent confidentiality and
the integrity of the parolee and jail
inmates samples. Field staff will coordinate with jail and parole office staff the
selection of the sample and the

Data collections and agents
y

Administrative Records Collection: Governments Division of the U.S. Census Bureau will conduct surveys in
2004, 2005, and 2006. The 2004 survey will collect data from the Federal Bureau of Prisons, all State prison and
juvenile systems, and a sample of local jails and private facilities. Detailed data will be collected on incidents
reported to State and Federal prison officials and State juvenile systems in 2005 and on incidents in sampled local
jails in 2006.

y

Self-reports by inmates in adult correctional facilities: Research Triangle Institute was awarded a 24-month
cooperative agreement to design, develop, and test audio-CASI in adult correctional facilities. The project includes
cognitive testing, conducting a national pretest, and delivering a national implementation plan in 2006.

y

Self-reports by persons in juvenile facilities: A data collection agent has been identified for an 18-month
cooperative agreement to design, develop, and test audio-CASI and other supplemental survey methodologies in
juvenile facilities. A national implementation plan is expected in 2006.

y

Self-reports by soon-to-be-released jail inmates and former State prisoners: BJS is finalizing an 18-month
interagency agreement with Demographic Surveys Division of the U.S. Census Bureau to design, develop, and test
PAPI and other CAI survey methods for use with inmates at time of discharge from local jails and for former State
prisoners on active parole supervision.

4

Data Collections for the Prison Rape Elimination Act of 2003

distribution and collection of survey
forms.
Cognitive tests are planned for late
2004 followed by a field test with up to
30 jails and 20 State parole offices in
mid-2005.
The data collected from former
inmates will permit BJS to evaluate
whether incarceration status has an
effect on self-reported victimization.
Should there be an effect, additional
methods will be developed to link
reports of former inmates to specific
facilities and prison systems.
BJS has hired experts to provide
support for data collection activities
BJS is currently finalizing agreements
with two individuals to provide expertise in sexual assault research and the
juvenile justice system.
Richard Tewksbury, Ph.D., is a professor at the University of Louisville in the
Department of Justice Administration.
Dr. Tewksbury has authored many
publications on sexual assault within
correctional facilities. He joined BJS
under the BJS Visiting Fellows
Program in June 2004 for an 18-month
term.

Howard Snyder, Ph.D., is the Director
of Systems Research for the National
Center for Juvenile Justice (NCJJ). Dr.
Snyder possesses a unique knowledge
of juvenile delinquents and their
movement within the juvenile justice
system. He will work as a consultant on
juvenile justice issues with BJS for an
18-month period. BJS may hire or
retain additional subject matter experts
as necessary to support data collection
activities.
National implementation has begun
and will be fully operational in 2006
To meet the requirements of the Act,
BJS will conduct an initial administrative records collection in 2004. This
collection will provide facility and
system-level estimates of sexual
assault for calendar year 2004. Similar
collections are planned for 2005 and
2006, with additional details from
official records on each reported
incident of sexual assault.
It is anticipated that the audio-CASI,
PAPI, and other CAI surveys of current
and former inmates will be fully operational for a national data collection by
the end of 2006. Test results of audioCASI, PAPI, and CAI surveys will be
made available in June 2006.

In 2006 BJS will issue solicitations
seeking data collection agents to
conduct the national implementation of
all four survey components. The first
facility-level measures of sexual
assault using victim self-reports from a
national sample of Federal and State
prisoners, local jail inmates, persons
held in juvenile facilities, and former
inmates will be collected for calendar
year 2006.
In meeting the requirements of the Act,
BJS is relying upon the support of
correctional administrators nationwide.
It will be difficult to obtain accurate and
reliable data on sexual assault,
especially in facilities that house
juveniles. Through extensive development and testing, BJS is seeking to
identify reliable methods in order to
begin national implementation.
Better information on the extent of the
problem will inform policy-makers in
their efforts to eliminate sexual victimization within correctional facilities. The
challenge is formidable, but with the
cooperation of corrections professionals, requirements of the Act will be met
and the objectives achieved.

Data Collections for the Prison Rape Elimination Act of 2003

5

Schedule of BJS activities for the Prison Rape Elimination Act
Start
Project Management

Finish

3/1/03

6/30/07

3/1/03

1/31/04

12/15/03

2/28/06

Develop implementation plan

3/1/03

1/30/04

Hire experts in sexual assault and juvenile justice

4/1/04

9/30/05

Obtain OMB clearances

6/1/04

1/15/07

Write report

4/1/04

6/30/07

4/30/03

4/15/07

Develop inter-agency agreements

4/30/03

7/30/04

Sample design and selection

10/1/03

5/15/05

Questionnaire development

10/15/03

6/30/06

Conduct testing

5/1/04

10/31/06

Data collection

1/1/05

2/28/07

Analysis and write final report

1/15/05

4/15/07

ACASI Development - Adult Facilities

9/1/03

2/15/06

Obtain data collection agent

9/1/03

3/29/04

Develop/test ACASI instrument

7/1/04

8/31/05

Develop/test alternative sampling designs

7/1/04

2/28/05

Develop national implemention plan

7/1/05

2/15/06

5/15/04

2/15/06

5/15/04

7/30/04

Develop/test ACASI instrument

8/1/04

3/30/05

Develop/test alternative sampling designs

5/1/05

7/30/05

11/1/05

2/15/06

5/15/04

1/15/06

5/15/04

7/30/04

Develop surveys and methodology

8/1/04

6/30/05

Conduct field test

6/1/05

10/31/05

Evaluate results

11/1/05

1/15/06

National data collection

1/1/06

6/30/07

Obtain data collection agent(s)

1/1/06

5/31/06

Obtain OMB clearance

1/1/06

5/31/06

Data collection

8/1/06

12/30/06

Analysis and write final report

1/1/07

6/30/07

Assemble materials
Convene panels of subject matter experts

Conduct Administrative Records Collection

ACASI Development - Juvenile Facilities
Obtain data collection agent

Develop national implemention plan
PAPI/CAI - Jail inmates and former prisoners
Obtain data collection agent

6

Data Collections for the Prison Rape Elimination Act of 2003

2003
2004
2005
2006
2007
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3