RAND Military Workplace Study Volume 1

RAND_RMWS_volume1.pdf

Workplace and Gender Relations Survey

RAND Military Workplace Study Volume 1

OMB: 0704-0615

⚠️ Notice: This form may be outdated. More recent filings and information on OMB 0704-0615 can be found here:

2024-05-28 - Revision of a currently approved collection

Document [pdf]

Download: pdf | pdf

SEXUAL ASSAULT AND
SEXUAL HARASSMENT
IN THE U.S. MILITARY
Volume 1. Design of the 2014
RAND Military Workplace Study

Andrew R. Morral, Kristie L. Gore, Terry L. Schell, editors

C O R P O R AT I O N

For more information on this publication, visit www.rand.org/t/RR870z1

Library of Congress Cataloging-in-Publication Data is available for this publication.
ISBN: 978-0-8330-8860-4

Published by the RAND Corporation, Santa Monica, Calif.
© Copyright 2014 RAND Corporation

R® is a registered trademark.

Limited Print and Electronic Distribution Rights
This document and trademark(s) contained herein are protected by law. This representation of RAND
intellectual property is provided for noncommercial use only. Unauthorized posting of this publication
online is prohibited. Permission is given to duplicate this document for personal use only, as long as it is
unaltered and complete. Permission is required from RAND to reproduce, or reuse in another form, any of
its research documents for commercial use. For information on reprint and linking permissions, please visit
www.rand.org/pubs/permissions.html.
The RAND Corporation is a research organization that develops solutions to public policy challenges to help
make communities throughout the world safer and more secure, healthier and more prosperous. RAND is
nonprofit, nonpartisan, and committed to the public interest.
RAND’s publications do not necessarily reflect the opinions of its research clients and sponsors.
Support RAND
Make a tax-deductible charitable contribution at
www.rand.org/giving/contribute

www.rand.org

The 2014 RAND Military Workplace Study Team

Principal Investigators
Andrew R. Morral, Ph.D.
Kristie L. Gore, Ph.D.
Instrument Design
Lisa Jaycox, Ph.D., team lead
Terry Schell, Ph.D.
Coreen Farris, Ph.D.
Dean Kilpatrick, Ph.D.*
Amy Street, Ph.D.*
Terri Tanielian, M.A.*

Study Design and Analysis
Terry Schell, Ph.D., team lead
Bonnie Ghosh-Dastidar, Ph.D.
Marc Elliott, Ph.D.
Craig Martin, M.A.
Mark Totten, M.S.
Q Burkhart, M.S.
Robin Beckman, M.P.H.

Project Management
Kayla M. Williams, M.A.
Caroline Epley, M.P.A.
Amy Grace Peele, M.P.P.

Survey Coordination
Jennifer Hawes-Dawson

Westat Survey Group
Shelley Perry, Ph.D., team lead
Wayne Hintze, M.S.
John Rauch
Bryan Davis
Lena Watkins
Richard Sigman, M.S.
Michael Hornbostel, M.S.

Project Communications
Steve Kistler
Jeffrey Hiday
Barbara Bicksler, M.P.P.

Scientific Advisory Board
Major General John Altenburg, Esq. (USA, ret.)
Captain Thomas A. Grieger, M.D. (USN, ret.)
Dean Kilpatrick, Ph.D.
Laura Miller, Ph.D.
Amy Street, Ph.D.
Roger Tourangeau, Ph.D.

David Cantor, Ph.D.
Colonel Dawn Hankins, USAF
Roderick Little, Ph.D.
Sharon Smith, Ph.D.
Terri Tanielian, M.A.
Veronica Venture, J.D.

* Three members of the Scientific Advisory Board were so extensively involved in the
development of the survey instrument that we list them here as full Instrument Design
team members.

iii

Preface

The Department of Defense (DoD) has assessed service member experiences with sexual
assault and harassment since at least 1996, when Public Law 104-201 first required a
survey of the “gender relations climate” experienced by active-duty forces. Since 2002,
four “Workplace and Gender Relations Surveys,” as they are known in 10 USC §481,
have been conducted with active-duty forces (in 2002, 2006, 2010, and 2012). DoD
conducted reserve-component versions of this survey in 2004, 2008, and 2012.
The results of the 2012 survey suggested that more than 26,000 active-duty service members had experienced unwanted sexual contacts in the prior year, an estimate
that received widespread public attention and concern. In press reports and congressional inquiries, questions were raised about the validity of the estimate, about what
unwanted sexual contact included, and about whether the survey had been conducted
properly. Although many of these concerns were unfounded, the leadership of the
Senate Armed Services Committee urged the Secretary of Defense to conduct an independent assessment of the number of service members who experience sexual assault
or sexual harassment.
The Sexual Assault Prevention and Response Office within the Office of the Secretary of Defense selected the RAND Corporation to provide a new and independent
evaluation of sexual assault, sexual harassment, and gender discrimination across the
military. As such, DoD asked the RAND research team to redesign the approach used in
previous DoD surveys if changes would improve the accuracy and validity of the survey
results for estimating the prevalence of sexual crimes and violations. In the summer of
2014, RAND fielded a new survey, called the RAND Military Workplace Study.
This report is the first of several that will, collectively, describe the study methodology and its main findings. This first volume describes the survey methods and the
rationale for those methods, covering topics related to the survey instrument, sampling,
recruitment, and analytic weighting. Reports to be released in spring 2015 will present
survey results for active-duty and reserve-component service members in the Army,
Navy, Air Force, and Marine Corps (Volume 2), and the Coast Guard (Volume 3).
This research was conducted within the Forces and Resources Policy Center of the
RAND National Defense Research Institute, a federally funded research and developv

Sexual Assault and Sexual Harassment in the U.S. Military

ment center sponsored by the Office of the Secretary of Defense, the Joint Staff, the
Unified Combatant Commands, the Navy, the Marine Corps, the defense agencies,
and the defense Intelligence Community.
For more information on the Forces and Resources Policy Center, see http://
www.rand.org/nsrd/ndri/centers/frp.html or contact the director (contact information
is provided on the web page).

Contents

The 2014 RAND Military Workplace Study Team.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Exhibits and Tables.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii
Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix
CHAPTER ONE

The 2014 RAND Military Workplace Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Background: The Workplace and Gender Relations Survey.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
The RAND Military Workplace Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
CHAPTER TWO

Measurement of Sexual Harassment and Sexual Assault. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Sexual Harassment in the American Civilian and Military Workforce . . . . . . . . . . . . . . . . . . . . 7
Sexual Assault of American Civilians and Military Service Members. . . . . . . . . . . . . . . . . . . . . 12
Challenges in Measuring Sexual Harassment and Sexual Assault.. . . . . . . . . . . . . . . . . . . . . . . . . 18
General Survey Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
CHAPTER THREE

Study Design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Sample Design and Selection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Fielding the Survey. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Study Reviews and Approval. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
CHAPTER FOUR

Questionnaire Development.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Development of the Sex-Based MEO Violations Module. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Development of the Sexual Assault Module. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

vii

viii

Sexual Assault and Sexual Harassment in the U.S. Military

CHAPTER FIVE

Analytic Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Final Case Disposition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Survey Weights. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
CHAPTER SIX

Summary of Major Improvements in the RMWS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
RMWS Modifications to Improve the Precision of Sex-Based MEO Violations
and Sexual Assault Estimates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Our Survey Modifications Will Likely Count Different People as Having
Experienced Sex-Based MEO Violations or Sexual Assault. . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Study Experiments Will Help Inform Future Administrations of the WGRA. . . . . . . . . . . 72
Timeline for Public Reporting of Study Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
APPENDIXES

A. Definitions of Measurement Constructs.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
B. Correspondence Between the RMWS Measure of Sexual Assault and
Title 10 USC § 920 (UCMJ Article 120) .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Survey Instrument.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
2014 RAND Military Workplace Study Forms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
Prior WGRA Form. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177

Exhibits

1.1. Trends in the Percentage of Service Men and Women Who Indicated
Experiencing Unwanted Sexual Contact in the Past Year .. . . . . . . . . . . . . . . . . . . . . . . . . . . 2
4.1. Categorization of Problematic Workplace Behaviors and Sex-Based MEO
Violations in the RMWS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.2. Introduction and Stem Question for 2012 WGRA Sexual Harassment
Question. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.3. RMWS Instructions Preceding the Sex-Based MEO Violations Module. . . . . . . . . . . . . 44
4.4. Categorization of Sexual Assault in the RMWS.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.5. 2012 WGRA Unwanted Sexual Contact Question. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.6. Preamble to 2014 RMWS Sexual Assault Assessment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
5.1. Predictors in WGRA Weighting Model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.2. Predictors in Outcome-Optimized RMWS Weights. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

Tables

1.1.
2 .1.
2.2.
2.3.

Survey Forms and Modules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Sexual Harassment Survey Comparison Chart. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Sexual Assault Victimization Survey Comparison Chart.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Sexual Harassment and Sexual Assault Measurement Decisions and Likely
Influence on Estimated Prevalence of These Experiences. . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.1. Size of the Active-Duty Sampling Frame and Sample, by Gender, Service,
and Pay Grade. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.2. Size of the Guard and Reserve Sampling Frame and Sample, by Gender and
Service. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.3. Sample Sizes for RMWS Survey Forms and Modules (Active-Duty Sample). . . . . . . . 34
5.1. Case Disposition Definitions.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
B.1. Types of Sexual Assault Under Article 120.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
B.2. Relation of Behavioral Screening Questions to Article 120. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
B.3. Relation of Offender Behavior Questions to Article 120. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
B.4. Relation of Additional Offender Behavior Questions I and J to Article 120. . . . . . . . . . 88
B.5. Relation of Additional Offender Behavior Question K to Article 120. . . . . . . . . . . . . . . . . . 89
ix

Summary

In early 2014, the Department of Defense (DoD) Sexual Assault Prevention and
Response Office asked the RAND National Defense Research Institute (NDRI) to
conduct an independent assessment of sexual assault, sexual harassment, and gender
discrimination in the military—an assessment last conducted in 2012 by the department itself through the Workplace and Gender Relations Survey. The 2014 RAND
Military Workplace Study (RMWS) fielded a new survey in the summer of 2014 to
a much larger sample of the military community than previous surveys—men and
women, both on active duty and in the reserve component, and including the four
DoD military services plus the Coast Guard. The objectives of the 2014 study were to
• establish precise and objective estimates of the percentage of service members who
experienced sexual assault, sexual harassment, and gender discrimination in the
past year
• describe the characteristics of these incidents, such as where and when they
occurred, who harassed or assaulted the member, whether the event was reported,
and what services the member sought
• identify barriers to reporting these incidents and barriers to the receipt of support
and legal services.
DoD asked the RAND research team to redesign the approach used previously
in the Workplace and Gender Relations Survey of Active Duty Personnel (WGRA)
surveys if changes would improve the accuracy and validity of the survey results for
counting sex crimes and sexual harassment. The approach developed for the RMWS
measurement of sexual assault aligns closely with the definitions in the Uniform Code
of Military Justice (UCMJ) for Article 120 and Article 80 crimes. The survey measures
sexual harassment and gender discrimination, which together we refer to as sex-based
military equal opportunity (MEO) violations, using criteria drawn directly from DoD
Directive 1350.2. Compared with past surveys that were designed to measure experiences associated with sexual misconduct, our approach offers greater precision in estimating the number of crimes and MEO violations that have occurred. Specifically, the
RMWS measures:
xi

xii

Sexual Assault and Sexual Harassment in the U.S. Military

• sexual assault, which captures three mutually exclusive categories of sexual assault:
penetrative, non-penetrative, and attempted penetrative crimes
• sex-based MEO violations, which consist of
–– sexually hostile work environment—a workplace characterized by persistent or
severe unwelcome sexual advances, or verbal or physical conduct that offends
service members
–– sexual quid pro quo—incidents in which someone uses his or her power or
influence within the military to attempt to coerce sexual behavior in exchange
for a workplace benefit
–– gender discrimination—incidents in which service members are subjected to
mistreatment on the basis of their gender that affects their employment conditions.
DoD, in consultation with the White House National Security Staff, stipulated
that the sample size for the RMWS was to include a census of all active-duty women
and 25 percent of active-duty men in the Army, Navy, Air Force, Marine Corps, and
Coast Guard. In addition, we were asked to include a smaller sample of National
Guard and reserve members sufficient to support comparisons of sexual assault and
harassment between the active-duty and reserve forces. Subsequently, the U.S. Coast
Guard also asked that RAND include a sample of their active-duty and reserve members. In total, therefore, RAND invited close to 560,000 service members to participate in the study, making it the largest study of sexual assault and harassment ever
conducted in the military.
The large sample for this study is particularly valuable for understanding the
experiences of relatively small subgroups in the population. For example, RAND’s
survey will provide more information about the experiences of men that have been sexually assaulted within each service and by rank. In the last WGRA, 117 men indicated
they had experienced unwanted sexual contacts, the term used to refer to events similar
to sexual assaults. This low number limits generalizations that can be made about the
sexual assault experiences of men in the military.
The large sample associated with the RMWS also gave RAND the opportunity to test how changing the questionnaire itself might have affected survey results.
Specifically, we were able to use a segment of our overall sample to draw direct comparisons between rates of sexual assault and sexual harassment as measured using the
2014 RMWS questionnaire and the 2012 WGRA questionnaire. To enable this comparison and others, we randomly assigned respondents to one of four different survey
questionnaires:
1. A “long form” consisting of a sexual assault module; a sex-based MEO violation module, which assesses sexual harassment and gender discrimination; and
questions on respondent demographics, psychological state, command climate,

Summary

xiii

attitudes and beliefs about sexual assault in the military and the nation, and
other related issues.
2. A “medium form” consisting of the sexual assault module, the sex-based MEO
violation module, and some demographic items.
3. A “short form” consisting of the sexual assault module, only the screening items
from the sex-based MEO violation module, and demographic items. Thus, these
respondents did not complete the full, sex-based MEO violation assessment.
4. A “prior WGRA form,” which included questions from the 2012 WGRA,
including the unwanted sexual contact, sexual harassment, and gender discrimination sections from that survey.
The RMWS survey is designed to provide a valid and precise estimate of the
number of service men and women who have experienced sexual assault, sexual harassment, or gender discrimination in the past year. Although many of our improvements
build on methods developed for the WGRA, the new survey collects more detailed
information related to whether the event is consistent with criminal offenses under the
UCMJ or violations of MEO. Other improvements in our survey approach include:
• Simplifying question syntax to improve respondent understanding. Earlier WGRA
surveys used complex questions for the sexual harassment and unwanted sexual
contact measure, questions that placed heavy demands on respondents’ reading
skills and comprehension. RAND’s approach presented a series of simpler questions asking about behaviorally specific experiences.
• Clarifying question terminology. The prior WGRA approach to measuring sexual
assault relied on respondents’ understanding of the complex concept of consent,
and did so without defining the term. The RAND questionnaire avoids use of
this term for most definitions of sexual assault. Instead, we substituted the behaviorally specific forms of coercion defined as criteria for sex crimes in Article 120
of the UCMJ. Similarly, we limited use of the term sexual in defining the events
that might qualify as sexual assault because sexual assaults that would qualify
as crimes in Article 120 need not be associated with sexual gratification if they
are designed to humiliate or debase the person who is assaulted. Instead, the
new RMWS survey inquires about sexual assaults using simple behavioral and
anatomical descriptions that make no reference to whether the behaviors were
“sexual” or not. Use of such behaviorally and anatomically specific language not
only better matches the similarly specific language of Article 120, it also has been
the standard approach for accurately assessing sexual assault in survey research
conducted with civilian populations for decades (National Research Council,
2014). We believe these changes (and many others like them) clarify the meaning
and intent of our survey questions and have improved the reliability and valid-

xiv

Sexual Assault and Sexual Harassment in the U.S. Military

ity of respondents’ answers. Pretesting of the survey indicated that respondents
found the items to be clear and easy to understand.
• Reducing overcounting of offenses due to telescoping. People often report crimes
as occurring more recently than they really did—a tendency that is referred to
as telescoping. To guard against this phenomenon, RAND implemented several
strategies in the RMWS survey that are designed to better orient respondents to
the specific timeframe under consideration in each section of the survey.
We took additional steps to increase response rates and reduce nonresponse bias:
• A shorter survey. The RMWS survey that most respondents received is shorter
than the prior WGRA and could be completed by most respondents in just eight
minutes.
• Maximizing responses to the key questions. We placed the sexual assault and sexual
harassment modules at the beginning of the survey to maximize the number of
respondents answering these questions, since historically there has been considerable survey break-off before reaching these core questions.
• Reaching junior enlisted members and others with limited access to computers. We
made the survey smartphone compatible and developed a communications plan
that promoted the survey through many channels, including social media, public
service announcements, and print news stories.
• Compensating for possible survey nonresponse bias. If sampled service members who
choose not to participate in the study have different experiences regarding sexual
harassment and assault than those who did participate, our estimates could misrepresent the experience of the full population of service members. Building off
the extensive analysis conducted during the 2012 WGRA survey, RAND developed nonresponse weights that allow us to account for a wider set of factors to
assess the representativeness of the sample and to reduce nonresponse bias on the
key study outcomes.
We cannot yet say how these improvements will affect our estimates of sexual
harassment and sexual assault. Some of the changes are likely to increase and others are
likely to decrease the number of individuals classified as experiencing sexual assault,
sexual harassment, or gender discrimination in the past year. But we believe that the
new questions should provide more precise and valid estimates of the offenses of greatest concern to the military and should perform better for tracking changes caused by
training and policies to reduce these offenses and violations.
This report is the first in a series that will describe the methodology and results
of the RAND Military Workplace Study for DoD active-duty forces, for guard and
reserve forces, and for Coast Guard active-duty and reserve forces. This volume covers
the study methodology—the study design, questionnaire development, and analytic

Summary

methods used in calculating estimates. Survey results will be presented in subsequent
reports, including a report of top-line estimates of sexual assault and sexual harassment
(by gender, service branch, and type of assault) provided to the DoD Sexual Assault
Prevention and Response Office and the U.S. Coast Guard, followed by a series of
publicly available reports on the main findings of the study scheduled for release in late
spring 2015.

Acknowledgments

We wish to thank the service men and women who took the time to complete the
RAND Military Workplace Study survey and share their experiences, even when those
experiences were painful to recount.
Many people assisted us with the development of the new survey instrument.
The leadership and staff in each of the services’ sexual assault prevention and response
(SAPR) offices provided many rounds of review, valuable suggestions, and feedback,
as did research staff from the Air Force and Army Research Institute. James Clark and
Teresa Scalzo consulted on the interpretation of Article 120 of the Uniform Code of
Military Justice. Jonathan Welch, RAND, and Tom Bush consulted on the National
Guard and reserve form of the survey. The questionnaire further benefited from comments of 24 anonymous service members and recent veterans who agreed to pre-test
the survey and provide us with their reactions to it. We express our appreciation to
John Boyle, Senior Vice President of ICF International; Richard Baskin, from Decipher Inc.; and Mary Koss, from the University of Arizona, for providing a survey
instrument they developed to measure sexual assault experiences among service members in the U.S. Air Force. This instrument was helpful to us in creating the new survey
instrument used in this study.
In addition to assisting us with the development of the survey instrument, the
members of our scientific advisory board have provided invaluable guidance on difficult decisions throughout the project.
We are grateful for the assistance and expert advice provided to us by the Defense
Manpower Data Center (DMDC), and especially to Elizabeth Van Winkle, who
shared DMDC experience from prior administrations of the Workplace Gender Relations surveys, and who served as a liaison between RAND and other parts of DMDC.
We also thank Paul Rosenfeld for his rapid and careful reviews of the survey licensing
materials submitted by RAND to the Office of the Secretary of Defense (OSD) and
Major Brandi Ritter in the Office of the Under Secretary of Defense for Personnel and
Readiness Research Regulatory Oversight Office for her review and oversight of study
human subjects protections.
Several people assisted us with the many technical challenges involved in fielding
such a large survey across DoD services. We thank Gary Lindeen at the Headquarters
xvii

xviii

Sexual Assault and Sexual Harassment in the U.S. Military

Marine Corps Manpower and Reserve Affairs, who assisted us to send survey invitations to Marines who lacked military email addresses; Colonel Scott K. Jackson,
who served as the J-6 liaison to the project, helping RAND troubleshoot technical
challenges encountered throughout the study; and Robert Flowe, Tanya Culbert, and
Jose Fernandez, who intervened when time was running out to arrange for RAND to
obtain DoD digital certificates for email invitations.
We have benefitted from a strong and critical set of internal and external quality
assurance reviewers, including Cynthia Cook, Greg Ridgeway, John Winkler, Bernie
Rostker, and Daniel Ginsberg, all of whom have provided valuable guidance throughout this effort. We also thank Lane Burgette for assistance with double-checking and
troubleshooting our statistical programming.
We also wish to thank the interviewers and helpdesk staff at Westat who supported the survey through its fielding period.
Finally, we wish to thank our sponsors in the OSD Sexual Assault Prevention
and Response Office. Nate Galbreath and Major General Jeffrey Snow have provided
strong support to the study team throughout the project.

Abbreviations

BAC

blood alcohol content

CDC

Centers for Disease Control and Prevention

DEERS

Defense Enrollment Eligibility Reporting System

DMDC

Defense Manpower Data Center

EEO

Equal Employment Opportunity

EEOC

Equal Employment Opportunity Commission

FBI

Federal Bureau of Investigation

GBM

Generalized Boosted Models

HRSAP

Human Resources Strategic Assessment Program

MEO

military equal opportunity

NCWSV

National College Women Sexual Victimization Survey

NCVS

National Crime Victimization Survey

NISVS

National Intimate Partner and Sexual Violence Survey

NRC

National Research Council

NVAWS

National Violence Against Women Study

NWS

National Women’s Study

NWS-R

National Women’s Study Replication

OMB

Office of Management and Budget
xix

Sexual Assault and Sexual Harassment in the U.S. Military

OSD

Office of the Secretary of Defense

OUSD/P&R Office of the Under Secretary of Defense for Personnel and Readiness
PTSD

posttraumatic stress disorder

RMWS

RAND Military Workplace Study

RSP

Response Systems to Adult Sexual Assault Crimes Panel

SAFE

Survey of Academic Field Experiences

SAPR

Sexual Assault Prevention and Response

SAPRO

Sexual Assault Prevention and Response Office

SEQ

Sexual Experiences Questionnaire

UCMJ

Uniformed Code of Military Justice

UCR

Uniform Crime Reporting

WDOT

Wisconsin Department of Transportation

WGRA

Workplace and Gender Relations Survey of Active Duty Members

WGRR

Workplace and Gender Relations Survey of Reserve Component
Members

CHAPTER ONE

The 2014 RAND Military Workplace Study
Andrew R. Morral, Kayla Williams,
Coreen Farris, and Kristie L. Gore

Background: The Workplace and Gender Relations Survey
The Department of Defense (DoD) Sexual Assault Prevention and Response Office
(SAPRO) released the results of the 2012 Workplace and Gender Relations Survey of
Active Duty Personnel (WGRA) in May 2013 (DMDC, 2013b). This report showed
a sharp increase in the numbers of service members experiencing past-year unwanted
sexual contacts in comparison to estimates last made in 2010. The 2012 estimate that
26,000 service members experienced unwanted sexual contacts in the past year were
widely reported in the news media, discussed in the blogosphere, cited by members of
Congress, and highlighted by advocacy organizations concerned with service members’ welfare.
Overall, the 2012 WGRA estimated that 6.1 percent of women and 1.2 percent of
men experienced unwanted sexual contact (i.e., rape, attempted rape, or nonconsensual
sexual contact) in the past year. The percentage of women had risen since 2010, but was
similar to estimates released in 2006 (see Exhibit 1.1); among men, the percentage who
experienced unwanted sexual contact had remained steady since 2006. The 2012 study
further detailed that, among women who experienced unwanted sexual contact, the
most serious event—or the event that had the “greatest effect” on them—was divided
evenly among those who had experienced penetrative assaults, attempted assaults, and
unwanted sexual touching. For men who experienced unwanted sexual contact, 51 percent indicated that the most serious event was unwanted sexual touching, 10 percent
said they experienced completed sex, 5 percent indicated attempted sex, and 34 percent could not be categorized either because the respondent said none of the response
options correctly described the event, or because they provided no response.
A confluence of factors around the time the 2012 results were released contributed to prolonged and intense media attention to the problem of sexual assaults in the
military. These included the ongoing critical and commercial success of The Invisible
War (2012) documentary about sexual assault in the military, allegations of sexual
misconduct levied against an Army general that received extensive media coverage
(Zucchino, 2013), the high-profile dismissal in February 2013 of a fighter pilot’s sexual
assault conviction by his commanding general (Montgomery, 2013), and the widely
1

Sexual Assault and Sexual Harassment in the U.S. Military

Exhibit 1.1
Trends in the Percentage of Service Men and Women Who Indicated Experiencing
Unwanted Sexual Contact in the Past Year
8

Percentage of population

7
6

Women

5
4
3
2
Men
1
0

2006

2008

2010

2012

Year
RAND RR870/1-1.1

publicized arrest for sexual battery of the Air Force Sexual Assault Prevention Chief,
which occurred just days before the release of the 2012 report (Shinkman, 2013).
The 2012 WGRA report discussed rapes and sexual assaults as forms of unwanted
sexual contact, but the public discussion that developed around the 26,000 estimate
often revealed confusion over just what unwanted sexual contact might include. In
some cases, reports explicitly or implicitly suggested that 26,000 service members
were raped, as opposed to raped or sexually assaulted in other ways, such as unwanted
touching. In other cases, questions were raised about whether unwanted sexual contact
might include sexual harassment or even events that are not illegal.
For instance, in a Senate Oversight Hearing in June 2013, one senator suggested
that “Unwanted sexual contact is everything from somebody looking at you sideways
when they shouldn’t to someone pushing you up against the wall and brutally raping
you.” This senator urged service chiefs to develop a metric that would disentangle
sexual assault and sexual harassment to better determine how many “women and
men are being raped and sexually assaulted on an annual basis” (U.S. Senate, 2013).
Although these suggestions were based on a misunderstanding—unwanted sexual contact does not include sexual harassment and focuses on rapes and sexual assaults exclusively—this fact did not come to light during the oversight hearing, and indeed those
providing testimony agreed with the criticism and recommendations.

The 2014 RAND Military Workplace Study

Confusion about the 2012 findings also arose from high-profile criticisms of the
2012 WGRA methodology, some of which were also unfounded. A widely cited opinion piece in the Wall Street Journal argued that the “26,000 figure is such bad math—
derived from an unscientific sample set and extrapolated military-wide—that no conclusions can be drawn from it” (Rodman, 2013). This essay noted that women were
disproportionately represented in the sample of 2012 WGRA respondents, making the
sample unrepresentative of the military as a whole. This conclusion, however, did not
account for the sampling and weighting procedures used to ensure the results from the
sample of respondents were representative of the larger population. Such procedures are
conventional in all professional survey research, and they are required by the Office of
Management and Budget (OMB) for all federal surveys (OMB, 2006). The weighting
procedures employed for the 2012 WGRA followed these standard procedures appropriately (DMDC, 2012, 2014).
Others asserted that the focus on sexual assault in the military was politically
motivated and that differences between the DoD results and Bureau of Justice Statistics surveys of civilian women called the methodology into question (Scarborough,
2014), despite the fact that the Bureau’s estimates of sexual assault are widely thought
to underestimate the extent of the problem (National Research Council, 2014).
Contrary to these critiques, the Defense Manpower Data Center (DMDC) had
relied on standard, well accepted, and scientifically justified approaches to survey sampling and derivation of survey results as reported for the 2012 WGRA. Nevertheless, the concern that the survey was flawed or unscientific resonated with many who
doubted the 26,000 number both inside and outside the Congress and military.
In the National Defense Authorization Act for Fiscal Year 2013, responding to
growing concerns about sexual assault in the military, Congress directed the Secretary
of Defense to establish the Response Systems to Adult Sexual Assault Crimes Panel to
conduct an independent review of the military’s systems to investigate, prosecute, and
adjudicate sexual assault and related offenses. Through hearings and expert testimony,
this panel closely examined the WGRA and methods for conducting crime victimization surveys to provide a more reliable estimate of the incidence of sexual assault in
the military for Congress and the Secretary of Defense. It recommended working with
experts in the field to develop a new crime victimization survey, and offered several
recommendations for how that survey could improve on the existing one.1
After discussions with leadership of the Senate Armed Services Committee in the
fall of 2013, DoD agreed to have the scheduled 2014 WGRA designed, administered,
and analyzed by an independent organization. DoD selected the RAND Corporation
to conduct the 2014 survey and additional studies to evaluate the validity of the survey
results. This report describes RAND’s approach to estimating the number of service
members who experienced sexual assault or sexual harassment in the past year.

Sexual Assault and Sexual Harassment in the U.S. Military

The RAND Military Workplace Study
To ensure the independence and validity of the 2014 WGRA, DoD asked the RAND
research team to redesign the approach used previously in the WGRA surveys if
changes would improve the accuracy and validity of the survey results for criminal and
policy violations. More generally, core objectives for the 2014 survey were to
• establish precise and objective estimates of the percentage of service members who
experienced sexual assault, sexual harassment, and gender discrimination in the
past year
• describe the characteristics of these incidents, such as where and when they
occurred, who harassed or assaulted the member, whether it was reported, and
what services the members sought
• identify barriers to reporting these incidents and barriers to the receipt of support
and legal services.
In consultation with White House staff, DoD stipulated that the survey would
go out to a much larger sample of active-duty members than in the past, to include a
census of all women and 25 percent of men. Moreover, a smaller sample of guard and
reserve members would be included sufficient to support comparisons of sexual harassment and assault in the active-duty and reserve components.
Because we intended to make extensive changes to the measurement of sexual
assault and harassment, we proposed that a random subsample of respondents receive
the questions used in prior WGRA administrations. Comparison of prevalence rates
assessed with the new and old versions of the items would provide important information on whether the two approaches differ in terms of the number or types of sexual
assault and harassment they assess. Moreover, overlapping administrations of the old
and new survey questions could provide information useful for transitioning between
earlier time-series data on these offenses and a new time series if future WGRA surveys
incorporate the questionnaire developed by RAND. In addition, we changed the name
of the survey from WGRA to the 2014 RAND Military Workplace Study to highlight
that it was being conducted independently of DoD and to reduce the association of the
survey with other gender-relations surveys that might appeal to a narrower segment of
the armed forces population than a more general military workplace construct.
The very large sample size stipulated for this study is particularly valuable for
understanding the experiences of relatively small subgroups in the population. For
instance, in the smaller 2012 WGRA, 117 men indicated that they had experienced
what the WGRA defined as unwanted sexual contact in the past year. This small
sample limited the generalizations that could be made about the sexual assault experiences of men in the military. By more than quadrupling the sample size, RAND’s data
will provide more-detailed information on men’s experiences, and the experiences of
other subpopulations, with the sexual assault response systems available to them.

The 2014 RAND Military Workplace Study

In contrast, it is not necessary to collect general experiences and attitudes from
such a large sample, and doing so would be wasteful of service members’ time (it would
not provide meaningful improvements in survey estimates). Therefore, we designed the
survey so that each question was posed to only as many service members as was necessary to provide the needed precision required for the question. We implemented this
plan by randomly assigning respondents to receive one of four different survey forms
(see Survey Instrument at the end of this report):
1. A “long form,” consisting of a sexual assault module (including screening and follow-up items); a sex-based military equal opportunity (MEO) violation module,
which assesses sexual harassment and gender discrimination; and questions on
demographics, psychological state, command climate, and attitudes and beliefs
about sexual assault. In addition, the long form included service-specific questions that were added at the request of three services. These questions were seen
only by members of the service that contributed them.
2. A “medium form,” consisting of the sexual assault module, the sex-based MEO
violation module, and some demographic questions.
3. A “short form,” consisting of the sexual assault module, only the screening items
from the sex-based MEO violation module, and demographic questions. Thus,
these respondents did not complete the full, sex-based MEO violation assessment.
4. The “prior WGRA form,” which, as described above, included questions from
the 2012 WGRA, including the unwanted sexual contact, sexual harassment,
and gender discrimination assessments from that survey.
Throughout the development of the survey instruments, the study design, and
the analysis plan, RAND worked in close consultation with a scientific advisory board
that included experts on civilian and military law, the assessment of sexual assault and
Table 1.1
Survey Forms and Modules
Items

Long Form Medium Form Short Form

Demographic questions

Sexual assault module

Sex-based MEO violation screening items

Sex-based MEO violation follow-up items

Attitude and climate questions

2012 WGRA questions

Prior WGRA Form

Sexual Assault and Sexual Harassment in the U.S. Military

sexual harassment, victim services, and survey methodology (see page iii for a list of
scientific advisory board members). Members of this board contributed their expertise
in formal meetings and as they reviewed many iterations of study plans and materials. In addition, RAND researchers consulted with other experts, advocacy groups,
and service members, including many who had experienced sexual assault or sexual
harassment, to ensure that each survey question assessed the legal construct it was
designed to measure as accurately as possible and to ensure that respondents could reliably understand the meaning of each question.
This report is the first in a series that will describe the methodology and results
of the RAND Military Workplace Study (RMWS) for DoD active-duty forces, for
National Guard and reserve forces, and for Coast Guard active-duty and reserve forces.
This volume covers the study methodology—the study design (Chapter Three), questionnaire development (Chapter Four), and analytic methods (Chapter Five) used in
calculating estimates—as well as a review of what is known about the prevalence of
sexual assault and sexual harassment in civilian and military populations, and some
of the challenges of measuring these crimes and violations (Chapter Two). The report
closes with a summary of key improvements of the RMWS designed to increase the
precision and validity of the survey results.

CHAPTER TWO

Measurement of Sexual Harassment and Sexual Assault
Coreen Farris, Amy Street, Andrew R. Morral,
Lisa Jaycox, and Dean Kilpatrick

After three decades of research on civilian and military sexual harassment and sexual
assault, there continues to be uncertainty about the true percentage of members of
the American civilian and military workforces who are sexually harassed. Likewise,
estimates of the percentage of American men and women who are sexually assaulted
each year or in their lifetimes vary substantially based on measurement differences
across surveys. In many ways, this variation serves to illustrate the challenges faced in
this research area and the careful decisionmaking that is required. To provide context
to the RAND research team’s design and methodological decisions presented in this
volume, this chapter describes variation in the existing epidemiological estimates of the
prevalence of sexual harassment and sexual assault in military and civilian populations,
and outlines the measurement challenges that all victimization surveys must address.
Sexual Harassment in the American Civilian and Military Workforce
Sexual harassment is defined by the Equal Employment Opportunity Commission
(EEOC) as unwelcome sexual advances, requests for sexual favors, and other verbal or
physical conduct of a sexual nature that (a) affects the terms, conditions, or employment decisions related to the individual’s job (quid pro quo harassment) or (b) creates an intimidating, hostile, or offensive working environment (“hostile workplace”
harassment; EEOC, 1980, 29 C.F.R. § 1604.11). It defines gender discrimination as a
separate equal employment opportunity (EEO) violation involving unfavorable treatment of an applicant or employee because of their gender. In contrast, the most widely
used questionnaires assessing sexual harassment typically subsume gender discrimination as one type of sexual harassment (e.g., Fitzgerald, Gelfand, and Drasgow, 1995).
As such, most published research makes no distinction between gender discrimination
and sexual harassment.
Although researchers have investigated sexual harassment in the workplace since
the 1980s, there is a paucity of high-quality epidemiological studies to provide national
estimates for the prevalence of workplace sexual harassment among American men and
women. In 2003, when Ilies and colleagues published a meta-analysis of sexual harass7

Sexual Assault and Sexual Harassment in the U.S. Military

ment rates across studies of smaller organizations or workplace fields, there had not
yet been a study of sexual harassment that relied on a nationally representative sample
(Ilies et al., 2003). Since 2003, we are aware of two national studies of sexual harassment among civilians (Das, 2009; Shannon, Rospenda, and Richman, 2007). Within
the military, estimates are derived from the biennial Workplace and Gender Relations Survey of Active Duty Members (WGRA) and Workplace and Gender Relations
Survey of Reserve Component Members (WGRR), and an independent investigation
of sexual harassment in a representative sample of former reservists (Street et al., 2008).
Ilies and colleagues (2003) identified 71 studies that reported an incidence rate of
sexual harassment in at least one sample of women.1 Here, we summarize results from
only those studies that relied on a random, stratified random, or complete census of an
organization. In studies that asked women if they had been “sexually harassed,” an average of 24 percent of surveyed women indicated that they had been sexually harassed
(95-percent confidence interval: 19–28 percent). Surveys that asked women to indicate
whether they had experienced a series of behaviors consistent with sexual harassment
(e.g., “Have you ever been in a situation where you felt you were being subtly bribed
with some sort of reward to engage in sexual behavior with a coworker?”; Fitzgerald,
Gelfand, and Drasgow, 1995) were associated with a higher average, 58 percent of
women (95-percent confidence interval: 52–63 percent; Ilies et al., 2003). For both
the direct query method (“have you ever been sexually harassed?”) and the series of
sexually harassing experiences, the average rate in military samples was larger than the
average of rate in civilian samples (36 percent and 69 percent, respectively; Ilies et al.,
2003). Despite the value of meta-analyses to summarize a broad literature on sexual
harassment, the rates reported by Ilies and colleagues are challenging to interpret as
they combine rates derived from non-comparable periods (e.g., annual, career to date)
to produce averages.
In the first sexual harassment study to employ a nationally representative sample
of both men and women, Shannon, Rospenda, and Richman (2007) used randomdigit dialing to reach a stratified random sample of working adults. Like many other
sexual harassment researchers, to assess sexual harassment experience in the past
12 months, they relied on a modified version of the Sexual Experiences Questionnaire
(SEQ; a questionnaire that combines sexual harassment and gender discrimination
into a single estimate of sexual harassment; Fitzgerald, Gelfand, and Drasgow, 1995).
Respondents completed the questionnaire twice, at time points separated by one year,
and were counted as sexually harassed if they indicated having experienced any one
type of sexually harassing behavior (e.g., unwanted romantic overtures) more than one
time (Shannon, Rospenda, and Richman, 2007). Among men, 44 percent indicated
that they had experienced incidents consistent with sexual harassment within the twoyear study period, and 46 percent of women indicated the same (Shannon, Rospenda,
and Richman, 2007). See Table 2.1 for a summary of nationally representative civilian
and military surveys assessing sexual harassment.

Table 2.1
Sexual Harassment Survey Comparison Chart
Source
Shannon,
Rospenda, and
Richman (2007)

Construct and Definition
Sexual Harassment: Unwelcome and
offensive behaviors based on one’s sex

Measurement Approach

Prevalence Estimates

1992 U.S. National Sexual Harassment: Sexual advances,
Health and Social propositions, or unwanted sexual
Life Survey (Das,
discussions from co-workers or supervisors.
2009)

Single survey item worded as follows: “Sometimes
at work, [men/women] find themselves the object
of sexual advances, propositions, or unwanted
sexual discussions from co-workers or supervisors.
The advances sometimes involve physical contact
and sometimes just involve sexual conversations.
Has this ever happened to you?”

Sexual Harassment over Career
41% of employed women
36% of employed men

Street et al. (2008) Sexual Harassment: Unwelcome sexual
advances, requests for sexual favors, and
other verbal or physical conduct of a sexual
nature that occur in a work setting

The full 24-item, DoD-specific SEQ (Fitzgerald et al., Sexual Harassment over Military
1999). To be classified as sexually harassed, former Career
reservists had to indicate that four different
60% of female reservists
potentially harassing experiences occurred or at
27% of male reservists
least one severe form of harassment occurred (i.e.,
quid pro quo exchanges, workplace-based sexual
assault).

Workplace and
Gender Relations
Survey of Active
Duty Service
Members (DMDC,
2013b)

Sexual Harassment: 12 behaviorally specific items
derived from the DoD SEQ (Fitzgerald et al., 1999).
Categorized as sexually harassed if respondent
indicates that one or more items occurred AND
indicates that they believe it constituted “sexual
harassment.”
Sexist Behavior: Four behaviorally specific items
derived from the DoD SEQ (Fitzgerald et al., 1999).
Categorized as experiencing “sexist behavior”
if respondent indicates that one or more items
occurred.
Gender Discrimination: 12 items assessing
evaluation, career, and assignment discrimination.
Respondents are classified as experiencing gender
discrimination if they indicate that a discrimination
event occurred in the past 12 months, believe their
gender was a factor, and believe at least one event
constituted sex discrimination.

Sexual Harassment: Crude/offensive
behavior (e.g., repeatedly told sexual stories
or jokes that are offensive); unwanted
sexual attention (e.g., unwanted attempts
to establish a romantic sexual relationship
despite efforts to discourage it); or sexual
coercion (e.g., treated badly for refusing to
have sex).
Sexist Behavior: Verbal or nonverbal
behaviors that convey insulting, offensive,
or condescending attitudes based on the
gender of the respondent.
Gender Discrimination: Treating individuals
differently in their employment specifically
because of their sex (e.g., unfair or unequal
access to professional development
resources and opportunities due to a
member’s gender).

Sexual Harassment (Annual)
23% of service women
4% of service men
Sexist Behavior (Annual)
47% of service women
15% of service men
Gender Discrimination (Annual)
12% of service women
2% of service men

Sexual Harassment over 2 years
46% of employed women
44% of employed men

Measurement of Sexual Harassment and Sexual Assault

9 behaviorally specific items derived from the
SEQ (Fitzgerald, Gelfand, and Drasgow, 1995).
Categorized as sexually harassed when any one
item is described as occurring “more than once” or
sexual coercion occurred at least once.

Workplace and
Gender Relations
Survey of ReserveComponent
Members (DMDC,
2013a)

Construct and Definition

Measurement Approach

Prevalence Estimates

Sexual Harassment: Crude/offensive
behavior (e.g., repeatedly told sexual stories
or jokes that are offensive); unwanted
sexual attention (e.g., unwanted attempts
to establish a romantic sexual
relationship despite efforts to discourage
it); or sexual coercion (e.g., treated badly for
refusing to have sex).
Sexist Behavior: Verbal or nonverbal
behaviors that convey insulting, offensive,
or condescending attitudes based on the
gender of the respondent.
Gender Discrimination: Treating individuals
differently in their employment specifically
because of their sex (e.g., unfair or unequal
access to professional development
resources and opportunities due to a
member’s gender).

Sexual Harassment (Annual)
18% of female reservists
2% of male reservists
Sexist Behavior (Annual)
34% of female reservists
10% of male reservists
Gender Discrimination (Annual)
12% of female reservists
2% of male reservists

Sexual Assault and Sexual Harassment in the U.S. Military

Source

Table 2.1—Continued

Measurement of Sexual Harassment and Sexual Assault

Relying on data from the 1992 U.S. National Health and Social Life Survey, Das
(2009) estimated that 36 percent of adult men and 41 percent of adult women who
have participated in the workforce are sexually harassed at some point during their
career. The sample was nationally representative of U.S. adults, and sexual harassment
was assessed with a single question. Respondents were queried as follows: “Sometimes
at work, [men/women] find themselves the object of sexual advances, propositions, or
unwanted sexual discussions from coworkers or supervisors. The advances sometimes
involve physical contact and sometimes just involve sexual conversations. Has this ever
happened to you?” Unlike the Shannon, Rospenda, and Richman (2007) estimate
above, this definition of sexual harassment excludes gender discrimination, but conceivably includes both quid pro quo harassment and hostile work environments.
For the military specifically, there are three sources of sexual harassment estimates
using a high-quality probability sample: the WGRA, the WGRR, and independent
work by Street and colleagues (2008). The Workplace and Gender Relations Surveys
rely on stratified, random samples of active-duty service members and reserve-component members. The report associated with the 2012 fielding included the estimate
that 4 percent of active-duty service men and 23 percent of active-duty service women
indicate that they were sexually harassed in the previous 12 months (DMDC, 2013b).
Among reserve and guard members, 2 percent of men and 18 percent of women indicate being sexually harassed in the previous 12 months (DMDC, 2013a). The lower
rates of sexual harassment relative to those in civilian samples likely reflect both the
difference in time periods and also the study’s more restrictive sexual harassment definition. Respondents are classified as sexually harassed only if they label their experience sexual harassment and only if they indicated that they have had experiences consistent with a hostile workplace or a sexual quid pro quo exchange. Labeling is generally
not required in civilian studies.
The active-duty and reserve-component surveys assess gender discrimination
with two different scales. The first is the four-item subscale of the SEQ, labeled “sexist
behavior.” Respondents are classified as having experienced sexist behavior if they indicate that they have experienced one or more sexist behaviors (e.g., a coworker “referred
to people of your gender in insulting or offensive terms”). Labeling the experience discrimination or harassment is not necessary. In 2012, based on this scale, 15 percent of
active-duty men and 47 percent of active-duty women indicated experiencing sexist
behavior (DMDC, 2013b). Among reserve and guard members, 10 percent of men and
34 percent of women indicated that they had experienced sexist behavior (DMDC,
2013a). A separate, 12-item gender discrimination scale includes items assessing negative career actions in which the respondent believes their gender was a factor (e.g., “you
were rated lower than you deserved on your last military evaluation” and “your gender
was a factor” [DMDC, 2013a]). To be classified as having experienced gender discrimination, a respondent had to both indicate experiencing an adverse career action
in which they believed their gender was a factor and explicitly label that experience “sex

Sexual Assault and Sexual Harassment in the U.S. Military

discrimination” (DMDC, 2013a). By these criteria, 2 percent of active-duty men and
12 percent of active-duty women were classified as having experienced gender discrimination in the past 12 months (DMDC, 2013b). Among reserve and guard members,
the results were the same: 2 percent of men and 12 percent of women were classified as
experiencing gender discrimination in the past year.
In an investigation of former members of the reserve forces of the U.S. military,
Street and colleagues (2008) accessed a stratified, random sample of recently separated
reservists. One strength of their study design was that by sampling from all recently
separated reservists (as opposed to current members), the sample includes both satisfied members (e.g., retirees) as well as those who might be more inclined to leave the
service (e.g., individuals who have been sexually harassed; Antecol and Cobb-Clark,
2006). The study relied on former reservists’ responses to the 24-item DoD-specific
SEQ (Fitzgerald et al., 1999). To be classified as sexually harassed, former reservists
had to indicate that they had experienced four different potentially harassing behaviors
or at least one severe form of harassment (i.e., quid pro quo exchanges). Sixty percent
of women and 27 percent of men indicated experiences that were consistent with these
criteria during their military service (Street et al., 2008). Controlling for age and race,
female former reservists were 5.5 times more likely to indicate that they experienced
sexually harassing behaviors during their time in the reserves than were males (Street
et al., 2007).
Sexual Assault of American Civilians and Military Service Members
DoD Directive 6495.01 defines sexual assault, consistent with Uniform Code of Military Justice (UCMJ) Article 120, as “intentional sexual contact characterized by use
of force, threats, intimidation, or abuse of authority or when the victim does not or
cannot consent. Sexual assault includes rape, forcible sodomy (oral or anal sex), and
other unwanted sexual contact that is aggravated, abusive, or wrongful (including
unwanted and inappropriate sexual contact), or attempts to commit these acts” (DoD,
2013b, 2012). To be clear, DoD defines consent to include “words or overt acts indicating a freely given agreement to the sexual conduct at issue by a competent person,” and
consent cannot be satisfied if the person is “sleeping or incapacitated, such as due to
age, alcohol or drugs, or mental incapacity” (DoD, 2013b, 2012). These definitions are
consistent with most U.S. state statutes on sexual assault, the majority of which include
both penetrative and non-penetrative sexual contact crimes, offender behaviors beyond
physical force such as threats and intimidation, and inclusion of situations in which the
victim is not legally capable of providing consent (Tracy et al., 2012).
According to the Uniform Crime Reporting (UCR) system, 84,376 forcible rapes
were reported to United States law enforcement in 2012 (FBI, 2012). This translates to
59.2 forcible rapes per 100,000 females. Unfortunately, for the purpose of document-

Measurement of Sexual Harassment and Sexual Assault

ing the extent of the problem, the UCR estimate is limited to only assaults by men
against women and only assaults perpetrated by force. (See Table 2.2 for definitions
and prevalence estimates for sexual assault in this and other sources.) Furthermore,
while the UCR data provides useful information about the number of female forcible
rape victims who report the assault to the criminal justice system, it provides little insight
into the total number of victims because it does not include unreported cases. Sexual
assault is a notoriously underreported crime, with investigators estimating that anywhere from 16–98 percent of sexual assaults are not reported to the police (Ahrens
et al 2007; Fisher et al., 2003; Jacques-Tiura et al., 2010; National Victim Center and
Crime Victims Research and Treatment Center, 1992; Langton et al., 2012; Tjaden and
Thoennes, 2000; Truman and Langton, 2014). Similarly, sexual assaults against service
members appear to be reported only rarely. For instance, whereas the 2012 WGRA
estimated that as many as 26,000 members may have experienced an unwanted sexual
contact in the past year, in 2012 only about 3,400 restricted or unrestricted reports of
sexual assaults were filed with military officials (DoD, 2013a). Victims face numerous
barriers to disclosure, including minimization of the severity of the assault, feeling
ashamed or embarrassed, fear that the police will not help them, and fear that the perpetrator will retaliate (Bachman, 1998; Black and Merrick, 2013; Cohn et al., 2013;
Fisher et al., 2003; Langton et al., 2012; Walsh et al., 2010).
Given that reporting to the police is not the norm among sexual assault victims,
it is generally recommended that UCR data not be used as an indicator of the prevalence of sexual assault victimization among U.S. residents (Bachman, 2012; Fisher and
Cullen, 2000). Instead, victimization surveys are the preferred approach to estimate
the extent of the crime. Briefly described, victimization surveys consist of contacting a
probability sample of the population, asking them questions that capture the types of
victimization experiences of interest, and using the data on sexual assault victimization
in the sample to estimate the overall prevalence in the population.
The only victimization survey that is conducted yearly is the National Crime
Victimization Survey (NCVS; Truman and Langton, 2014). Sponsored by the Bureau
of Justice Statistics, the NCVS produces annual estimates of criminal victimization,
including sexual assaults. At this writing, the most recent fielding was 2013, which
produced an estimated annual prevalence rate of sexual assault of 0.1 percent of the
U.S. population (Truman and Langton, 2014). This estimate is almost certainly an
underestimate; the NCVS has been widely criticized for employing poor survey design
that is expected to bias downward the estimated annual rate of sexual assault (Fisher,
2004; Fisher and Cullen, 2000; Kilpatrick, 2004; Koss, 1996; National Research
Council, 2014). For instance, the survey does not provide the respondent with a definition of sexual assault, and it does not query incidents in which the victim was not
legally capable of consent (Bureau of Justice Statistics, 2008). Indeed, these and other
perceived shortcomings of the NCVS were judged sufficiently severe that a 2014 report
from the National Research Council recommended against further use of NCVS esti-

Construct and Definition

Measurement Approach

Prevalence Estimates
Rape
Academic Yeara: 2.5% of college
women
Lifetime: 21.0% of college women
Unwanted Sexual Contact
Academic Year: 1.9% of college
women
Lifetime: 35.5% of college women

National College
Women Sexual
Victimization Study
(NCWSVS; Fisher,
Cullen, and Turner,
2000)

Rape (completed or attempted): Unwanted penetration
(vaginal, oral, or anal) by force or the threat of force.
Unwanted Sexual Contact (completed or attempted):
Unwanted sexual contact (not penetration) with force
or the threat of force. Sexual contact includes touching;
grabbing or fondling of the breasts, buttocks, or genitals,
either under or over clothes; kissing; licking or sucking; or
some other form of unwanted sexual contact.

12 behaviorally specific
screening questions
adapted from the National
Women’s Study (NWS)
followed by an incident
report

National College
Women Sexual
Victimization Survey
(Kilpatrick et al.,
2007)

Rape: An unwanted sexual act involving oral, anal, or
vaginal penetration that occurs (a) by force or threat of
force or (b) when the victim is passed out or awake but too
drunk or high to know what she is doing or to control her
behavior.

Four behaviorally specific
Rape
items assessing forcible
Annual: 5.2% of women
rapes and two behaviorally
Lifetime: 11.5% of women
specific items assessing
incapacitated rapes

National Crime
Victimization Survey
(NCVS; Truman and
Langton, 2014)

Rape (completed or attempted): Includes both psychological
coercion as well as physical force. Forced sexual intercourse
means vaginal, anal, or oral penetration by the offenders(s).
Sexual Assault (completed or attempted): Attacks or
attempted attacks generally involving unwanted sexual
contact between the victim and offender. May or may
not involve force and include such things as grabbing or
fondling. Includes verbal threats.

Four screening items
that query attacks, rape,
attempted rape, sexual
attacks, and forced or
coerced unwanted sexual
activity. Respondents do
not receive a definition of
these constructs.

Rape or Sexual Assault
Annual: 0.1% of U.S. residents

National Intimate
Partner and Sexual
Violence Survey
(NISVS; Breiding
et al., 2014)

Rape (completed or attempted): Unwanted vaginal, oral,
or anal penetration through the use of physical force or
threats to physically harm and includes times when the
victim was drunk, high, drugged, or passed out and unable
to consent.

21 sexual violence items,
13 of which measured
attempted or completed
rapes

Attempted/Completed Rape
Annual: 1.6% of women
Lifetime: 19.3% women; 1.7% of
men

National Violence
Against Women
Study (NVAWS;
Tjaden and
Thoennes, 1998)

Rape (completed or attempted): An event that occurred
without the victim’s consent that involved the use or threat
of force to penetrate the victim’s vagina or anus by penis,
tongue, fingers, or object, or the victim’s mouth by penis.

Five behaviorally specific
screening questions
(adapted from the NWS)

Attempted/Completed Rape
Annual: 0.3% of women; 0.1% of
men
Lifetime: 17.6% of women; 3.0%
of men

Sexual Assault and Sexual Harassment in the U.S. Military

Source

Table 2.2
Sexual Assault Victimization Survey Comparison Chart

Table 2.2—Continued
Source

Construct and Definition

Measurement Approach

Prevalence Estimates

National Women’s
Study (NWS;
National Victim
Center and Crime
Victims Research and
Treatment Center,
1992)

Rape (completed): An event that occurred without the
woman’s consent, involved the use of a force or threat
of force, and involved penetration of the victim’s vagina,
mouth, or rectum.

Four behaviorally specific
items assessing forcible
rapes

Forcible Rape
Annual: 0.7% of women
Lifetime: 13% of women

National Women’s
Study Replication
(NWS-R; Kilpatrick
et al., 2007)

Four behaviorally specific
Rape
items assessing forcible
Annual: 0.9% of women
rapes and two behaviorally
Lifetime: 18% of women
specific items assessing
incapacitated rapes
Compilation of official
reports filed with law
enforcement

Forcible Rape
Annual: 0.05% of women

Workplace and
Gender Relations
Survey of Active
Duty Service
Members (DMDC,
2013b)

Unwanted Sexual Contact (attempted and completed):
Intentional sexual contact that was against a person’s
will or which occurred when the person did not or could
not consent, and includes completed or attempted sexual
intercourse, sodomy (oral or anal sex), penetration by an
object, and the unwanted touching of genitalia and other
sexually-related areas of the body.

One gate question with
five follow-up items to
assess the type of sexual
assault

Unwanted Sexual Contact
Annual: 6.1% of women; 1.2% of
men
Military Career: 23% of women;
4% of men
Before Military Service: 30% of
women; 6% of men

Workplace and
Gender Relations
Survey of Reserve
Component
Members (DMDC,
2013a)

One gate question with
five follow-up items to
assess the type of sexual
assault

Unwanted Sexual Contact
Annual: 2.8% of women; 0.5%of
men
Military Career: 18% of women;
2% of men
Before Military Service: 25% of
women; 4% of men

a Incidents that occurred “Since school began in fall 1996.” On average, a 6.9-month reporting period (Fisher, Cullen, and Turner, 2000).

Measurement of Sexual Harassment and Sexual Assault

Uniform Crime
Forcible Rape (attempted and completed): Carnal
Reports (UCR)
knowledge of a female forcibly and against her will (i.e.,
Summary Reporting forcible vaginal penetration by a penis).
System (FBI, 2012)

Sexual Assault and Sexual Harassment in the U.S. Military

mates of sexual assault (National Research Council, 2014). Instead, the expert panel
advised the Bureau of Justice Statistics to develop a new, stand-alone survey that relies
on state-of-the-art approaches, as identified in the scientific literature, to sexual assault
measurement (National Research Council, 2014).
In 2010, the Centers for Disease Control and Prevention (CDC) began fielding a
recurrent victimization survey of a nationally representative sample of men and women
living in the United States. The National Intimate Partner and Sexual Violence Survey
(NISVS) definition of rape includes attempted and completed assaults perpetrated
either by physical force or facilitated by alcohol or drug intoxication (Black et al., 2011).
At this writing, the 2011 fielding had been released with estimates that 1.6 percent of
women had been raped in the past year and 19.3 percent in their lifetime (Breiding et
al., 2014). For men, the sample was not adequate to estimate a 12 month prevalence
rate; the lifetime prevalence of rape among men was 1.7 percent (Breiding et al., 2014).
During the 2010 fielding of the NISVS, and with joint support from DoD, the
CDC added an oversample of active-duty service women, of whom 36.3 percent indicated that they had been sexually assaulted in their lifetime, and 5.6 percent indicated
that they had experienced a sexual assault in the previous 12 months (Black and Merrick, 2013). Although these estimates appear larger than among civilians, after controlling for demographic differences between active-duty women and civilian women, the
annual prevalence of sexual assault among military and non-military women in the
NISVS sample was not significantly different (Black and Merrick, 2013). Moreover,
the NISVS estimate of the percentage of active-duty women sexually assaulted in the
past year was similar to a 2010 DoD estimate (DMDC, 2011).
Three additional notable epidemiological studies have estimated the rate of sexual
assault among a nationally representative sample, and two have provided recent, highquality estimates of sexual victimization among college women. The National Women’s
Study (NWS) limited measurement to forcible rapes only, a deliberately conservative measure of assaults (National Victim Center and Crime Victims Research and
Treatment Center, 1992). Like the NCVS, the survey’s longitudinal design permitted
bounded interviews, a strategy used to limit telescoping—a recall bias in which respondents indicate events occurred more recently than they really did (Lynn et al., 2005;
National Research Council, 1984).2 According to NWS results, 13 percent of women
indicated experiencing a forcible rape in their lifetime, and 0.7 percent of women were
forcibly raped in the previous 12 months (National Victim Center and Crime Victims
Research and Treatment Center, 1992).
The second nationally representative study, conducted in 1996, was the National
Violence Against Women Survey (NVAWS; Tjaden and Thoennes, 1998), results of
which indicated that 17.6 percent of women experienced an attempted or completed
rape in their lifetime and 0.3 percent experienced an attempted or completed rape in
the previous 12 months. Despite the survey title, the NVAWS also included a nationally representative sample of men and estimated that 3.0 percent of men experienced

Measurement of Sexual Harassment and Sexual Assault

an attempted or completed rape in their lifetime, and 0.1 percent of men had been
assaulted in the past year (Tjaden and Thoennes, 1998). Note that this survey limited
incidents to those perpetrated using physical force or threat of force, excluding incidents in which the victim was not legally capable of providing consent.
The National Women’s Study Replication (NWS-R; Kilpatrick et al., 2007) was
conducted in the fall of 2005 with a national household probability sample of adult
women using similar methodology to that used in the previously described NWS.
However, in addition to forcible rapes, this survey also measured completed rapes that
resulted from an inability to consent due to incapacitation associated with alcohol or
drug consumption. Results indicated that 18 percent of U.S. adult women had experienced a lifetime completed rape and that 1 percent had experienced a completed rape
during the past year. The percentage of women who had experienced a lifetime forcible rape was 16 percent, in contrast to 13 percent found in the first NWS, which was
conducted 15 years previously (National Victim Center and Crime Victims Research
and Treatment Center, 1992; Kilpatrick et al., 2007). It is noteworthy that the lifetime
and past-year estimates for any rape from the NWS-R were essentially replicated by
the NISVS findings.
Two nationally representative studies have assessed sexual violence specifically
among college students. The National College Women Sexual Victimization Survey
(NCWSV; Fisher, Cullen, and Turner, 2000) surveyed a nationally representative
sample of college women attending two- and four-year colleges and universities in the
fall of 1996. Nearly 3 percent of college women indicated that they had experienced
an attempted or completed rape in the previous academic year (a time frame limited to
7 months on average), and one in five had a lifetime history of attempted or completed
rape (Fisher, Cullen, and Turner, 2000). The National College Women Study (NCWS;
Kilpatrick et al., 2007) surveyed a representative sample of college women attending
two- and four-year colleges and universities in the fall of 2005. The sampling and other
methodology were quite similar to that used in the study by Fisher and colleagues
described above, but this survey also measured completed rapes involving inability to
consent due to incapacitation associated with alcohol and other drug consumption.
Among college women, 11.5 percent had experienced a forcible or incapacitated rape
in their lifetime and 5.2 percent reported experiencing a rape in the past year. These
estimates are not directly comparable with the NCWSV because that study included
attempted rapes, whereas the NCWS did not, and because the NCWSV excluded
incapacitated rapes. Nevertheless, the findings from these surveys, conducted ten years
apart, confirm the high prevalence of sexual assault among college women.
The biennially fielded WGRA provides regular estimates of the rates of “unwanted
sexual contact” among military service members (DMDC, 2013a, 2013b). Unwanted
sexual contact is defined, consistent with the UCMJ, to include both penetrative and
non-penetrative sexual assaults (DMDC, 2013c). However, the survey items lack the
detail necessary to definitively label the experience a sexual crime; hence, DoD termi-

Sexual Assault and Sexual Harassment in the U.S. Military

nology unwanted sexual contact. Survey results were used to estimate that 6.1 percent of
active-duty women and 1.2 percent of active-duty men experienced unwanted sexual
contact (i.e., may have been sexually assaulted) in 2012, and 23 percent of active-duty
women and 4 percent of active-duty men experienced unwanted sexual contact during
their military career (DMDC, 2013b). Within the reserve component, an estimated
2.8 percent of women and 0.5 percent of men experienced unwanted sexual contact
in 2012, and an estimated 18 percent of women and 2 percent of men experienced
unwanted sexual contact during their military careers (DMDC, 2013a).
Across national studies, regardless of the estimate selected, these numbers reveal
that sexual assault is a significant and widespread problem among both U.S. civilians and military service members. Given differences in survey methodology, time
frames, and the crimes included under the sexual assault umbrella, it is difficult to
compare estimates across studies to establish a definitive estimate. Nonetheless, collectively, the results illustrate that large numbers of men and women are victimized,
and that women have particularly elevated risk of lifetime and past-year victimization.
Although continuing to improve measurement precision will provide the data necessary to best target prevention and response resources to those in need, there is no longer
a question that an unconscionable number of Americans are survivors of sexual assault.
Challenges in Measuring Sexual Harassment and Sexual Assault
Measuring any stigmatizing behavior or experience in a survey poses methodological
challenges (Tourangeau and Yan, 2007). Respondents may be reluctant to participate
fully and candidly, the researcher and respondent may not share the same definitions
of the event, and seemingly trivial methodological changes can substantially influence
survey results. Below we review the primary challenges associated with designing a
high-quality survey to measure the prevalence of sexual harassment and sexual assault.
The reviewed measurement choice points and their likely effect on prevalence estimates
are summarized in Table 2.3.
What Events Are Included and How They Are Defined

In any victimization survey, decisions are made about what the survey is attempting to
measure (e.g., what types of unwanted sexual events will be included in the survey) and
how included events should be defined (e.g., should sexual harassment or sexual assault
events be defined using legal definitions or should these events be defined more broadly
to include the events that are unpleasant but not illegal). Once decisions have been
made about which events to include and how they should be defined, the next critical
issue is how exposure to those events should be defined for the survey respondent and
measured (e.g., what specific questions will be asked to measure whether respondents

Measurement of Sexual Harassment and Sexual Assault

Table 2.3
Sexual Harassment and Sexual Assault Measurement Decisions and Likely Influence on
Estimated Prevalence of These Experiences
Measurement Decision

Prevalence Estimate

Require respondent to self-label the event as “sexual harassment” or “sexual
assault”

Depressed

Define sexual assault narrowly (e.g., to include only forceful assaults or only
vaginal assaults)

Depressed

Single gating question to assess entire class of experiences with further
assessment only of respondents who respond “yes” to the gating question

Depressed

Sexual assault questions embedded in a survey context that assesses crimes

Depressed

Experience defined consistently with the local jurisdiction or to capture legal
requirements that are shared across most jurisdictions

Varies

Nonresponse bias: Invited respondents’ choices to participate are non-random

Varies

Provide a list of explicit and behaviorally specific questions that assess the
relevant events

Increased

Define sexual assault inclusively to include all types of sexual assault (e.g.,
assaults that occur while the victim is incapacitated)

Increased

Comprehensive list of screening questions administered to all respondents

Increased

Sexual assault questions embedded in a public health or neutral survey context

Increased

Respondent completes survey in a private setting

Increased

have experienced the events that will be included in the survey). All of these decisions
are important, and how they are decided can substantially influence survey results.
There has been limited variation in measurement of sexual harassment, partially
due to the reliance on one widely disseminated measure of sexual harassment—the
SEQ (Fitzgerald, Gelfand, and Drasgow, 1995). Nonetheless, researchers do use different approaches to defining sexual harassment via this questionnaire. While it is
common to categorize any respondent who indicates experiencing at least one sexually
harassing behavior as “sexually harassed” (e.g., Shannon, Rospenda, and Richman,
2007), some researchers select more conservative thresholds, such as requiring that the
respondent also label their experience as “sexual harassment” (DMDC, 2013b). Others
require respondents to indicate that they experienced four or more events to establish
that the harassment was persistent (Street et al., 2007). As noted above, rates of sexual
harassment derived from studies that only require respondents to indicate experiencing sexually harassing behaviors typically are higher than rates of sexual harassment
that require the victim to indicate that they believe the experience constituted sexual
harassment.
In contrast, sexual assault measures vary substantially both in the events being
measured and the definitions of the construct. For example, some surveys define sexual

Sexual Assault and Sexual Harassment in the U.S. Military

assault narrowly (e.g., forceful, penetrative assaults only), whereas others use a broad
definition that captures penetrative and non-penetrative assaults, attempted and completed assaults, and those perpetrated by force, threats, or when the victim was not
capable of providing legal consent. Not surprisingly, these definitional differences are
associated with dramatic differences in the estimated rates and make it difficult to synthesize or compare sexual assault estimates across studies. (See Table 2.2 for review of
sexual assault definitions across surveys.)
The lack of consensus in research definitions of sexual assault may be due, in part,
to differences in the legal definitions of sexual assault across states. State law varies
as to whether injury is required to substantiate that force was used, whether proof of
consent requires words or action (i.e., disallowing inaction as an indicator of consent),
the minimum age of consent (ranges from 10 to 18), the extent to which intoxication
due to alcohol or drugs indicates inability to consent, and whether non-penetrative
sexual contact crimes must include skin-to-skin touching or whether touching through
clothing is included (Fisher and Cullen, 2000; Tracy et al., 2012). Some researchers
have sought to match the definition of sexual assault within their jurisdiction (Koss
and Gidycz, 1985), whereas others capture legal requirements that are relatively stable
nationwide (Black et al., 2011; Fisher and Cullen, 2000; Kilpatrick et al., 2007).
Behavioral Specificity of Survey Questions

Sexual assault and sexual harassment laws are complex and often poorly understood
by the general public. Many people, for example, are hesitant to label as rape assaults
that do not produce injury or assaults in which the victim was intoxicated even if all
legally required elements are present (Schwartz and Leggett, 1999; Bondurant, 2001;
Layman, Gidycz, and Lynn, 1996). Moreover, it cannot be assumed respondents know
or understand the legal definition of rape, sexual assault, and other sexual contact
crimes in their state and use the definitions consistently when describing their personal experiences. Because of these known misunderstandings about what constitutes
sexual assault, survey items that ask individuals directly about whether they have ever
been raped, sexually assaulted, or sexually harassed underestimate the true extent of
these experiences (Fisher and Cullen 2000; National Research Council, 2014). To
avoid these problems, published recommendations for measurement of sexual assault
and harassment typically endorse the use of behaviorally specific questions (Bachman, 1998; Fisher and Cullen, 2000; Kilpatrick, 2004; Koss, 1996; National Research
Council, 2014).
Behaviorally specific questions provide a detailed, lay description of the legal elements necessary to categorize the incident. Instead of asking a respondent whether
she has ever been raped, a behaviorally specific item asks whether she has ever had
unwanted vaginal sex because the person used physical force to make her (for example).
Instead of asking whether the respondent has ever been sexually harassed, a behaviorally specific approach would ask questions about specific examples of sexual harass-

Measurement of Sexual Harassment and Sexual Assault

ment, such as whether the respondent has ever heard repeated offensive sexual jokes in
the workplace. Structuring items in clear, specific terms helps to minimize subjective
interpretations of the constructs. Provided the questions include all the attributes necessary to categorize the event within the legal framework, affirmative responses can be
used to estimate rates of rape, sexual assault, or sexual harassment.
Given that many different kinds of experiences can be categorized as sexual
harassment and sexual assault, measurement of these constructs requires multiple,
behaviorally specific questions. Comprehensive lists of questions that ask respondents to consider different forms of sexual assault or harassment seem to aid recall and
increase disclosure about unwanted events (Bachman, 2000; Fisher, 2004; Tjaden and
Thoennes, 1998). Although 30 years have elapsed since the development of behaviorally specific measures of sexual assault, there is still no universally accepted gold-
standard measure of sexual assault (Bachman, 2012). Each of the national surveys used
different measures of sexual victimization.
Gating Questions

Administration of large epidemiological studies, particularly those designed to estimate experiences that have a low probability of occurring, can be a costly undertaking.
To both reduce cost and limit the time burden on respondents, survey designers strive
to minimize the length of survey instruments. One strategy to reduce survey length
has been to ask a single “gate” question to assess an entire class of events. Respondents
who respond “yes” to the item receive additional questions to clarify the nature of the
event(s), whereas those who respond “no” to the item skip past these additional questions. For example, the WGRA uses a single question to assess whether the respondent may have been sexually assaulted in any way in the past year. Respondents who
responded “yes” to the gate question receive additional questions that gather the necessary detail about the experience (DMDC, 2013c). The use of a single gating question—no matter how broad—tends to result in lower recall and disclosure of sexual
assault (Crowell and Burgess, 1996; Fisher and Cullen, 2000). One item may lack the
detail necessary to cue a victim’s memory of all events in a class. Even when the gating
question is detailed (as is the case with the 2012 WGRA item), respondents may process only some elements of the definition and fail to recognize that their experience
could be included in other elements of the item (National Research Council, 2014).
Generally, it is recommended that surveys assessing sensitive behaviors and experiences
avoid gating questions, and rely instead on a series of questions for all respondents
(Bachman, 2012; Fisher and Cullen, 2000; National Research Council, 2014).
Reading Level

It is generally recommended that surveys designed for the American public be targeted at a high school reading level or below. Both vocabulary and sentence complexity
contribute to reading level, and it has proved challenging to design sexual assault and

Sexual Assault and Sexual Harassment in the U.S. Military

sexual harassment screening questions that are targeted to an appropriate reading level.
Part of the challenge is that simple, easy-to-read words like “sex” or “rape” can be difficult to interpret. Readers may believe they understand the meaning of these terms,
but in fact have personal meanings that differ from the intent of the survey designers
(National Research Council, 2014). Young adults, for example, often exclude oral sex
from their personal interpretation of the word sex (Sanders and Reinisch, 1999), which
can create error if the survey designer interprets their responses as inclusive rather than
exclusive of this form of sexual intercourse.
The solution to this problem has been to create survey items that explicitly define
the targeted event so as to eliminate any interpretational errors. An unfortunate side
effect of this approach is that it introduces long, complex sentence structures that
increase the required reading level. For example, the sexual assault gating question
used in the 2012 WGRA (see Exhibit 4.2 in Chapter Four) requires a minimum of 12
years of formal education to read, and most reading-ease calculators consider the question to require substantially greater reading skills than those associated with college
graduates.3 Although the item includes the necessary elements to define the crime, to
do so the sentence becomes long, unwieldy, and difficult to parse for readers with lower
literacy levels. Investigators working with populations known to have high literacy,
such as college students, have been able to rely on these items given their respondents’
reading skill. However, investigators seeking to survey broad, general populations are
cautioned against using an approach that attempts to reduce interpretational error by
substituting error due to limited literacy. Attention to reading level must not be lost in
pursuit of other goals.
Survey Context

The context in which survey questions are asked can be as influential as the questions
themselves. For example, the NCVS has been widely criticized for embedding questions about sexual assault within a survey of criminal behavior (National Research
Council, 2014; Fisher and Cullen, 2000). Critics argue that the low rate of sexual
assault estimated by this survey relative to other national surveys is due in part to the
fact that respondents are cued to consider “crimes” (National Research Council, 2014;
Fisher and Cullen, 2000). Given that so few sexual assault victims personally label the
event as a crime (Bondurant, 2001; Fisher, Cullen, and Turner, 2000; Layman, Gidycz,
and Lynn, 1996), it is perhaps unsurprising that they fail to disclose their experiences
on a survey that directs respondents to consider crimes committed against them. To
correct this bias, it is generally recommended that questioning about sexual harassment
and sexual assault be conducted either in a neutral context or in the context of public
health topics (Bachman, 2012; National Research Council, 2014).

Measurement of Sexual Harassment and Sexual Assault

Mode of Survey Administration and Privacy

The mode in which a survey is administered and privacy while completing the survey
can significantly influence estimated rates of sensitive behaviors and experiences
(National Research Council, 2014; Tourangeau and Yan, 2007; Yu, Stasny, and Li,
2008). Modes of administration include face-to-face interviews in which a trained
interviewer reads survey items, telephone interviews, and self-administered surveys in
which the respondent completes the survey either on paper or via computer. Survey
modes that permit more privacy (e.g., computer-based assessments) tend to produce
higher estimates for sensitive behaviors than do other survey modes (e.g., telephone
interviews; Hussain et al., 2013; Percy and Mayhew, 1997; Tourangeau and Smith,
1996; Turner et al., 1998; Yu, Stasny, and Li, 2008). With this in mind, it is preferable
that surveys that measure sensitive topics select among feasible administration techniques the one that affords the greatest degree of privacy (National Research Council, 2014; Tourangeau and Yan, 2007). If interviewers are the only feasible way to
administer a survey, guidelines recommend that the interviewer remind respondents
about the confidentiality of their responses and also suggest that respondents complete the interview in an area where their family members, coworkers, or others are
unlikely to overhear their responses (National Research Council, 2014; Yu, Stasny,
and Li, 2008).
General Survey Problems
Two general survey methodology challenges are particularly important to consider for
surveys of sexual harassment and sexual assault. First, generally and across national studies, fewer and fewer people have been willing to take part in research surveys. The Pew
Research Center for the People and the Press, for instance, has reported that its typical
response rate for telephone surveys in 1997 was 36 percent. By 2000, that number had
fallen to 28 percent. In 2006 it was 21 percent, and by 2012 telephone surveys had just
a 9 percent response rate (Kohut et al., 2012). Similar trends are seen in large government surveys, such as the National Household Education Survey (National Research
Council, 2013), across survey administration modalities (in person, web-based, and
telephone; National Research Council, 2013), and in surveys of military populations.
The DMDC at DoD has documented declines in response rates across the military
populations it surveys, including active-duty and reserve-component members, DoD
civilians, and spouses of military members (Falk, 2012).
Declining participation in surveys likely results from multiple factors, including growth in the number of households with cell phones but no landlines; passage
of the Do-Not-Call Implementation Act in 2003 (U.S.C. § 6101); “survey fatigue,” or
the growing sense among many in the population that they are asked to complete too

Sexual Assault and Sexual Harassment in the U.S. Military

many surveys; and a growing interest among many in protecting their time and privacy
(Kohut et al., 2012; National Research Council, 2013).
Low participation rates present a clear threat to the validity of surveys. When
those who participate in the survey are a true random sample of the population,
their responses can be taken to represent those of the population from which they
are drawn. That is, because they are selected at random, there should be no systematic differences between the findings from the sample and those that a survey of the
entire population would yield. However, when individuals in a random sample of the
population self-select out of the survey at high rates, there is a genuine possibility that
the remaining survey participants are no longer representative of the population from
which they were drawn. For instance, if people who have been sexually harassed are
more likely to participate in a survey about workplace gender relations than those
who have no such experiences, then as the survey response rate declines, those with
sexual harassment experiences will make up a growing share of survey respondents.
Alternatively, if people who have been sexually harassed are more likely to not participate (because the topic is too upsetting for them), then the estimated prevalence
rate will be biased downward. If only 10 percent of the sample do not participate, the
effect of this nonresponse bias might be small. But with nonresponse rates typical of
recent WGRA and other studies (76 percent nonresponse for the 2012 WGRA and
66–70 percent in earlier versions), the risk of nonresponse biases entering survey estimates is considerable.
Because low response rates threaten the validity of survey estimates, it is important to maximize response rates in all surveys of sexual harassment and sexual assault.
In addition, surveys must adopt nonresponse weighting techniques that can reduce the
impact of nonresponse bias (Heeringa, West, and Berglund, 2010; Little and Rubin,
2002; Lohr, 2010).4 This is standard practice in all professional survey research and
is a required element of most government survey research (OMB, 2006). It has been
used to reduce nonresponse bias in all prior WGRA administrations (DMDC, 2012;
DMDC, 2014). Nonresponse weighting in military populations can be an especially
powerful tool because DoD maintains detailed personnel records that offer a wealth of
information about every service member, just the kind of information that can identify
factors on which the responders and nonresponders may differ from one another. Once
those variables are identified, the respondents can be weighted so that the analytic
sample matches the true population on those variables.
Finally, even when nonresponse weights are used to reduce biases introduced by
low participation rates, studies need to examine whether nonresponse bias remains in
survey estimates. Even a comparatively small nonresponse rate could, in theory, lead
to biased estimates that are not adequately corrected with nonresponse weights. Therefore, the Office of Management and Budget (OMB), for instance, provides guidance
to federal agencies that nonresponse bias must be investigated when participation rates
fall below 80 percent (OMB, 2006). This kind of study has rarely been done in surveys

Measurement of Sexual Harassment and Sexual Assault

of sexual harassment and assault. An exception is that as part of its analysis of the 2012
WGRA results, DMDC conducted several nonresponse bias studies, none of which
provided strong evidence that such bias presented a problem for the 2012 estimates
(DMDC, 2014).

CHAPTER THREE

Study Design
Terry L. Schell, Andrew R. Morral,
and Bonnie Ghosh-Dastidar

The 2014 RAND Military Workplace study (RMWS) was implemented as a confidential, web-based survey of active-duty, guard, and reserve service members. The survey
was designed to meet the overarching project objectives of estimating the percentage
of service members who experienced sexual assault and sexual harassment, as well as
describing those experiences. Specifically, it was designed to accomplish the following:
• Maximize response rates subject to time and budget constraints by making the
survey smartphone compatible, promoting it actively with the help of the Office
of the Secretary of Defense (OSD) and the services, keeping it as short as possible for most respondents, and increasing the number of email invitations relative
to the prior Workplace and Gender Relations Survey of Active Duty Personnel
(WGRA).
• Provide respondents with assurances of higher levels of confidentiality, in particular the assurance that their survey responses would not be shared with the
Department of Defense (DoD) in a way that could identify them even if subpoenaed by a court of law.
• Minimize respondent burden by keeping the number of people answering each
survey question to the minimum required for precise population estimates on
that measure. As a result, sexual assault questions were asked of the full sample of
all active-duty women and 25 percent of men, while other types of measures were
assessed on smaller samples.
• Compare prevalence estimates for unwanted sexual contact and sexual harassment using the earlier WGRA surveys with estimates from new, more-detailed
measures of sexual assault and sex-based military equal opportunity (MEO) violations (sexual harassment and gender discrimination).
• Broaden the representativeness of respondents by reducing emphasis in the project
title and recruitment materials on “gender relations,” “sexual assault,” and “sexual
harassment.” Using these terms could introduce unwanted associations between
an individual’s decision to respond to the survey and their personal experiences
with sexual assault or harassment that could bias the results.
27

Sexual Assault and Sexual Harassment in the U.S. Military

In this chapter, we describe the survey and sampling design, the methods used
to conduct the fieldwork, and the review process leading to the approval of this study
design.
Sample Design and Selection
Active Component

Sample frame. The active-duty population included all Army, Navy, Air Force, Marine
Corps and Coast Guard active-duty members listed in the May 2014 Defense Enrollment Eligibility Reporting System (DEERS) database maintained by the Defense
Manpower Data Center (DMDC) who are not members of the reserve component, a
population of 1,371,880.1 For continuity with earlier WGRA surveys we matched the
exclusion criteria previously used to define WGRA sampling frames.2 Specifically, we
excluded from the total population all members
1. with fewer than 6 months of service as of August 1, 2014 (12,469)3
2. who were listed in the May 2014 DMDC data as general officers or flag officers
(941)
3. who were under the age of 18 as of August 1, 2014 (72).
An additional 171 records were duplicates and 1,554 were missing an entry date,
making verification of their eligibility impossible. These additional records were also
removed from the sampling frame. These exclusion criteria resulted in a sampling frame
of 1,356,673 active-duty service members, which serves as the inferential population
for this study.
Sample selection. DoD asked RAND to ensure that the primary active-duty
sample included all active-duty women in the sample frame and 25 percent of activeduty men. This was designed to provide enough respondents who had experienced
a sexual assault in the past year so that the characteristics of those assaults could be
analyzed with sufficient statistical precision. To ensure proportionate representation in
the sample across services and pay grades, men were grouped into 20 sampling strata
defined by the intersection of the five services and four pay grade categories (E1-E4,
E5-E9, O1-O3, O4-O6). Warrant officers were included within the E5-E9 stratum
during sampling. Selection probabilities in each of the 20 strata were equivalent (.25).
The resulting sample included 491,680 active-duty members, of whom 41 percent were
women. The distribution of the sampling frame and the drawn sample is listed in
Table 3.1.

Study Design

Table 3.1
Size of the Active-Duty Sampling Frame and Sample, by Gender, Service, and
Pay Grade
Totals

Women
Frame

Sample

Frame

Sample

491,680

203,343

1,153,330

288,337

37.1%

36.2%

34.2%

37.6%

E1-E4

15.8%

15.6%

15.2%

15.9%

E5-E9

15.3%

14.2%

11.9%

15.9%

O1-O3

3.6%

4.0%

4.8%

3.4%

O4-O6

2.3%

23.1%

24.3%

27.0%

22.4%

E1-E4

9.3%

10.6%

13.6%

8.5%

E5-E9

10.0%

9.8%

9.2%

10.2%

O1-O3

2.3%

2.5%

2.9%

2.1%

O4-O6

1.5%

1.3%

1.5%

Air Force

23.2%

25.1%

29.2%

22.1%

E1-E4

8.2%

8.7%

9.9%

8.0%

E5-E9

10.3%

11.2%

13.2%

9.8%

O1-O3

2.6%

3.0%

4.0%

2.3%

O4-O6

2.1%

2.0%

13.7%

11.6%

6.8%

15.0%

E1-E4

8.1%

6.9%

4.3%

8.8%

E5-E9

4.2%

3.5%

1.9%

4.6%

O1-O3

0.9%

0.8%

0.5%

1.0%

O4-O6

0.5%

0.4%

0.1%

0.5%

Total number

Frame

Sample

1,356,673

Men

Column percentages:
Army

Navy

Marines

Sexual Assault and Sexual Harassment in the U.S. Military

Table 3.1—Continued
Totals
Frame

Women

Sample

Frame

Sample

Men

Frame

Sample

Coast Guard

2.9%

E1-E4

0.9%

1.0%

1.2%

0.8%

E5-E9

1.5%

1.3%

1.0%

1.6%

O1-O3

0.3%

0.4%

0.3%

O4-O6

0.2%

NOTE: Warrant officers were included in the E5-E9 groups for the purposes of sampling.

Reserve Component

Sample frame. The reserve-component population included all members of the Selected
Reserves in the Army, Navy, Air Force, Marine Corps, and Coast Guard, including
both National Guard and reserve members, listed in the May 2014 DMDC dataset—
a population of 823,524.4 Exclusion criteria were similar to those for the active-duty
sample. Specifically, we excluded from the total population those
1. with fewer than 6 months of reserve-component service as of August 1, 2014
(16,760)5
2. who were listed in the May 2014 DMDC data as general officers or flag officers
(628)
3. who were under the age of 18 as of August 1, 2014 (4,462).
Also, we could not determine active-duty or reserve status for 24 records; another
7 individuals were missing service affiliation. These additional records were excluded
from the sampling frame. These exclusion criteria resulted in a sampling frame of
801,643 reserve members, which serves as the inferential population for this study.
Sample selection. The objectives for the reserve sample differ from those for the
active-duty sample. Specifically, the primary objectives for including reserves are: (1)
to compare overall rates of sexual assault and sex-based MEO violations between the
active-duty and reserve components for men and women, and (2) to establish whether
the new survey questions developed by RAND work in the selected reserve population.
Because we do not intend to produce separate prevalence estimates within each reserve
component by service, pay grade, or other detailed reporting categories, we required a
much smaller sample size for guard and reserve compared to active duty.
We sampled about 60,000 guard and reserve members from the four DoD services using stratified random sampling. The sample includes 27,004 women and 33,003
men to ensure that estimates for reserve-component men and women (the primary

Study Design

reporting categories) have statistical precision similar to the reporting categories for
the active-duty sample (e.g., female Marines, female junior officers). The DoD sample
sizes above correspond to selection probabilities of 5.1 percent and 18.3 percent for
men and women, respectively. These sampling probabilities were applied to the six
DoD reserve-component strata to ensure the same distribution in the sample as in
the population across components. Survey wording was adjusted to be appropriate for
reserve-component survey participants, as is described in more detail in the questionnaire development section.
We sampled from the Coast Guard reserve at different rates than the DoD reserve
component members. Whereas the reserve-component analyses for those services was
planned to span all reserve components, the Coast Guard reserve analysis was planned
as a stand-alone comparison to Coast Guard active-duty prevalence estimates. That
is, Coast Guard reserve is a primary reporting category for this study. Because of the
comparatively small size of the Coast Guard reserve (7,592 members) and its historically low rate of sexual assault, we included all Coast Guard reserves in the sample6 for
a combined total of 67,599 reserve members across all five services (Table 3.2).
Assignment to Survey Forms

To compare rates of sexual assault and sex-based MEO violations (sexual harassment
and gender discrimination) as measured on the new instrument with rates of unwanted
sexual contact and sexual harassment as measured on the WGRA surveys, we assigned
individuals within our overall sample to one of several survey forms. Specifically, we
randomly assigned 100,000 individuals from the overall DoD active-duty sample to
the prior WGRA form survey, consisting of questions taken from the 2012 WGRA.
The prior WGRA form sample was designed to yield statistical precision that is similar
Table 3.2
Size of the Guard and Reserve Sampling Frame and Sample, by Gender and Service
Totals

Women

Frame

Sample

Frame

Sample

801,643

28,271

652,964

39,328

Army

64.4%

62.0%

66.0%

55.9%

Navy

7.0%

8.7%

8.3%

7.1%

6.0%

21.1%

20.0%

25.0%

24.1%

20.2%

17.1%

Marines

4.8%

3.3%

1.1%

5.7%

4.8%

Coast Guard

0.9%

11.2%

0.9%

4.5%

1.0%

16.1%

Total number:

Men

Frame

Sample

67,599

148,679

65.7%

58.5%

7.4%

Column percentages:

Air Force

Sexual Assault and Sexual Harassment in the U.S. Military

to, or slightly better than, estimates based on the 2012 WGRA using the same questions.7 Members of the Coast Guard were not given the prior WGRA form survey
because they were not included in the 2012 WGRA.
In addition to having different forms of the survey that correspond to the new
survey versus 2012 WGRA measures, we also created three slightly different versions
of the new survey form for random assignment among those who received the new
instrument. The overall goal of these different forms was to increase response rates and
to decrease survey burden without compromising statistical precision. This approach
is based on the observation that different types of measures require different sample
sizes to ensure similar precision levels, depending on the rate at which we expect service
members to select them.
The total active-duty sample size of those who received the new survey instrument is approximately 400,000 individuals. Such a large sample is especially valuable
for gaining a clear understanding of the characteristics of rare events. For example, the
2012 WGRA survey (with a sample size of approximately 100,000) yielded only 832
respondents who reported an unwanted sexual contact (DMDC, 2013d).8 Because
many important questions on the WGRA survey were necessarily restricted to respondents who experienced an unwanted sexual contact, it was difficult to provide precise
estimates of the characteristics of these events—particularly when broken down into
multiple reporting categories (e.g., branch of service). To address that limitation, the
new sexual assault measure was given to a much larger number of active-duty service
members than that in the 2012 WGRA.
However, such a large sample size—400,000—is not required to establish precise
estimates for other phenomena, such as service member opinions about the military,
and there are good reasons to avoid surveying more people than necessary to provide
an estimate with adequate precision for a particular question. Specifically, surveying
more people than needed
• wastes resources and service members’ time
• may contribute to “survey fatigue,” thereby reducing respondent willingness to
participate in other surveys where their responses are essential for good estimates
• may result in longer than necessary surveys, potentially reducing sampled members’ willingness to participate.
In designing the RMWS, therefore, we sought to ensure sufficient numbers of
respondents to support detailed analyses of sexual assaults, the primary outcome with
lowest prevalence. We needed fewer respondents to produce precise estimates for the
prevalence and experiences of those reporting sex-based MEO violations in the past
year, because these events have historically had higher prevalence compared to sexual
assault. For example, the 2012 WGRA contained about four times as many individuals who experienced sexual harassment as experienced unwanted sexual contact. Simi-

Study Design

larly, we need even fewer respondents to establish precise estimates of service members’
beliefs on topics such as whether sexual harassment is increasing or decreasing in the
military, because all service members are likely to respond to these questions, in comparison to the smaller percentage of respondents who provide information about experiencing a sexual assault.
Therefore, to ensure that we surveyed sufficient numbers to measure low-rate
events while reducing the overall respondent burden, we eliminated some questions
from the survey for some individuals in our sample. Specifically, we randomly assigned
those receiving the new RAND instrument to one of three versions:
1. a “long form” that includes the sexual assault module, the sex-based MEO violation module, demographic questions, and supplementary items that can be
answered by all respondents
2. a “medium form” that includes the sexual assault module, the sex-based MEO
violation module, and demographic questions
3. a “short form” that includes the sexual assault module, the screening questions
from the sex-based MEO violation module without the follow up questions,
and demographic questions.
Individuals in the overall sample who had not been assigned to receive the prior
WGRA form were randomly assigned to one of the three versions of the new questionnaire. The number of people assigned to each of the forms was determined by the
sample size requirements for the various types of measures.
• Approximately 60,000 of the active-duty DOD sample were randomly assigned
to receive the long form. In addition, 6,250 from the active-duty Coast Guard
sample were randomly assigned to receive the long form.9
• Of the remaining members in the active-duty sample, about one-half were randomly assigned to receive the short form, and half were assigned the medium
form.
This results in approximately 400,000 sampled individuals assigned to receive the
new sexual assault module, 230,000 assigned to receive the full sex-based MEO violation module, and 66,000 assigned to receive the questions about military climate and
attitudes (see Table 3.3). While the entire sample received the sexual assault module,
we reduced the number who received the full sex-based MEO violation module. This
was done after ensuring that there would be an estimated 6,000 respondents or more
who will have experienced a sex-based MEO violation (estimated under the assumptions that the response rate and rates of sexual harassment would be similar to that of
the 2012 WGRA). Even with a large portion of the sample assigned to a form that does
not include the full sex-based MEO violation module, we expected twice as many service members to indicate that they experienced sexual harassment as to indicate they

Sexual Assault and Sexual Harassment in the U.S. Military

Table 3.3
Sample Sizes for RMWS Survey Forms and Modules (Active-Duty Sample)
Items

Long Form

Medium Form

Short Form

Sample Size

Demographic questions

391,680

Sexual assault module

391,680

Sex-based MEO violation screening items

391,680

Sex-based MEO violation follow-up items

Supplementary questions

Sample size

66,133

229,047
66,133

162,914

162,633

391,680

experienced sexual assault, so we will have a sufficient number to achieve the levels of
precision needed for sexual harassment analyses.
The more limited objectives for reserve-component analyses did not require either
the long form or the prior WGRA form. Instead, equal numbers of the reserve sample
were assigned to the short or the medium form.
Differences in Sampling from the 2012 WGRA Survey

Our approach to sampling differs from that used by DMDC for the 2012 WGRA,
primarily because of the much larger sample size in 2014 and some different objectives.
In 2012, DMDC used more than 250 sampling strata, many with differing sampling probabilities. The purpose of this sampling plan was twofold: to minimize total
sample size and to ensure sufficient numbers of respondents in each of 74 reporting
domains (listed in Appendix B of DMDC, 2012). Because separate prevalence estimates are required of some groups with lower response rates than others (e.g., junior
enlisted), DMDC oversampled low response-rate groups to ensure enough respondents
to produce estimates with adequate precision. In addition, Marine Corps leadership
asked DMDC to oversample Marines in the 2012 WGRA at a rate almost four times
greater than other services.
For the 2014 RMWS, we sampled all active-duty women, so no oversampling
of women is possible. It would, however, be feasible to oversample certain groups of
men. As was the case for the 2012 WGRA, oversampling would be a prudent decision if we would not otherwise have enough respondents in a key reporting domain to
provide prevalence estimates with acceptable precision. However, if our response rates
are comparable to those of the 2012 WGRA, we expect to have larger sample sizes in
all the 2012 WGRA reporting domains even without oversampling.10 Moreover, the
use of different sampling probabilities across reporting domains involves trade-offs in
precision: while it increases precision within the smaller reporting categories that are
oversampled, it reduces precision for larger categories and can reduce precision for the

Study Design

overall study. Given (a) the large numbers expected in each reporting category planned
for the 2014 RMWS, (b) the larger overall sample size in 2014, and (c) the inability to
oversample subgroups of women, the 2014 sample design does not need to oversample
on the basis of pay grade, race, or service, as was done in 2012.
Fielding the Survey
The RMWS was a web-based survey fielded between August 13, 2014, and September 24, 2014, by Westat under subcontract to RAND. This period was preceded by
a one-week soft rollout in which 3,000 sampled individuals were invited to take the
survey. The purpose of the soft rollout was to test whether all systems were working as
designed, including questionnaire programming, “whitelisting” of email invitations so
that they were not blocked by DoD spam or security filters, survey helpdesk services,
and the compatibility of the web survey with the wide range of browsers and devices
that respondents were expected to use (the survey interface was designed to be compatible with smartphones and tablets, in addition to computers).
All sampled members received up to three letters by mail and nine emails requesting their participation in the study.11 Personalized letters included (a) copies of a signed
letter from the service member’s Chief of Service explaining the study and its importance, (b) instructions for accessing the survey with a unique login credential, and (c) a
list of frequently asked questions about the study. Email and letter invitations were discontinued when members completed the survey or asked not to be contacted further.12
During the survey fielding period, RAND and DoD leadership sought to draw
attention to the survey and to emphasize its importance in public communications
using several means. These included messages from military officials on Facebook and
Twitter, DoD News stories, public service announcements that ran on Armed Forces
Network and could be linked to on YouTube, and messages from RAND on Twitter
and Facebook, as well as other means. In an effort to recruit junior enlisted members
who were expected to have especially low response rates, RAND purchased advertising
on Facebook and Duffleblog (a satirical news website with a military audience) and ran
public service announcements through those venues as well.
Study Reviews and Approval
RAND’s Institutional Review Board reviewed and approved the study procedures and
survey instrument to ensure that it met all human subjects’ protection protocols. The
Office of the Under Secretary of Defense for Personnel and Readiness (OUSD/P&R)
and the Coast Guard’s Institutional Review Board conducted second-level review of
human subjects’ protections. The study procedures, or portions of them, also received

Sexual Assault and Sexual Harassment in the U.S. Military

reviews and approvals by the OSD Office of General Counsel, the Chief Privacy Officer of OSD and the Joint Staff, the DMDC Chief Privacy Officer, DMDC survey
protocol review, OUSD/P&R Records management, and the Joint Chiefs of Staff. The
project received licensing approval from the Washington Headquarters Service after
receiving approvals from the OUSD/P&R. In addition, we solicited multiple rounds of
reviews and comments with researchers and leadership from each service’s headquarters sexual assault prevention office.
The sample design, questionnaire, survey methods, and statistical analyses were
developed in collaboration with the scientific advisory board. This 12-member board
included statisticians, survey methodologists, experts in relevant civilian and military
law, and scientists who study sexual assault in civilian, military, or veteran populations.

CHAPTER FOUR

Questionnaire Development
Lisa H. Jaycox, Terry L. Schell, Coreen Farris, Amy Street,
Dean Kilpatrick, Andrew R. Morral, and Terri Tanielian

The overarching objective for instrument development was to create questions that
would yield valid estimates of respondents’ experiences of past-year sexual assault,
sexual harassment, and gender discrimination, as these are defined by law and Department of Defense (DoD) policies. The module assessing sexual assault was designed to
use the definitions and criteria listed in the Uniform Code of Military Justice (UCMJ)
for Article 120 crimes. Our survey questions measuring sexual harassment and gender
discrimination align closely with the definitions of those violations as described in
DoD directives, which themselves are patterned on federal civil rights law.1 In developing the instrument, we sought to
• address criticisms of the 2012 WGRA when possible
• retain 2012 WGRA questions, when highly relevant to policy
• follow research-based recommendations for the assessment of sexual assault and
sexual harassment
• aid the respondent in focusing their attention on the time frame covered by each
question
• minimize distress among those who experienced sexual harassment or assault
• ask the fewest possible questions to obtain the data necessary to make precise
estimates.
Thus, in the measurement of sexual assault and sexual harassment, we sought
to develop simple sets of questions that could be used to correctly classify respondents’ experiences according to the complex criteria set out in DoD regulations and
the UCMJ defining military equal opportunity (MEO) violations and sexual assault
crimes. In addition to breaking down complex laws and regulations into questions
amenable to a self-administered survey format, we also sought to introduce technical
changes to improve respondents’ understanding and therefore to enhance the validity
of their answers. Each of these domains will be discussed in turn below.
To achieve these goals, the instrument development team worked through multiple rounds of discussion and input from the scientific advisory board for this project
37

Sexual Assault and Sexual Harassment in the U.S. Military

and from the Sexual Assault Prevention and Response Office (SAPRO) and the service
sexual assault prevention and response (SAPR) leads.
Since several parts of the survey differed substantially from the 2012 WGRA,
we conducted interviews to test the new items. We recruited 24 service members and
recent veterans by convenience sampling through social and professional networks and
volunteers from advocacy organizations. These individuals were diverse with respect to
gender; branch of service; component; rank; prior experiences of sexual harassment,
gender discrimination, or sexual assault; and sexual orientation. Using cognitive interviewing techniques (Sirken et al., 1999), we asked participants to review the survey
questions and reflect on what the question meant; identify any ambiguous, confusing, or service-specific wording; determine whether the questions captured the relevant
experiences; and determine whether the questions created any distress. The survey was
further modified based on this feedback, and reviewed again by the scientific advisory
board and SAPR service leads and researchers. At the end of this process, we consulted with individual members of the scientific advisory board and outside experts on
specific questions that posed special problems of interpretation or formulation.
The version of the survey delivered to National Guard and reserve members was
based on the active-component survey with modifications to reflect the reserve component. Several experts reviewed the instrument to adjust wording to ensure the questions were applied only to their military workplace, and to use appropriate terms for
the guard and reserve context.
Development of the Sex-Based MEO Violations Module
Clarification of Military Equal Opportunity Violations

We sought to improve alignment between the survey measures of sexual harassment
and gender discrimination, and MEO definitions of these offenses.2 DoD Directive
1350.2 defines sexual harassment as “unwelcome sexual advances, requests for sexual
favors, and other verbal or physical conduct of a sexual nature” when either (a) “submission to such conduct is made either explicitly or implicitly a term or condition of
a person’s job, pay, or career,” or (b) “submission to or rejection of such conduct by a
person is used as a basis for career or employment decisions affecting that person,” or
(c) “such conduct has the purpose or effect of unreasonably interfering with an individual’s work performance or creates an intimidating, hostile, or offensive working environment.” DoD regulations specify that to be actionable, sexual harassment “need not
result in concrete psychological harm to the victim, but rather need only be so severe
or pervasive that a reasonable person would perceive, and the victim does perceive,
the work environment as hostile or offensive” (DoD, 2003). It does not suffice that an
unusually sensitive person took offense at the conduct, if a reasonable person would
not have found the conduct especially offensive and the behavior was not pervasive. In

Questionnaire Development

addition to sexual harassment, MEO violations also include gender discrimination as
a form of unlawful discrimination “that is not otherwise authorized by law or regulation” (DoD, 2003).
In the new sex-based MEO violation module, we include questions to assess
gender discrimination and two forms of sexual harassment: quid pro quo harassment
and hostile workplace harassment. Quid pro quo harassment under federal civil rights
law refers to “any use . . . of any form of sexual behavior to control, influence, or affect
the career, pay, or job of a member of the armed forces” (10 U.S.C 1561[e][2]; United
States Code, 2006). Hostile workplace harassment includes unwelcome sexual conduct
or comments that are “explicitly or implicitly a term or condition of a person’s job,
pay, or career” or “unreasonably [interferes] with an individual’s work performance or
creates an intimidating, hostile, or offensive working environment” (10 U.S.C 1561[e]
[1] ; United States Code, 2006). Similar definitions exist in DoD Directive 1350.2
(Enclosure 2, Item 1.15) and apply specifically to military personnel. The DoD Directive differs from the more broadly applicable federal civil rights law in that the sexual
conduct “need not result in concrete psychological harm to the victim, but rather need
only be so severe or pervasive that a reasonable person would perceive, and the victim
does perceive, the work environment as hostile or offensive” (DoD, 2003, Enclosure 2,
Item 1.15). The RAND Military Workplace Study (RMWS) attempts to assess whether
a sexually hostile work environment rises to this level of severity of persistence, but not
whether the harms from the actions rise to a violation of federal civil rights law.
In the 2014 RMWS, respondents were asked if in the past year they had any
of a range of experiences that might qualify as a sexually hostile work environment
(11 screening questions), gender discrimination (2 screening questions), or sexual quid
pro quo (2 screening questions). Affirmative answers to these screening questions are
categorized as problematic workplace behaviors. Such behaviors can be problematic—
and interfere with well-being and work performance—even if they do not rise to the
level of an MEO violation. However, as described in the next section, when participants indicated that they experienced these problematic workplace behaviors, a series
of follow-up questions were asked to clarify whether these experiences also included
elements that would meet the more stringent standards required to rise to the level of
an MEO violation. The flow of questions and definitions used for each of these types
of sex-based MEO violations can be seen in Exhibit 4.1.
Assessment of Sexually Hostile Work Environment

If a respondent indicates they experienced any of the 11 hostile workplace screening
questions, he or she is categorized as having experiences consistent with a sexually hostile work environment if and only if the offensive behavior either (a) persisted after the
person exhibiting the behavior knew the respondent or other people wanted it to stop,
or (b) was sufficiently severe that, in the respondents’ opinion, most other military
members of the respondents’ gender would have found it offensive. These two criteria

Sexual Assault and Sexual Harassment in the U.S. Military

Exhibit 4.1
Categorization of Problematic Workplace Behaviors and Sex-Based MEO Violations in the
RMWS
Sexually hostile
work environment

Sexual quid pro quo

Offended by a coworker’s
workplace behavior

Perceived a quid pro quo
from a coworker

Yes

Experienced comments
and behaviors related
to gender

Yes

Offensive
workplace
behavior was
persistent

Gender discrimination

Gender
discrimination
harmed or
limited career

Quid pro quo
offer was
directly
communicated

Yes

Workplace
behavior meets
reasonable person
standard for
offensiveness

None

Experienced
problematic
workplace
behaviors

Yes

Experienced
sexually
hostile
workplace
environment

None

Experienced
problematic
workplace
behaviors

Experienced
sexual
quid pro quo

None

Experienced
problematic
workplace
behaviors

Experienced
gender
discrimination

RAND RR870/1-4.1

are designed to rule out those instances in which respondents themselves realized that
the offender had no way of knowing that his or her behavior could create a hostile work
environment. It is worth noting that some of the specific behaviors assessed might also
rise to the level of a sex crime under UCMJ Article 120. This part of the instrument
does not assess whether these experiences would meet the legal definition of an Article
120 offense, but instead determines if these experiences involving someone from their
workplace would constitute a hostile work environment.
Assessment of Sexual Quid Pro Quo

Two sexual quid pro quo questions assessed the respondents’ belief that a workplace
benefit (or avoidance of a workplace punishment) was contingent on their sexual
behavior. Responding “yes” to either question is categorized as problematic workplace

Questionnaire Development

behavior. The instrument uses follow-up questions to rule out instances in which the
respondent reports no direct evidence that a benefit (or avoidance of punishment) was
being offered in exchange for sexual behavior (e.g., instances in which their perception
was based on rumor). Those experiences in which the respondent indicates direct evidence of a quid pro quo are categorized as sexual quid pro quo. If an exchange for sexual
behavior actually took place, the experience might have risen to the level of a sex crime
under UCMJ Article 120. Even without an exchange, the offer may be prosecutable
as a crime under UCMJ Article 133 or Article 134. Here again, this portion of the
instrument does not pursue the question of whether a criminal offense occurred, but
instead determines if these experiences involving someone from their workplace represent sexual quid pro quo harassment.
Assessment of Gender Discrimination

If a respondent answered “yes” to either of the two gender discrimination screening
questions, he or she received a follow-up question to assess whether there were any
work-related negative outcomes associated with the experience. If so, the experience
was categorized as gender discrimination.
Assessment of Sexual Harassment

If a respondent was categorized as having experienced either a sexually hostile work environment or sexual quid pro quo, then they also were categorized as having experienced
sexual harassment.
Assessment of Sex-Based Military Equal Opportunity Violations

All respondents who were categorized as having experienced gender discrimination, a
sexually hostile work environment, or a sexual quid pro quo in the past year were also
included in a broader category capturing any sex-based MEO violation. Anyone who
experienced a sex-based MEO violation received additional questions about the situation, the person who committed the sexual harassment or gender discrimination, and
the respondents’ reporting decision.
Comparison to the 2012 WGRA

In 2012 and earlier WGRA surveys, sexual harassment was assessed using modified
versions of the Sexual Experiences Questionnaire (SEQ) designed for DoD (SEQDoD; Fitzgerald, Swan, and Fischer, 1995; Fitzgerald, Gelfand, and Drasgow, 1995;
Fitzgerald et al., 1999). The module includes a series of questions that assess unpleasant
workplace experiences. For each question, respondents are asked to consider how often
during the past 12 months they have been in workplace situations where someone
engaged in sex- or gender-related talk or behavior that was unwanted (see Exhibit 4.2
for full text of the introduction and stem question from the 2012 WGRA). Example
questions assessing these behaviors include “repeatedly told sexual stories or jokes that
were offensive to you” or “made offensive remarks about your appearance, body, or

Sexual Assault and Sexual Harassment in the U.S. Military

sexual activities.” Respondents indicated the frequency with which they
experienced the behavior on a 5-point
In this question you are asked about sex/
scale (“never” to “very often”).
gender-related talk and/or behavior that was
The 2012 WGRA used a modiunwanted, uninvited, and in which you did not
fied version of the SEQ-DoD-Short
participate willingly.
(Stark et al., 2002), which adapted
How often during the past 12 months have
the SEQ-DoD-Short by omitting two
you been in situations involving
questions, adding two new questions,
• Military Personnel (Active Duty or
retaining two questions from the
National Guard/Reserve)
original SEQ-DoD scale (Fitzgerald
–– on- or off-duty
–– on- or off-installation or ship; and/or
et al., 1999), and altering the scor• DoD/Service Civilian Employees and/or
ing. SEQ questions were summarized
Contractors
with four subscales: crude/offensive
–– in your workplace or on your installation/ship
behavior, unwanted sexual attention,
sexual coercion, and sexist behaviors.
Where one or more of these individuals (of
A respondent who indicated that the
either gender)… Mark one answer for each
item.
behavior described in any subscale
question had happened at least one
time was categorized as having experienced the unwanted workplace behavior captured by that subscale. Respondents were then asked a single question: “How many of
these behaviors that you marked as happening to you, do you consider to have been
sexual harassment?” Respondents were classified as having experienced sexual harassment in the previous 12 months if they answered “yes” to any questions in the crude/
offensive behavior, sexual coercion, or unwanted sexual attention categories and also
indicated that they perceived at least one of the broader set of 19 behaviors to be sexual
harassment. The four questions assessing sexist behaviors—among the most commonly
endorsed questions—and the two questions assessing attempted or completed sexual
assaults by coworkers, were not included in the DoD definition of sexual harassment.
It is not possible to know if the behavior considered to be sexual harassment by the
respondent was among the 12 items used to define sexual harassment in scoring or was
one of the six excluded items. This scoring procedure departs from the conventional
approach to scoring the SEQ, which treats sexist behaviors as indicative of a climate of
sexual harassment.
The SEQ-DoD was designed as a broad workplace climate measure, not to count
the frequency of EEO or MEO violations. The developers refer to the measure as capturing psychological sexual harassment (Fitzgerald, Swan, and Magley, 1997). Although
the SEQ has been found to have high internal consistency and corresponds to a range
of organizational climate, job satisfaction, mental health, and work performance measures in predicted ways (Willness, Steel, and Lee, 2007), it has not been shown to be a
valid assessment of sexual harassment EEO or MEO violations.3
Exhibit 4.2
Introduction and Stem Question for 2012 WGRA
Sexual Harassment Question

Questionnaire Development

Moreover, while the SEQ has been shown to have good measurement properties
(Fitzgerald et al., 1999; Stark et al., 2002), the standard version of the scale was not
implemented on prior WGRA surveys. Instead, a modified set of questions was fielded
and the scoring scheme was substantially altered from the scheme on which validity was established. In the validated SEQ, scoring is continuous, which means that
the more experiences reported by respondents—the higher their score on the questionnaire—the greater the likelihood that they have experienced psychological sexual
harassment. As used in the WGRA, a single binary outcome (sexual harassment or
no sexual harassment) is scored if the respondent indicates experience in the past year
with at least one of 12 unwanted behaviors, and they answer yes to a supplementary
question: “How many of these behaviors that you marked as happening to you, do you
consider to have been sexual harassment?” Every respondent who says he or she considered at least one of the unwanted experiences to be sexual harassment is subsequently
counted as having experienced sexual harassment in the past year.
A shortcoming of this approach is that victims of sexual harassment often do not
label their experiences as sexual harassment. Indeed, the developers of the SEQ have
made this argument, noting that many respondents who experience sexual harassment based on their SEQ scores say they have never been sexually harassed (Fitzgerald,
Swan, and Fischer, 1995). Similarly, in 1995, DoD fielded an older sexual harassment
scale asking respondents to list which offensive behaviors they had experienced in the
past year. This survey, which did not require respondents to label their experiences as
“sexual harassment,” found that 53 percent of Navy women reported at least one such
experience of sexual harassment. At about the same time, the Navy fielded a similar
survey, but to reach the list of objectionable behaviors, respondents had to first complete a gating question asking if they had been sexually harassed in the past year. The
Navy study found a rate of sexual harassment of women that was approximately half
that of the DoD survey (27 percent) (DoD, 2002). Both studies suggest that requiring
respondents to label unwanted workplace behaviors as sexual harassment will undercount sexual harassment, leading to lower estimated rates of exposure than would be
established by the SEQ using standardized SEQ scoring procedures.
Another problematic implication of this procedure is that it could cause reported
rates of sexual harassment to increase artificially over time. As DoD implements new
and more rigorous training and prevention efforts, and service members become better
informed about what constitutes sexual harassment, they will be more likely to label
their qualifying experiences as sexual harassment. In other words, the unconventional
scoring procedure used in the prior WGRA to categorize sexual harassment could lead
to the paradoxical result that as training and prevention programs succeed in reducing
the incidence of sexual harassment, the WGRA counts of sexual harassment could rise.
The 2014 RMWS questionnaire includes a question assessing whether the respondent considered any of the selected incidents to be sexual harassment, but it does not
use that question as part of the definition of sex-based MEO violations. Instead, scor-

Sexual Assault and Sexual Harassment in the U.S. Military

ing procedures in the 2014 RMWS rely on meeting the specific criteria spelled out in
the DoD regulations defining sexual harassment and gender discrimination (DoD,
2003). The respondents’ assessment of the events will be used to explore which types of
experiences are related to labeling MEO violations as sexual harassment.
Simplification of Complex Constructs

The version of the SEQ used in prior WGRA administrations began each question
with a complex definition of the workplace and time period (see Exhibit 4.2). The
grammatical structure of this stem question, and its many conditions, make parsing
its meaning and remembering it a complex cognitive task. On the 2014 RMWS, we
define the workplace and the identity of potential offenders (i.e., “someone at work”)
in a preamble to the section (see Exhibit 4.3 for instructions). For the questions that
follow, the relatively simple term “someone from work” is used in each question, with a
definition appearing on the computer screen to remind the respondent what that term
means. This allows for much simpler sentence structure and greater clarity within each
question. The web survey was also designed to use appropriate gender pronouns for
each respondent.
In addition, the questions are relatively simple yes/no questions that ask whether
specific events occurred in the past year. In contrast, the SEQ asks respondents to
indicate how many times in the past year the behavior, or series of behaviors, occurred.
In the RMWS, relevant follow-up questions are only asked of those who responded
“yes” to earlier screening questions. This minimizes the number of questions seen
by most respondents, and makes each question easier to answer. For example, the
first question reads, “Since [date one year prior], did someone from work repeatedly
tell sexual ‘jokes’ that made you uncomfortable, angry, or upset?” and the response
options are “yes” or “no.”
Exhibit 4.3
RMWS Instructions Preceding the Sex-Based MEO Violations Module
In this section, you will be asked about several things that someone from work might have
done to you that were upsetting or offensive, and that happened AFTER [date one year
prior to survey administration date].
When the questions say “someone from work,” please include any person you have contact
with as part of your military duties. “Someone from work” could be a supervisor, someone
above or below you in rank, or a civilian employee / contractor. They could be in your unit
or in other units.
These things may have occurred on-duty or off-duty, on-base or off-base. Please include
them as long as the person who did them to you was someone from work.

Questionnaire Development

Technical Changes to Improve Respondent Precision

People often do not remember exactly when life events happened. In attempting to
measure how many crimes happened in a specific time, forward telescoping, or the tendency for respondents to report crimes as occurring more recently than they actually
occurred, represents an important potential threat to the validity of crime-victimization estimates using survey data. For instance, studies using data from surveys of criminal victimization have shown that telescoping can lead to increased prevalence rates
for some types of crimes (Andersen, Frankel, and Kasper, 1979; Cantor, 1989; Lehnen
and Skogan, 1984; National Research Council, 1984). Whereas the best approach to
minimizing forward telescoping involves the use of bounded interviews (i.e., conducting a longitudinal survey with at least two assessments and asking about events that
happened since the last assessment), several alternative strategies exist for reducing telescoping on cross-sectional surveys like the 2014 RMWS. We adopted the following
procedures to reduce telescoping:
• We begin the survey with a series of questions designed to remind respondents
of where they lived, their job, paygrade, and other specific details from the date
exactly 12 months ago.
• Instead of referring to period of interest as “the past 12 months,” we provide
respondents with a specific date, and ask that they report only events that occurred
after that date.
• We inform respondents that the questions assessing unwanted sexual experiences
are just about events that occurred since the precise date one year before survey
administration, but assured them that we will ask them about earlier, similar
events later in the survey.
• Instead of stipulating the date range just once in a stem question, and requiring
respondents to remember to apply that date range to subsequent questions, we
incorporate the date 12 months ago into each question.
• If a respondent indicates they have experienced a sexual assault, we ask them a
series of questions about the details of that event. At the end of those questions,
we ask respondents to reconsider if the events described actually happened in the
past 12 months. At that point, they can indicate that they are certain that the
events described actually happened more than one year ago. We exclude those
sexual assaults from our total 12-month counts.
Development of the Sexual Assault Module
Our approach to measuring criminal sexual assault strives to achieve three specific
goals: to focus the assessment on events that are defined as sex crimes under UCMJ
Article 120, to simplify measurement of complex concepts and make them more ame-

Sexual Assault and Sexual Harassment in the U.S. Military

nable to a self-administered survey format, and to introduce technical changes to
improve precision in determining when these events occurred. As a self-report survey
of victim experiences, the survey does not provide information about whether these
events could be prosecuted successfully in court, but rather whether the victim’s experience corresponds to the UCMJ definition of sexual assault crimes.
Clarifications of UCMJ Sexual Assault Offenses

UCMJ rape and sexual assault offenses include unwanted penetration, however slight,
of the vulva, anus, or mouth, or touching of body parts, with intent either to gratify
a sexual desire or to abuse, humiliate, or degrade (except for experiences involving
penetration with a penis, in which case evidence of intent is not required). The UCMJ
provides a list of coercive offender behaviors that are sufficient to demonstrate that
the sexual contact was criminal, including the use of force, threats, by drugging the
victim or assaulting them while unconscious, by fraudulently claiming the contact
served a legitimate professional purpose, by falsely claiming to be someone else, or by
having sexual contact with someone who is incapable of providing consent (e.g., due to
impairment by any drug or intoxicant or due to mental disease or defect).
Since the UCMJ definition of sexual assault is too lengthy, nuanced, and complex to itself serve as a survey question, RAND took the approach of extracting the
core elements establishing sexual assault and presenting questions to respondents in
as simple a form as possible. RAND’s sexual assault questions follow current UCMJ
language more closely than does the 2012 WGRA question, and they do so in a nested
three-stage process to reduce demands placed on the respondent’s memory and reading
level. The survey is designed to first assess whether an unwanted experience occurred
that matches one of the possibly criminal acts described in the UCMJ, then whether it
satisfies the UCMJ intent criteria (i.e., the contact occurred “with an intent to abuse,
humiliate, harass, or degrade any person or to arouse or gratify the sexual desire of any
person”), and finally to assess whether one of the UCMJ coercive offender behaviors
occurred (e.g., use of force, threats, deception, drugs).
Screening questions have been designed to avoid the terms “sexual” and “sex.”
This decision ensures that respondents who have experienced sexual assaults under
UCMJ are included in the survey estimates, even if the respondent does not consider
the crime to have been sexual. This could happen, for example, with penetration or
genital touching that occurs during hazing incidents, or violent events that are not perceived as sex acts but as physical assaults. Such events may not be thought of as sexual
touching, anal sex, or sexual intercourse by the respondent, and so those terms are not
used in the RMWS module assessing sexual assault.
Our approach begins with six behaviorally specific screening questions that are
designed to include all sexual assaults in the past 12 months. However, these questions
may also include behaviorally similar unwanted experiences that are not sexual assaults
under the UCMJ definition (e.g., sexual contact that was accidental or penetration that

Questionnaire Development

occurred for a legitimate medical purpose). These unwanted, but not criminal, events
are identified using responses to subsequent questions, and are not included in the final
count of sexual assaults. The flow of questions and resulting definitions are illustrated
in Exhibit 4.4.
When a respondent indicates that they have experienced a specific unwanted
behavior, they are asked follow-up questions about that experience. First, they are
asked if, in the respondents’ opinion, the behavior occurred for one of the purposes or
offender intents that the UCMJ defines as indicative of sexual assault. If the respondent
indicates that the purpose or offender intent is consistent with a UCMJ sexual assault,
they are then asked about the behaviors by which the offender caused the experience to
occur. These coercive offender behaviors track the UCMJ definitions of sexual assault.
Exhibit 4.4
Categorization of Sexual Assault in the RMWS

Unwanted experience consistent
with UCMJ Article 120

Yes

Event was intended to be
abusive or humiliating

Yes

Event was intended
for sexual gratification

Yes

Offender’s behavior aligns
with UCMJ code

No unwanted
experience

RAND RR870/1-4.4

Unwanted
experience

Yes

Sexual
assault

Sexual Assault and Sexual Harassment in the U.S. Military

The RAND questions, like the UCMJ offense descriptions, are designed to avoid most
ambiguities around the concept of “consent” that have been cited as criticisms of the
WGRA 2012.4 The questions included in the list of possible mechanisms or behaviors
correspond to the specific definition of a sexual assault under the UCMJ and for the
most part map directly onto UCMJ language, as in the instances of the use of force
producing injuries and the use of threats.
Because the UCMJ offers clear behavioral descriptions of most Article 120 crimes,
we can say with some confidence that our survey identifies criminal offenses with a
high probability if the events occurred in the way the respondent reported them on the
survey. However, we also include questions exploring two ways in which sexual assaults
might occur that are more difficult to assess with a self-administered survey. The first
is the possibility of tonic immobility, or the tendency of a victim to freeze in response to
an attack. In such situations, the victim may not be overpowered, overtly threatened,
or drugged, yet at the same time, he or she is not capable of providing consent. The
second assessment concerns situations in which the victim is sufficiently intoxicated by
alcohol at the time of the assault as to sustain a subsequent alcohol-induced blackout
state.5 Such a state could indicate a level of intoxication reflecting an inability to legally
provide consent. For both of these more complex circumstances, the UCMJ has sections that can be interpreted as defining such events as crimes, but more detail would
be needed to verify that the event qualifies as a crime under UCMJ Article 80 or Article 120. Finally, a general non-consent question delivered at the end of the list (“I did
not consent”) is included in the survey (as it is in Article 120) to catch other instances
of nonconsensual, unwanted experiences, such as unwanted touching of private parts
that occurred so suddenly or quickly that there was not time to indicate non-consent.
Details on the correspondence between the 2014 RMWS questions and UCMJ Article
120 can be found in Appendix B.
For our analytic purposes, responding “yes” to any of the six screening questions
is classified as having had an unwanted sexual experience. Subsequently, a positive
response to either of the intent questions and to one of the UCMJ mechanisms that
are aligned with Article 120 results in the unwanted sexual experience being classified
as a sexual assault.
Responding “yes” to the tonic immobility or blackout questions that might meet
the UCMJ definition is categorized as “probable” sexual assault (unless the respondent also responds “yes” to another clear non-consent question). These incidents are
included in the overall sexual assault rate and also reported separately if they are sufficiently common to influence the estimates we provide.
Because many of the behaviorally specific sexual assault screening questions are
highly associated (e.g., those experiencing unwanted penetration with a penis will
also report having contact with someone else’s genitals), it is likely that people who
say “yes” to one question may say “yes” to several. To reduce respondent burden, the
purpose/intent and the offender behavior follow-up questions are not repeated once

Questionnaire Development

sexual assault has been established. This results in respondents being placed into three
mutually exclusive, hierarchical categories: (1) Penetrative Sexual Assault, designed to
assess rape or sexual Assault under Article 120; (2) Sexual Assault Without Penetration, designed to assess aggravated or abusive sexual contact under Article 120; and (3)
Attempted Penetrative Sexual Assault, designed to assess attempted rapes or attempted
sexual assaults under Article 80 that did not also include behaviors that could be classified as aggravated or abusive sexual contact.
One implication of this ordering of assault categories is that if a respondent experiences an attempted penetrative assault that also included any unwanted sexual touching, they are categorized as having experienced “sexual assault without penetration”
rather than “attempted penetrative sexual assault.” The rationale for this decision is
that, provided the attempt included physical contact, an assault that fits clearly within
Article 120 occurred. This is a much more definitive classification that can occur for
attempted sexual assault (i.e., a combination of Article 80 and Article 120), which
depends on more-detailed assessments of offender motivation and behaviors. This is
a departure from the 2012 WGRA prioritization of incidents, in which the attempted
sexual intercourse, anal, or oral sex category had precedence over the unwanted sexual
touching only classification. In other words, some events that likely would have been
categorized as attempted sexual intercourse, anal, or oral sex given the 2012 WGRA
coding rules will be categorized as nonpenetrative sexual assault in the 2014 RMWS.
Returning to the questionnaire structure, if respondents answered affirmatively
to more than one screening question, they were asked to indicate on how many separate occasions in the past year these experiences occurred. Individuals with more than
one sexual assault experience were asked to select the “worst” or most serious experience. All respondents with sexual assaults were then asked detailed follow-up questions
about either their only or their “worst” experience.
Differences from Prior WGRA Measurement

While the 2012 WGRA unwanted sexual contact question was designed to identify
experiences consistent with UCMJ Article 120 definitions of sexual assault, Article
120 has been revised since the WGRA questions were developed in 2006. Further,
the wording of questions in the 2012 WGRA survey required respondents to make
judgments around complicated legal concepts (e.g., consent) not defined in the survey,
which may have negatively impacted the validity of their responses for measuring
UCMJ assaults.
The 2012 WGRA assessed these events by asking a single gating question. This
question asked if respondents were made to engage in any of five behaviorally specific
classes of sexual contact in the past 12 months that were described as “intentional
sexual contacts that were against your will or occurred when you did not or could not
consent.” Whereas the UCMJ describes coercive offender behaviors to define sexual
assault, the 2012 WGRA definition leaves it to the survey respondent to decide what

Sexual Assault and Sexual Harassment in the U.S. Military

“did not or could not” consent means, and what it means to be “made to” engage in an
activity. This leaves the unwanted sexual contact questions open to the interpretation
that “holding a mental reservation about sexual activity is sufficient without any manifestation of lack of consent” for the 2012 WGRA to count the activity as an unwanted
sexual contact (Schenck, 2014). If respondents interpret “did not consent” as meaning
that no explicit statement of approval was provided (even if nonverbal approval cues
were provided), then unwanted sexual contacts measured in the 2012 WGRA could
include sexual contacts that may not be sexual assault as defined in the UCMJ. Alternatively, if a respondent believes that they must have verbally expressed non-consent,
they may fail to classify a forceful rape in which they were too afraid to say “no” as a
sexual assault, leading to an undercount.
The UCMJ considers many types of touching and penetration to be sexual
assault, even when the perpetrator’s intent is not explicitly sexual (provided the purpose is to “abuse, humiliate, harass, or degrade any person”). In contrast, the 2012
WGRA unwanted sexual contact questions limit respondents to consider only “sexual
contacts” involving sexual touching, sexual intercourse, oral sex, anal sex, or penetration by a finger or object. As such, the questions place a greater emphasis on the sexual
nature of the unwanted contact than does the UCMJ. Victims of hazing, bullying, or
other harassment that could qualify as sexual assault in the UCMJ may fail to report
their experiences on the unwanted sexual contact question because no one involved in
the assault considered it sexual. This would lead the 2012 WGRA unwanted sexual
contact questions to undercount the true prevalence of UCMJ sexual assaults against
service members.
Simplification of Complex Constructs

The 2012 WGRA unwanted sexual contact question presents respondents with a complex cognitive challenge. Respondents read a stem question in which multiple inclusion criteria are provided for an unwanted sexual contact to be considered: it occurred
in the past 12 months, it was intentional, it was against your will or it occurred when
you did not or could not consent, and it occurred in the context of five different types
of physical contact (each with a definition, see Exhibit 4.5). Respondents must then
indicate whether any of five types of sexual contact—some of which include multiple
alternative types of contact—meet all of the inclusion criteria. This type of complex
gating question does not meet the recommended measurement standards delineated in
the sexual assault assessment literature (Fisher and Cullen, 2000; National Research
Council, 2014).
RAND’s 2014 sexual assault questions are designed to reduce cognitive burden
by breaking the complex components of sexual assault down into a series of specific,
behaviorally defined questions and follow-up questions, each of which contains all
the information necessary to arrive at a yes/no response (i.e., there is no obligation to
understand or remember a complex stem question).

Questionnaire Development

It begins with instructions to
describe the kinds of experiences that
the survey is designed to capture (see
Exhibit 4.6).
Then, the six screening questions are presented with a yes/no
response format:

Exhibit 4.5
2012 WGRA Unwanted Sexual Contact Question
In the past 12 months, have you experienced
any of the following intentional sexual
contacts that were against your will or
occurred when you did not or could not
consent where someone...
• Sexually touched you (e.g., intentional
touching of genitalia, breasts, or buttocks) or made you sexually touch them?
• Attempted to make you have sexual
intercourse, but was not successful?
• Made you have sexual intercourse?
• Attempted to make you perform or
receive oral sex, anal sex, or penetration by a finger or object, but was not
successful?
• Made you perform or receive oral sex,
anal sex, or penetration by a finger or
object?

1. Since [date one year prior to
survey administration], did
you have any unwanted experiences in which someone put
his penis into your [if female,
then display “vagina,”] anus or
mouth?
2. Since [date one year prior to
survey administration], did
you have any unwanted experiences in which someone put any object or any
body part other than a penis into your [if female, then display: “vagina,”] anus
or mouth? The body part could include a finger, tongue or testicles.
3. Since [date one year prior to survey administration], did anyone make you put
any part of your body or any object into someone’s mouth, vagina, or anus when
you did not want to? A part of the body could include your [if male, then display: “penis, testicles,”] tongue or fingers.
4. Since [date one year prior to survey administration], did you have any unwanted
experiences in which someone intentionally touched private areas of your body
(either directly or through clothing)? Private areas include buttocks, inner thigh,
breasts, groin, anus, vagina, penis, or testicles.
5. Since [date one year prior to survey administration], did you have any unwanted
experiences in which someone made you touch private areas of their body or
someone else’s body (either directly or through clothing)? This could involve the
person putting their private areas on you. Private areas include buttocks, inner
thigh, breasts, groin, anus, vagina, penis, or testicles.
6. Since [date one year prior to survey administration], did you have any unwanted
experiences in which someone attempted to put a penis, an object, or any body
part into your [if female, then display: “vagina,”] anus or mouth, but no penetration actually occurred?
Technical Changes to Improve Respondent Precision

As described above in the sexual harassment section, forward telescoping, or the tendency for victims to report crimes as more recent than they actually were, represents an
important threat to the validity of crime-victimization estimates using survey data. As

Sexual Assault and Sexual Harassment in the U.S. Military

Exhibit 4.6
Preamble to 2014 RMWS Sexual Assault Assessment
Questions in this next section ask about unwanted experiences of an abusive, humiliating,
or sexual nature. These types of unwanted experiences vary in severity. Some of them
could be viewed as an assault. Others could be viewed as hazing or some other type of
unwanted experience.
They can happen to both women and men.
Some of the language may seem graphic, but using the names of specific body parts is the
best way to determine whether or not people have had these types of experiences.
When answering these questions, please include experiences no matter who did it to you
or where it happened. It could be done to you by a male or female, service member or
civilian, someone you knew or a stranger.
Please include experiences even if you or others had been drinking alcohol, using drugs, or
were intoxicated.
The following questions will ask you about events that happened AFTER [date one year
prior to survey administration]. You will have an opportunity to describe experiences that
happened BEFORE [date one year prior to survey administration] later in the survey.
Remember, all the information you share will be kept confidential. RAND will not give your
identifiable answers to the DoD.

with the sex-based MEO violations module, in the sexual assault module we attempted
to guard against telescoping in the 2014 RMWS by refreshing memory for life experiences that occurred one year prior to the survey administration date at the beginning
of the survey and reminding the respondent, in each question, by using the specific
date one year prior to survey administration.
In addition, at the end of the past-year sexual assault module, respondents are
asked to verify that the assault did occur in the last 12 months. If a respondent is
certain that the sexual assault occurred prior to the past 12-month period, and had
no other experiences during the 12-month period that might be classified as a sexual
assault, the service member was not counted as experiencing a sexual assault in the
past year.
A second precision improvement offered by the RAND sexual assault module
is greater detail on the types of sexual assaults experienced by service members. As
noted above, the WGRA unwanted sexual contact section begins with a single yes/
no gating question that asks whether the respondent experienced any of five types of
sexual assault in the past year. Research on survey design shows omnibus questions
about rape do not cue memories of relevant experiences as effectively as do a series of
behaviorally specific questions (Sabol and Beck, 2014; Koss, 1993; National Research
Council, 2014). In the 2014 RMWS, each type of unwanted experience is assessed
separately, until a qualifying crime in the past year is detected.

Questionnaire Development

Assessment of Lifetime Sexual Assault

Assessment of sexual assaults prior to the past year was a secondary aim of the survey,
but a thorough assessment of all lifetime sexual assaults would have lengthened the
survey excessively. Therefore, we simplified the assessment of prior lifetime sexual
assault exposure. It was assessed with the same screening questions used for the prioryear assessment, except that two questions capturing penetrative sexual assaults were
collapsed into one, resulting in five questions. In addition, no follow-up questions were
used to assess the UCMJ-specific criteria regarding the purpose and mechanisms of
sexual assault. However, to improve precision, the questions were introduced as events
that are of an abusive, humiliating, or sexual nature, and are situations in which “you
did not want it and did not consent.” Further, “did not consent” is defined and kept
visible when the five screening questions are asked. Specifically, the definition reads,
“‘Did not consent’ means that you told or showed them that you were unwilling, that
they used physical force or threats to make you do it, or that they did it to you when
you were unconscious, asleep or so high or drunk that you could not understand what
was happening.” Thus, specific aspects of the nature of the event and the way in which
it was coerced are included in the questions. Respondents who indicated that they
experienced a lifetime sexual assault were further categorized as having experienced
sexual assault(s) prior to joining the military, during their military career, or both.
Development of Demographic Questions

Most demographic information for this study was available in the DMDC data delivered to RAND for purposes of sampling and weighting. Thus, it was only necessary
to include a few questions on the survey that were (1) not available in the DMDC
dataset, (2) needed for purposes of validating that the person invited was the person
who completed the survey, or (3) likely to change between the time the DMDC data
was drawn and the time the survey was fielded. We needed to ask two questions about
characteristics that might have changed since the DMDC data were drawn: deployment in the prior year and, for guard or reserve members, the number of days spent in
a compensated guard or reserve status. In addition, we asked respondents to confirm
their month of birth.
In addition to the information available in administrative records, we included
questions on the survey to assess factors known to be associated with increased risk for
sexual harassment or sexual assault. We asked one question about current relationship
status, assessing statuses that would not be included in DMDC data, such as being
in a committed romantic relationship but unmarried (“single” in DMDC data). We
also planned to ask about the respondent’s sexual orientation because surveys of civilian populations have shown that sexual minorities are at substantially elevated risk
of sexual assault and sexual harassment, and these groups may be vulnerable in military settings (e.g., Burks, 2011; Walters et al., 2013). However, the Office of General
Council at OSD determined that while legally defensible, inclusion of this question

Sexual Assault and Sexual Harassment in the U.S. Military

would violate current DoD policy against collecting information on service members’
sexual orientation. Given reluctance on the part of service representatives to support
an exemption to DoD policy for purposes of including this question in the survey,
we elected not to include it in the 2014 survey. We recommend, however, that it be
included in future assessments.
Development of Long Form Questions

Beyond the assessment of sexual harassment and sexual assault, the 2012 WGRA
included questions on workplace climate, sense of safety, interactions with supervisors
and peers, mental health status, and experiences with trainings and policies related to
sexual harassment and assault. We sought to include a subset of these questions if they:
• Were part of an important time series used to evaluate DoD policies. These
included questions assessing opinions about DoD efforts to address sexual harassment and sexual assault, workplace climate, and likelihood of retention in the
military. In some cases these questions were edited for clarity and precision, and
in other cases they were kept exactly the same as the 2012 WGRA for complete
comparability.
• Provided useful data that could be used to better understand the 2014 RMWS
questions on sex-based MEO violations and sexual assault. These included questions on general health status and current depression and posttraumatic stress
disorder (PTSD) symptoms. In some cases we used updated measures for these
constructs. Specifically, we used
–– one question to assess general health status drawn from the SF-36 (Ware and
Sherbourne, 1992). This is a new question and replaced a lengthier scale.
–– the PHQ-8 (Kroenke et al., 2009) to assess depression symptoms (same as on
the 2012 WGRA)
–– the PC-PTSD updated for the Diagnostic and Statistical Manual of Mental
Disorders: Fifth Edition (DSM-V). This is a new measure of PTSD recently
updated from a primary care PTSD screening tool (Prins et al., 2004). This
measure is substantially shorter than the one used on the 2012 WGRA and
conforms to the more recent diagnostic criteria as defined in the DSM-V.
Inclusion of Service-Specific Questions

Each service was invited to suggest up to ten additional questions for inclusion on the
survey to be presented only to respondents from that service. RAND worked with each
service to edit questions for precision and clarity. The Marine Corps added four questions, the Coast Guard added five questions, and the Air Force added one question.
These questions were about training, programs, or policy questions relevant to sexual
assault prevention in those services (see the Survey Instrument for specific items).

Questionnaire Development

Development of Prior WGRA Form Questions

We developed the prior WGRA form to be very similar to the 2012 WGRA, but
included fewer questions to improve response/completion rates. Specifically, the prior
WGRA form
• Included 2012 WGRA questions about workplace relations in the first section of
the survey to help set context for questions about sexual harassment and gender
discrimination.
• Omitted items in the first section of the survey that were not explicitly related to
workplace relations and were shown in the 2012 WGRA to be associated with
high rates of survey break-off (i.e., where respondents discontinued survey participation). For instance, the questions assessing current PTSD and depression
were omitted due to extensive break-offs at the beginning of this section in 2012.
• Included the full set of items related to workplace discrimination, sexual harassment, and sexual assault.
• Added one additional item after the unwanted sexual contact questions to assess
forward telescoping.
• Omitted items following the unwanted sexual contact section to decrease the
overall length of the survey.6
For data drawn from responses to the prior WGRA form questionnaire, unwanted
sexual contact and sexual harassment will be calculated in the same way as the 2012
WGRA.

CHAPTER FIVE

Analytic Methods
Bonnie Ghosh-Dastidar, Terry L. Schell, and Andrew R. Morral

This chapter provides details of the steps necessary to produce population-representative estimates of the percentage of service members who experienced sexual assault or
a sex-based military equal opportunity (MEO) violation in the past year. The key steps
are as follows:
• Sampled individuals must be assigned a case disposition to reflect the status of
their survey. This step involves determining who provided sufficient survey data
to be counted as a respondent.
• Data collected from respondents will be weighted to produce population-representative estimates.
• Population estimates of the rates of sexual assault and other study outcomes
are computed in a manner that correctly incorporates the effects of the analytic
weights.
Final Case Disposition
Table 5.1 summarizes the case disposition categories and rules, which follow survey
research standards for documentation (American Association for Public Opinion
Research, 2011). Our approach differs from earlier WGRA case disposition definitions
in several ways. First, we consider every service member in the 2014 sampling frame
to be eligible if they were alive at the end of the survey field period. In the 2012 Workplace and Gender Relations Survey of Active Duty Personnel (WGRA), the majority
of individuals whose sampled records were ruled ineligible were classified as Disposition 1: Sample ineligible—deceased or no address available in DEERS (DMDC, 2012).
Those who were hospitalized and those who recently separated from the military were
also counted as ineligible for the survey. Our view is that these categories of individuals should not be classified as ineligible, because they have served a major part of the
past year in the military, which makes them part of the population that we are trying
to measure. Hospitalizations and separation from the military may also be associated
with the primary outcomes of this study, so the exclusion of these groups from the
survey could bias the survey results. Those who separated from the military must have
57

Sexual Assault and Sexual Harassment in the U.S. Military

Table 5.1
Case Disposition Definitions
Case Disposition

Data source

Description

Eligible complete
Complete

Survey response file

Any Sexual Assault (RAND form) or
Unwanted Sexual Contact (prior WGRA
form) is coded as non-missing

Partial complete,
insufficient information

Survey response file

Survey started, but will not be used in
analyses because Any Sexual Assault
(RAND form) or Unwanted Sexual Contact
(prior WGRA form) were coded as missing

Refusal

Survey management system

Sampled member refuses to take the
survey or requests to be removed from
future contact

Survey management system

No response from sampled member

DMDC personnel records

Sampled member died before the end of
fielding period

Eligible incomplete

No response
Ineligible
Deceased

left recently—they were part of the Defense Manpower Data Center (DMDC) dataset of employed service members provided just prior to the start of fieldwork—so any
reports of sexual assault or harassment over the past year experienced by these individuals are likely to describe their experiences as a service member. Treating these
hard-to-reach groups of service members as ineligible has the potential to increase the
nominal response rate for the study, but ultimately makes the eligible population a
less accurate representation of the population of interest for this study. As a result, our
study has just one category of ineligible—anyone who could not be surveyed because
he or she had died. All other individuals in the sample who did not respond to the
survey are treated as eligible non-respondents for the response rate and included in the
nonresponse adjustments.
A second important difference involves the definition of a “partial complete with
sufficient information”; such cases have some number of unanswered survey questions
but are included in analyses. In 2012, DMDC required WGRA respondents to complete at least 50 percent of all survey questions to be included as a respondent. Our
definition is motivated by the primary aim of this study, which is to estimate the prevalence of sexual assault. Therefore, we do not exclude from our analyses respondents
who completed the sexual assault items. We have defined a partial complete to be those
members who have provided sufficient information to allow us to determine whether
they were sexually assaulted in the past year.1

Analytic Methods

Survey Weights
Once respondents and nonrespondents have been identified, the respondents’ data
need to be weighted. Survey weighting makes the analytic sample more representative of the population (Heeringa, West, and Berglund, 2010; Little and Rubin, 2002;
Schafer and Graham, 2002). Under current federal guidelines, all major federal surveys are directed to use weights (or comparable methods) to improve the representativeness of the analytic sample (OMB, 2006). Specifically, analyses should incorporate weights that adjust for sampling probabilities and nonresponse, and nonresponse
weights should “make use of the most relevant data available” to insure a representative
analytic sample (Guideline 3.2.12; OMB, 2006).
Weighting is used by all major surveys conducted by the Census Bureau, the Centers for Disease Control and Prevention, the National Institutes of Health, the Bureau
of Justice Statistics, the Department of Labor, and the Federal Reserve. All reputable
private polling and research firms also use similar nonresponse adjustments when generating population estimates, including research conducted by Gallup, Pew, SurveyUSA, WESTAT, NORC, Abt SRBI, and RTI. The prior WGRA surveys (DMDC,
2012; DMDC, 2014), as well as other major surveys conducted for the Department of
Defense (DoD; e.g., Status of Forces), use survey weights to reduce biases that would
otherwise be present in the population estimates. Consistent with federal guidelines
on survey analysis, this study uses analytic weights that adjust for both sample design
and survey nonresponse. Similar to prior WGRA surveys, analytic weights are developed as three components: design weights, nonresponse weights, and post-stratification weights.2
For active-duty service members, we sampled every woman and one out of four
men—that is, women were selected with certainty (sampling probability of 1) while
25 percent of men were selected for the study. An unweighted average of the respondents’ survey reports would not correctly represent population results: It would overrepresent the opinions and experiences of women, relative to their share of the activeduty population. Thus, design weights are needed to account for the different sampling
probabilities for men versus women.
In addition, surveys typically have some amount of nonresponse (eligible incompletes). Because some types of people may be less likely to complete a survey than
others (e.g., junior enlisted, men, Marines), the unweighted respondent data may not
accurately reflect the responses expected from the full population. Thus, nonresponse
weights are required to make the respondents accurately represent all of those originally
sampled for survey participation. This study, like the prior WGRA surveys, has access
to detailed data from the military describing the characteristics of all of those who were
sampled, even if they did not respond. Thus, we can describe in detail the characteristics on which the respondents do not match the nonrespondents to the survey, and
adjust for those differences.

Sexual Assault and Sexual Harassment in the U.S. Military

Finally, there is a post-stratification step to ensure that the weighted respondent
sample precisely matches the full population within each of our reporting categories
(e.g., gender, branch of service, pay grade). This step ensures that the proportions in our
analytic dataset perfectly match those in the population on key features such as gender,
pay grade, service, and related characteristics.
Primary Objectives for Weighting

We identified a weighting method that met multiple primary objectives:
• Produce unbiased population estimates of key survey outcomes across and within
various reporting categories
• Make full use of the socio-demographic and occupational information available
in the administrative data on the full population
• Enable a direct comparison of the WGRA form estimates in 2014 with the estimates from the 2012 WGRA
• Ensure that the analytic weights do not inadvertently identify individual respondents when they are provided to DMDC.
To meet these objectives, we needed two sets of weights:
• WGRA weights. To produce estimates using the prior WGRA form that are as
comparable as possible to the WGRA estimates from prior years, we developed
one set of weights using the same procedures DMDC used in 2012. These weights
ensure that any differences in survey results between 2012 and 2014 are not due
to changes in analytic methods. We use these WGRA weights only for analysis of
responses from the prior WGRA form survey.
• RAND Military Workplace Study (RMWS) weights. DMDC and RAND have
identified several types of administrative information about service members that
were not used when developing the 2012 weights. Including that information
in the development of the RMWS weights allows the RMWS weights to better
ensure that the analytic sample is representative of the population. We intend to
apply these weights to all estimates generated from the new RAND forms, not
the prior WGRA form.
Both weighting approaches used the same design weights, since the design weights
adjust for the disproportionate sampling of women relative to men for the RMWS.3
Similarly, the post-stratification weights are derived in comparable ways. Specifically,
in the WGRA weight’s post-stratification step, the weighted respondents (applying
both sampling and nonresponse weights) were further adjusted to precisely match the
population subgroups defined by the cross-categorization of service, pay grade, gender,
and minority status. For example, the proportion of male, Army, junior officers who
are caucasian in the weighted sample is set exactly to that of the population. The

Analytic Methods

RMWS weight’s post-stratification step is essentially the same, though we will balance
40 reporting categories (defined by pay grade, service, and gender) within each RAND
form type, and will not include the minority status indicator because detailed racial
and ethnic categories were included in our nonresponse weights.4
The two methods differ primarily in the derivation of nonresponse weights, which
we describe next.
Differences in the WGRA and RMWS Nonresponse Weights

The WGRA nonresponse weighting approach attempts to account for (or explain)
survey nonresponse on the basis of the factors listed in Exhibit 5.1. Variables that differed substantially between respondents and sampled nonrespondents were identified
through statistical modeling, and model-based weights were derived to compensate for
those differences.5
For the RMWS nonresponse weights, we included more of the information available for the entire sample, including both respondents and nonrespondents, to help
identify possible sources of nonresponse bias in the population estimates of our key
outcomes. To do this, we assembled a very broad range of information. These characteristics, listed in Exhibit 5.1, include socio-demographic information (e.g., age, ethnicity, marital status), occupation-related information (e.g., pay grade, service, unit
size, occupation code, percent male in one’s occupation, deployment history, length of
time served in the military), and fieldwork information (e.g., missing email address,
missing mail address, number of letters returned as undeliverable, percentage of emails
that were bounced back).
In general, adding variables to the nonresponse model can only reduce nonresponse bias to the extent that those variables are associated with both (a) one of the outcomes of interest and (b) the probability of responding to the survey (Little and Rubin,
2002). For example, if junior enlisted personnel are at higher-than-average risk for
sexual assault, but have a lower-than-average response rate, we would need to give junior
Exhibit 5.1
Predictors in WGRA Weighting Model
Variables
Gender (men, women)
Race/ethnicity (5 levels)
Service (Air Force, Army, Marine Corps, Navy)
Pay grade (E1-E3, E4, E5-E6, E7-E9, W1-W5, O1-O3, O4-O6)
Deployment status (never deployed, not deployed past year, deployed past
year)
Combat occupation flag (combat, combat support)
Family status (single, married)

Sexual Assault and Sexual Harassment in the U.S. Military

enlisted respondents more weight in the analyses to ensure unbiased estimates of the
rates of sexual assault in the military. For this reason, we included every available variable that is plausibly associated with either survey nonresponse or with one of our key
outcomes (Exhibit 5.2).
Exhibit 5.2
Predictors in Outcome-Optimized RMWS Weights
Demographic Factors

Military Environment Factors

Gender (male/female)

Duty unit location (CONUS/OCONUS)

Date of birth

Percent male at military installation**

Race code (32 categories)

Number of people at military installation**

Ethnic affinity code (23 categories)

Percent male in military unit***

Marital status code (8 categories)

Number of people in military unit***

Total number of dependents

Survey Fielding Factors

Education level code (21 categories)

Change-of-address entered in DMDC records after
sampling (Y/N)

Armed Forces Qualification Test score

Change of station after sampling (Y/N)

Military Career Factors

Change of station, past year (Y/N)

Service (5 levels, including Coast Guard)

No mailing address at time of sampling (Y/N)

Pay grade (20 levels)

No email address at time of sampling (Y/N)

Days of active-duty service, past year

First letter returned as postal nondeliverable (Y/N)

Cumulative months of active federal military
service

Email sent by Marine Corps (Y/N)

Projected end date for current term

Percentage of sent emails that bounced back

Date of entry into military services
Military accession type (25 categories)
Strength accounting code (8 categories)
Separated or retired after sampling (Y/N)
Months deployed since 9/11/2001
Months deployed since 7/01/2013
DoD occupational group (20 categories)
Percent male within members’ specific occupation*
Number of people within members’ specific occupation*
NOTES: Categories containing fewer than 40 cases among survey respondents were combined in the
statistical models. There are slightly different military factors used when developing weights for the
reserve component relative to the active component; those additional variables are not included in this
exhibit.
* Derived from 302 DoD occupational categories
** Derived as two separate variables for each member’s assigned installation (N=3031) and their duty
installation (N=3147).
*** Derived as two separate variables for each member’s Assigned Unit (N=24,496) and their Duty Unit
(N=24,517).

Analytic Methods

Whereas including all variables that could plausibly explain nonresponse has the
advantages described above for reducing bias, it can also have the unwelcome effect
of reducing the precision of estimates. This is because weighting on variables that are
not associated with any survey outcomes cannot remove any nonresponse bias in those
outcomes, but can increase the variance of the weights, with resulting reductions in the
precision of estimates (Little and Vartivarian, 2005). Thus, the ideal weights are based
on a nonresponse model that includes only those factors that are associated with the
key outcomes and nonresponse. Such a model would remove the maximum amount
of nonresponse bias, while limiting the variance in the weights to just the amounted
needed to eliminate nonresponse bias. Thus, the best weighting approach is one that
has been optimized for the specific outcomes that the study is designed to measure.
For this reason, the RMWS nonresponse weights are designed as a two-step process. First, we identified factors contained in Exhibit 5.2 that are statistically associated with one or more of the six primary outcomes for our study (three types of sexual
assault and three types of sex-based MEO violations). Second, we estimated a nonresponse model in which the factors in Exhibit 5.2 are included only to the extent that
they are predictive of one or more of the primary outcomes. This first step creates six
new variables, each of which is a weighted combination of the factors in Exhibit 5.2
that best predict one of the six outcomes. Specifically, a separate regression model
was estimated for each of the six primary outcomes among survey respondents using
the Exhibit 5.2 variables as predictors.6 The second step creates nonresponse weights
that balance the respondents to the full sample on these six combination variables.7
In this way, the nonresponse weights are optimized to remove as much nonresponse
bias as possible in the primary survey outcomes, while avoiding the inclusion of additional variables in the nonresponse adjustment that would reduce precision without
any reduction in bias.
Evaluation of the WGRA and RMWS Weighting Approaches

Under both the WGRA and the RMWS weighting approaches, the sample is precisely
balanced to the eligible population on the cross-classification of gender, service, and
pay grade group. For example, both weights yield the right number of female, Marine,
senior enlisted service members. The only difference between the two weights is how
they balance the sample to the population within each of those 40 categories. Within
each reporting category, the WGRA weights balance the respondents to the population on (a) the racial distribution within reporting categories, (b) the number of E7-E9
within the senior enlisted categories, and (c) the number of E4 within junior enlisted
categories.
In contrast, the RMWS weights attempt to achieve balance within each of the
40 reporting categories on a weighted combination of the most important predictors
of sexual assault and sex-based MEO violations. While the WGRA weights precisely
balance on minority versus nonminority status, the RMWS weights only attempt to

Sexual Assault and Sexual Harassment in the U.S. Military

achieve balance on minority status to the extent that it predicts one of the six sexual
assault or sex-based MEO violation outcomes. This allows the RMWS weights to
incorporate a much wider range of factors that might explain nonresponse bias without unnecessarily eroding precision. For example, it is plausible that women in occupations that are predominately male experience more discrimination than women in
occupations that include more women (e.g., Moore, 2010). The RMWS system of
weights will therefore seek to improve the representativeness of the sample in each of
the 40 reporting categories on the gender distribution of occupations. If the sample in
some reporting category does not have enough respondents in jobs that are 95-percent
male, it will give more weight to those in the sample who are in such jobs to be more
representative on that dimension.
RAND’s main findings report, which will be published in spring 2015, will present the balance achieved between weighted respondents and the population on each of
the reporting categories, and on variables in Exhibit 5.2 that prove to be key predictors
of sexual assault or sex-based MEO violations.

CHAPTER SIX

Summary of Major Improvements in the RMWS
Andrew R. Morral, Kristie L. Gore, and Terry L. Schell

Sexual harassment, gender discrimination, and sexual assault against service members
are matters of intense concern within the Department of Defense (DoD), the Congress, and among members of the public. For this reason, prior efforts to quantify
the extent of the problem have been closely reviewed and critiqued, as we expect our
efforts will be. RAND researchers were given the opportunity to take a fresh look at
this survey challenge to establish independent estimates of the number of service members experiencing sexual crimes and violations. Surveys can always be improved, and
new approaches to measuring complex, sometimes hidden phenomena like sex-based
military equal opportunity (MEO) violations and sexual assault can yield important
insights.
Although we have developed a new approach to measuring these phenomena,
ours draws heavily on the methods and analysis developed within DoD’s own survey
research department in the Defense Manpower Data Center (DMDC), which has
conducted research on sexual harassment and assault in the military for more than two
decades. Indeed, several of the modifications we incorporated into our approach were
suggested by analyses conducted by DMDC.
In this section, we highlight the key improvements we have contributed to the
measurement of sex-based MEO violations and sexual assault, and describe some of the
experiments we are conducting that may help improve the accuracy of future surveys.
Clearly, however, the science of accurately estimating the rates at which these offenses
occur is still developing. In future administrations of the Workplace and Gender Relations Survey of Active Duty Personnel (WGRA), survey researchers at DoD will need
to continue to revise and improve our methods, their own, or some combination to
arrive at the most valid results possible.
RMWS Modifications to Improve the Precision of Sex-Based MEO
Violations and Sexual Assault Estimates
All of the key modifications we implemented for the 2014 RAND Military Workplace
Study (RMWS) were designed to improve the validity and precision of our estimated
65

Sexual Assault and Sexual Harassment in the U.S. Military

rates of exposure to sex-based MEO violations and sexual assault. We describe these as
“improvements,” though we note that the WGRA was not designed to produce valid
and precise counts of crimes and violations. Instead, it included a screening item for
unwanted sexual contact that was patterned on the Uniform Code of Military Justice
(UCMJ) definition of sexual assault, but never presented by DMDC as an estimate
of crimes, per se. Similarly, as discussed earlier, the WGRA measurement of sexual
harassment used a scale designed to capture a psychological sense of harassment, not
one designed to estimate true violations of the law or DoD sexual harassment policy.
Therefore, many of our “improvements” should be considered as advances for the purposes of estimating the prevalence of crimes and violations, a somewhat different objective than that of prior WGRA surveys.
Enlarged Sample Size

The sample size DoD established for this study is five times larger than recent WGRA
samples. There are good reasons to avoid such large surveys, but large samples are
especially valuable for evaluating the experiences of small subpopulations that cannot
be readily targeted on their own, such as men who have experienced a sexual assault.
By sampling more than one-half of a million service members, we ensure sufficient
numbers of men who have experienced a sexual assault—and many other small subgroups—to draw reasonably precise estimates and generalizations about their experiences. This large sample also allowed us to give respondents different sets of questions,
a procedure that will provide important information on the differences between the
old and new methods of assessing sexual crimes and violations.
Clarifying the Interpretation of the Measured Constructs

The WGRA was designed to measure sexual harassment, sexist behavior, and
unwanted sexual contacts, but how one should interpret these measures may not be
sufficiently clear. In some DoD communications, unwanted sexual contact was equated
with sexual assault, while others emphasized that the precise legal standard of sexual
assault could not be established through the unwanted sexual contact questions. Similarly, DoD leadership appeared unclear on whether unwanted sexual contact included
sexual harassment or even just annoying attention with sexual overtones. Neither the
WGRA’s sexual harassment nor the sexist behavior measures correspond directly to
MEO violations as defined by DoD directives or federal civil rights law. Each of these
uncertainties raises questions about what meaning should be attached to the specific
behaviors assessed with the WGRA, and the severity of the problems they represent.
In the RMWS, we address these problems by designing measures that explicitly
assess the legal criteria necessary for determining whether a sex-based MEO violation
occurred or whether a UCMJ Article 120 or Article 80 offense occurred. Of course,
these measures are based on respondents’ perceptions of events. Independent investigation might find that some of these reports do not meet legal criteria for the offense.

Summary of Major Improvements in the RMWS

Nevertheless, if the events occurred as described by the respondent, then an Article 120
or Article 80 offense occurred. For sex-based MEO violations, we can make the same
strong claim for one of the three violations we assess, sexual quid pro quo. Hostile workplace and gender discrimination arguably require information that an individual may
in some cases not be able to provide, such as knowledge of how a “reasonable person”
might react to the events, and knowledge of why, for instance, someone else got a job
assignment the respondent sought. Nevertheless, we incorporate questions that elicit
the best information available from the respondent on these questions, which advances
the identification of likely sex-based MEO violations well beyond prior methods that
did not correspond to the regulations defining these violations.
A second improvement in the RMWS sex-based MEO violations assessment is
that we do not require respondents to correctly label their discriminatory experiences
as “sexual harassment” or “gender discrimination.” Because most people are not familiar with the details of Equal Employment Opportunity law and MEO regulations,
many who experience sexual harassment do not recognize it as such (Fitzgerald, Swan,
and Fischer, 1995). Moreover, since this labeling requirement makes the sexual harassment measure sensitive to how knowledgeable the workforce is about regulations and
law, there is a danger that effective prevention and training programs will appear to be
causing more sexual harassment over time, merely because victims become better able
to recognize and label harassment as such.
Simplifying Question Syntax to Improve Respondent Understanding

As discussed in Chapter Four, the primary questions used to measure sexual harassment
and unwanted sexual contact in the earlier WGRA surveys impose heavy demands on
respondents because of their syntactic complexity, because of the number of exclusion
and inclusion criteria they ask respondents to simultaneously consider, and because
respondents must retain each of these conditions in their memories while they apply
them to the subsequent series of questions or descriptions—some of which also entail
multiple parts. This type of complexity is likely to generate imprecise and ambiguous
survey results (Tourangeau, Rips, and Rasinski, 2000). The likelihood of respondent
error and resulting imprecision in survey estimates of these constructs is high, as some
respondents will miss or ignore some criteria, not understand their place in the sentence syntax, or not remember and apply each condition on the questions that follow.
Question complexity is one of the cardinal threats to survey reliability and precision
(Saris and Gallhofer, 2014).
In lieu of a complex gating question, such as the unwanted sexual contact question in the prior WGRA surveys, we break the complex concept of sexual assault into
a series of separate, simpler questions. Research on survey design shows omnibus questions about rape do not cue memories of relevant experiences as effectively as do a
series of behaviorally specific questions (Cook et al., 2011; National Research Council,
2014; Koss, 1993; Sabol and Beck, 2014). In the RMWS, we present six separate sexual

Sexual Assault and Sexual Harassment in the U.S. Military

assault screening questions, selecting any one of which could lead to a finding of pastyear sexual assault, and test each of the legally required elements in separate follow-up
questions if a respondent answers “yes” to a screening question.
Clarifying Question Terminology

Instead of relying on respondents’ intuition about what consent entails, to the extent
possible we have substituted the specific offender behaviors defined under the UCMJ
as sex crimes. Consent is a complex issue that has multiple definitions across legal systems and over time, but it is central to the WGRA unwanted sexual contact question
and was presented without a definition. As such, it is likely that people differed in their
interpretation of the unwanted sexual contact question. Our approach follows that of
the UCMJ, which attempts, to the extent possible, to eliminate the concept of consent
from definitions of Article 120 crimes and replace it with specific offender actions that
indicate a crime. We ask if the offender used specific coercive behaviors to accomplish
the unwanted sexual contact, such as use of force, threats, drugging, or other means.
By changing the way consent is assessed, we expect respondents to better understand
the questions and to answer them more reliably.
The WGRA unwanted sexual contact question also emphasized that the events
it references are sexual. That is, the question emphasizes sexual contacts, sexual intercourse, anal sex, oral sex, and sexual touching. The UCMJ does not, however, require
that Article 120 crimes arouse or gratify sexual desires, though this is certainly one
motive recognized in the law. The law also recognizes assaults that are designed only
for the purpose of abusing, humiliating, harassing, or demeaning the victim. The latter
such assaults—hazing and bullying incidents, for instance—may not be considered by
the respondent in the context of a question that focuses so centrally on sexual contacts.
We attempted to clarify the scope of Article 120 crimes by referring only once to sexuality in the introduction to the screening questions (“Questions in this next section ask
about unwanted experiences of an abusive, humiliating, or sexual nature”). Each subsequent screening question asks about a specific type of physical contact or attempted
physical contact that could qualify as an Article 120 offense, but without suggestion
that the contact was sexual.
Reducing Overcounting of Offenses Due to Telescoping

As discussed in Chapter Four, people often report crimes as occurring more recently
than they really did—a bias that is referred to as telescoping (National Research
Council, 1984). This can cause survey estimates of past-year crime rates to be overestimated by 30–50 percent (Andersen et al., 1979; Cantor, 1989; Lehnen and Skogan,
1984). We have incorporated a number of techniques into the RMWS to reduce the
effects of telescoping. We begin the survey by noting that many questions ask about
events within the past year. We orient respondents to this timeframe with a series of
questions designed to help them remember where they were and what they were doing

Summary of Major Improvements in the RMWS

exactly one year ago. Before the sexual assault module, respondents are told they are
first going to be asked about events during the past year and then will be asked about
events prior to the past year. Instead of mentioning the past-year timeframe once in an
introduction or stem question, each of the core questions in our sex-based MEO violations and sexual assault modules asks whether the respondent had such an experience
since the date exactly one year ago. The web survey is programmed to present the exact
date in the text of the questions. Finally, at the end of the sexual assault module, those
who have described a sexual assault are asked to verify that it occurred in the past year,
and not earlier. Each of these reminders and emphases is designed to help limit telescoping biases in the RMWS.
Improving Survey Response Rates

As discussed in Chapter Two, survey response rates have declined precipitously over
the past two decades for all types of government and private-sector surveys, including
those of military personnel. Low response rates threaten the validity of survey results.
That is, as more and more people choose not to participate, the risk that participants
differ systematically from the population they were drawn from increases. Therefore, we
designed the survey and the recruitment effort to optimize response rates. For instance,
we made the survey shorter than in the past. We were able to tell most respondents
that it would take them just eight minutes to complete. Relatedly, we placed the key
sex-based MEO violations and sexual assault measures at the beginning of the survey
to maximize the number of respondents on these questions, given that we know many
respondents will break off their participation before completing the survey.
We also made the survey smartphone-compatible for the first time, since many
junior enlisted members do not have frequent or convenient access to computers, but
do have personal smartphones. We developed a communications plan that provided
senior leadership in each of the services with advice on how to promote and encourage survey participation without coercing participation. We developed public service
announcement videos that played on the Armed Forces Network and were posted on
YouTube. We bought Facebook advertising that targeted hundreds of thousands of
junior enlisted members, and posted advertising on a website popular with this group.
Also, we supported and encouraged print news stories that appeared in a wide range of
DoD and independent news services.
Reducing Distortions in Estimates Caused by Survey Nonresponse

Despite our efforts to increase response rates, a large proportion of sampled members
will choose not to participate in the RMWS. To the extent their experiences differ systematically on key outcome variables in comparison with participants, our estimates
could misrepresent the experience of the full population of service members. For past
WGRA surveys, DMDC Human Resources Strategic Assessment Program (HRSAP)
researchers have used powerful nonresponse weighting techniques to limit the distor-

Sexual Assault and Sexual Harassment in the U.S. Military

tion that could result from nonresponse. Moreover, after the 2012 WGRA, HRSAP
conducted a battery of nonresponse analyses that suggested more factors that could be
used to better control this potential problem. We build on their analyses, incorporating
the factors previously included, the new factors suggested by HRSAP’s later research,
and additional factors we suspect may be relevant for these models.
In all nonresponse weighting, the goal is to reduce bias in the estimate that
can be identified by variables that are associated with both the survey outcomes and
survey nonresponse. For example, the weights ensure the analytic sample has the correct, representative proportion of women, because women are more likely to experience assault and more likely to respond to the survey. However, directly including too
many predictor variables in the nonresponse weights may actually increase the error in
the estimates. This is because weighting on variables that are not associated with any
survey outcomes cannot remove any nonresponse bias in those outcomes, but it can
increase the imprecision of the estimates. Our approach to nonresponse weighting in
the RMWS is optimized to our key outcomes, allowing us to account for the widest set
of predictive factors, while limiting the deterioration of estimate precision.
We also used a more speculative strategy to reduce the biasing effects of nonresponse—changing the name of the survey to eliminate the concept of “gender relations” from the survey title, and ensuring that communications about the survey
avoided reference to sexual harassment or assault. Instead, letters, emails, public service
announcements, and other communications referred to the survey as concerning safety
and professionalism.1 A concern in prior research on sexual harassment and assault has
been that knowledge of the study’s focus will attract respondents with particular types
of experiences that may not be representative of the experiences of the population at
large (see, e.g., Clancy et al., 2014; Sabol and Beck, 2014). For instance, a survey about
sexual assault may attract the participation of those with sexual assault experiences so
that they can make their experiences known. Alternatively, it could discourage their
participation if they believe the topic will be uncomfortable or violate their privacy.
Either way, this kind of purposive nonparticipation could introduce distorting effects
in the survey estimates. This concern, coupled with our impression that many service
members feel they are trained and surveyed excessively about sexual offenses, led us to
suspect that vague and neutral titles and descriptions for the RMWS could improve
the representativeness of respondents.
Finally, we emphasized with the project title and all survey promotion materials
that the study was being conducted independently of DoD, and that survey responses
would not be shared with DoD in a way that would reveal the identity of any respondent. Furthermore, we advertised the fact that we had obtained a Federal Certificate
of Confidentiality that would provide RAND additional protections against any effort
by DoD or others to compel RAND to release confidential information. Each of these
assurances was designed to appeal to service members who would choose not to participate if they believed that their answers would become known to colleagues at DoD,

Summary of Major Improvements in the RMWS

or would be considered in promotion or fitness reviews. Of course, researchers at DoD
also assure respondents that their answers will be kept in confidence, but we speculated
that the independence of the RMWS could provide an extra layer of confidentiality in
the minds of some members. By encouraging the participation of this group, we sought
to further reduce the distorting effects of nonresponse.
Our Survey Modifications Will Likely Count Different People as Having
Experienced Sex-Based MEO Violations or Sexual Assault
In advance of data collection we could not say whether the RMWS would estimate
a higher or lower percentage of service members who experienced sex-based MEO
violations or sexual assault than would the prior WGRA form. Many of the changes
we made have an uncertain effect on the number of cases that would be counted. For
instance, our new method for assessing lack of consent for the past-year sexual assault
assessment could have eliminated some cases that would have been counted as experiencing unwanted sexual contacts on the WGRA form if some respondents incorrectly
interpret that question to mean that failure to provide verbal consent (even if nonverbal consent was conveyed) was sufficient in unwanted sexual contacts, even those that
would not meet the legal threshold for a crime. On the other hand, some respondents
who answered “no” to the unwanted sexual contact item because they did “agree” to
sex, but under duress due to threats, might now be included as having experienced a
sexual assault.
There are, however, several changes we made that we predicted would increase the
number of reports of past-year sexual crimes and violations. For instance:
• eliminating the “labeling” requirement for sexual harassment and gender discrimination
• improved cuing of respondent memories by using more-explicit descriptions of
these events, and presenting them serially rather than embedded within a single
gating question
• expanding the scope of the sexual assault questions to explicitly include not just
sexually motivated contacts, but also qualifying experiences that were not sexually motivated, but designed to humiliate or degrade the service member (e.g.,
some hazing or bullying experiences).
Other changes we made seemed likely to lower numbers of service members
meeting criteria for sexual harassment and assault than earlier WGRA questions. For
instance:
• counting only those experiences that meet UCMJ legal standards
• reducing overcounting due to telescoping.

Sexual Assault and Sexual Harassment in the U.S. Military

Each of the six changes listed above could have large effects on estimates, but they
work in opposite directions. Therefore, we expected the full extent of the improvements
in survey estimates would not be revealed as a sizable difference between estimates
from the RWMS and WGRA measures. Therefore, similar prevalence estimates from
the old and new versions of the questions would not mean they are counting the same
types of experiences. As we have detailed here, there were many reasons to believe the
new questions would provide precise and valid estimates of the offenses of greatest
concern, and that these measures could therefore be more easily interpreted and communicated to policymakers and the public. As such, we expected the new questions to
perform better for tracking changes in service members’ experiences and the effectiveness of DoD’s prevention and response systems. (Indeed, as discussed above, some of
the old measures, like sexual harassment, may track in the opposite direction of progress; that is, they could show increases when sexual harassment is actually decreasing.)
Study Experiments Will Help Inform Future Administrations of the
WGRA
In addition to the innovations described above, we have implemented several experiments in the RMWS, the results of which could provide valuable guidance for future
administrations of the WGRA. Specifically:
1. The effects of question wording (old versus new) in the number of service members
counted as experiencing sexual harassment and sexual assault. By randomly assigning sampled members to receive either the prior WGRA form or the new forms,
we have ensured that there should be no systematic differences between the
rates of exposure to sexual harassment and assault between members receiving
the different forms. Therefore, any difference in measured rates of exposure can
be attributed to the survey instruments themselves. That is, we can compare
rates of exposure on the prior and new questionnaires—and the types of crimes
reported on each—and draw strong inferences about how the two approaches to
counting service members who have experienced sexual harassment and sexual
assaults differ. This will be helpful for explaining why, for instance, the new
measures may reveal different rates of exposure to these offenses. Moreover, it
will provide a firm basis on which to link prior trend data using the old forms,
to a new set of time trends using the new questionnaire.
2. The effect of survey length on study participation and survey breakoff. In the
RMWS, we randomized a subset of the sample to either the long-, medium-, or
short-form conditions. Recruitment materials described the survey time commitment correspondingly as requiring approximately 20, 12, or 8 minutes. This
allows two tests that will be useful in developing future surveys:

Summary of Major Improvements in the RMWS

a. By comparing rates of starting the survey across conditions, we will infer
the effect of reported length on willingness to participate in these types of
web-based military surveys.
b. By comparing conditions on the rates at which respondents quit the survey
before completing it (breakoff rates), we will shed light on the trade-off in
survey design between length and completion rates for this population.
3. The effectiveness of techniques designed to reduce the effects of response telescoping on
the measurement of sexual offenses. If the many techniques we have used to discourage response telescoping are effective, then on the new form fewer people
whose most recent sexual assault experience occurred more than a year ago will
be counted as having a past-year sexual assault. Because the prior WGRA form
uses none of these techniques, we hypothesize that a larger number of service
members were counted as experiencing past-year unwanted sexual contact than
had such an experience within that specific timeframe.
As an indirect test of this hypothesis, we compare the proportion of people
who describe events that meet criteria for unwanted sexual contact (on the
prior WGRA form) and sexual assault (on the new RMWS form) in the past
year who later indicate that they are certain the event they were describing actually occurred more than a year ago. We can do this because we add a question
about the respondent’s confidence about the timing of the assault at the end of
the sexual assault questions in both the prior WGRA form and the new RMWS
form. This allows us to test the effectiveness of our efforts to limit telescoping.
Specifically, we hypothesized that fewer RMWS form participants would revise
their estimate of the event timing relative to prior WGRA form participants.
Such a difference will provide compelling evidence that the WGRA questionnaire overestimated the number of service members who experienced pastyear sexual assaults, due to the inclusion of cases that actually occurred more
than a year earlier.
4. The effects of survey nonparticipation on sexual harassment and assault estimates. As
noted throughout this report, survey nonparticipation can undermine the validity of survey estimates if those who do and do not participate differ in terms of
their experiences of sexual harassment and assault. One way to clarify whether
survey estimates suffer from this type of bias is to learn more about the experiences of those who have chosen not to participate. For the RMWS, we conducted a follow-up study of this question. Specifically, after the field period of
the main study closed, we conducted an intensive follow-up study of a random
sample of RMWS nonrespondents. In an effort to encourage them to complete
the survey, we offered them alternative modes of administration (paper forms or
telephone interviews) and incentives.
We expected the results of this follow-on study to provide vital information
on whether those who are nonparticipants in the RMWS differ in terms of their

Sexual Assault and Sexual Harassment in the U.S. Military

exposure to sexual harassment and assault from those who participated in ways
that are not captured by our nonresponse weights. If they do, and their rates of
exposure are higher than in the RMWS, this will provide strong evidence that
the RMWS estimates (and likely those of prior WGRA as well) underestimate
the true rates of exposure in the population of service members. If their rates are
lower, then the RMWS and WGRA estimates have likely overestimated rates of
exposure across the military.
5. The effects of alternate approaches to sample weighting. Our main published estimates from the prior WGRA form use sample weighting procedures that match
those used for the 2012 WGRA. This will improve the direct comparability of
the 2012 WGRA and 2014 prior WGRA form estimates. For the new forms,
however, our primary published estimates will use weights optimized for producing unbiased estimates of the key study outcomes and differ in two respects
from the method used for the 2012 WGRA. The new approach accounts for
many more factors that could improve bias reduction, and it limits the degradation in estimate precision that can occur when trying to make the analytic
sample representative of the population on factors that are not associated with
the outcomes. By comparing estimates made using the two sets of weights we
expect to be able to investigate the effects of these analytic decisions, and to
provide guidance for future studies that will improve efforts to limit nonresponse biases.
Timeline for Public Reporting of Study Results
This is the first of a multivolume series of reports on the RMWS. This volume introduces the background, rationale, and methods used to arrive at findings that will be
reported in later volumes. Prior to release of the second volume, RAND will provide
SAPRO and the U.S. Coast Guard top-line estimates of sex-based MEO violations and
sexual assault for inclusion in a report requested by the White House on the progress
DoD has made addressing sexual crimes and violations against service members. That
report is due no later than December 1, 2014. Top-line estimates will include only
estimates of sexual harassment and sexual assault by service and gender, with moredetailed information released in the full RAND report in spring 2015.
The top-line results are likely to generate many questions about differences in
estimates produced by the prior WGRA form and the new questionnaire, about the
comparability of the 2014 findings and those from prior years, and about the relative
merits of the different approaches to measurement and survey weighting. The RAND
team will analyze these and other topics through the winter of 2014–2015, and we will

Summary of Major Improvements in the RMWS

provide these more-detailed analyses, along with public reports on the main findings,
in the late spring of 2015. These reports will include findings on
• percentage of service members experiencing sexual assault, sexual harassment,
and gender discrimination in the past year among
–– DoD active-duty and reserve-component members
–– Coast Guard active-duty and reserve-component members
• characteristics of sex-based MEO violations and sexual assault, such as where and
when an event occurred, who harassed or assaulted the member, and whether the
event was reported
• perceived barriers to sexual assault and harassment reporting
• service members’ experiences with support and prosecution systems available to
those who report sexual assaults or sex-based MEO violations
• results from a detailed investigation into survey nonresponse and possible nonresponse bias
• recommendations for future surveys of sexual harassment and sexual assault in
the military.

APPENDIX A

Definitions of Measurement Constructs

2014 RMWS Constructs
Gender discrimination. Individuals are counted as having experienced gender discrimination in the prior year if they respond “yes” to:
• either of the 2 screener items for gender discrimination, and
• a follow-up question indicating that the discriminatory behavior harmed or limited their career.
Problematic workplace behaviors. Individuals are counted as having experienced
problematic workplace behaviors in the prior year if they respond “yes” to any of the
15 screener items on the RMWS sex-based EEO violation module, regardless of their
responses to the follow-up questions.
RMWS sexual harassment. Individuals are counted as having experienced sexual
harassment in the prior year if they were classified as experiencing either of these two
types of EEO violation:
• Sexually hostile work environment, or
• Sexual quid pro quo.
Sex-based MEO violation. Individuals are counted as having experienced a sex-based
MEO violation in the prior year if they were classified as experiencing any of these
violations:
• Sexually hostile work environment
• Sexual quid pro quo), or
• Gender discrimination.

Sexual Assault and Sexual Harassment in the U.S. Military

Sexual assault. Individuals are counted as having experienced a sexual assault in the
prior year if they:
• responded “yes” to any of the 6 screener items for unwanted sexual experiences, and
• indicated the experience was intended to be either (a) abusive/humiliating or (b)
for the offender’s sexual gratification, and
• responded “yes” to at least one of the offender’s behavior questions indicating a
type of coercion consistent with a sex crime under UCMJ Article 120.encounter
This measure of Sexual Assault is divided into three mutually exclusive types
depending on which initial screener question led to classification:
–– Penetrative sexual assault. Individuals are counted as having experienced
penetrative sexual contact in the prior year if they:
◦◦ Were counted as having experienced a sexual assault on the basis of any of
the first three screener items that assess penetration of the vagina, anus or
mouth.
–– Non-penetrative sexual assault. Individuals are counted as having experienced non-penetrative sexual assault in the prior year if they:
◦◦ Were not previously counted as having experienced a penetrative sexual
assault
◦◦ Were counted as having experienced a sexual assault on the basis of either of
the two screener items that assess unwanted contact with their private parts
(#4 or #5 above)
–– Attempted penetrative sexual assault. Individuals are counted as having
experienced an attempted penetrative assault in the prior year if they
◦◦ Were not previously counted as having experienced either a penetrative or
non-penetrative sexual assault.
◦◦ Were counted as having experienced an attempted sexual assault on the basis
of the screener item assessing attempted penetration (#6 above).
Sexual quid pro quo. Individuals are counted as having experienced a sexual quid pro
quo (a workplace benefit or punishment that was contingent on sexual behavior) in the
prior year if they respond “yes” to:
• either of the 2 screener items for sexual quid pro quo, and
• follow-up probes indicating that the quid pro quo was directly communicated to
them.
Sexually hostile work environment. Individuals are counted as having experienced a
sexually hostile work environment in the prior year if they respond “yes” to:
• any of the 11 screener items for sexually hostile work environment, and

Definitions of Measurement Constructs 79

• follow-up probes indicating that at least one of these experiences either:
–– persisted after the offender knew that someone wanted the behavior to stop, or
–– was sufficiently severe that a reasonable person would find it offensive.
Unwanted sexual experience. Individuals are counted as having had an unwanted
sexual experience in the prior year if they responded “yes” to any of 6 screener questions:
1. Did you have any unwanted experiences in which someone put his penis into
your [if female, “vagina,”] anus or mouth?
2. Did you have any unwanted experiences in which someone put any object or
any body part other than a penis into your [if female, “vagina,”] anus or mouth?
The body part could include a finger, tongue or testicles.
3. Did anyone make you put any part of your body or any object into someone’s
mouth, vagina, or anus when you did not want to? A part of the body could
include your [if male, “penis, testicles,”] tongue or fingers.
4. Did you have any unwanted experiences in which someone intentionally touched
private areas of your body (either directly or through clothing)? Private areas
include buttocks, inner thigh, breasts, groin, anus, vagina, penis, or testicles.
5. Did you have any unwanted experiences in which someone made you touch private areas of their body or someone else’s body (either directly or through clothing)? This could involve the person putting their private areas on you. Private
areas include buttocks, inner thigh, breasts, groin, anus, vagina, penis or testicles
6. Did you have any unwanted experiences in which someone attempted to put
a penis, an object, or any body part into your [if female, “vagina,”] anus or
mouth, but no penetration actually occurred?

Prior WGRA Form Constructs
Unwanted sexual contact. Individuals are counted as having experienced unwanted
sexual contact in the prior year if they respond “yes” to the single unwanted sexual
contact item on the prior WGRA form, exactly as presented and coded on the 2012
WGRA. Those who qualify as experiencing an unwanted sexual contact were asked
additional questions to assess the type of behavior that occurred as part of the “one
event that had the greatest effect on you.” This was used to divide these individuals into
four mutually exclusive types:
• Completed sexual intercourse, anal or oral sex: Individuals are counted as
having experienced completed sexual intercourse, anal or oral sex if they answered
“did this” to one of the following two behaviors with regard to the “one event”:

Sexual Assault and Sexual Harassment in the U.S. Military

–– Made you have sexual intercourse
–– Made you perform or receive oral sex, anal sex, or penetration by a finger or
object.
• Attempted sexual intercourse, anal or oral sex Individuals are counted as
having experienced attempted sexual intercourse, anal or oral sex if they:
–– Had not been previously classified as having a “completed” event, and
–– Answered “did this” to one of the following two behaviors with regard to the
“one event”:
◦◦ Attempted to make you have sexual intercourse but was not successful
◦◦ Attempted to make you perform or receive oral sex, anal sex, or penetration
by a finger or object, but was not successful.
• Unwanted sexual touching only: Individuals are counted as having experienced
unwanted sexual touching if they:
–– Had not been previously classified as having a “completed” or “attempted”
event, and
–– Answered “did this” to “Sexually touched you (e.g., intentional touching of
genitalia, breasts, or buttocks) or made you sexually touch them.”
• Specific behaviors not specified/none of the above: Individuals who were classified as experiencing an unwanted sexual contact, but who did not indicate any
of the listed behaviors for the “one event that had the greatest effect on them”
were counted as “specific behaviors not specified.” This includes a mix of respondents who answered “did not happen” to all five behaviors, as well as those who
skipped these questions.
WGRA sexual harassment. Individuals are counted as having experienced sexual
harassment in the prior year if they:
• Indicated the occurrence of any item on the Sexual Experiences Questionnaire from the following subscales (a response of “once or twice,” “sometimes,”
“often,”or “very often”):
–– crude/offensive behaviors, or
–– unwanted sexual attention, or
–– sexual coercion (not including unwanted touching of private parts or unwanted
sexual intercourse)
• Indicated on a follow-up item in regard to these indicated items that either: “some
were sexual harassment; some were not sexual harassment” or “all were sexual
harassment.”

APPENDIX B

Correspondence Between the RMWS Measure of Sexual
Assault and Title 10 USC § 920 (UCMJ Article 120)
Terry L. Schell, Coreen Farris, Lisa H. Jaycox,
Dean G. Kilpatrick, Amy E. Street

One of the primary goals of the RAND revision to the DoD WGRS was to more
closely align the survey assessment of sexual assault with the legal definitions of these
crimes in the UCMJ. These definitions are set out by Congress in Title 10 USC §
920, which is Article 120 in the Military’s Manual for Courts-Martial (MCM; DoD,
2012). This MCM Article, titled “Rape and sexual assault generally,” was substantially
revised, with the new definitions going into effect on June 28, 2012 (see Clark, 2012,
for a review of the changes).
Overview of the Article 120 Classification of Sexual Assault
To qualify as a crime under 2012 Article 120, the incident must be classified as either a
Sexual Act or a Sexual Contact. To be classified as a Sexual Act or Sexual Contact, it must
meet two criteria: one based on the characteristics of the physical contact and another
based on the intent of the offender. In addition, the act/contact had to occur by means
of one of the specified types of coercion to be classified as a crime. Thus, an assessment
of sexual assault under this statute has three components: (1) Does the type of physical contact meet the definition of a Sexual Act or Sexual Contact? (2) Did the intent of
offender meet the definition of either a Sexual Act or Sexual Contact? and (3) Did the
offender compel or coerce the contact by one of the specified means or mechanisms?
Does the Physical Contact Meet the Definition of a Sexual Act or Sexual Contact?

A physical contact is deemed a Sexual Act if it involves the penetration of the vulva,
anus, or mouth by any body part or any object. Sexual Contact is a more general category that involves either (a) contact of any specified private body part, even through
clothing; or (b) any physical contact, if it was done for a sexual (not just an abusive)
purpose.

Sexual Assault and Sexual Harassment in the U.S. Military

Table B.1
Types of Sexual Assault Under Article 120
Method Used to Compel

Contact Involving Penetration

Contact Without Penetration

Physical force or a high level of
coercion

Article 120(a): Rape

Article 120(c): Aggravated Sexual
Contact

Lesser types of coercion or
non-consent

Article 120(b): Sexual Assault

Article 120(d): Abusive Sexual
Contact

Did the Intent of Offender Meet the Definition of Either a Sexual Act or Sexual
Contact?

The intent criterion is met if the act was done either with intent (a) “to abuse, humiliate, harass, or degrade any person” or (b) “to arouse or gratify the sexual desire of any
person.” The first element allows for incidents that are not overtly sexual (e.g., as part of
hazing or punishment) to qualify for the same “sexual assault” charge as if it had been
done for a sexual purpose. The intent of the offender is determined as a finding during
court martial, unless the act involved penetration with a penis. Such acts are defined as
sex acts without additional intent findings.
Did the Offender Achieve the Contact by One of the Specified Means or
Mechanisms?

Behaviors by which the offender compels the assault or contact to happen are defined
in two levels, with a higher-level offense occurring when physical force, threat of physical force, or forced intoxication was used. A lesser offense occurs when the offender
used other types of threats, misrepresentations, impairments, or bodily harm. Bodily
harm, in the revision to Article 120, is now a category that includes any contact without consent. This definitional structure makes it possible to prosecute many sexual
assaults without requiring any testimony about whether consent was given (e.g., if force
was used). Evidence of objective defendant behavior is sufficient in most cases. However, an offender could still be charged on the basis of testimony about a lack of victim
consent in those cases in which the prosecution lacks evidence of the other behaviors
specified in Article 120.
RAND Behavioral Screening Questions for Sexual Assault Generally
One of the challenges of assessing sexual assault in a survey of victims is that it encompasses a wide range of specific behaviors, and these different behaviors are not universally known by a common set of labels. The measurement approach used in the RMWS
is consistent with recommendations outlined in a recent National Research Council
report (2014). To ensure that respondents attempt to recall the full range of experiences that might constitute sexual assault, they are first asked six behaviorally specific

Correspondence Between the RMWS Measure of Sexual Assault and Title 10 USC § 920

screening questions that define a range of unwanted experiences based on the specific
body parts involved. These experiences may or may not be rapes or sexual assaults, but
affirmative responses are followed by a series of questions to assess other required elements of the definition of rapes or sexual assault under Article 120. Detailed discussion
of these screening questions is provided in Table B.2.
Table B.2
Relation of Behavioral Screening Questions to Article 120
Behavioral Screening Question

Justification

1. Since [X date], did you have any unwanted
experiences in which someone put his penis into
your [vagina,] anus or mouth?

This is defined as a Sexual Act under Article 120(g)
(1)(A). Vulva in the code is replaced with “vagina”
to improve comprehension. This type of act is
separated out from other penetration (Item 2)
because penile penetration is defined as a Sexual
Act, and therefore, does not require additional
intent questions.

2. Since [X date], did you have any unwanted
experiences in which someone put any object
or any body part other than a penis into your
[vagina,] anus, or mouth? The body part could
include a finger, tongue, or testicles.

This is defined as a behavior that might qualify as a
Sexual Act under Article 120(g)(1)(B) provided the
intent was abusive/degrading/humiliating/harassing
or sexual. The specific body part examples do not
appear in the code, but are given to better cue
respondents’ memories of events.

3. Since [X date], did anyone make you put any
part of the body or any object into someone’s
mouth, vagina, or anus when you did not want
to? A part of the body could include your [penis,
testicles,] tongue, or fingers.

This is defined as a behavior that might qualify as a
Sexual Act under Article 120 (g)(1)(A) or (B). Article
120 does not distinguish whether the victim is being
penetrated or being made to penetrate, so long as
the act meets the other criteria for sexual assault
under the code. The instrument separates out these
experiences to better cue respondents’ memories.

4. Since [X date], did you have any unwanted
experiences in which someone intentionally
touched private areas of your body (either directly
or through clothing)? Private areas include
buttocks, inner thigh, breasts, groin, anus, vagina,
penis, or testicles.

This is defined as a behavior that might qualify as
a Sexual Contact under Article 120 (g)(2)(A) or (B).
The term “genitalia” in the code has been replaced
with “vagina, penis, or testicles” for improved
comprehension. To narrow the scope of what needs
to be recalled, we restrict the question to events
that are “intentional.” However, the detailed
assessment of intent is assessed in follow-up
questions

5. Since [X date], did you have any unwanted
experiences in which someone made you touch
private areas of their body or someone else’s body
(either directly or through clothing). This could
involve the person putting their private areas on
you. Private areas include buttocks, inner thigh,
breasts, groin, anus, vagina, penis, or testicles.

This is defined as a behavior that might qualify as
a Sexual Contact under Article 120 (g)(2)(A) or (B),
and is parallel to question 4. Article 120 includes
both “touching” and “causing another person to
touch”; however, these are asked as two separate
screening questions to better cue respondents’
memories of events.

6. Since [X date], did you have any unwanted
experiences in which someone attempted to put
a penis, an object, or any body part into your
[vagina,] anus or mouth, but no penetration
actually occurred?
If Yes, then:
6a. As part of this attempt, did the person touch
you anywhere on your body? This includes
grabbing your arm, hair or clothes, or pushing
their body against yours.

The second question of this series, Item 6a, is a
behavior that might qualify as a Sexual Contact
under Article 120(g)(2)(B). Even if the contact is
not with private body parts (which would have
been captured earlier in Item 4 and Item 5), if that
contact is being made as part of an attempted
sexual act, the survey will classify this as a possible
Sexual Contact.

Sexual Assault and Sexual Harassment in the U.S. Military

There are two discrepancies between the behavioral acts referred to in Article 120
and the acts captured by these screening questions. The largest discrepancy is that we
restrict these behavioral screening questions to “unwanted experiences.” Article 120
makes no such restriction and explicitly classifies certain acts as sexual assault regardless of whether they were wanted (see, for example, Article 120[g][8][B]). The RAND
view is that these “wanted” sexual acts are not a significant policy concern, even if
they might be technically illegal, and such “wanted” acts are not likely to meet criteria
for a sexual assault based on subsequent questions about offender intent and coercive
offender behaviors. Thus, there is no significant loss by excluding these experiences
in the screening questions. On the other hand, restricting the memory cues in the
behavioral screening questions to “unwanted experiences” reduces the personal experiences to be recalled to a cognitively manageable number so that respondents can better
answer the detailed follow-up questions.
Secondly, the screening questions do not attempt to comprehensively assess a new
type of Sexual Contact that was introduced in the 2012 version of the code. Specifically,
contact for a sexual purpose that does not involve the designated private body areas
(see Article 120[g][2][B]). This instrument only counts such instances if they occurred
as part of an attempted penetrative Sexual Act. Thus the instrument may miss some
unusual types of sexual assaults (e.g., sexual practices involving only those body parts
that are not usually seen as private areas). RAND has omitted this class because such
behaviors cannot be measured without a highly detailed and lengthy series of questions.
Classification of Attempts
In addition to assessing sex crimes under Article 120, the instrument assesses Attempted
Penetrative Sexual Assault. These are unsuccessful attempts to commit a rape or sexual
assault (as defined in Article 120[a] or Article 120[b]). The assessment first tries to
classify such incidents as sexual assaults under Article 120, which is possible if they
involved any physical contact with the respondent. The category of Attempted Penetrative Sexual Assault is reserved for only those incidents that could not be classified more
concretely as Abusive Sexual Contact under Article 120. Such attempts might be crimes
under Article 80; however, it is difficult to determine if they include all the elements
of this crime given the lack of concrete behaviors demonstrating specific intent. Article
80(a) defines an attempt as “specific intent to commit an offense under this chapter,
amounting to more than mere preparation . . .” The instrument counts instances as
an Attempted Penetrative Sexual Assault when respondents answer “yes” to screening
question 6 and “no” to 6a, so long as the respondent then confirms in the follow up
questions that the offender’s intended behavior would have met the criteria for rape
or sexual assault. The instrument does not ensure that the respondents’ definitions
of “attempt” are consistent with the definition in Article 80. Thus, it is possible that

Correspondence Between the RMWS Measure of Sexual Assault and Title 10 USC § 920

some of these experiences would not count as “more than mere preparation” on further
investigation. However, our view is that it is unlikely that a respondent would answer
affirmatively to this question unless the offender exhibited some specific verbal or nonverbal behavior that indicated intent to commit these acts.
The instrument does not assess Attempted Aggravated or Abusive Sexual Contact
although such acts might also be crimes under Article 80 (for example, unsuccessful
attempts to touch genitals). It seems likely that most concerted attempts to inappropriately touch a service member’s body would result in some type of physical contact
that could be charged directly under Article 120, and so the number of incidents that
qualify only as attempts is likely small. In addition, it is difficult to define these acts
behaviorally, given the range of actions involved. It is also difficult to assess whether
they rise to the level of an attempt under Article 80. For these reasons, the instrument
omits assessments of these attempts.
Assessment of the Intent of the Contact
After the respondent indicates a type of unwanted experience that might qualify as
either a Sexual Act or a Sexual Contact under Article 120, the respondent is asked
whether the behavior was consistent with the intent requirement under Article 120(g)
(1) or (2). Specifically, respondents are first asked: “Was this unwanted experience (or
any experiences like this if you had more than one) abusive or humiliating, or intended
to be abusive or humiliating? If you aren’t sure, choose the best answer.” If the respondents answer “yes” to that question, they are classified as meeting the intent requirement. If they answer “no,” they are then asked: “Do you believe the person did it for
a sexual reason? For example, they did it because they were sexually aroused or to get
sexually aroused. If you aren’t sure, choose the best answer.” If they then answer “yes”
to that question, they are classified as meeting the intent requirement. If they answer
“no,” the incident is classified as not meeting the intent requirement under Article
120(g)(1) or (2). The latter question about “a sexual reason” is very closely aligned with
the wording of the code, although the code’s wording “intent to arouse or gratify the
sexual desire of any person” has been simplified for better comprehension. The question about “abusive or humiliating” has been modified in two ways from the criteria in
the code on the basis of cognitive testing that indicated problems interpreting the legal
wording. The code’s wording “abuse, humiliate, harass, or degrade any person” has
been simplified to “abusive or humiliating.” In addition, participants’ experience of the
act as being “abusive or humiliating” is sufficient for classification. Thus the respondent
indicates that either the perpetrator intended to be “abusive or humiliating” or this
intent is inferred because their behavior was experienced as “abusive or humiliating.”
The RAND team believes this is a reasonable basis on which to infer intent, particularly combined with the fact that screening items 4 and 5 have already ruled out non-

Sexual Assault and Sexual Harassment in the U.S. Military

penetrative contact that was accidental. (Note: Consistent with Article 120[g][1][A],
sexual intent is inferred and does not need to be assessed if the act involved penetration
by a penis, i.e., screening Item 1).
Assessment of the Means Used by the Offender to Compel the Victim
On the basis of the behavioral screening questions and the offender intent questions,
these experiences may be classified as either a Sexual Act or a Sexual Contact under
Article 120. We refer to these legal categories as Penetrative and Nonpenetrative, to
better explain this distinction to lay readers. For these acts/contacts to rise to the level
of a crime, the offender must also have used one of the specified means or mechanisms
to compel the act or contact. To assess this, respondents are asked to indicate which of
the items in Table B.3 occurred during the incident.
If the respondent indicates that none of the offender behaviors A–H were present
in the incident, they are asked about three additional situations that might describe the
incident (Table B.4). The first two of these incidents are likely to meet the criteria of a
crime under Article 120, but are less clear-cut than the criteria embodied by questions
A–H above.
The final item delivered to respondents who did not indicate that they experienced items A–H was placed last in the survey to catch any instances of nonconsent
that were not captured in the earlier items (see Table B.5). It was explicitly included
to capture instances where an event happened so suddenly that explicit refusals (as in
offender behavioral item A) were not possible and threats or force was not used. This
could occur, for example, with a sudden groping of genitals that would not be well
described in offender behaviors A through J.
The set of options (A–K) describing the means by which the offender achieved the
sexual act or sexual contact omits one category that is included in Article 120, specifically Article 120(b)(3)(B): “commits a sexual act upon another person when the other
person is incapable of consenting to the sexual act due to a mental disease or defect, or
physical disability, and that condition is known or reasonably should be known by the
person.” This was omitted as a separate category because it is thought to be rare among
currently serving U.S. service members, and it is difficult to ask the respondent to apply
this criterion to him or herself.
The offender behavior criteria described in items I and J do not align as closely
to the Article 120 language as the other criteria (A–H and K) but capture additional
situations that are likely to represent sexual assaults. More generally, they are often
included in the legal definitions of rape/sexual assault/sexual battery/etc. in other jurisdictions. While the current instrument attempts to carefully replicate the Article 120
legal criteria, a substantial fraction of sexual assaults of service members cannot be
prosecuted under the UCMJ because the offender is not under military jurisdiction.
Those cases must be charged under whatever code applies in that local jurisdiction.

Correspondence Between the RMWS Measure of Sexual Assault and Title 10 USC § 920

Table B.3
Relation of Offender Behavior Questions to Article 120
Offender Behavior Question

Justification

a. They continued even
when you told them or
showed them that you were
unwilling.

Indicative of bodily harm under Article 120(g)(3): “any offensive
touching of another, however slight, including any nonconsensual
sexual act or nonconsensual sexual contact.” Qualifies as meeting
criteria under Article 120(b)(1)(B).

b. They used physical force
to make you comply. For
example, they grabbed
your arm or used their body
weight to hold you down.

Indicative of force under Article 120(g)(5)(B): “the use of such physical
strength or violence as is sufficient to overcome, restrain, or injure a
person.” Force is assumed to be unlawful, meeting criteria under Article
120(a)(1)

c. They physically injured you. Indicative of force under Article 120(g)(5)(B) or (C): “inflicting physical
harm sufficient to coerce or compel submission by the victim.” Force is
assumed to be either unlawful, meeting criteria under Article 120(a)(1);
or serious, meeting criteria under Article 120(a)(2).
d. They threatened to
physically hurt you (or
someone else).

Indicative of a threat or placing in fear under Article 120(g)(7). Note that
due to Article 120(e) we do not need evidence that the offender had the
ability to carry out the threat. By itself, it qualifies under Article 120(b)
(1)(A), but it also indicates the possibility of coercion that is serious
enough to qualify as rape, which is assessed in two follow-up questions:

If “yes” to d, then ask:
d1. Did they threaten you
(or someone else) with a
weapon?

Indicative of force under Article 120(g)(5)(A): “the use of a weapon.”
Force is assumed to be unlawful, meeting criteria under Article 120(a)(1)

If “yes” to d, then ask:
Indicative of a threat or placing in fear that rises to the level of Article
d2. Did they threaten to
120(a)(3): “fear that any person will be subjected to death, grievous
seriously injure, kill, or kidnap bodily harm, or kidnapping.”
you (or someone else)?
e. They threatened you (or
someone else) in some other
way? For example, by using
their position of authority,
by spreading lies about you,
or by getting you in trouble
with authorities.

Indicative of a threat or placing in fear under Article 120(g)(7): “a
communication or action that is of sufficient consequence to cause
a reasonable fear that non-compliance will result in the victim or
another person being subjected to the wrongful action contemplated
by the communication or action.” The examples are included to cue
memories about the types of wrongful actions that might have been
contemplated. This qualifies under criteria Article 120(b)(1)(B). Note that
due to Article 120(e) we do not need evidence that the offender had the
ability to carry out the threat.

f. They did it when you
were passed out, asleep, or
unconscious.

Indicative of either rendering unconscious under Article 120(a)(4),
or asleep/unconscious/unaware under Article 120(b)(2). Note that
this question does not distinguish whether the offender caused the
unconsciousness. Thus, these incidents cannot be divided into Rape/
Aggravated Contact versus Sexual Assault/Abusive Contact. However,
both subcategories are sexual assaults under Article 120.

g. They did it when you
were so drunk, high or
drugged that you could
not understand what was
happening or could not
show them that you were
unwilling.

Indicative of either coerced intoxication under Article 120(a)(5), or
incapable of consent under Article 120(g)(3)(A): “the other person is
incapable of consenting to the sexual act due to impairment by any
drug, intoxicant, or other similar substance, and that condition is known
or reasonably should be known by the person.” Note that this question
does not distinguish whether the offender caused the unconsciousness.
Thus, these incidents cannot be divided into Rape/Aggravated Contact
versus Sexual Assault/Abusive Contact. However, both subcategories are
sexual assaults under Article 120.

Sexual Assault and Sexual Harassment in the U.S. Military

Table B.3—Continued
Offender Behavior Question

Justification

h. They tricked you into
thinking that they were
someone else or that they
were allowed to do it for a
professional purpose (like a
person pretending to be a
doctor).

Indicative of either fraudulent representation under Article 120(b)(1)
(C): “fraudulent representation that the sexual act serves a professional
purpose”; or mistaken identity under Article 120(b)(1)(D): “a belief
by any artifice, pretense, or concealment that the person is another
person”

Table B.4
Relation of Additional Offender Behavior Questions I and J to Article 120
Additional Offender Behavior
Question

Justification

i. They made you so afraid
that you froze and could not
tell them or show them that
you were unwilling.

This item captures tonic immobility, a psychological phenomenon that
results in temporary fear-based impairment and disability. It does not
correspond directly to a particular criterion under the code. Depending
on the circumstances it might be charged under:
• threat or placing in fear under Article 120(g)(7). This could be
done under a theory that the respondent’s extreme fear is evidence of offender behavior that constituted an implicit threat.
• incapable of consent under Article 120(b)(3)(B): “commits a sexual
act upon another person when the other person is incapable of
consenting to the sexual act due to a mental disease or defect,
or physical disability, and that condition is known or reasonably
should be known by the person.”
• bodily harm under Article 120(g)(3): “any offensive touching of
another, however slight, including any nonconsensual sexual act
or nonconsensual sexual contact.”

j. They did it after you had
consumed so much alcohol
that the next day you
could not remember what
happened.

This question addresses problems inherent is assessing sexual assault
entirely from victim’s memory when alcohol is instrumental to the
assault. Even in instances in which a crime under Article 120 clearly
occurred (e.g., there is strong physical evidence or witnesses), it is
possible that the victim will not remember the events sufficiently to
indicate that they experienced items F and G above. Such events may
be common, and excluding them could undercount the true number of
sexual assault victims.
The existence of a “blackout” due to drinking is, itself, strong evidence
of substantial impairment due to intoxication. Alcohol research suggests
that blackouts typically occur after approximately 10 alcoholic drinks,
and with blood alcohol content (BAC) of 0.16% or greater (Goodwin
et al., 1970; White et al., 2004). Individuals in a blackout state typically
can remember only the prior 3–5 minutes. This level of intoxication
routinely causes obvious speech impediments, impairments in gross
motor function, loss of balance, disorientation, nausea/vomiting, and is
twice the legal limit for motor vehicle operation. Finally, blackouts are
made much more common when alcohol is combined with a “date-rape
drug” (e.g., Rohypnal, Valium; see White, Simson, and Best, 1997; Saum
and Inciardi, 1997). The victim may commonly experience a blackout in
rapes under those circumstances.
Because of the high level of alcohol required, and the potential for the
use of other drugs, the existence of the blackout is treated as evidence
that the act meets criteria under Article 120(b)(3)(A): “when the other
person is incapable of consenting to the sexual act due to impairment
by any drug, intoxicant, or other similar substance, and that condition is
known or reasonably should be known by the person.”

Correspondence Between the RMWS Measure of Sexual Assault and Title 10 USC § 920

Table B.5
Relation of Additional Offender Behavior Question K to Article 120
Final Offender Behavior
Question
k. It happened without your
consent.

Justification
While the text of Article 120 has been revised to minimize the need
to classify incidents as crimes based entirely on victim consent, it
continues to classify all nonconsensual sexual acts or sexual contacts as
crimes under Article 120. This is done under Article 120(b)(1)(B) via the
definition of bodily harm provided in Article 120(g)(3): “any offensive
touching of another, however slight, including any nonconsensual
sexual act or nonconsensual sexual contact.” Thus even when the
respondent cannot indicate that they experienced items A–H above,
Article 120 classifies the Sexual Act or Sexual Contact as a crime if it was
nonconsensual.

In addition, prosecutors in the military justice system can incorporate local or federal
statutes into their charges to include charges not articulated in Article 120. Thus, while
our classification system is primarily concerned with capturing crimes under Article
120, it should not be exclusively concerned with those criteria since that is not always
a legal standard that could be applied to these incidents.
Scoring for Estimation of Prevalence
To be counted as type of sexual assault in the survey, respondents must have responded
“yes” to at least one behavioral screening question that indicates that the act was
unwanted, have indicated that this act (or acts) was done with abusive or sexual intent,
and have indicated that the incident(s) involved at least one element of coercion consistent with Article 120. When the instrument is scored, respondents are placed into
three mutually exclusive categories based on the events that they experienced: (1) Penetrative Sexual Assault, designed to correspond to rape or sexual assault under Article
120; (2) Sexual Assault Without Penetration, designed to correspond to aggravated or
abusive sexual contact under Article 120; and (3) Attempted Penetrative Sexual Assault,
designed to assess attempted rape or sexual assault under Article 80 when the person
could not be classified as experiencing a sexual contact in Article 120. Thus individuals are first placed into a category under Article 120 when the behaviors can be clearly
classified as meeting the legal definitions for rape or sexual assault. Otherwise, the
experiences are classified under the less-concrete terms of Article 80.
The instrument is not designed to separate rape from sexual assault, or aggravated from abusive sexual contact. While the instrument does contain the information
required to make these severity distinctions in most cases, in other instances these
distinctions require fine-grained information that is difficult to assess. This includes
details about the nature of the threat, the severity of the injury (e.g., Article 120[g][4])],
or the role of the offender in facilitating intoxication.

Notes

Chapter One: The 2014 RAND Military Workplace Study
1

The findings and recommendations concerning military surveys of sexual harassment and assault
were released by the Comparative Systems Subcommittee of the Response Systems Panel in draft or
interim form on or around May 5, 2014. That was too late in our development of the RMWS for us
to incorporate all of the Panel’s suggestions. Nevertheless, nearly all of the recommendations made by
the Panel accorded with our plans and decisions for improving the measurement of sexual harassment
and assault in the military.

Chapter Two: Measurement of Sexual Harassment and Sexual Assault
1 Sexual

harassment of men was not included in the meta-analysis.

2 Bounded

interviews refer to longitudinal surveys with at least two assessments when the respondent
is asked about events that happened since the last assessment. Chapter Four includes a more complete
discussion of bounded interviews and the influence of telescoping on crime estimates.

3 To evaluate the reading level of the unwanted sexual contact question we formatted it as the following paragraph:

In the past 12 months, have you experienced any of the following intentional sexual contacts that
were against your will or occurred when you did not or could not consent where someone: sexually
touched you (e.g., intentional touching of genitalia, breasts, or buttocks) or made you sexually touch
them; attempted to make you have sexual intercourse, but was not successful; made you have sexual
intercourse; attempted to make you perform or receive oral sex, anal sex, or penetration by a finger
or object, but was not successful; made you perform or receive oral sex, anal sex, or penetration by
a finger or object?

We then submitted this paragraph to five online readability assessment tools, each of which provide
six measures of readability. This produced results suggesting the minimum reading level was at a 12th
grade level, though most scores were much higher:
Calculator 1

Calculator 2

Calculator 3

Calculator 4

Calculator 5

Average

Flesch Kincaid
Reading Ease

−35

−41

−26

−33

−34

Flesch Kincaid
Grade Level

Sexual Assault and Sexual Harassment in the U.S. Military
Calculator 1

Calculator 2

Calculator 3

Calculator 4

Calculator 5

Average

Gunning Fog
Score

SMOG Index

Coleman Liau
Index

Automated
Readability Index

Calculator 1

http://read-able.com/

Calculator 2

http://www.readabilityformulas.com/free-readability-formula-tests.php

Calculator 3

http://www.online-utility.org/english/readability_test_and_improve.jsp

Calculator 4

http://www.editcentral.com/gwt1/EditCentral.html

Calculator 5

https://readability-score.com/

NOTE: For Calculator 4, we replaced semicolons with commas to indicate clauses are not complete
sentences

Arguably the formatting of the original question aids readability, but this is lost when the entire
question is formatted as a long paragraph. To establish a minimum bound for the readability of the
unwanted sexual contact question, we considered just the first complete sentence. That is, we included
just the stem portion of the question, and the first optional response:
In the past 12 months, have you experienced any of the following intentional sexual contacts that
were against your will or occurred when you did not or could not consent where someone: sexually touched you (e.g., intentional touching of genitalia, breasts, or buttocks) or made you sexually
touch them?

The average reading levels improve, but remain at a 12th-grade level for the lowest estimate, and
significantly higher for most others:

Calculator 1

Calculator 2

Calculator 3

Calculator 4

Calculator 5

Average

Flesch Kincaid
Reading Ease

Flesch Kincaid
Grade Level

Gunning Fog
Score

SMOG Index

Coleman Liau
Index

Automated
Readability Index

4 The

nonresponse weighting methods used in this study are discussed in Chapter Five of this report.

Notes

Chapter Three: Study Design
1

Coast Guard active-duty service members were included in the active-duty sampling frame and
sampled as a separate stratum. Results for the Coast Guard samples (active duty and reserves), will,
however, be analyzed separately from the results of the four DoD services.
2 Those with fewer than six months of service have historically been excluded from WGRA surveys
for logistical and substantive reasons. In terms of survey logistics, the development of a sample frame
and survey fielding historically have taken several months, so it has not been possible to enter the field
pursuing a sample that has fewer than several months of service. In addition, those still in basic training or transitioning to their first assigned units are difficult to reach, as their addresses and even email
addresses are likely to have changed between the time the sample is drawn and the field date of the
survey. Substantively, those with less than six months of service can provide only a partial estimate for
the main past-year measures in the WGRA. Alternative sampling and survey methods would need to
be employed to get accurate population estimates of newer service members.

General and flag officers have been excluded in the past (and in the RMWS) because, as the leaders and decisionmakers in the services, their experience is not expected to be comparable to others, yet
their numbers are too small to satisfactorily analyze separately.
3 Approximately

5,200 enlisted active-duty personnel had missing entry dates, representing 0.45 percent of the active-duty enlisted records. For these records, we assume that any member with a paygrade
of E2 or above meets the inclusion criterion of six months in the service, and anyone with a paygrade
of E1 does not. There were 1,554 records for E1 service members excluded through this procedure.
For officers, we used commission dates to determine length of service because, as with the WGRA,
we did not intend to include active-duty members in service academies, as they are subject to a separate SAPRO workplace gender relations survey. Approximately 1,600 officers had missing commission dates, representing 0.72 percent of all active-duty officer records. For these records, we assumed
that any officer with a paygrade of O2 or above met the six months of service inclusion criterion. For
the 839 officers with paygrades below O2 and no commission date, we had no means of determining
whether they met the six-month inclusion criterion. Therefore, we include this small number of officers
in our sample frame.

Coast Guard reservists were included in the reserve sampling frame and sampled within their own
strata. Results for the Coast Guard samples (active duty and reserves), will, however, be analyzed separately from the results of the other four services.

5 Time in service used reserve entry date to calculate the six-month period. About 2 percent of the
population was missing a reserve entry date. When drawn into our sample, these cases were assumed
to meet the six-month service criterion.
6

Even when including the entire Coast Guard reserve sample, prevalence estimates of sexual assault
will be imprecise because of the low numbers of assaults likely to be detected. For instance, assuming Coast Guard reserves have rates of assault comparable to the rates of unwanted sexual contact
experienced by reserve-component members of the other services in 2012 (2.8 percent for women and
0.5 percent for men; DMDC, 2013c), then even if the entire Coast Guard reserve is sampled and they
have a 40-percent response rate, we would expect to detect only about 17 assaults against women and
12 assaults against men.

7 For

the prior WGRA form condition, we originally proposed that a smaller sample would suffice to
establish whether sexual assault and sexual harassment prevalence rates differed on the new form from
the unwanted sexual contact and sexual harassment rates from the prior WGRA form. Some members of Congress expressed concern that the RAND survey would not permit sufficient comparisons
of prior WGRA results with 2014 results. This increased the importance of ensuring a sufficient prior

Sexual Assault and Sexual Harassment in the U.S. Military

WGRA form sample. In response, we substantially increased the sample size. While the 2014 prior
WGRA form sample (N=100,000) is slightly smaller than the 2012 WGRA sample (N=108,478), estimates based on the 2014 prior WGRA form have slightly greater precision than the 2012 WGRA. This
is primarily because the 2012 sample design was less efficient than the 2014 design due to the inclusion
of a very large oversample of Marines that was made by request of command, rather than for statistical
reasons (DMDC, 2012; stats methods report).
8 Because

of the need to weight the sample to account for the sample design and nonresponse, the statistical precision of estimates based within this sample of victims is substantially lower than a simple
random sample of 832. It offers precision comparable to a simple random sample of approximately
300 individuals. (See DMDC, 2012, for details on the design effects produced by the sampling and
weighting plans.)
9 The long form was the only form that included service-specific questions and a range of policy-relevant items favored by service sexual assault prevention and response leads. To ensure that the Army,
Navy, and Air Force had at least as many responses (and statistical precision) on these items as were
available with the 2012 WGRA, we set the sample size for this condition to be 60,000. We chose not to
offer the Marine Corps the same numbers as they had in 2012, because the 2012 WGRA substantially
oversampled the Marine Corps.
10 The

only exception is reporting domain #68 (Marines*Male*E1-E3), because the 2012 survey oversampled Marines. For this domain, we projected it was possible we would end up with lower precision
than available in the 2012 WGRA.
11 The

schedule of contacts is listed below. Because of the large volume of mail and email, most postal
mailings required three to four days to complete. Emails often required two days to complete.
Task
Soft rollout test start

12 Because

Start
8/7/2014

Survey field date

8/13/2014

Invitation (email #1 & postal letter #1)

8/13/2014

Reminder (email #2)

8/19/2014

Reminder (email #3)

8/27/2014

Reminder (email #4)

9/2/2014

Reminder (postal letter #2)

9/3/2014

Reminder (email #5)

9/5/2014

Reminder (email #6)

9/9/2014

Reminder (email #7)

9/15/2014

Reminder (postal letter #3)

9/18/2014

Reminder (email #8)

9/19/2014

Reminder (email #9)

9/23/2014

the Marine Corps sample included a significant number records containing no valid email
address (15,908), the Marine Corps offered to send invitations on RAND’s behalf to those sampled

Notes

members for whom they had personal or other email addresses. A subgroup of 6,791 Marines therefore
received invitations from RAND sent by the Marine Corps. Because RAND promised respondents
that information on who participated in the survey would not be shared with DoD or the services,
RAND could not remove names from this list of Marines when they completed the survey. Therefore,
for just this group of 6,791, sampled members received a total of four email invitations whether they
had completed the survey or not.

Chapter Four: Questionnaire Development
1 DoD

Directive 1350.2 defines military equal opportunity (MEO) violations with respect to sex and
other protected characteristics, and the survey questions closely align with these definitions. This DoD
Directive is based on federal civil rights laws (e.g., Title VII of the Civil Rights Act of 1964). However,
one difference between MEO and the federal definitions of equal employment opportunity (EEO)
violations is that MEO defines all persistent or severe harassment based on sex as an unfair condition
of military employment.

2 The development of the sexual harassment and gender discrimination portions of the survey benefited

from prior unpublished RAND work in survey development commissioned by the UniHealth Foundation and conducted by Terry L. Schell, Coreen Farris, Terri Tanielian, and Lisa H. Jaycox.
3 The

scale has been criticized on these grounds, with one review stating “It does not seem to measure
anyone’s definition of sexual harassment, including that of its own developers” (Gutek, Murphy, and
Douma, 2004).

4 For instance, in testimony presented to the Response Systems Panel, Lisa Schenck has argued that the
definition of unwanted sexual contact and its perceived affirmative consent standard could result in
non-criminal activity being counted as unwanted sexual contact: “The wording of the question implies
that holding a mental reservation about sexual activity is sufficient without any manifestation of lack
of consent.”
5 Alcohol-induced

blackouts (i.e., memory loss for events that occurred during intoxication) typically
occur after approximately 10 alcoholic drinks, and with blood alcohol content (BACs) of 0.16 percent
or greater (Goodwin et al., 1970; White et al., 2004). In addition to the potential for alcohol-induced
blackout, individuals with BACs at this level experience gross motor impairment, loss of balance, and
may require assistance to walk. Vomiting is common, and the gag reflex is impaired, raising the risk
for asphyxiation. Judgment, reaction time, vision and hearing are impaired. Speech is slurred. The
individual may be disoriented to time and place.
6 Since most of the eliminated WGRA items appeared after the assessment of sexual harassment and
sexual assault in the WGRA, we do not believe these changes could alter the context or interpretation
of those key items. We did, however, eliminate the lengthy PTSD assessment from the WGRA—which
occurred before the sexual offenses questions—because of high breakoff rates observed on this scale in
the 2012 survey. This could have an effect on who makes it to the sex-based MEO and sexual assault
questions, and it could alter the context of those questions for respondents. We view both of these
possible differences with earlier administrations of the WGRA as likely small by comparison to, for
instance, the much larger context differences of having RAND rather than DoD conduct the survey.

Sexual Assault and Sexual Harassment in the U.S. Military

Chapter Five: Analytic Methods
1

For the prior WGRA form measure of unwanted sexual contact, a case is considered usable if the
respondent answered either “yes” or “no” to the single question assessing unwanted sexual contact. For
respondents randomized to one of the RAND instruments, they were considered usable for analyses if
they had a nonmissing value on the any sexual assault variable.
Because the RAND any sexual assault measure is based on a series of questions, the determination of
when a case is missing is more complex. The coding of missingness has two steps. First, we determine if
the series of questions that starts with a behaviorally specific screener should be counted as “yes” “no”
or “missing.”
1. If the respondent answers “no” to a screening question, they are classified as not experiencing
that type of event
2. If the responded chooses to skip the screening question, they are classified as missing on that
type of event
3. If a respondent answers “yes” to a screening question, and “yes” to enough follow ups to meet our
definitional criteria, they are coded as experiencing that type of event regardless of how many
follow-up questions they skipped.
4. If the respondent answers “yes” to a screening question, skips fewer than half of the follow-up
questions, and responds “no” to those follow-up questions they did answer, they are coded as not
experiencing that type of event.
5. If the respondent answers “yes” to a screening question and skips half or more of the follow-up
questions, they are coded as missing.
The six types of events corresponding to the screening questions are then combined into the higherlevel variable, Any Sexual Assault in the Past 12 Months:
1. If the respondent was coded “yes – experienced event” to any of the six sexual assault events, they
are classified as experiencing a sexual assault
2. If the respondent is coded as both (a) missing on half of the corresponding sexual assault variables or fewer, and (b) was coded “no – did not experience event” to all of the corresponding
nonmissing sexual assault events, they are classified as not experiencing a sexual assault
3. If the respondent was coded as missing on more than half of the six sexual assault events, they
are classified as missing on Any Sexual Assault.

(The final step for deriving Any Sexual Assault removes instances of sexual assaults from the past
12-month total if the respondent indicated—after completing the detailed sexual assault module—
that the assault[s] occurred more than 12 months ago.)
A similar process is used to derive each of the key sexual assault and sex-based MEO violation
outcomes. However, only missingness on the Any Sexual Assault measure determines when a survey is
considered too incomplete to be usable.
2 In the 2012 WGRA, DMDC included one additional step in which individuals who had left the mili-

tary were inferentially excluded from the sample’s nonrespondents. Because we did not have the same
eligibility requirements, this was not done in 2014. However, the effect is trivial and cannot affect our
estimates; less than 1 percent of the sample was inferred to be ineligible by this criterion.

3 Proportionate sampling (sampling with the same selection probabilities) was carried out for all other
sampling strata, such as branch of service, component of service, and pay grade, and therefore we need
not factor these into the design weights. The design effect for the active-duty sample is 1.32.
4 The statistical models used to derive nonresponse weights included members of the active-duty Coast
Guard. Although the weighting analyses are conducted jointly, each group is balanced to its own population and the survey results are presented in separate reports. Thus the analyses included 40 reporting

Notes

categories, with 32 reporting categories corresponding to DoD active-duty members and 8 additional
reporting categories corresponding to gender and pay grade within the Coast Guard.
5 More

specifically, logistic regression models were used to predict responding within each form type.
These models were weighted by the design weights and included main effects and two-way interactions among the variables listed in Exhibit 5.1. To mimic the model selection used in the 2012
WGRA, model selection was done in SAS PROC LOGISTIC (SAS Institute, Cary, N.C.) using the
Forward selection option to conduct automated selection of predictors until the Bonferroni-adjusted
chi-squared statistic was not significant. Interactions were allowed to enter even when main effects had
not entered the model. This approach is similar to the automated selection of predictors in the CHAID
(Chi-squared Automated Interaction Detector) routine used in the 2012 WGRA. This method results
in variables being entered in the model only if they have a strong association with response propensity. As with the 2012 WGRA, the nonresponse weighting adjustment was computed as the inverse of
predicted probability of response obtained from this logistic regression model. One minor difference
between the 2012 WGRA weights and the 2014 WGRA weights is that, in 2012, a five-level race variable was included in the nonresponse model and a dichotomous race variable was used in poststratification. In 2014, a dichotomous version was used for both steps. This difference does not affect study
prevalence estimates.

6 These regression models were run separately for men and women because the relationship between risk

factors and outcomes were hypothesized to differ across gender. These models included as predictors
the items listed in Exhibit 5.2. The regression models were estimated using a machine learning algorithm, Generalized Boosted Models (GBM; Ridgeway, 2012), to best capture the relationship between
the predictors and the outcomes. GBM is a general, automated, data-adaptive modeling a lgorithm that
can estimate the relationship between a variable of interest and a large number of covariates of mixed
type, while also allowing for flexible non-linear relationships between the covariates and the response
propensity (Friedman, 2001; Ridgeway, 1999). These routines were run in the R software package.
During the GBM model estimation the complexity of each model was optimized using 10-fold cross
validation, i.e., parameters were added to the model until the model maximized out-of-training-sample
prediction. This procedure prevents over-fitting the data.
The resulting model was used to create predicted values for each person in the full sample (both
respondents and nonrespondents) on each of our outcomes. By definition, these predicted values are
weighted combinations of the variables contained in Exhibit 5.2. In particular, they are the weighted
combinations that are most associated with the outcomes. The original variables are no longer associated with the outcomes when controlling for these particular weighted combinations.
7 The

response propensity model was estimated with a binary indicator of survey response (respondent
versus nonrespondent) as the dependent variable and the predicted outcome variables from the initial
step as the independent variables, along with 40 indicators of the reporting categories (service x pay
grade x gender), form type (long, medium, or short), and two-way interactions among the predictors.
Including the reporting categories and form types in this model insures that any nonresponse bias
identified in this process is removed from both the aggregate DoD estimate and from estimates within
various reporting categories.
The nonresponse model was estimated using GBM (Ridgeway, 2012) to best capture the relationship between the various predictors and survey responding. This approach allows for flexible modeling
and has been shown to improve upon the performance of logistic regression (McCaffrey, Ridgeway,
and Morral, 2004; Ridgeway and McCaffrey, 2007). The GBM modeling approach also permits inclusion of a large number of predictors, without a need to collapse them, to preserve the most information.
Unlike ordinary prediction models, in which parameters are added to the model until out-of-trainingsample prediction is maximized, we wished to optimize the model to achieve the best weights. Specifically, we added parameters to the model until the resulting weights maximized the similarity between
the responders and the full sample. The similarity was assessed using maximum Kolmogorov–Smirnov

Sexual Assault and Sexual Harassment in the U.S. Military

statistic among all predictors in the model. Thus, the GBM model stopped when the weights achieved
the best balance between the cumulative distributions of responders and nonresponders on all of the
predictor variables in the model.

Chapter Six: Summary of Major Improvements in the RMWS
1

Although public communications about the survey described it in terms of these broad generalizations about safety and professionalism, all participants who went to the website to take the survey were
provided a detailed informed consent statement that clearly explained that the survey would cover
topics including sexual assault and sexual harassment, that some of the questions might make people
uncomfortable, that some of the language in the survey might be considered to be explicit, and that
they could skip any question they wished.

References

Ahrens, Courtney E., Rebecca Campbell, N. Karen Ternier‐Thames, Sharon M. Wasco, and Tracy
Sefl, “Deciding Whom to Tell: Expectations and Outcomes of Rape Survivors’ First Disclosures,”
Psychology of Women Quarterly, Vol. 31, No. 1, 2007, pp. 38–49.
American Association for Public Opinion Research, “Standard Definitions: Final Dispositions of
Case Codes and Outcome Rates for Surveys,” 7th ed., 2011. As of November 25, 2014:
http://aapor.org/Content/NavigationMenu/AboutAAPOR/StandardsampEthics/
StandardDefinitions/StandardDefinitions2011.pdf
Andersen, Roland, Martin R. Frankel, and Judith Kasper, Total Survey Error: Applications to Improve
Health Surveys, San Francisco: Jossey-Bass Publishers, 1979.
Antecol, Heather, and Deborah A. Cobb-Clark, “The Sexual Harassment of Female Active-Duty
Personnel: Effects on Job Satisfaction and Intentions to Remain in the Military,” Journal of Economic
Behavior and Organization, Vol. 61, No. 1, 2006, pp. 55–80.
Bachman, Ronet, “The Factors Related to Rape Reporting Behavior and Arrest New Evidence from
the National Crime Victimization Survey,” Criminal Justice and Behavior, Vol. 25, No. 1, 1998,
pp. 8–29.
———, “A Comparison of Annual Incidence Rates and Contextual Characteristics of IntimatePartner Violence against Women from the National Crime Victimization Survey (NCVS) and the
National Violence against Women Survey (NVAWS),” Violence Against Women, Vol. 6, No. 8, 2000,
pp. 839–867.
———, “Measuring Rape and Sexual Assault: Successive Approximations to Consensus,” paper
presented at National Academy of Sciences, Washington, D.C., June 6, 2012.
Black, Michele C., Kathleen C. Basile, Matthew J. Breiding, Sharon G. Smith, Mikel L. Walters,
Melissa T. Merrick, Jieru Chen, and Mark R. Stevens, National Intimate Partner and Sexual Violence
Survey: 2010 Summary Report, Atlanta, Ga.: National Center for Injury Prevention and Control,
2011. As of October 3, 2014:
http://www.cdc.gov/violenceprevention/pdf/nisvs_report2010-a.pdf
Black, Michele C., and Melissa T. Merrick, Prevalence of Intimate Partner Violence, Sexual Violence
Among Active Duty Women and Wives of Active Duty Men—Comparisons with Women in the U.S.
General Population, 2010, Technical Report, Atlanta, Ga.: National Center for Injury Prevention and
Control, 2013. As of October 3, 2014:
http://www.sapr.mil/public/docs/research/2010_National_Intimate_Partner_and_Sexual_Violence_
Survey-Technical_Report.pdf
Bondurant, Barrie, “University Women’s Acknowledgment of Rape Individual, Situational, and
Social Factors,” Violence Against Women, Vol. 7, No. 3, 2001, pp. 294–314.

100

Sexual Assault and Sexual Harassment in the U.S. Military

Breiding, Matthew J., Sharon G. Smith, Kathleen C. Basile, Mikel L. Walters, Jieru Chen, and
Melissa T. Merrick, Prevalence and Characteristics of Sexual Violence, Stalking, and Intimate Partner
Violence Victimization—National Intimate Partner and Sexual Victimization Survey, United States,
2011, Atlanta, Ga.: Centers for Disease Control, 2014.
Bureau of Justice Statistics, “National Crime Victimization Survey: NCVS-1 Basic Screen
Questionnaire,” 2008. As of October 3, 2014:
http://www.bjs.gov/content/pub/pdf/ncvs1_2012.pdf
Burks, Derek J., “Lesbian, Gay, and Bisexual Victimization in the Military: An Unintended
Consequence of ‘Don’t Ask, Don’t Tell’?,” American Psychologist, Vol. 66, No. 7, 2011, p. 604.
Cantor, David, “Substantive Implications of Longitudinal Design Features: The National Crime
Survey as a Case Study,” in D. Kasprzyk, G. Duncan, G. Kalton, and M. P. Singh, eds., Panel
Surveys, New York: John Wiley, 1989, pp. 25–51.
Clancy, Kathryn B., Robin G. Nelson, Julienne N. Rutherford, and Katie Hinde, “Survey of
Academic Field Experiences (SAFE): Trainees Report Harassment and Assault,” PLoS ONE, Vol. 9,
No. 7, e102172, 2014.
Clark, Jim, “Analysis of Crimes and Defenses 2012, UCMJ Article 120, effective 28 June 2012,”
Emerging Issues Analysis, No. 6423, 2012.
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, pp. 455–473.
Cook, Sarah L., Christine A. Gidycz, Mary P. Koss, and Megan Murphy, “Emerging Issues in the
Measurement of Rape Victimization,” Violence Against Women, Vol. 17, 2011.
Crowell, Nancy A., and Burgess, Ann W., eds., Understanding Violence Against Women, Washington,
D.C.: National Academies Press, 1996.
Das, Aniruddha, “Sexual Harassment at Work in the United States,” Archives of Sexual Behavior,
Vol. 38, No. 6, 2009, pp. 909–921.
Defense Manpower Data Center, 2010 Workplace and Gender Relations Survey of Active Duty
Members: Overview Report on Sexual Assault, DMDC Report No. 2010-025, Washington, D.C.,
2011.
———, 2012 Workplace and Gender Relations Survey of Active Duty Members: Statistical Methodology
Report, DMDC Report No. 2012-067, December 2012.
———, “Survey Note No. 2013-002: 2012 Workplace and Gender Relations Survey of Reserve
Component Members,” Washington, D.C., January 18, 2013a. As of October 3, 2014:
http://www.sapr.mil/public/docs/research/2012_Workplace_and_Gender_Relations_Survey_of_
Reserve_Component_Members-Survey_Note_and_Briefing.pdf
———, “Survey Note No. 2013-007: 2012 Workplace and Gender Relations Survey of Active Duty
Members,” Washington, D.C., March 15, 2013b. As of October 3, 2014:
http://www.sapr.mil/public/docs/research/2012_Workplace_and_Gender_Relations_Survey_of_
Active_Duty_Members-Survey_Note_and_Briefing.pdf
———, 2012 Workplace and Gender Relations Survey of Active Duty Members: Tabulations of
Responses, DMDC Report No. 2012-065, Washington, D.C., April 2013c. As of October 3, 2014:
http://www.dod.mil/pubs/foi/Personnel_and_Personnel_Readiness/Personnel/WGRA1201_
TabVolume.pdf

References

101

———, 2012 Workplace and Gender Relations Survey of Active Duty Members: Administration,
Datasets, and Codebook, DMDC Report No. 2012-068, Washington, D.C., November 4, 2013d.
———, 2012 Workplace and Gender Relations Survey of Active Duty Members: Nonresponse Bias
Analysis Report, DMDC Report No. 2013-059, Washington, D.C., 2014. As of October 3, 2014:
www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA593110
DMDC—See Defense Manpower Data Center.
DoD—See U.S. Department of Defense.
EEOC—See Equal Employment Opportunity Commission.
Equal Employment Opportunity Commission Guidelines on Discrimination Because of Sex,
29 C.F.R § 1604.11, 1980.
Falk, Eric T., “Improving Response Rates in Military Surveys,” Presentation at Joint Statistical
Meetings, San Diego, Calif., 2012.
FBI—See Federal Bureau of Investigation.
Federal Bureau of Investigation, “Crime in the United States 2012: Forcible Rape,” 2012. As of
October 3, 2014:
http://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2012/crime-in-the-u.s.-2012/violent-crime/rape
Fisher, Bonnie S., “Measuring Rape against Women: The Significance of Survey Questions,” in
Violence Against Women and Family Violence: Developments in Research, Practice, and Policy, 2004,
p. I–4–1.
Fisher, Bonnie S., and Francis T. Cullen, “Measuring the Sexual Victimization of Women: Evolution,
Current Controversies, and Future Research,” Criminal Justice, Vol. 4, 2000, pp. 317–390.
Fisher, Bonnie S., Francis T. Cullen, and Michael G. Turner, The Sexual Victimization of College
Women, Washington, D.C.: National Institute of Justice, 2000.
Fisher, Bonnie S., Leah E. Daigle, Francis T. Cullen, and Michael G. Turner, “Reporting Sexual
Victimization to the Police and Others Results from a National-Level Study of College Women,”
Criminal Justice and Behavior, Vol. 30, No. 1, 2003, pp. 6–38.
Fitzgerald, Louise F., Michele J. Gelfand, and Fritz Drasgow, “Measuring Sexual Harassment:
Theoretical and Psychometric Advances,” Basic and Applied Social Psychology, Vol. 17, No. 4, 1995,
pp. 425–445.
Fitzgerald, Louise F., Vicki J. Magley, Fritz Drasgow, and Craig R. Waldo, “Measuring Sexual
Harassment in the Military: The Sexual Experiences Questionnaire (SEQ—DoD),” Military
Psychology, Vol. 11, No. 3, 1999, pp. 243–263.
Fitzgerald, Louise F., Suzanne Swan, and Karla Fischer, “Why Didn’t She Just Report Him? The
Psychological and Legal Implications of Women’s Responses to Sexual Harassment,” Journal of Social
Issues, Vol. 51, No. 1, 1995, pp. 117–138.
Fitzgerald, Louise F., Suzanne Swan, and Vicki J. Magley, “But Was It Really Sexual Harassment?:
Legal, Behavioral, and Psychological Definitions of the Workplace Victimization of Women,” in
William O’Donohue, ed., Sexual Harassment: Theory, Research, and Treatment, Needham Heights,
Mass.: Allyn & Bacon, 1997.
Friedman, Jerome H., “Greedy Function Approximation: A Gradient Boosting Machine,” Annals of
Statistics, Vol. 29, No. 5, 2001, pp. 1189–1232. As of November 6, 2014:
http://www.jstor.org/stable/2699986

102

Sexual Assault and Sexual Harassment in the U.S. Military

Goodwin, Donald W., Ekkehard Othmer, James A. Halika, and Frank Freemon, “Loss of ShortTerm Memory as a Predictor of the Alcoholic ‘Blackout,’” Nature, Vol. 227, 1970, pp. 201–202.
Gutek, B. A., R. O. Murphy, and B. Douma, “A Review and Critique of the Sexual Experiences
Questionnaire (SEQ),” Law and Human Behavior, Vol. 28, No. 4, 2004, pp. 457–482.
Heeringa, Steven G., Brady T. West, and Patricia A. Berglund, Applied Survey Data Analysis, Boca
Raton, Fl.: CRC Press, April 5, 2010.
Hussain, Nasir, Sheila Sprague, Kim Madden, Farrah N. Hussain, Bharadwaj Pindiprolu, and Mohit
Bhandari, “A Comparison of the Types of Screening Tool Administration Methods Used for the
Detection of Intimate Partner Violence: A Systematic Review and Meta-Analysis,” Trauma Violence
Abuse, December 15, 2013. As of November 19, 2014:
http://tva.sagepub.com/content/early/2013/12/13/1524838013515759.full.pdf
Ilies, Remus, Nancy Hauserman, Susan Schwochau, and John Stibal, “Reported Incidence Rates of
Work‐Related Sexual Harassment in the United States: Using Meta‐Analysis to Explain Reported
Rate Disparities,” Personnel Psychology, Vol. 56, No. 3, 2003, pp. 607–631.
The Invisible War, dir. Kirby Dick, Cinedigm Entertainment Group, 2012. As of October 3, 2014:
http://invisiblewarmovie.com/
Jacques-Tiura, Angela J., Rifky Tkatch, Antonia Abbey, and Rhiana Wegner, “Disclosure of Sexual
Assault: Characteristics and Implications for Posttraumatic Stress Symptoms among African
American and Caucasian Survivors,” Journal of Trauma & Dissociation, Vol. 11, No. 2, 2010,
pp. 174–192.
Kilpatrick, Dean G., “What Is Violence against Women Defining and Measuring the Problem,”
Journal of Interpersonal Violence, Vol. 19, No. 11, 2004, pp. 1209–1234.
Kilpatrick, Dean G., Heidi S. Resnick, Kenneth J. Ruggiero, Lauren M. Conoscenti, and Jenna
Mccauley, Drug-Facilitated, Incapacitated, and Forcible Rape: A National Study, Medical University of
South Carolina, National Crime Victims Research and Treatment Center Charleston, S.C., 2007. As
of November 25, 2014:
http://www.niccsa.org/downloads/elders/
DRUGFACILITATEDINCAPACITATEDANDFORCIBLERAPE.pdf
Kohut, Andrew, Scott Keeter, Carroll Doherty, Michael Dimock, and Leah Christian, Assessing
the Representativeness of Public Opinion Surveys, Pew Research Center for the People and the Press,
Washington, D.C., 2012.
Koss, Mary P., “Detecting the Scope of Rape: A Review of Prevalence Research Methods,” Journal of
Interpersonal Violence, 1993, pp. 198–222.
———, “The Measurement of Rape Victimization in Crime Surveys,” Criminal Justice and Behavior,
Vol. 23, No. 1, 1996, pp. 55–69.
Koss, Mary P., and Christine A. Gidycz, “Sexual Experiences Survey: Reliability and Validity,”
Journal of Consulting and Clinical Psychology, Vol. 53, No. 3, Jun, 1985, pp. 422–423.
Kroenke, K., T. W. Strine, R. L. Spitzer, J. B. Williams, J. T. Berry, and A. H. Mokdad, “The
PHQ-8 as a Measure of Current Depression in the General Population,” Journal of Affective
Disorders, Vol. 114, No. 1, 2009, pp. 163–173.
Langton, Lynn, Marcus Berzofsky, Christopher Krebs, and Hope Smiley-Mcdonald, Victimizations
Not Reported to the Police, 2006–2010, Special Report, NCJ 238536, Bureau of Justice Statistics,
August 9, 2012. As of October 3, 2014:
http://www.bjs.gov/content/pub/pdf/vnrp0610.pdf

References

103

Layman, Melissa J., Christine A. Gidycz, and Steven J. Lynn, “Unacknowledged Versus
Acknowledged Rape Victims: Situational Factors and Posttraumatic Stress,” Journal of Abnormal
Psychology, Vol. 105, No. 1, February 1996, pp. 124–131.
Lehnen, Robert, and Wesley Skogan, eds., The National Crime Survey: Working Papers, Vol. 2,
Methodological Studies, Washington, D.C.: U.S. Department of Justice, Bureau of Justice Statistics
NCJ-90307, October 1984.
Little, Roderick J., and Donald B. Rubin, Statistical Analysis with Missing Data (2nd ed.), New York:
Wiley-Interscience, 2002.
Little, Roderick J., and Sonya Vartivarian, “Does Weighting for Nonresponse Increase the Variance
of Survey Means?,” Survey Methodology, Vol. 31, No. 2, 2005, pp. 161–168.
Lohr, Sharon L., Sampling: Design and Analysis, 2nd ed., Cengage Learning, 2010.
Lynn, Peter, Nicholas Buck, Jonathan Burton, Annette Jäckle, and Heather Laurie, “A Review of
Methodological Research Pertinent to Longitudinal Survey Design and Data Collection,” Working
Papers of the Institute for Social and Economic Research, paper 2005-29, Colchester, UK: University of
Essex, 2005.
McCaffrey, D. F., G. Ridgeway, and A. R. Morral, “Propensity Score Estimation with Boosted
Regression for Evaluating Causal Effects in Observational Studies,” Psychological Methods, Vol. 9,
2004, pp. 403–425.
Moore, Brenda L., Effects of Sexual Harassment on Job Satisfaction, Retention, Cohesion, Commitment
and Unit Effectiveness: The Case of the Air Force,” Defense Equal Opportunity Management Institute
Directorate of Research, Technical Report (Executive Summary) #08-10, 2010.
Montgomery, Nancy, “Case Dismissed Against Aviano IG Convicted of Sexual Assault,” Stars and
Stripes, February 27, 2013. As of October 3, 2014:
http://www.stripes.com/news/air-force/case-dismissed-against-aviano-ig-convicted-of-sexualassault-1.209797
National Research Council, Cognitive Aspects of Survey Methodology: Building a Bridge Between
Disciplines, Washington, D.C.: The National Academies Press, 1984.
———, “Nonresponse in Social Science Surveys: A Research Agenda,” in Roger Tourangeau and
Thomas J. Plewes, eds., Panel on a Research Agenda for the Future of Social Science Data Collection,
Committee on National Statistics. Division of Behavioral and Social Sciences and Education,
Washington, D.C.: The National Academies Press, 2013.
———, “Estimating the Incidence of Rape and Sexual Assault,” in Candace Kruttschnitt, William
D. Kalsbeek, and Carol C. House, eds., Panel on Measuring Rape and Sexual Assault in Bureau of
Justice Household Surveys, Washington, D.C.: The National Academies Press, 2014.
National Victim Center and Crime Victims Research and Treatment Center, Rape in America: A
Report to the Nation, Arlington, Va.: National Victim Center, 1992.
Office of Management and Budget, Standards and Guidelines for Statistical Surveys, Washington,
D.C., September 2006. As of October 3, 2014:
http://www.whitehouse.gov/sites/default/files/omb/inforeg/statpolicy/standards_stat_surveys.pdf
Percy, A., and P. Mayhew, “Estimating Sexual Victimization in a National Crime Survey: A New
Approach,” Studies on Crime & Crime Prevention, Vol. 6, No. 2, 1997, pp. 125–150.
Prins, Annabel, Paige Ouimette, Rachel Kimerling, Rebecca P. Cameron, Daniela S. Hugelshofer,
Jennifer Shaw-Hegwer, Ann Thrailkill, Fred D. Gusman, and Javaid I. Sheikh, “The Primary Care
PTSD Screen (PC-PTSD): Development and Operating Characteristics,” International Journal of
Psychiatry in Clinical Practice, Vol. 9, No. 1, 2004, pp. 9–14.

104

Sexual Assault and Sexual Harassment in the U.S. Military

Ridgeway, Greg, “The State of Boosting,” in K. Berk and M. Pourahmadi, eds., Computing Science
and Statistics, Fairfax, Va.: Interface Foundation of North America, 1999, pp. 172–181.
———, Generalized Boosted Models: A Guide to the Gbm Package, May 23, 2012. As of October 3,
2014:
http://gradientboostedmodels.googlecode.com/git/gbm/inst/doc/gbm.pdf
Ridgeway, Greg, and Daniel F. McCaffrey, “Comment: Demystifying Double Robustness: A
Comparison of Alternative Strategies for Estimating a Population Mean from Incomplete Data,”
Statistical Science: A Review Journal of the Institute of Mathematical Statistics, Vol. 22, No. 4, 2007,
p. 569. As of October 3, 2014:
http://www.jstor.org/stable/27645859
Rodman, Lindsay L., “The Pentagon’s Bad Math on Sexual Assault,” The Wall Street Journal, 2013.
As of October 3, 2014: http://online.wsj.com/news/articles/SB1000142412788732358290457848494
1173658754
Sabol, William J., and Allen Beck, Testimony, Appearance before the Comparative Systems
Subcommittee of the Response Systems to Adult Sexual Assault Crimes Panel (RSP), April 11, 2014.
As of October 3, 2014:
http://responsesystemspanel.whs.mil/Public/docs/meetings/Sub_Committee/20140411_CSS/25_
BJS_Presentation_20140411.pdf
Sanders, Stephanie A., and June M. Reinisch, “Would You Say You ‘Had Sex’ If...?,” Journal of the
American Medical Association, Vol. 281, No. 3, January 20, 1999, pp. 275–277.
Saris, William E., and Irmtraud N. Gallhofer, Design, Evaluation, and Analysis of Questionnaires for
Survey Research, Hoboken, N.J.: John Wiley & Sons, 2014.
Saum, Christine A., and James A. Inciardi, “Rohypnol Misuse in the United States,” Substance Use &
Misuse, Vol. 32, No. 6, 1997, pp. 723–731.
Scarborough, Rowan, “Doubts on Military’s Sex Assault Stats as Numbers Far Exceed Those for the
U.S.,” The Washington Times, April 6, 2014. As of October 3, 2014:
http://www.washingtontimes.com/news/2014/apr/6/
doubts-on-militarys-sex-assault-stats-as-numbers-f/?page=all
Schafer, Joseph L., and John W. Graham, “Missing Data: Our View of the State of the Art,”
Psychological Methods, Vol. 7, No. 2, 2002, pp. 147–177.
Schenck, Lisa, “Testimony Before the Response Systems to Adult Sexual Assault Crimes Panel
(RSP),” Public Meeting of the Response Systems to Adult Sexual Assault Crimes Panel, May 5–6, 2014.
As of October 3, 2014:
http://responsesystemspanel.whs.mil/Public/docs/meetings/20140505/
20140506_Transcript_Final.pdf
Schwartz, Martin D., and Molly S. Leggett, “Bad Dates or Emotional Trauma? The Aftermath of
Campus Sexual Assault,” Violence Against Women, Vol. 5, No. 3, 1999, pp. 251–271.
Shannon, Candace A., Kathleen M. Rospenda, and Judith A. Richman, “Workplace Harassment
Patterning, Gender, and Utilization of Professional Services: Findings from a US National Study,”
Social Science & Medicine, Vol. 64, No. 6, March 2007, pp. 1178–1191.
Shinkman, Paul D., “Air Force Sexual Assault Prevention Chief Arrested for Sexual Assault,” U.S.
News & World Report, May 7, 2013. As of October 3, 2014:
http://www.usnews.com/news/articles/2013/05/07/
air-force-sexual-assault-prevention-chief-arrested-for-sexual-assault

References

105

Sirken, Monroe, Douglas J. Herrmann, Susan Schechter, Norbert Schwarz, Judith M. Tanur, and
Roger Tourangeau, eds.,Cognition and Survey Research, New York: Wiley, 1999.
Stark, Stephan, Oleksandr S. Chernyshenko, Anita R. Lancaster, Fritz Drasgow, and Louise F.
Fitzgerald, “Toward Standardized Measurement of Sexual Harassment: Shortening the SEQ-DoD
Using Item Response Theory,” Military Psychology, Vol. 14, No. 1, 2002, p. 49.
Street, Amy E., Jamie L. Gradus, Jane Stafford, and Kacie Kelly, “Gender Differences in Experiences
of Sexual Harassment: Data from a Male-Dominated Environment,” Journal of Consulting and
Clinical Psychology, Vol. 75, No. 3, 2007, pp. 464–474.
Street, Amy E., Jane Stafford, Clare M. Mahan, and Ann Hendricks, “Sexual Harassment and
Assault Experienced by Reservists During Military Service: Prevalence and Health Correlates,”
Journal of Rehabilitation Research and Development, Vol. 45, No. 3, 2008, pp. 409–419.
Tjaden, Patricia, and Nancy Thoennes, “Prevalence, Incidence, and Consequences of Violence
Against Women: Findings from the National Violence Against Women Survey—Research in Brief,”
Washington, D.C.: National Institute of Justice and Centers for Disease Control and Prevention,
1998.
———, Full Report of the Prevalence, Incidence, and Consequences of Violence Against Women: Findings
from the National Violence Against Women Survey, Research Report, Washington, D.C.: National
Institute of Justice and Centers for Disease Control and Prevention, NCJ 172837, November 2000.
Tourangeau, R., L. J. Rips, and K. Rasinski, The Psychology of Survey Response, Cambridge, UK:
Cambridge University Press, 2000.
Tourangeau, Roger, and Tom W. Smith, “Asking Sensitive Questions the Impact of Data Collection
Mode, Question Format, and Question Context,” Public Opinion Quarterly, Vol. 60, No. 2, 1996,
pp. 275–304.
Tourangeau, Roger, and Ting Yan, “Sensitive Questions in Surveys,” Psychological Bulletin, Vol. 133,
No. 5, 2007, pp. 859–883.
Tracy, Carol E., Terry L. Fromson, Jennifer G. Long, and Charlene Whitman, Rape and Sexual
Assault in the Legal System, Washington, D.C.: U.S. Department of Justice, Bureau of Justice
Statistics, 2012.
Truman, Jennifer L., and Lynn Langton, Criminal Victimization 2013, Bulletin, Washington, D.C.:
U.S. Department of Justice, Bureau of Justice Statistics,. NCJ 247648, September 2014.
Turner, C. F., L. Ku, S. M. Rogers, L. D. Lindberg, J. H. Pleck, and F. L. Sonenstein, “Adolescent
Sexual Behavior, Drug Use, and Violence: Increased Reporting with Computer Survey Technology,”
Science, Vol. 280, No. 5365, 1998, pp. 867–873.
United States Code, Title 10, Subtitle A, Part II, Chapter 80, Section 1561, Complaints of Sexual
Harassment: Investigation by Commanding Officers, 2006.
U.S. Department of Defense, Standardized Survey Measure for Sexual Harassment, Memorandum
from the Under Secretary for Personnel and Readiness to the Assistant Secretaries of Army, Navy and
Air Force (Manpower and Reserve Affairs), Office of the Under Secretary of Defense for Personnel
and Readiness, March 12, 2002.
———, “Department of Defense Military Equal Opportunity (MEO) Program,” DoD Directive
1350.2, August 18, 1995, incorporating Change 1, May 7, 1997, certified current as of November 21,
2003.
———, Manual for Courts-Martial United States, 2012 ed., Washington, D.C., 2012. As of October
3, 2014:
http://www.loc.gov/rr/frd/Military_Law/pdf/MCM-2012.pdf

106

Sexual Assault and Sexual Harassment in the U.S. Military

———, Department of Defense Annual Report on Sexual Assault in the Military: Fiscal Year 2012,
Vol. 1, Washington, D.C.: Sexual Assault Prevention and Response Office, 2013a. As of October 3,
2014:
http://www.sapr.mil/public/docs/reports/FY12_DoD_SAPRO_Annual_Report_on_Sexual_AssaultVolume_One.pdf
———, Sexual Assault Prevention and Response (SAPR) Program, DoD Directive 6495.01, 2013b. As
of October 3, 2014:
http://www.nationalguard.mil/Portals/31/Documents/J1/SAPR/Policy/DoD%20Directive%20
6495.01p.pdf
U.S. Senate, Oversight Hearing to Receive Testimony on Pending Legislation Regarding Sexual Assaults in
the Military: Hearing Before the Committee on Armed Services, Washington, D.C., June 4, 2013. As of
October 3, 2014:
http://www.armed-services.senate.gov/imo/media/doc/13-44%20-%206-4-13.pdf
Walsh, Wendy A., Victoria L. Banyard, Mary M. Moynihan, Sally Ward, and Ellen S. Cohn,
“Disclosure and Service Use an a College Campus after an Unwanted Sexual Experience,” Journal of
Trauma & Dissociation, Vol. 11, No. 2, 2010, pp. 134–151.
Walters, Mikel L., Jieru Chen, and Matthew J. Breiding, The National Intimate Partner and Sexual
Violence Survey (NISVS): 2010 Findings on Victimization by Sexual Orientation, Atlanta, Ga.:
National Center for Injury Prevention and Control, January 2013.
Ware John E., Jr., and Cathy D. Sherbourne, “The MOS 36-Item Short-Form Health Survey (SF36): I. Conceptual Framework and Item Selection,” Medical Care, 1992, pp. 473–483.
White, Aaron M., Matthew L. Signer, Courtney L. Kraus, and H. Scott Swartzwelder, “Experiential
Aspects of Alcohol–Induced Blackouts Among College Students, American Journal of Drug and
Alcohol Abuse, Vol. 30, No. 1, 2004, pp. 205–224.
White, Aaron M., Peter E. Simson, and Phillip J. Best, “Comparison Between the Effects of Ethanol
and Diazepam on Spatial Working Memory in the Rat” Psychopharmacology, Vol. 133, No. 3, 1997,
pp. 256–261.
Willness, Chelsea R., Piers Steel, and Kibeom Lee, “A Meta‐Analysis of the Antecedents and
Consequences of Workplace Sexual Harassment,” Personnel Psychology, Vol. 60, No. 1, 2007,
pp. 127–162.
Yu, Qingzhao, Elizabeth A. Stasny, and Bin Li, “Bayesian Models to Adjust for Response Bias in
Survey Data for Estimating Rape and Domestic Violence Rates from the NCVS,” The Annals of
Applied Statistics, Vol. 2, No. 2, 2008, pp. 665–686.
Zucchino, David, “Army General to Face Court-Martial on Sexual Misconduct Charges,” Los Angeles
Times, January 21, 2013. As of October 20, 2014:
http://articles.latimes.com/2013/jan/21/nation/la-na-nn-fort-bragg-sexual-assault-court-martial130121

Survey Instrument

2014 RAND Military Workplace Study Forms

The RAND research team designed the RMWS survey so that each question was
posed to only as many service members as was necessary to provide the needed precision required for each question. The team assigned those receiving the new RAND
questionnaire to one of three different versions: a short form, a medium form, and a
long form. The tables below show the survey questions assigned to the different RMWS
forms and to each survey module.
RMWS Forms
Form

Question Numbers

Short

1 to 22,
117 to 269

Description
• Sexual assault module
• Screening items from the sex-based MEO violation module
• Demographic questions

Medium

1 to 269

• Sexual assault module
• Sex-based MEO violation module
• Demographic questions

Long

1 to 376

•
•
•
•
•

Sexual assault module
Sex-based MEO violation module
Demographic questions
Attitude and climate questions
Service-specific questions

Survey Modules
Items

Question Numbers

Timeframe

1 to 5

Sex-based MEO violation module

6 to 116

Sexual assault module

117 to 265

Demographic, attitude, climate, and servicespecific questions

266 to 376

109

110

Sexual Assault and Sexual Harassment in the U.S. Military

Programming notes are in blue
Variables and response values are in red
Limit screen width of question text for readability
No scrolling pages
Unless specified otherwise, all Respondents get the question
Questions and response options in regular type
Example text uses slightly different color throughout the survey, dark grey instead of
black.
One question per page or more if the stems are the same.
Universal Rules:
97 = Do not know
0 = Unchecked checkboxes in choose all that apply questions, clean data, or
response option
-5 = Non-Response
-7 = Appropriately Skipped

Administrative Variables:

[Programming note: variables defined in the SMS are assumed to be initialized as
missing and will be assigned other values only as required for the skip pattern. These
are not the analytic variables.]
X Date: month, day and year that is one year prior to the date the person began the
survey - Set “X Date” when the Respondent answers the consent question and keep
that date no matter when they return during fielding
Times
PerceivedHostileWorkEnvironment
PerceivedSexualQuidProQuo
PerceivedGenderDiscrimination
PerceivedEqualOpportunityViolation
sexualAssault_12m
penetrativeSA_12m
attemptedSA_12m
contactSA_12m

Survey Instrument

[Landing Page START]

2014 RAND Military Workplace Study

Please sign in

Continue
Technical Assistance: support@randmilitarystudy.org, 1-855-365-5914
[Landing Page END]

111

112

Sexual Assault and Sexual Harassment in the U.S. Military

[Consent START]
Welcome to the 2014 RAND Military Workplace Study

Dear [Soldier/Sailor/Airman/Marine/Coast Guardsman]:

RCS # DD-P&R(QD)1947
Expires: 7/25/2015

Before you begin this short web survey, please read the following information about
the purpose of the RAND Survey and why it’s important for you and for the Services,
and the informed consent statement that follows.
Press the Frequently Asked Questions (FAQs) button at the bottom of this page if
you want to read more details about the study.
PRIVACY ADVISORY

The Defense Manpower Data Center has provided certain information about you
to allow RAND to conduct this survey. Your name and contact information have
been used to send you notifications and information about this survey. The Defense
Manpower Data Center has provided certain demographic information to reduce
the number of questions in the survey and minimize the burden on your time. Your
response and demographic data are linked by RAND to allow for a thorough analysis
of the responses by the demographics. RAND has not been authorized by DoD to
identify or link survey response and demographic information with your name and
contact information. The resulting reports will not include analysis of groupings of less
than 15.
INFORMED CONSENT STATEMENT

Introduction: You are being asked to complete a survey being conducted by the
RAND Corporation and Westat. The survey asks about whether or not you have experienced harassment, discrimination, or inappropriate sexual behavior. We need your
responses whether or not you have had these experiences. RAND is a private, nonprofit organization that conducts research and analysis to help improve public policy
and decisionmaking. RAND’s research partner is Westat, an internationally known
research and statistical survey organization.
Purpose: The Department of Defense (DoD) and Congress are working to understand
the full extent of harassment and assault in the military, and whether current efforts
to reduce them are helping. The DoD has funded RAND to conduct an independent
assessment of the military work environment during the past year. You and other Service members, including all women and approximately 25% of men, are being urged
to participate in order to ensure that DoD and Congress have a full understanding of
Service member experiences. The survey results will have a direct impact on training,
military justice, and services that affect you and other Service members.

Survey Instrument

113

Survey Length: This web survey will take approximately [TimetoComplete] minutes
to complete. Depending on your responses, it may take you more or less time.
Voluntary Participation: Your participation is completely voluntary and you may
stop at any time. If you need to break off before completing the survey, you may log on
again later to complete it.
Privacy: RAND/Westat will not give the DoD information about who participated
in the study, nor will RAND link your individual responses on this survey with your
name or identity. DoD has agreed to this condition to protect your privacy. RAND
has also received a federal “Certificate of Confidentiality” that provides RAND with
additional protection against any attempt to subpoena confidential survey records.
Added Protection Procedures: Only members of the RAND/Westat study team will
have access to your individual responses and we will take great care to protect your privacy and data. For example, RAND will collapse some categories or ranges of potentially identifying variables to prevent identification by inference. Study staff members
have been trained to deidentify data to protect your identity and are subject to civil
penalties for violating your confidentiality. Our research team has a number of safeguarding procedures in place to ensure that survey data are protected from accidental
disclosure.
Risks of Participation: For most respondents, the survey involves no risks of participation. However, if you have ever experienced sexual harassment or assault, some questions may cause discomfort or distress. Some questions may be explicit. Therefore, you
may prefer to take the survey in a private setting.
Reporting Harassment or Assault: It is important to note that this survey is not a
means of making a formal complaint or report that you wish to have DoD act upon.
The survey will not collect the identity of any perpetrators of assault or harassment.
Instead, we provide information below and at the end of the survey about how you can
make a formal report of harassment or assault.
Resources Available to You: If you need resources or assistance, the DoD Safe Helpline
(https://www.safehelpline.org/) provides worldwide live, confidential support, 24/7.
You can initiate a report and search for your nearest Sexual Assault Response Coordinator (SARC). You can find links to Service-specific reporting resources and access
information about the prevention of and response to sexual assault on their website or
by calling the hotline at 1-877-995-5247.
Some questions in the survey may ask about upsetting experiences. If you feel distressed, for confidential support and consultation, you can contact the Military Crisis
Line (http://veteranscrisisline.net/ActiveDuty.aspx) or call them at 1-800-273-8255
(then press 1).

114

Sexual Assault and Sexual Harassment in the U.S. Military

Who do you contact if you have questions or concerns about the survey?
• Questions about the overall study or RAND: Contact the RAND team by
email: WGRS2014@rand.org or go to the RAND website link: www.WGRS2014.
rand.org.
• Questions about computer, technical or survey problems: Contact the Westat
Survey Helpdesk toll free number: 1-855-365-5914 (OCONUS please call collect: 240-453-2620) or by email: support@randmilitarystudy.org.
• Questions about your rights as a participant in this study: Contact the
RAND Human Subjects Protection Committee: 310-393-0411, ext. 6369 in
Santa Monica, California.
• Questions about the licensing of the survey: Information about DoD surveys
can be found at http://www.dtic.mil/whs/directives/corres/intinfocollections/iic_
search.html; this survey’s RCS # is DD-P&R(QD)1947 and the expiration date
is 7/25/2015.
Some of the questions in this survey will be personal. For your privacy, you may want
to take this survey where other people won’t see your screen. It is okay to forward the
survey to a personal email address or you may complete it on a smartphone.
Press the Continue button if you agree to do the survey.
You can print a copy of this Informed Consent Statement by pressing the following
button:
Print Informed
Consent
Continue
[Consent END]

Survey Instrument

115

Thank you for agreeing to participate in this important study.
Please answer each question thoughtfully and truthfully. This will allow us to provide
an accurate picture of the different experiences of today’s military members. If you
prefer not to answer a specific question for any reason, just leave it blank.
Some of the questions in this survey will be personal. For your privacy, you may want
to take this survey where other people won’t see your screen.
• Are you… Intro1
–– Male 1
–– Female 2
[Intro1 will determine wording in items – [brackets] indicate alternative forms. If R
does not provide gender then grab sample gender.]
Most of this survey asks about experiences that have happened within the past 12
months. When answering these questions, please do NOT include any events that
occurred before [Day_of_Week, X date].
Please try to think of any important events in your life that occurred near [X date] such
as birthdays, weddings, or family activities. These events can help you remember which
things happened before [X date] and which happened after as you answer the rest of
the survey questions.
The following questions will help you think about your life one year ago.
1. Do you currently live in the same house or building that you did on
[X Date]? P1
a. Yes 1
b. No 2
c. Do not remember 97
2. Are you the same rank today that you were on [X Date]? P2
a. Yes 1
b. No 2
c. Do not remember 97
3. Are you in the same military occupation today as you were on [X Date]? P3
a. Yes 1
b. No 2
c. Do not remember 97
4. Were you on vacation or leave on [X Date]? P4
a. Yes 1

116

Sexual Assault and Sexual Harassment in the U.S. Military

b. No 2
c. Do not remember 97
5. Were you married or dating someone on [X Date]? P5
a. Yes 1
b. No 2
c. Do not remember 97
[If reservecomp = True, show]
The first part of this survey asks about experiences that happened while you were
on military duty, including National Guard or Reserve duty such as weekend drills,
annual training, and any period in which you were on active duty. Do not include
experiences that happened in your non-military job.
Next Page
In this section, you will be asked about several things that someone from work might
have done to you that were upsetting or offensive, and that happened AFTER [X date].
When the questions say “someone from work,” please include any person you have
contact with [as part of your military duties / If reservecomp = True, replace with: “as
part of your military duties”]. “Someone from work” could be a supervisor, someone
above or below you in rank, or a civilian employee / contractor. They could be in your
unit or in other units.
These things may have occurred on-duty or off-duty, on-base or off-base. Please include
them as long as the person who did them to you was someone from [If reservecomp =
True, insert here: “your military”] work.
Remember, all the information you share will be kept confidential.
[Programming note: the “Someone from work” box appears for SH1-SH15. Please
place the definition box above the question text for all items SH1-SH15:]
“Someone from work” includes any person you have contact with [as part of your military duties / If reservecomp = True, replace with: “as part of your military duties”].
“Someone from work” could be a supervisor, someone above or below you in rank, or
a civilian employee /contractor. They could be in your unit or in other units. These
things may have occurred off-duty or off-base. Please include them as long as the
person who did them to you was someone from [If reservecomp = True, insert here:
“your military”] work.

Survey Instrument

117

[Programming note: Use gender questions asked at the beginning of the survey to
branch into parallel forms. Brackets within items show which words will be used by
gender of respondent.]
6. Since [X Date], did someone from work repeatedly tell sexual “jokes” that
made you uncomfortable, angry, or upset? SH1
a. Yes 1
b. No 2
[Programming note: Same sex as respondent]
7. Since [X Date], did someone from work embarrass, anger, or upset you by
repeatedly suggesting that you do not act like a [man/woman] is supposed
to? For example, by calling you [male respondents: “a woman, a fag, or gay”;
female respondents: “a dyke, or butch”]. SH2
a. Yes 1
b. No 2
8. Since [X Date], did someone from work repeatedly make sexual gestures
or sexual body movements (for example, thrusting their pelvis or grabbing
their crotch) that made you uncomfortable, angry, or upset? SH3
a. Yes 1
b. No 2
9. Since [X Date], did someone from work display, show, or send sexually
explicit materials like pictures or videos that made you uncomfortable,
angry, or upset? SH4
a. Yes 1
b. No 2
10. Since [X Date], did someone from work repeatedly tell you about their
sexual activities in a way that made you uncomfortable, angry, or upset?
SH5
a. Yes 1
b. No 2
11. Since [X Date], did someone from work repeatedly ask you questions about
your sex life or sexual interests that made you uncomfortable, angry, or
upset? SH6
a. Yes 1
b. No 2

118

Sexual Assault and Sexual Harassment in the U.S. Military

12. Since [X Date], did someone from work make repeated sexual comments
about your appearance or body that made you uncomfortable, angry, or
upset? SH7
a. Yes 1
b. No 2
13. Since [X Date], did someone from work either take or share sexually suggestive pictures or videos of you when you did not want them to? SH8
a. Yes 1
b. No 2
[If SH8=2 (No) then skip to SH9]
14. Did this make you uncomfortable, angry, or upset? SH8a
a. Yes 1
b. No 2
15. Since [X Date], did someone from work make repeated attempts to establish an unwanted romantic or sexual relationship with you? These could
range from repeatedly asking you out for coffee to asking you for sex or a ‘hookup’. SH9
a. Yes 1
b. No 2
[If SH9=2 (No) then skip to SH10]
16. Did these attempts make you uncomfortable, angry, or upset? SH9a
a. Yes 1
b. No 2
17. Since [X Date], did someone from work intentionally touch you in a sexual
way when you did not want them to? This could include touching your genitals, breasts, buttocks, or touching you with their genitals anywhere on your
body. SH10
a. Yes 1
b. No 2
[If SH10=1 (Yes) then Skip to SH12 and PerceivedHostileWorkEnvironment = TRUE]
18. Since [X Date], did someone from work repeatedly touch you in any other
way that made you uncomfortable, angry, or upset? This could include

Survey Instrument

119

almost any unnecessary physical contact including hugs, shoulder rubs, or
touching your hair, but would not usually include handshakes or routine uniform adjustments. SH11
a. Yes 1
b. No 2
19. Since [X Date], has someone from work made you feel as if you would get
some [If reservecomp = True, insert here: “military”] workplace benefit in
exchange for doing something sexual? For example, they might hint that they
would give you a good evaluation/fitness report, a better assignment, or better
treatment at work in exchange for doing something sexual. Something sexual
could include talking about sex, undressing, sharing sexual pictures, or having
some type of sexual contact. SH12
a. Yes 1
b. No 2
20. Since [X Date], has someone from work made you feel like you would get
punished or treated unfairly in the [If reservecomp = True, insert here: “military”] workplace if you did not do something sexual? For example, they
hinted that they would give you a bad evaluation/fitness report, a bad assignment, or bad treatment at work if you were not willing to do something sexual.
This could include being unwilling to talk about sex, undress, share sexual pictures, or have some type of sexual contact. SH13
a. Yes 1
b. No 2
21. Since [X Date], did you hear someone from work say that [men/women] are
not as good as [women/men] at your particular [If reservecomp = True, insert
here: “military”] job, or that [men/women] should be prevented from having
your job? SH14
a. Yes 1
b. No 2
22. Since [X Date], do you think someone from work mistreated, ignored,
excluded, or insulted you because you are a [man/woman]? SH15
a. Yes 1
b. No 2
[If PM_filter <.57 AND FormType=”short”, skip to SA1, else continue to SH1b. ]
[Ask if SH1=1 (Yes) else skip to SH2b]

120

Sexual Assault and Sexual Harassment in the U.S. Military

23. You indicated that, after [X date], someone from work made you uncomfortable, angry, or upset by repeatedly telling sexual “jokes.”
Do you think they knew that you or someone else wanted them to stop? If
it happened more than once or by more than one person, do you think any
of them ever knew? SH1b
a. Yes 1
b. No 2
[If SH1b=2 (No) then skip to SH1d]
24. You indicated that, after [X date], someone from work made you uncomfortable, angry, or upset by repeatedly telling sexual “jokes.”
Did they continue this unwanted behavior even after they knew that you
or someone else wanted them to stop? SH1c
a. Yes 1
b. No 2
[If SH1c=1 (Yes) then skip to SH2b, AND PerceivedHostileWorkEnvironment =
TRUE]
[Same sex as respondent for all SHXd questions]
25. Do you think that this was ever severe enough that most [men/women] in
the military would have been offended by these jokes if they had heard
them? If you aren’t sure, choose the best answer. SH1d
a. Yes 1
b. No 2
[If SH1d=1 (Yes) then PerceivedHostileWorkEnvironment = TRUE]
[Ask if SH2=1 (Yes) else skip to SH3b]
26. You indicated that, after [X date] someone from work made you embarrassed,
angry, or upset by repeatedly suggesting that you do not act like a [man/woman]
is supposed to. For example, by calling you [male respondents: “a woman, a fag,
or gay”; female respondents: “a dyke, or butch”].
Do you think they knew that you or someone else wanted them to stop? If
it happened more than once or by more than one person, do you think any
of them ever knew? SH2b

Survey Instrument

121

a. Yes 1
b. No 2
[If SH2b=2 (No) then skip to SH2d]
27. You indicated that, after [X date] someone from work made you embarrassed,
angry, or upset by repeatedly suggesting that you do not act like a [man/woman]
is supposed to. For example, by calling you [male respondents: “a woman, a fag,
or gay”; female respondents: “a dyke, or butch”].
Did they continue this unwanted behavior even after they knew that you
or someone else wanted them to stop? SH2c
a. Yes 1
b. No 2
[If SH2c=1 (Yes) then skip to SH3b, AND PerceivedHostileWorkEnvironment =
TRUE]
28. Do you think that this was ever severe enough that most [men/women] in
the military would have been offended if someone had said these things to
them? If you’re not sure, choose the best answer. SH2d
a. Yes 1
b. No 2
[If SH2d=1 (Yes) then PerceivedHostileWorkEnvironment = TRUE]
[Ask if SH3=1 (Yes) else skip to SH4b]
29. You indicated that, after [X date], someone from work made you uncomfortable, angry, or upset by repeatedly making sexual gestures or sexual body movements.
Do you think they knew that you or someone else wanted them to stop? If
it happened more than once or by more than one person, do you think any
of them ever knew? SH3b
a. Yes 1
b. No 2
[If SH3b=2 (No) then skip to SH3d]

122

Sexual Assault and Sexual Harassment in the U.S. Military

30. You indicated that, after [X date], someone from work made you uncomfortable, angry, or upset by repeatedly making sexual gestures or sexual body movements.
Did they continue this unwanted behavior even after they knew that you
or someone else wanted them to stop? SH3c
a. Yes 1
b. No 2
[If SH3c=1 (Yes) then skip to SH4b, AND PerceivedHostileWorkEnvironment =
TRUE]
31. Do you think that this was ever severe enough that most [men/women] in
the military would have been offended by these gestures? If you’re not sure,
choose the best answer. SH3d
a. Yes 1
b. No 2
[If SH3d=1 (Yes) then PerceivedHostileWorkEnvironment = TRUE]
[Ask if SH4=1 (Yes) else skip to SH5b]
32. You indicated that, after [X date], someone from work made you embarrassed,
angry, or upset by displaying, showing, or sending sexually explicit materials
like pictures or videos.
Do you think they knew that you or someone else wanted them to stop? If
it happened more than once or by more than one person, do you think any
of them ever knew? SH4b
a. Yes 1
b. No 2
[If SH4b=2 (No) then skip to SH4d]
33. You indicated that, after [X date], someone from work made you embarrassed,
angry, or upset by displaying, showing, or sending sexually explicit materials
like pictures or videos.
Did they continue this unwanted behavior even after they knew that you
or someone else wanted them to stop? SH4c
a. Yes 1
b. No 2

Survey Instrument

123

[If SH4c=1 (Yes) then skip to SH5b, AND PerceivedHostileWorkEnvironment =
TRUE]
34. Do you think that this was ever severe enough that most [men/women] in
the military would have been offended by seeing these sexually explicit
materials? If you’re not sure, choose the best answer. SH4d
a. Yes 1
b. No 2
[If SH4d=1 (Yes) then PerceivedHostileWorkEnvironment = TRUE]
[Ask if SH5=1 (Yes) else skip to SH6b]
35. You indicated that, after [X date], someone from work made you uncomfortable, angry, or upset by repeatedly telling you about their sexual activities.
Do you think they knew that you or someone else wanted them to stop? If
it happened more than once or by more than one person, do you think any
of them ever knew? SH5b
a. Yes 1
b. No 2
[If SH5b=2 (No) then skip to SH5d]
36. You indicated that, after [X date], someone from work made you uncomfortable, angry, or upset by repeatedly telling you about their sexual activities.
Did they continue this unwanted behavior even after they knew that you
or someone else wanted them to stop? SH5c
a. Yes 1
b. No 2
[If SH5c=1 (Yes) then skip to SH6b, AND PerceivedHostileWorkEnvironment =
TRUE]
37. Do you think that this was ever severe enough that most [men/women] in
the military would have been offended by hearing about these sexual activities? If you’re not sure, choose the best answer. SH5d
a. Yes 1
b. No 2
[If SH5d=1 (Yes) then PerceivedHostileWorkEnvironment = TRUE]

124

Sexual Assault and Sexual Harassment in the U.S. Military

[Ask if SH6=1 (Yes) else skip to SH7b]
38. You indicated that, after [X date], someone from work made you embarrassed,
angry, or upset by asking you questions about your sex life or sexual interests.
Do you think they knew that you or someone else wanted them to stop? If
it happened more than once or by more than one person, do you think any
of them ever knew? SH6b
a. Yes 1
b. No 2
[If SH6b=2 (No) then skip to SH6d]
39. You indicated that, after [X date], someone from work made you embarrassed,
angry, or upset by asking you questions about your sex life or sexual interests.
Did they continue this unwanted behavior even after they knew that you
or someone else wanted them to stop? SH6c
a. Yes 1
b. No 2
[If SH6c=1 (Yes) then skip to SH7b, AND PerceivedHostileWorkEnvironment =
TRUE]
40. Do you think that this was ever severe enough that most [men/women] in
the military would have been offended if they had been asked these questions? If you’re not sure, choose the best answer. SH6d
a. Yes 1
b. No 2
[If SH6d=1 (Yes) then PerceivedHostileWorkEnvironment = TRUE]
[Ask if SH7=1 (Yes) else skip to SH8b]
41. You indicated that, after [X date], someone from work made you uncomfortable, angry, or upset by making repeated sexual comments about your appearance or body.
Do you think they knew that you or someone else wanted them to stop? If
it happened more than once or by more than one person, do you think any
of them ever knew? SH7b

Survey Instrument

125

a. Yes 1
b. No 2
[If SH7b=2 (No) then skip to SH7d]
42. You indicated that, after [X date], someone from work made you uncomfortable, angry, or upset by making repeated sexual comments about your appearance or body.
Did they continue this unwanted behavior even after they knew that you
or someone else wanted them to stop? SH7c
a. Yes 1
b. No 2
[If SH7c=1 (Yes) then skip to SH8b, AND PerceivedHostileWorkEnvironment =
TRUE]
43. Do you think that this was ever severe enough that most [men/women] in
the military would have been offended if these remarks had been directed
to them? If you’re not sure, choose the best answer. SH7d
a. Yes 1
b. No 2
[If SH7d=1 (Yes) then PerceivedHostileWorkEnvironment = TRUE]
[Ask if SH8a=1 (Yes) else skip to SH9b]
44. You indicated that, after [X date], someone from work made you embarrassed,
angry, or upset by taking or sharing sexually suggestive pictures or videos of you
when you did not want them to.
Do you think that this was ever severe enough that most [men/women] in
the military would have been offended if it had happened to them? If you’re
not sure, please choose the best answer. SH8d
a. Yes 1
b. No 2
[If SH8d=1 (Yes) then PerceivedHostileWorkEnvironment = TRUE]
[Ask if SH9a=1 (Yes) else skip to SH11b]
45. You indicated that, after [X date], someone from work made you uncomfortable, angry, or upset by making repeated attempts to establish an unwanted
romantic or sexual relationship with you.

126

Sexual Assault and Sexual Harassment in the U.S. Military

Do you think they knew that you or someone else wanted them to stop? If
it happened more than once or by more than one person, do you think any
of them ever knew? SH9b
a. Yes 1
b. No 2
[If SH9b=2 (No) then skip to SH9d]
46. You indicated that, after [X date], someone from work made you uncomfortable, angry, or upset by making repeated attempts to establish an unwanted
romantic or sexual relationship with you.
Did they continue this unwanted behavior even after they knew that you
or someone else wanted them to stop? SH9c
a. Yes 1
b. No 2
[If SH9c=1 (Yes) then skip to SH11b, AND PerceivedHostileWorkEnvironment =
TRUE]
47. Do you think that this was ever severe enough that most [men/women] in
the military would have been offended by these unwanted attempts? If
you’re not sure, choose the best answer. SH9d
a. Yes 1
b. No 2
[If SH9d=1 (Yes) then PerceivedHostileWorkEnvironment = TRUE]
[Ask if SH11=1 (Yes) else skip to SH12a]
48. You indicated that, after [X date], someone from work made you uncomfortable, angry, or upset by touching you unnecessarily.
Do you think they knew that you or someone else wanted them to stop? If
it happened more than once or by more than one person, do you think any
of them ever knew? SH11b
a. Yes 1
b. No 2
[If SH11b=2 (No) then skip to SH11d]

Survey Instrument

127

49. You indicated that, after [X date], someone from work made you uncomfortable, angry, or upset by touching you unnecessarily.
Did they continue this unwanted behavior even after they knew that you
or someone else wanted them to stop? SH11c
a. Yes 1
b. No 2
[If SH11c=1 (Yes) then skip to SH12a, AND PerceivedHostileWorkEnvironment =
TRUE]
50. Do you think that this was ever severe enough that most [men/women] in
the military would have been offended by this unnecessary touching? If
you’re not sure, choose the best answer. SH11d
a. Yes 1
b. No 2
[If SH11d=1 (Yes) then PerceivedHostileWorkEnvironment = TRUE]
[Ask if SH12=1 (Yes) else skip to SH13a]
You indicated that, after [X date], someone from work made you feel as if you would
get some [If reservecomp = True, insert here: “military”] workplace benefit in exchange
for doing something sexual.
What led you to believe that you would get a [If reservecomp = True, insert here:
“military”] workplace benefit if you agreed to do something sexual? Select “Yes” or
“No” for each item
51. They told you that they would give you a reward
or benefit for doing something sexual. SH12a

Yes 1

No 2

52. They hinted that you would get a reward or benefit for doing something sexual. For example
they reminded you about your evaluation/fitness
report about the same time that they expressed
sexual interest. SH12b

Yes 1

No 2

53. Someone else told you they got benefits from this
person by doing sexual things SH12c

Yes 1

No 2

54. You heard rumors from other people that this
person treated others better in exchange for
doing sexual things. SH12d

Yes 1

No 2

55. Based on what you knew about their personality,
you thought you could get a benefit. SH12e

Yes 1

No 2

[If SH12a=1 (Yes) then PerceivedSexualQuidProQuo = TRUE]
[If SH12b=1 (Yes) then PerceivedSexualQuidProQuo = TRUE]
[If SH12c=1 (Yes) then PerceivedSexualQuidProQuo = TRUE]

128

Sexual Assault and Sexual Harassment in the U.S. Military

[Ask if SH13=1 (Yes) else skip to SH14a]
You indicated that, after [X date], someone from work made you feel as if you would
get punished or treated unfairly in the [If reservecomp = True, insert here: “military”]
workplace if you did not do something sexual.
What led you to believe that you would get punished or treated unfairly in the [If
reservecomp = True, insert here: “military”] workplace if you did not do something
sexual? Select “Yes” or “No” for each item
56. They told you that you would be punished or
treated unfairly if you did not do something
sexual. SH13a

Yes 1

No 2

57. They hinted that you would be punished or
treated unfairly if you did not do something
sexual. For example they reminded you about your
evaluation/fitness report near the same time that
they expressed sexual interest. SH13b

Yes 1

No 2

58. Someone else told you they were punished or
treated unfairly by this person for not doing something sexual. SH13c

Yes 1

No 2

59. You heard rumors from other people that this
person treated others badly unless they were willing to do sexual things. SH13d

Yes 1

No 2

60. Based on what you knew about their personality,
you thought you might get punished or treated
unfairly. SH13e

Yes 1

No 2

[If SH13a=1 (Yes) then PerceivedSexualQuidProQuo = TRUE]
[If SH13b=1 (Yes) then PerceivedSexualQuidProQuo = TRUE]
[If SH13c=1 (Yes) then PerceivedSexualQuidProQuo = TRUE]

[Ask if SH14=1 (Yes) else skip to SH15a]
You indicated that, after [X date], someone from work said that [men/women] are not
as good as [women/men] at your particular [If reservecomp = True, insert here: “military”] job, or that they should be prevented from having your job.
61. Do you think their beliefs about [men/women] ever harmed or limited your
[If reservecomp = True, insert here: “military job/”]career? For example, did
they hurt your evaluation/fitness report, affect your chances of promotion or
your next assignment? SH14a
a. Yes 1
b. No 2
[If SH14a=1 (Yes) then PerceivedGenderDiscrimination = TRUE]

Survey Instrument

129

[Ask if SH15=1 (Yes) else skip to next section]
You indicated that, after [X date], someone from work mistreated, ignored, excluded,
or insulted you because you are a [man/woman].
62. Do you think this treatment ever harmed or limited your [If reservecomp =
True, insert here: “military job/”]career? For example, did it hurt your evaluation/fitness report, affect your chances of promotion or your next assignment?
SH15a
a. Yes 1
b. No 2
[If SH15a=1 (Yes) then PerceivedGenderDiscrimination = TRUE]
[If any: SH1=1 (Yes) SH2=1 (Yes) SH3=1 (Yes) SH4=1 (Yes) SH5=1 (Yes) SH6=1 (Yes)
SH7=1 (Yes) SH8=1 (Yes) SH9=1 (Yes) SH10=1 (Yes) SH11=1 (Yes) SH12=1 (Yes)
SH13=1 (Yes) SH14=1 (Yes) SH15=1 (Yes) then continue. Else skip to SA1]
63. Earlier you answered questions about upsetting or offensive things that someone from work did since [X date]. How many of these behaviors that you
selected as happening to you, do you consider to have been sexual harassment? SHFU1
a. None were sexual harassment 1
b. Some were sexual harassment; some were not sexual harassment 2
c. All were sexual harassment 3
[If PerceivedSexualQuidProQuo = TRUE, OR PerceivedHostileWorkEnvironment =
TRUE, OR PerceivedGenderDiscrimination = TRUE then PerceivedEqualOpportunityViolation = TRUE]
[If PerceivedEqualOpportunityViolation ≠ TRUE), skip to SA1. Else continue.]
Based on your answers earlier, it appears that at least one person you worked with in the
last 12 months acted in a way that created an upsetting or offensive work environment.
The following section includes additional questions about the upsetting situation(s)
you experienced, including those situations in which someone from work:
[List all items below that were endorsed earlier (SH1-SH15)]
• Repeatedly told sexual jokes
• Repeatedly suggested that you do not act like a [man/woman] is supposed to

130

Sexual Assault and Sexual Harassment in the U.S. Military

• Repeatedly made sexual gestures or sexual body movements
• Displayed, showed you, or sent you sexually explicit materials like pictures or
videos
• Repeatedly told you about their sexual activities
• Repeatedly asked you questions about your sex life or sexual interests
• Made repeated sexual comments about your appearance or body
• Took or shared sexually suggestive pictures or videos of you
• Made repeated attempts to establish an unwanted romantic or sexual relationship
with you
• Touched you in a sexual way
• Touched you in any way that made you uncomfortable, angry, or upset
• Made you feel like you would get some workplace benefit in exchange for doing
something sexual
• Made you feel like you would get punished or treated unfairly if you refused to
do something sexual
• Said that [men/women] are not as good as [women/men] at your job, or that they
should be prevented from having that job.
• Mistreated, ignored, or insulted you because you were a [man/woman]
[Plural if more than one item endorsed SH1-SH15]
64. [Was this upsetting experience / Were these upsetting experiences] the result of
behavior by: SHFU2
a. One person 1
b. A group of people in the same setting 2
c. Different people in different situations 3
[If SHFU2=1 (One person) then SH_number = “Ind”]
[If SHFU2=2 (A group of people in the same setting) then SH_number = “Grp”]
[If SHFU2=3 (Different people in different situations) then ask]
65. You indicated that you had several situations like this happen to you. For the
next series of questions that refer to the “upsetting situation” please think about
the situation since [X date] that had the biggest effect on you – the one you consider to be the worst or the most serious.
Was the upsetting behavior in that situation due to: SHFU2_1
a. One person 1
b. A group of people 2

Survey Instrument

131

[If SHFU2_1=1 (One person) then SH_number = “Ind”]
[If SHFU2_1=2 (A group of people) then SH_number = “Grp”]
[If SH_number = “Ind” then ask]
66. Was this person a… SHFU3a
a. Man 1
b. Woman 2
[If SH_number = “Grp” then ask]
67. Were these people… SHFU3b
a. Men 1
b. Women 2
c. A mix of men and women 3
[If SH_number = “Ind” then ask]
68. Was this person…select one: SHFU4
a. One of your work supervisors or one of your unit leaders? 1
b. One of your peers at about the same level? 2
c. One of your subordinates or someone you managed? 3
[If SH_number = “Grp” then ask]
Were any of the individuals who acted this way:
69. One of your work supervisors or one of your unit
leaders? SHFU4a

Yes 1

No 2

70. One of your peers at about the same level?
SHFU4b

Yes 1

No 2

71. One of your subordinates or someone you managed? SHFU4c

Yes 1

No 2

[If SH_number = “Ind” then ask]
72. At the time of the upsetting situation was this person…Select one SHFU5
a. A Service member of higher rank than you? 1
b. A Service member of about the same rank as you? 2
c. A Service member of lower rank than you? 3

132

Sexual Assault and Sexual Harassment in the U.S. Military

d. A civilian / contractor working for the military? 4
e. Do not know 97
[If SH_number = “Grp” then ask]
At the time of the upsetting situation, were any of these individuals…
73. Service members of higher rank than you? SHFU5a

Yes 1

No 2

Do not
know 97

74. Service members of about the same rank as you?
SHFU5b

Yes 1

No 2

Do not
know 97

75. Service members of lower rank than you? SHFU5c

Yes 1

No 2

Do not
know 97

76. Civilians / contractors working for the military?
SHFU5d

Yes 1

No 2

Do not
know 97

77. Thinking about this situation, about how long did these upsetting behaviors continue? If the situation is still happening, indicate how long it has been
going on. SHFU6
a. It happened one time 1
b. About one week 2
c. About one month 3
d. A few months 4
e. A year or more 5
Thinking about this upsetting behavior:
78. Did it ever occur on a military installation/ship
[If reservecomp = True, insert here: “, armory, or
Reserve unit site”]? SHFU7a

Yes 1

No 2

79. Did it ever occur while you were on TDY/TAD, at
sea, or during field exercises/alerts? SHFU7b

Yes 1

No 2

80. Did it ever occur while you were deployed to a
combat zone or to an area where you drew imminent danger pay or hostile fire pay? SHFU7c

Yes 1

No 2

81. Did it ever occur during recruit training/basic
training? SHFU7d

Yes 1

No 2

82. Did it ever occur in a civilian location? SHFU7e

Yes 1

No 2

83. Did you request a transfer or other change of
assignment as a result of the situation? SHFU8a

Yes 1

No 2

84. Did it make you want to leave the military?
SHFU8b

Yes 1

No 2

Thinking about this upsetting situation:

Survey Instrument
85. Did this situation ever make it hard to do your job
or complete your work? SHFU8c

Yes 1

No 2

86. Did this situation ever make your workplace either
less productive or compromise your unit’s mission?
SHFU8d

Yes 1

No 2

87. Did you take a sick call day or any other type of
leave because of this situation? SHFU8e

Yes 1

No 2

88. Did this situation negatively affect your evaluation/fitness reports or promotions? SHFU8f

Yes 1

No 2

89. Did this situation either cause arguments in the
workplace or damage unit cohesion? SHFU8g

Yes 1

No 2

90. Did this situation damage your relationships with
coworkers? SHFU8h

Yes 1

No 2

91. Did this situation damage your other personal
relationships, for example, with your spouse or a
friend? SHFU8i

Yes 1

No 2

92. Did you discuss this situation with your friends,
family, or co-workers? SHFU9a

Yes 1

No 2

93. Did you discuss this situation with a chaplain,
counselor, or medical person? SHFU9b

Yes 1

No 2

94. Did you discuss this situation with a work supervisor or anyone up your chain of command? SHFU9c

Yes 1

No 2

95. Did you officially report this situation as possible
harassment to any person tasked with enforcing
sexual harassment or Equal Opportunity regulations? SHFU9d

Yes 1

No 2

133

Thinking about this upsetting situation:

[IF SHFU9c=1 (Yes) OR SHFU9d=1 (Yes) continue. Else skip to SHFU12]

[Use “reporting the situation” if SHFU9d=1 (Yes).]
[Person vs People is based on SH_number Ind vs Grp]
What actions were taken in response to your [discussing the situation with a supervisor or anyone up your chain of command/reporting the situation]? Select “Yes,” “No,”
or “Do not know” for each item.
96. No action was taken because you asked for the
discussion to be kept private. SHFU10a

Yes 1

No 2

Do not
know 97

97. You discussed the situation, but no action was
taken because you chose not to give enough
details about the situation. SHFU10b

Yes 1

No 2

Do not
know 97

98. The person you told took no action. SHFU10c

Yes 1

No 2

Do not
know 97

134

Sexual Assault and Sexual Harassment in the U.S. Military
99. The rules on harassment were explained to everyone in the workplace. SHFU10d

Yes 1

No 2

Do not
know 97

100. Someone talked to the [person/people] to ask
them to change their behavior. SHFU10e

Yes 1

No 2

Do not
know 97

101. Your work station or duties were changed to help
you avoid [that person/those people]. SHFU10f

Yes 1

No 2

Do not
know 97

102. The [person was, people were] moved or reassigned so that you did not have as much contact
with them. SHFU10g

Yes 1

No 2

Do not
know 97

103. There was some official career action taken
against [the person/the people] for their upsetting
behavior (for example, a negative evaluation/fitness report). SHFU10h

Yes 1

No 2

Do not
know 97

104. The [person/people] stopped their upsetting
behavior. SHFU10i

Yes 1

No 2

Do not
know 97

105. You were encouraged to drop the issue. SHFU10j

Yes 1

No 2

Do not
know 97

106. You were discouraged from filing a formal complaint. SHFU10k

Yes 1

No 2

Do not
know 97

107. The [person/people] who did this retaliated
against you for complaining. For example, their
upsetting behavior became worse or they threatened you. SHFU10l

Yes 1

No 2

Do not
know 97

108. Your coworkers treated you worse, avoided you,
or blamed you for the problem. SHFU10m

Yes 1

No 2

Do not
know 97

109. Your supervisor punished you for bringing it up
(e.g., loss of privileges, denied promotion/training, transferred to less favorable job). SHFU10n

Yes 1

No 2

Do not
know 97

How satisfied were/are you with the following aspects of how the discussion or
report was handled? Select one answer for each item.
110. Availability of information
about how to file a complaint SHFU11a

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

111. How you were treated by
personnel handling your
situation SHFU11b

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

112. The action taken by the
personnel handling your
situation SHFU11c

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

Survey Instrument

135

113. The current status of the
situation SHFU11d

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

114. Amount of time it took
to address your situation
SHFU11e

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

115. Availability of information
Very
or updates on the status of
dissatisfied
your report or complaint
1
SHFU11f

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

[IF SHFU9c=1 (Yes) OR SHFU9d=1 (Yes) skip to SA1, else continue]
116. W hat were your reasons for not discussing it with someone above you in
your chain of command and not reporting it to a person who enforces
sexual harassment regulations? Select all the reasons that apply to you.
a. The offensive behavior stopped on its own. SHFU12a
b. Someone else already reported it. SHFU12b
c. You thought it was not serious enough to report. SHFU12c
d. You did not want more people to know. SHFU12d
e. You did not want people to see you as weak. SHFU12e
f. You did not want people to think you were gay/lesbian/bisexual/transgender. SHFU12f
g. You wanted to forget about it and move on. SHFU12g
h. You did not know how to report it. SHFU12h
i. Someone told you not to report it. SHFU12i
j. You did not think anything would be done. SHFU12j
k. You did not think you would be believed. SHFU12k
l. You did not trust that the process would be fair. SHFU12l
m. You felt partially to blame. SHFU12m
n. You thought other people would blame you. SHFU12n
o. You thought you might get in trouble for something you did. SHFU12o
p. You thought a supervisor would make too big of a deal out of it. SHFU12p
q. You thought you might be labeled as a troublemaker. SHFU12q
r. You thought it might hurt your performance evaluation/fitness report.
SHFU12r
s. You thought it might hurt your career. SHFU12s
t. You did not want to hurt the person’s career or family. SHFU12t
u. You were worried about retaliation by the person(s) who did it. SHFU12u
v. You were worried about retaliation by a supervisor or someone in your chain
of command. SHFU12v

136

Sexual Assault and Sexual Harassment in the U.S. Military

w. You were worried about retaliation by your military co-workers or peers.
SHFU12w
x. You took other actions to handle the situation. SHFU12x
Please read the following special instructions before continuing the survey.
Questions in this next section ask about unwanted experiences of an abusive, humiliating, or sexual nature. These types of unwanted experiences vary in severity. Some of
them could be viewed as an assault. Others could be viewed as hazing or some other
type of unwanted experience.
They can happen to both women and men.
Some of the language may seem graphic, but using the names of specific body parts is
the best way to determine whether or not people have had these types of experiences.
When answering these questions, please include experiences no matter who did it to
you or where it happened. It could be done to you by a male or female, Service member
or civilian, someone you knew or a stranger.
Please include experiences even if you or others had been drinking alcohol, using drugs,
or were intoxicated.
The following questions will ask you about events that happened AFTER [X date].
You will have an opportunity to describe experiences that happened BEFORE [X date]
later in the survey.
Remember, all the information you share will be kept confidential. RAND will not
give your identifiable answers to the DoD.
<>
117. Since [X Date], did you have any unwanted experiences in which someone put his penis into your [If Intro1=2 (Female), display: “vagina,”] anus or
mouth? SA1
a. Yes 1
b. No 2
[If SA1=1 (Yes) ask “OB1a”, else continue]
118. Since [X Date], did you have any unwanted experiences in which someone
put any object or any body part other than a penis into your [If Intro1=2
(Female), display: “vagina,”] anus or mouth? The body part could include a
finger, tongue or testicles. SA2

Survey Instrument

137

a. Yes 1
b. No 2
[If SA2=1 (Yes) and sexualAssault_12m ≠ “True”, ask “PF2a”, else continue]
119. Since [X Date], did anyone make you put any part of your body or any
object into someone’s mouth, vagina, or anus when you did not want to?
A part of the body could include your [If Intro1=1 (Male) display: “penis, testicles,”] tongue or fingers. SA3
a. Yes 1
b. No 2
[If SA3=1 (Yes) and sexualAssault_12m ≠ “True”, ask “PF3a”, else continue]
[Programming note: If sexualAssault_12m = “TRUE” on the basis of follow ups to
SA1-SA3 then penetrativeSA_12m = “TRUE” else penetrativeSA_12m = “FALSE”]
120. Since [X Date], did you have any unwanted experiences in which someone intentionally touched private areas of your body (either directly or
through clothing)?
Private areas include buttocks, inner thigh, breasts, groin, anus, vagina, penis,
or testicles. SA4
a. Yes 1
b. No 2
[If SA4=1 (Yes) and sexualAssault_12m ≠ “True”, ask “PF4a”, else continue]
121. Since [X Date], did you have any unwanted experiences in which someone
made you touch private areas of their body or someone else’s body (either
directly or through clothing)? This could involve the person putting their
private areas on you.
Private areas include buttocks, inner thigh, breasts, groin, anus, vagina, penis
or testicles. SA5
a. Yes 1
b. No 2
[If SA5=1 (Yes) and sexualAssault_12m ≠ “True”, ask “PF5a”, else continue]
[Programming note: If SexualAssault_12m = “TRUE” on the basis of follow ups to
SA4-SA5 then contactSA _12m = “TRUE” else contactSA_12m= “FALSE”]

138

Sexual Assault and Sexual Harassment in the U.S. Military

122. Since [X Date], did you have any unwanted experiences in which someone
attempted to put a penis, an object, or any body part into your [If Intro1=2
(Female), display: “vagina,”] anus or mouth, but no penetration actually
occurred? SA6
a. Yes 1
b. No 2
[If SA6=2 (No) skip to SAFU1]
123. As part of this attempt, did the person touch you anywhere on your body?
This includes grabbing your arm, hair or clothes, or pushing their body against
yours. SA6a
a. Yes 1
b. No 2
[If sexualAssault_12m ≠ “True”, ask “PF6a”, else continue]
[Programming note: If sexualAssault_12m = “TRUE” on the basis of follow ups to
SA6 then attemptedSA_12m = “TRUE” else attemptedSA_12m = “FALSE”]
[Purpose Follow Up Module START]
[Purpose Follow up module: “X” in the question number refers to appropriate SA
screener number (2-6)]
124. Was this unwanted experience (or any experiences like this if you had
more than one) abusive or humiliating, or intended to be abusive or humiliating? If you aren’t sure, choose the best answer. PFXa
a. Yes 1
b. No 2
[If PFXa=1 (Yes) skip to OBX item]
125. Do you believe the person did it for a sexual reason? For example, they did it
because they were sexually aroused or to get sexually aroused. If you aren’t sure,
choose the best answer. PFXb
a. Yes 1
b. No 2
[If PFXb=1 (Yes) continue to OBX item]
[If PFXb=2 (No) skip to next SA_screener question (SA3 –SA6)]

Survey Instrument

139

[Purpose Follow Up Module END]
[Offender Behavior Module START]
[Offender Behavior Module: “X” in the question number refers to appropriate screener
number (1-6)]
[Programming Note: Please keep the instructions on the screen for all items OBXaOBXk. Please present each item alone on the screen]
The following statements are about things that might have happened to you when you
had this experience. In these statements, ‘they’ means the person or people who did
this to you.
Please indicate which of the following happened.
126. They continued even when you told them or showed them that you were
unwilling. OBXa
a. Yes 1
b. No 2
[If OBXa=1 (Yes) sexualAssault_12m= “TRUE”]
127. They used physical force to make you comply. For example, they grabbed
your arm or used their body weight to hold you down. OBXb
a. Yes 1
b. No 2
[If OBXb=1 (Yes) sexualAssault_12m= “TRUE”]
128. They physically injured you. OBXc
a. Yes 1
b. No 2
[If OBXc=1 (Yes) sexualAssault_12m= “TRUE”]
129. They threatened to physically hurt you (or someone else). OBXd
a. Yes 1
b. No 2
[If OBXd=1 (Yes) sexualAssault_12m= “TRUE”]

140

Sexual Assault and Sexual Harassment in the U.S. Military

[IF OBXd=1 (Yes) then ask]
130. Did they threaten you (or someone else) with a weapon? OBXd_1
a. Yes 1
b. No 2
[IF OBXd=1 (Yes) then ask]
131. Did they threaten to seriously injure, kill, or kidnap you (or someone else)?
OBXd_2
a. Yes 1
b. No 2
132. They threatened you (or someone else) in some other way. For example, by
using their position of authority, by spreading lies about you, or by getting you
in trouble with authorities. OBXe
a. Yes 1
b. No 2
[If OBXe=1 (Yes) sexualAssault_12m= “TRUE”]
133. They did it when you were passed out, asleep, or unconscious. OBXf
a. Yes 1
b. No 2
[If OBXf=1 (Yes) sexualAssault_12m= “TRUE”]
134. They did it when you were so drunk, high, or drugged that you could not
understand what was happening or could not show them that you were
unwilling. OBXg
a. Yes 1
b. No 2
[If OBXg=1 (Yes) sexualAssault_12m= “TRUE”]
135. They tricked you into thinking that they were someone else or that they
were allowed to do it for a professional purpose (like a person pretending
to be a doctor). OBXh
a. Yes 1
b. No 2

Survey Instrument

141

[If OBXh=1 (Yes) sexualAssault_12m= “TRUE”]
[If sexualAssault_12m = TRUE, then skip to next screening item SA2-SA6, Else
continue.]
136. They made you so afraid that you froze and could not tell them or show
them that you were unwilling. OBXi
a. Yes 1
b. No 2
[If OBXi=1 (Yes) sexualAssault_12m= “TRUE”]
137. They did it after you had consumed so much alcohol that the next day you
could not remember what happened. OBXj
a. Yes 1
b. No 2
[If OBXj=1 (Yes) sexualAssault_12m= “TRUE”]
138. It happened without your consent. OBXk
a. Yes 1
b. No 2
[If OBXk=1 (Yes) sexualAssault_12m= “TRUE”]
[If SA2=1 (Yes) AND (any item OB2a-OB2k=1 (Yes)), then O2. Else continue]
O2. In the previous item, you indicated that this unwanted experience involved someone putting an object or part of the body into your [If Intro1=2 (Female), display:
“vagina, or”] anus or mouth. Did it include putting the body part or object into
[If Intro1=2 (Female), display:]
139. Your vagina? O2a
a. Yes 1
b. No 2
140. Your anus? O2b
a. Yes 1
b. No 2

142

Sexual Assault and Sexual Harassment in the U.S. Military

141. Your mouth? O2c
a. Yes 1
b. No 2
[Offender Behavior Module END]
[After OB6 series questions continue to SAFU1. After OB1-OB5, continue to next
screening item SA2-SA6]
[If SexualAssault_12m = “TRUE” then ask SAFU1, else skip to Prior_SA]
Thank you for answering the questions so far. Remember that your answers are
confidential.
Based on your answers earlier, you indicated that you had at least one of these unwanted
experiences since [X date]. The items that follow will ask for additional information
about the unwanted event(s) in which someone:
[Present any endorsed items SA1-SA6]
• Put their penis into your [If Intro1=2 (Female), display: “vagina,”] anus or mouth.
• Put any object or any body part other than a penis into your [If Intro1=2 (Female),
display: “vagina,”] anus or mouth.
• Made you put any part of your body or any object into someone’s mouth, vagina
or anus
• Intentionally touched private areas of your body.
• Made you touch private areas of their body or someone else’s body.
• Attempted to put a penis, an object, or any body part into your [If Intro1=2
(Female), display: “vagina,”] anus or mouth, but no penetration actually occurred.
142. Please give your best estimate of how many different times (on how many
separate occasions) during the past 12 months, you had these unwanted
experiences? SAFU1
a. 1 time since [X date] 1
b. 2 times since [X date] 2
c. 3 times since [X date] 3
d. 4 times since [X date] 4
e. 5 or more times since [X date] 5
f. More than once, but not sure the number of times it happened since [X
date] 6

Survey Instrument

143

[times=’missing’. If SAFU1=1 (Yes) then times=’one’. If SAFU1=(2 or 3 or 4 or 5 or 6),
then times=’multiple’.]
[If times=one, skip to SAFU5, else continue]
You indicated that you had more than one unwanted event since [X date].
143. Were all these events done by the same person? SAFU2
a. Yes 1
b. No, more than one person. 2
c. Not sure 3
The following questions ask about the unwanted event that had the biggest effect on
you. Before you continue, please choose the one unwanted event since [X date] that
you consider to be the worst or most serious.
[If more than one endorsed SA screener (SA1-SA6), continue. Else skip to SAFU5]
[Programming pulls in any SA screening item to which the respondent answered ‘yes’]
Which of the following experiences happened during the event you chose as the
worst or most serious? Select “Yes” or “No” for each item
144. Put their penis into your [If Intro1=2 (Female), display: “vagina,”] anus or mouth? SAFU3a

Yes 1

No 2

145. Put any object or any body part other than a penis
into your [If Intro1=2 (Female), display: “vagina,”]
anus or mouth? SAFU3b

Yes 1

No 2

146. Made you put any part of your body or any object
into someone’s mouth, vagina or anus? SAFU3c

Yes 1

No 2

147. Intentionally touched private areas of your body?
SAFU3d

Yes 1

No 2

148. Made you touch private areas of their body or
someone else’s body? SAFU3e

Yes 1

No 2

149. Attempted to put a penis, an object, or any body
part into your [If Intro1=2 (Female), display:
“vagina,”] anus or mouth, but no penetration
actually occurred? SAFU3f

Yes 1

No 2

[If the behavior that caused sexualAssault12_m to equal TRUE is not selected, then
continue, else skip to SAFU5]

144

Sexual Assault and Sexual Harassment in the U.S. Military

Earlier in the survey you answered questions about several events. Please focus on the
worst or most serious event in the questions that follow.
[Show items as a table. A-H can be shown on a single screen or blocked into sets based
on Westat’s recommendation. I-K need to be on their own screen]
Please indicate which of the following happened to you when you had this most
serious experience. In these statements, ‘they’ means the person or people that did
this to you.
150. They continued even when you told them or
showed them that you were unwilling. SAFU4a

Yes 1

No 2

151. They used physical force to make you comply. For
example, they grabbed your arm, or used their
body weight to hold you down. SAFU4b

Yes 1

No 2

152. They physically injured you. SAFU4c

Yes 1

No 2

153. They threatened to physically hurt you (or someone else). SAFU4d

Yes 1

No 2

154. They threatened you (or someone else) in some
other way. For example, by using their position of
authority, by spreading lies about you, or by getting you in trouble with authorities. SAFU4e

Yes 1

No 2

155. They did it when you were passed out, asleep, or
unconscious. SAFU4f

Yes 1

No 2

156. They did it when you were so drunk, high, or
drugged that you could not understand what was
happening or could not show them that you were
unwilling. SAFU4g

Yes 1

No 2

157. They tricked you into thinking that they were
someone else or that they were allowed to do it
for a professional purpose (like a person pretending to be a doctor). SAFU4h

Yes 1

No 2

[If SAFU4a – SAFU4h=2 (No) then continue. Else skip to SAFU5]
158. They made you so afraid that you froze and could
not tell them or show them that you were unwilling. SAFU4i

Yes 1

No 2

159. They did it after you had consumed so much alcohol that the next day you could not remember all
or part of what happened. SAFU4j

Yes 1

No 2

160. It happened without your consent. SAFU4k

Yes 1

No 2

[If times=one, ask to SAFU5, else skip to SAFU6]

Survey Instrument

145

161. How many people did this to you? SAFU5
a. One person 1
b. More than one person 2
c. Not sure 3
[If SAFU1=(2 3 4 5 or 6) (more than one time) and SAFU2=1 (Yes)
OR
SAFU1=1 (One time) and SAFU5=1 (One person)
then SA_offenders= “one”
Else if SAFU1=(2 3 4 5 or 6) (more than one time) and SAFU2=2 (No, more than one
person)
OR
SAFU1=1 (One time) and SAFU5=2 (More than one person)
then SA_offenders= “multiple”]
[If SA_offenders=”one”, then ask]
Please continue to focus on this worst or most serious event in the questions that follow.
162. Was the person… SAFU6a
a. A man 1
b. A woman 2
c. Not sure 3
[If SA_offenders=”multiple”, then ask]
Please continue to focus on this worst or most serious event in the questions that follow.
163. Were these people… SAFU6b
a. Men 1
b. Women 2
c. A mix of men and women 3
d. Not sure 4
[Singular vs plural in question text and response is based on one vs multiple SA_offenders]

146

Sexual Assault and Sexual Harassment in the U.S. Military

164. At the time of the event, did you know or had you previously met [the
person / any of the people] who did this to you? SAFU7.
a. Yes 1
b. No, [it was / they were] [a stranger / all strangers] 2
[If SAFU7=2 (No, [it was / they were] [a stranger / strangers]) skip to SAFU8g, else
continue]
[Singular vs plural in question text and response is based on one vs multiple SA_offenders]
165. At the time of the event, [was the person / were the people] who did this to
you…
a. Your spouse? SAFU8a
b. Your boyfriend or girlfriend? SAFU8b
c. Someone you had divorced or broken up with? SAFU8c
d. A friend or acquaintance? SAFU8d
e. Someone who you have a child with (your child’s mother or father)? SAFU8e
f. A family member or relative? SAFU8f
g. None of the above. SAFU8l
[Singular vs plural in question text and response is based on one vs multiple SA_offenders]
At the time of the event, [was the person / were the people] who did this to you…
166. Someone in the military? SAFU8g

Yes 1

No 2

Do not
know 97

167. Civilian employee(s) or contractor(s) working for
the military? SAFU8h

Yes 1

No 2

Do not
know 97

168. Person(s) in the local community? SAFU8i

Yes 1

No 2

Do not
know 97

169. A foreign national? SAFU8j

Yes 1

No 2

Do not
know 97

170. [If reservecomp = True, ask this question]: Someone you work with in your civilian job or school?
SAFU8k

Yes 1

No 2

Do not
know 97

[If SAFU8g=1 (Yes) then ask SAFU9. Else skip to SAFU10]

You said a person / people in the military did this to you. Were any of them:
171. A lower rank than you? SAFU9a

Yes 1

No 2

Do not
know 97

Survey Instrument

172. A similar rank as you? SAFU9b

Yes 1

No 2

Do not
know 97

173. A higher rank than you? SAFU9c

Yes 1

No 2

Do not
know 97

174. Officers? SAFU9d

Yes 1

No 2

Do not
know 97

147

[If SAFU9c=1 (Yes) then continue, else skip to SAFU10]

175. Was the higher ranked person your unit leader or someone above them in
your chain of command? SAFU9e
a. Yes 1
b. No 2
c. Do not know 97
Did the unwanted event occur... Select “Yes” or “No” for each item.
176. At a military installation/ship [If reservecomp
= True insert: “, armory, or Reserve unit site”]?
SAFU10a

Yes 1

No 2

177. During your [work day/duty hours / If reservecomp
= True, replace with: “National Guard or Reserve
duties, including active duty, active duty for training, full-time National Guard duty, and performing in active duty training/drills”]? SAFU10b

Yes 1

No 2

178. While you were on TDY/TAD, at sea, or during
field exercises/alerts? SAFU10c

Yes 1

No 2

179. While you were deployed to a combat zone or to
an area where you drew imminent danger pay or
hostile fire pay? SAFU10d

Yes 1

No 2

180. While you were in a delayed entry program?
SAFU10e

Yes 1

No 2

181. While you were in recruit training/basic training?
SAFU10f

Yes 1

No 2

182. While you were in any type of military combat
training? SAFU10g

Yes 1

No 2

183. While you were in Officer Candidate or Training
School/Basic or Advanced Officer Course? SAFU10h

Yes 1

No 2

184. While you were completing military occupational
specialty school/technical training/advanced individual training/professional military education?
SAFU10i

Yes 1

No 2

185. W hich of the following best describe the situation when this unwanted
event occurred? Select all that apply
a. You were out with friends or at a party. SAFU11a

148

Sexual Assault and Sexual Harassment in the U.S. Military

b.
c.
d.
e.
f.
g.
h.
i.
j.
k.

You were on a date. SAFU11b
You were being intimate with the other person. SAFU11c
You were at work. SAFU11d
You were alone in a public place. SAFU11e
You were in your home or quarters. SAFU11f
You were in someone else’s home or quarters. SAFU11g
You were at a military function. SAFU11h
You were in temporary lodging/hotel. SAFU11i
None of the above. SAFU11j
Do not recall. SAFU11k

186. Would you describe this unwanted event as hazing? Hazing refers to things
done to humiliate or “toughen up” people prior to accepting them into a group.
SAFU12
a. Yes 1
b. No 2
Did the offender(s)...
187. Sexually harass you before the situation? SAFU13a

Yes 1

No 2

188. Stalk you before the situation? SAFU13b

Yes 1

No 2

189. Sexually harass you after the situation? SAFU13c

Yes 1

No 2

190. Stalk you after the situation? SAFU13d

Yes 1

No 2

191. At the time of this unwanted event had you been drinking alcohol? Even if
you had been drinking, it does not mean that you are to blame for what happened. SAFU14
a. Yes 1
b. No 2
c. Not sure 3
[If SAFU14=1 (Yes)]

192. Just prior to this unwanted event, did the person(s) who did this to you
buy or give you alcohol to drink? SAFU15
a. Yes 1
b. No 2
c. Do not know 97

Survey Instrument

149

193. Just prior to this unwanted event, do you think that you might have been
given a drug without your knowledge or consent? SAFU16
a. Yes 1
b. No 2
c. Do not know 97
194. At the time of this unwanted event, had the person(s) who did it been
drinking alcohol? SAFU17
a. Yes 1
b. No 2
c. Do not know 97
After this unwanted event:
195. Did you request a transfer or other change of
assignment as a result of the event? SAFU18a

Yes 1

No 2

196. Did it make you want to leave the [military / If
reservecomp = True, replace with: “leave the
Guard or Reserve, or transfer to another Reserve
component”]? SAFU18b

Yes 1

No 2

197. Did it ever make it hard to do your job or complete your work? SAFU18c

Yes 1

No 2

198. Did you take a sick day or any other type of leave
because of the event? SAFU18d

Yes 1

No 2

199. Did the event damage your personal relationships, for example with your spouse or a friend?
SAFU18e

Yes 1

No 2

200. Did you tell anyone about the unwanted event? SAFU19
a. Yes 1
b. No 2
[IF SAFU19=2 (No) SKIP to SAFU30]
Who have you talked to about this event?
201. A friend or family member SAFU20a

Yes 1

No 2

202. The Sexual Assault Response Coordinator (SARC)
SAFU20b

Yes 1

No 2

203. A Sexual Assault Prevention and Response Victim
Advocate (SAPR VA or VA) SAFU20c

Yes 1

No 2

204. A Safe Helpline or other hotline counselor
SAFU20d

Yes 1

No 2

150

Sexual Assault and Sexual Harassment in the U.S. Military
205. A medical professional (for example, a doctor or
nurse) SAFU20e

Yes 1

No 2

206. A chaplain or religious leader SAFU20f

Yes 1

No 2

207. A counselor, therapist, or psychologist SAFU20g

Yes 1

No 2

208. A Special Victims’ Counsel or Victims’ Legal
Counsel SAFU20h

Yes 1

No 2

209. Some other military lawyer (for example, a JAG
officer) SAFU20i

Yes 1

No 2

210. A supervisor or someone above you in your chain
of command SAFU20j

Yes 1

No 2

211. An officer or NCO outside of your chain of command SAFU20k

Yes 1

No 2

212. Someone in military law enforcement SAFU20l

Yes 1

No 2

213. Civilian law enforcement authority SAFU20m

Yes 1

No 2

214. Civilian rape crisis center or other sexual assault
advocacy group SAFU20n

Yes 1

No 2

[If penetrativeSA_12m = True then continue, else skip to ask SAFU22b]
215. Did you receive a sexual assault forensic exam or “rape exam”? This is often
given by medical personnel to collect evidence about a sexual assault and could
be either civilian or military. SAFU21
a. Yes 1
b. No 2
[If SAFU20b=1 (Yes) then ask SAFU22b]
[If SAFU20c=1 (Yes) then ask SAFU22c]
[If SAFU20d=1 (Yes) then ask SAFU22d]
[If SAFU20e=1 (Yes) then ask SAFU22e]
[If SAFU20f=1 (Yes) then ask SAFU22f]
[If SAFU20g=1 (Yes) then ask SAFU22g]
[If SAFU20h=1 (Yes) then ask SAFU22h]
[If SAFU20i=1 (Yes) then ask SAFU22i]
[If SAFU20j=1 (Yes) then ask SAFU22j]
[If SAFU20k=1 (Yes) then ask SAFU22k]
[If SAFU20l=1 (Yes) then ask SAFU22l]

Survey Instrument

151

[If SAFU20m=1 (Yes) then ask SAFU22m]
[If SAFU20n=1 (Yes) then ask SAFU22n]
[If SAFU21=1 (Yes) then ask SAFU22o]
How satisfied have you been with your treatment by:
216. The Sexual Assault
Response Coordinator
(SARC)? SAFU22b

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

217. Sexual Assault Prevention and Response Victim
Advocates (SAPR VA or
VA)? SAFU22c

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

Very
218. Safe Helpline or other hotdissatisfied
line counselors? SAFU22d
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

219. Medical professional(s)?
For example, a doctor or
nurse. SAFU22e

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

220. Chaplains or religious
leaders? SAFU22f

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

221. Counselors, therapists or
psychologists? SAFU22g

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

222. Special Victims’ Counsel
or Victims’ Legal Counsel?
SAFU22h

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

Very
223. O ther military lawyers (for
example, a JAG officer)?
dissatisfied
SAFU22i
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

224. Supervisors or people in
your chain of command?
SAFU22j

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

225. Officers or NCOs outside
of your chain of command? SAFU22k

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

226. Military law enforcement
personnel? SAFU22l

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

152

Sexual Assault and Sexual Harassment in the U.S. Military

Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

228. Civilian rape crisis center
Very
or other sexual assault
dissatisfied
advocacy group? SAFU22n
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

Very
229. The sexual assault forensic
dissatisfied
exam? SAFU22o
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very
satisfied
5

227. Civilian law enforcement
personnel? SAFU22m

[Keep this definition on screen for items SAFU23, SAFU24, and SAFU25]
DoD provides two types of sexual assault reports.
• Restricted reports allow people to get information, collect evidence, and receive
medical treatment and counseling without starting an official investigation of the
assault.
• Unrestricted reports start an official investigation in addition to allowing the services available in restricted reporting.
230. Did you officially report this unwanted event to the military? This could
have been either a restricted or unrestricted report. SAFU23
a. Yes 1
b. No 2
[If SAFU23=2 (No) skip to SAFU30]
231. Did you make... Select one SAFU24
a. Only a restricted report? 1
b. Only an unrestricted report? 2
c. A restricted report that turned into an unrestricted report? 3
d. Or were you not sure what type of report it was? 4
[If SAFU24=1 (Only a restricted report?) skip to SAFU29]
[If SAFU24=4 (Not sure what type of report it was), skip to SAFU26]
232. Was an unrestricted report what you preferred? SAFU25
a. Yes, that’s what you wanted. 1

Survey Instrument

153

b. No, you wanted a restricted report, but could not. For example, someone
you talked to filed the unrestricted report. 2
We have a few questions about the status of the case you reported.
233. Have military police or criminal investigators interviewed you about the
case? SAFU26.
a. Yes 1
b. No 2
c. Do not know 97
234. Has a suspect been arrested or charged with a crime? SAFU27a
a. Yes 1
b. No 2
c. Do not know 97
[If SAFU27a=2 or 97, skip to SAFU27d]
[If SAFU27a=1 (Yes)]
235. Were they arrested or charged in… Select one SAFU27b
a. The military justice system? 1
b. The civilian justice system? 2
c. Do not know 97
[If SAFU27a=1 (Yes)]
236. To the best of your knowledge, which one of the following best describes
the current status of the case? Select one SAFU27c
a. The legal case was dropped because the prosecutor or a Commander decided
not to pursue it. 1
b. The legal case was dropped because you chose not to pursue it. 2
c. A defendant has already pled guilty. 3
d. A defendant was convicted of the crime at a trial or court martial. 4
e. A defendant was tried at a trial or court martial but acquitted. 5
f. The trial or court martial has not happened yet, but it is supposed to happen
in the future. 6
g. Do not know/None of the above. 97
237. Have they been punished through an administrative action, for example,
discharged from the military or reduced in rank? SAFU27d

154

Sexual Assault and Sexual Harassment in the U.S. Military

a. Yes 1
b. No 2
c. Do not know 97
238. To the best of your knowledge, are the military police or criminal investigators still investigating the case? SAFU27e
a. Yes 1
b. No 2
c. Do not know 97
[If SAFU27a=1 (Yes) then ask SAFU28b and SAFU28c]
Based on your experience with this case...
239. How satisfied are
Neither
you with the way you
Very
Dissatisfied satisfied nor Satisfied
have been treated by dissatisfied
2
dissatisfied
4
the criminal justice
1
3
system? SAFU28a

Very
satisfied
5

Not
Applicable
6

240. How satisfied have
you been with your
treatment by the
Neither
offender(s)’ lawyer?
Very
Dissatisfied satisfied nor Satisfied
For example, the
dissatisfied
2
dissatisfied
4
Military Trial Defense
1
3
Office or a civilian
defense attorney.
SAFU28b

Very
satisfied
5

Not
Applicable
6

241. How satisfied have
you been with your
treatment by the legal
Neither
Very
prosecutor? For examDissatisfied satisfied nor Satisfied
dissatisfied
ple the Military Legal
2
dissatisfied
4
1
Office (prosecution)
3
personnel or a civilian
prosecutor. SAFU28c

Very
satisfied
5

Not
Applicable
6

242. What were your reasons for reporting the event to a military authority?
Select all the reasons that apply to you.
a. Someone else made you report it or reported it themselves. SAFU29a
b. To stop the offender(s) from hurting you again. SAFU29b
c. To stop the offender(s) from hurting others. SAFU29c
d. It was your civic/military duty to report it. SAFU29d
e. To punish the offender(s). SAFU29e
f. To discourage other potential offenders. SAFU29f
g. To get medical assistance. SAFU29g
h. To get mental health assistance. SAFU29h
i. To stop rumors. SAFU29i

Survey Instrument

155

j. Someone you told encouraged you to report. SAFU29j
k. You wanted to document the incident, so that you could get help or benefits
from the Department of Veterans Affairs (VA) in the future. SAFU29k
[All respondents who see SAFU29, skip to SAFU32]
243. What were your reasons for not reporting the event to a military authority? Select all the reasons that apply to you.
a. You reported it to civilian authorities / law enforcement. SAFU30a
b. Someone else already reported it. SAFU30b
c. You thought it was not serious enough to report. SAFU30c
d. You did not want more people to know. SAFU30d
e. You did not want people to see you as weak. SAFU30e
f. You did not want people to think you were gay/lesbian/bisexual/transgender. SAFU30f
g. You wanted to forget about it and move on. SAFU30g
h. You did not know how to report it. SAFU30h
i. Someone told you not to report it. SAFU30i
j. You did not think your report would be kept confidential. SAFU30j
k. You did not think anything would be done. SAFU30k
l. You did not think you would be believed. SAFU30l
m. You did not trust the process would be fair. SAFU30m
n. You felt partially to blame. SAFU30n
o. You thought other people would blame you. SAFU30o
p. You thought you might get in trouble for something you did (for example,
underage drinking or fraternization). SAFU30p
q. You thought you might be labeled as a troublemaker. SAFU30q
r. You thought it might hurt your performance evaluation/fitness report.
SAFU30r
s. You thought it might hurt your career. SAFU30s
t. You did not want to hurt the person’s career or family. SAFU30t
u. You were worried about retaliation by the person(s) who did it. SAFU30u
v. You were worried about retaliation by your supervisor or someone in your
chain of command. SAFU30v
w. You were worried about retaliation by your military co-workers or peers.
SAFU30w
x. You took other actions to handle the situation. SAFU30x
[If more than one response SAFU30a-x is selected, continue. Else skip to SAFU32].

156

Sexual Assault and Sexual Harassment in the U.S. Military

244. Which was the main reason for not reporting the event? Choose one. [Display all endorsed probes from SAFU30]. SAFU31
a. You reported it to civilian authorities / law enforcement. 1
b. Someone else already reported it. 2
c. You thought it was not serious enough to report. 3
d. You did not want more people to know. 4
e. You did not want people to see you as weak. 5
f. You did not want people to think you were gay/lesbian/bisexual/transgender. 6
g. You wanted to forget about it and move on. 7
h. You did not know how to report it. 8
i. Someone told you not to report it. 9
j. You did not think your report would be kept confidential. 10
k. You did not think anything would be done. 11
l. You did not think you would be believed. 12
m. You did not trust the process would be fair. 13
n. You felt partially to blame. 14
o. You thought other people would blame you. 15
p. You thought you might get in trouble for something you did (for example,
underage drinking or fraternization). 16
q. You thought you might be labeled as a troublemaker. 17
r. You thought it might hurt your performance evaluation/fitness report. 18
s. You thought it might hurt your career. 19
t. You did not want to hurt the person’s career or family. 20
u. You were worried about retaliation by the person(s) who did it. 21
v. You were worried about retaliation by your supervisor or someone in your
chain of command. 22
w. You were worried about retaliation by your military co-workers or peers. 23
x. You took other actions to handle the situation 24
245. In retrospect, would you make the same decision about reporting if you
could do it over? SAFU32
a. Yes 1
b. No 2
As a result of the unwanted event, did you... Select “Yes,” “No,” or “Do not know” for
each item.
246. Experience any professional retaliation? For example, loss of privileges, denied promotion/training,
transferred to less favorable job. SAFU33a

Yes 1

No 2

Do not
know 97

247. Experience any social retaliation? For example,
ignored by coworkers, being blamed for what
happened. SAFU33b

Yes 1

No 2

Do not
know 97

Survey Instrument
248. Experience any administrative actions that you
did not want? For example, placed on a legal or
medical hold, denied a deployment opportunity,
transferred to a different assignment. SAFU33c

Yes 1

No 2

Do not
know 97

249. Experience any punishments for infractions/violations such as underage drinking or fraternization?
SAFU33d

Yes 1

No 2

Do not
know 97

157

[IF SAFU33b=1 (Yes)]

250. Who socially retaliated against you? Select all that apply
a. Someone who outranks me SAFU33b_1
b. Someone who is a similar rank or below you in rank SAFU33b_2
c. Non-military personnel SAFU33b_3
d. Do not know who they were SAFU33b_4
251. It can be difficult to remember the exact date when events occurred. In this
study, it is important to know which events happened in the last 12 months,
and which events happened earlier.
Thinking about when the event occurred, how certain are you that it
occurred in the last 12 months? If the event occurred over a long time, think
about whether it ever happened after [X date]. SAFU34
a. Definitely occurred AFTER [X date]. 1
b. Not sure if it occurred BEFORE OR AFTER [X date]. 2
c. Definitely occurred BEFORE [X date]. 3
[If SAFU34=1 (Definitely occurred AFTER [X date]) skip to SAFU36.]
[If SAFU34=2 (Not sure if it occurred BEFORE OR AFTER [X date]) skip to
SAFU36.]
[If SAFU34=3 (Definitely occurred BEFORE [X date]) AND times= “one”, then
sexualAssault_12m = “FALSE” and skip to SAFU36, else if SAFU34=3 (Definitely
occurred BEFORE [X date]) AND times=”multiple”, then continue to SAFU35]
[Programming pulls endorsed sexual assault screening items SA1-SA6]
Earlier in the survey you indicated that you experienced more than one unwanted
event in which someone:
• Put their penis into your [If Intro1=2 (Female), display: “vagina,”] anus or mouth.
• Put any object or any body part other than a penis into your [If Intro1=2 (Female),
display: “vagina,”] anus or mouth.

158

Sexual Assault and Sexual Harassment in the U.S. Military

• Made you put any part of your body or any object into someone’s mouth, vagina
or anus.
• Intentionally touched private areas of your body.
• Made you touch private areas of their body or someone else’s body.
• Attempted to put a penis, an object, or any body part into your [If Intro1=2
(Female), display: “vagina,”] anus or mouth, but no penetration actually occurred.
252. What was the date of your MOST RECENT unwanted event like this?
SAFU35Month SAFU35Year
[MM pull down / YYYY pull down. YYYY=2009 through 2014]
[IF response is before [X date], THEN sexualAssault_12m = “FALSE”]
[If times=’multiple’, then display full instructions. Else skip to “Since [X date] paragraph]
253. Thank you for sharing these details about the unwanted event you chose as
the worst or most serious. For the next question, please consider any unwanted
event that happened to you.
Since [X Date], did you initial and sign a form labeled VICTIM REPORTING PREFERENCE STATEMENT (DD Form 2910 or CG Form 6095)?
This form allows you to decide whether to make a restricted or unrestricted
report of sexual assault. A Sexual Assault Response Coordinator (SARC) or
Victim Advocate (VA) would have assisted you with completing this form. To
see a version of this form, click here. [hyperlink on “here” to an image of DD
Form 2910] SAFU36
a. Yes 1
b. No 2
c. Not sure 3
[If SAFU36=2 (No) skip to SAFU38]
When you reported the event, were you offered... Select “Yes” or “No” for each item.
254. Sexual assault advocacy services (e.g., referrals
or offers to accompany/transport you to appointments)? SAFU37a

Yes 1

No 2

255. Counseling services? SAFU37b

Yes 1

No 2

256. Medical or forensic services? SAFU37c

Yes 1

No 2

257. Legal services? SAFU37d

Yes 1

No 2

258. Chaplain services? SAFU37e

Yes 1

No 2

Survey Instrument

159

[Prior_SA here]
The questions so far have been about things that occurred in the past year. For the next
questions, please think about events that happened more than one year ago, BEFORE
[X date]. These are all experiences that you did not tell us about earlier in the survey.
These questions assess experiences of an abusive, humiliating, or sexual nature, and
that occurred even though you did not want it and did not consent.
Please include an experience regardless of who did it to you or where it happened.
“Did not consent” means that you told or showed them that you were unwilling, that
they used physical force or threats to make you do it, or that they did it to you when
you were unconscious, asleep, or so high or drunk that you could not understand what
was happening.
259. Before [X Date], had anyone put a penis, an
object, or any body part into your [If Intro1=2
(Female), display: “vagina,”] anus or mouth when
you did not want it and did not consent? SAFU38a

Yes 1

No 2

260. Before [X Date], had anyone made you insert [If
Intro1=1 (Male) display: “your penis,”] an object or
body part into someone’s mouth, vagina or anus
when you did not want to and did not consent?
SAFU38b

Yes 1

No 2

261. Before [X Date], had anyone tried to put a penis,
an object, or any body part into your [vagina,]
anus or mouth, against your will but it did not
happen? SAFU38c

Yes 1

No 2

262. Before [X Date], had anyone intentionally
touched private areas of your body (either
directly or through clothing) when you did not
want it and did not consent? Private areas include
buttocks, inner thigh, breasts, groin, anus, vagina,
penis, or testicles. SAFU38d

Yes 1

No 2

263. Before [X Date], had anyone made you touch private areas of their body or someone else’s body
(either directly or through clothing) when you did
not want it and did not consent? This might have
involved the person pressing their private areas on
you. Private areas include buttocks, inner thigh,
breasts, groin, anus, vagina, penis, or testicles.
SAFU38e

Yes 1

No 2

[If all SAFU38a-e=2 (No) skip to DEMO1. Else continue]

264. Did any of these unwanted experiences happen after you joined the military? SAFU39
a. Yes 1
b. No 2

160

Sexual Assault and Sexual Harassment in the U.S. Military

265. Did any of these unwanted experiences happen before you joined the military? SAFU40
a. Yes 1
b. No 2
266. Thank you for sharing your experiences. This information will help improve
the climate and safety of the US military. We have a few more general questions.
What is the month of your birth date? DEMO1
a. January 1
b. February 2
c. March 3
d. April 4
e. May 5
f. June 6
g. July 7
h. August 8
i. September 9
j. October 10
k. November 11
l. December 12
[If reservecomp = True, then ask:]
267. In the past 12 months, how many days (days, not drill periods) did you
spend in a compensated (pay or points only) National Guard/Reserve
status? Select one answer RGSF1
a. 0 days 0
b. 1-24 days 1
c. 25-47 days 2
d. 48-180 days 3
e. 181 days or more 4
268. Since [X Date], have you been deployed to an area where you drew hostile
fire pay or imminent danger pay? Select one answer DEMO2
a. Yes, and you are still deployed. 1
b. Yes, but you are not currently deployed. 2
c. No. 3
269. What is your current relationship status? Select one answer DEMO3
a. Married 1

Survey Instrument

b.
c.
d.
e.

Living with a boyfriend or girlfriend 2
In a committed romantic relationship, but not living together 3
Single 4
Other or prefer not to say 5

[If shortform, present:]
Thank you for completing the survey. You may have found that the
questions did not completely cover your experiences. Nonetheless, the
answers you provided are very important to this study.
Sometimes answering questions like the ones on this survey can be
upsetting. If you feel you need support or would like to talk to someone, you can call:
DoD Safe Helpline number (877-995-5247)
Military Crisis Line (1-800-273-8255)
RAINN (1-800-656-HOPE)
A SAFE helpline counselor can also explain how to report a sexual
assault and how to find out the current status of a sexual assault report.

161

162

Sexual Assault and Sexual Harassment in the U.S. Military

[LONGFORM begins below. It consists of the items above PLUS the items below]
270. To what extent do/would you feel safe from being sexually assaulted at
your home duty station? Longform1
a. Very safe 1
b. Safe 2
c. Neither safe nor unsafe 3
d. Unsafe 4
e. Very unsafe 5
271. To what extent do/would you feel safe from being sexually assaulted during
military operations, training, or exercises away from your home duty station? Longform2
a. Very safe 1
b. Safe 2
c. Neither safe nor unsafe 3
d. Unsafe 4
e. Very unsafe 5
272. How common is sexual harassment in the military? Longform3
a. Very common 1
b. Common 2
c. Rare 3
d. Very rare 4
273. How common is discrimination against women in the military? Longform4
a. Very common 1
b. Common 2
c. Rare 3
d. Very rare 4
274. In the military, how likely is it that an instance of sexual harassment would
be reported? Longform5
a. Very likely 1
b. Likely 2
c. Neither likely nor unlikely 3
d. Unlikely 4
e. Very unlikely 5

Survey Instrument

163

275. In the military, how likely is it that something would be done to try to
stop the sexual harassment after it is reported? Longform6
a. Very likely 1
b. Likely 2
c. Neither likely nor unlikely 3
d. Unlikely 4
e. Very unlikely 5
276. In the military, how likely is it that an instance of sexual assault would be
reported? Longform7
a. Very likely 1
b. Likely 2
c. Neither likely nor unlikely 3
d. Unlikely 4
e. Very unlikely 5
277. In the military, how likely is it that there would be an investigation after
an unrestricted report of sexual assault? Longform8
a. Very likely 1
b. Likely 2
c. Neither likely nor unlikely 3
d. Unlikely 4
e. Very unlikely 5
278. In the military, how likely is it that a person who sexually assaulted someone would be held accountable or punished? Longform9
a. Very likely 1
b. Likely 2
c. Neither likely nor unlikely 3
d. Unlikely 4
e. Very unlikely 5
How likely would you be to... Select one answer for each item.
279. Encourage someone who
has experienced sexual
harassment to tell a supervisor? Longform10a

Very likely
1

Likely
2

Neither likely
nor unlikely
3

Unlikely
4

Very unlikely
5

280. Encourage someone
who has experienced
sexual assault to seek
counseling? Longform10b

Very likely
1

Likely
2

Neither likely
nor unlikely
3

Unlikely
4

Very unlikely
5

164

Sexual Assault and Sexual Harassment in the U.S. Military

281. Encourage someone who
has experienced sexual
assault to report it?
Longform10c

Very likely
1

Likely
2

Neither likely
nor unlikely
3

Unlikely
4

Very unlikely
5

282. Tell a supervisor about
sexual harassment if
it happened to you?
Longform10d

Very likely
1

Likely
2

Neither likely
nor unlikely
3

Unlikely
4

Very unlikely
5

283. Report a sexual assault
if it happened to you?
Longform10e

Very likely
1

Likely
2

Neither likely
nor unlikely
3

Unlikely
4

Very unlikely
5

[If service=USCG, ask]
Compared to this time last year, rate your likeliness to do the following should
an event occur.
284. Report a sexual assault
incident USCG3a

Much more
likely
5

More Likely
4

Same
3

Less Likely
2

Much less
likely
1

285. Report a sexual harassment incident USCG3b

Much more
likely
5

More Likely
4

Same
3

Less Likely
2

Much less
likely
1

286. Seek out health care
Much more
provider services following
likely
a sexual assault USCG3c
5

More Likely
4

Same
3

Less Likely
2

Much less
likely
1

287. Engage in bystander inter- Much more
vention to prevent an
likely
assault USCG3d
5

More Likely
4

Same
3

Less Likely
2

Much less
likely
1

288. In the past 12 months, did you observe a situation that you believed was,
or could have led to, a sexual assault? Longform11
a. Yes 1
b. No 2
[If Longform11= “no”, skip to Longform12]
289. Select the one response that most closely resembles your actions. Longform11_1
a. You stepped in and separated the people involved in the situation. 1
b. You asked the person who appeared to be at risk if they needed help. 2
c. You confronted the person who appeared to be causing the situation. 3
d. You created a distraction to cause one or more of the people to disengage
from the situation. 4
e. You asked others to step in as a group and diffuse the situation. 5
f. You told someone in a position of authority about the situation. 6

Survey Instrument

165

g. You considered intervening in the situation, but you could not safely take
any action. 7
h. You decided to not take action. 8
Please indicate how well your unit leadership… Select one answer for each item.
290. Makes it clear that
sexual assault has no
place in the military.
Longform12a

Very well
1

Well
2

Neither well
nor poorly
3

Poorly
4

Very poorly
5

291. Promotes a unit climate
based on mutual respect
and trust. Longform12b

Very well
1

Well
2

Neither well
nor poorly
3

Poorly
4

Very poorly
5

292. Leads by example (e.g.,
refrains from sexist comments and behaviors).
Longform12c

Very well
1

Well
2

Neither well
nor poorly
3

Poorly
4

Very poorly
5

293. Catches and immediately corrects incidents of
sexual harassment (e.g.,
inappropriate jokes, comments, and behaviors).
Longform12d

Very well
1

Well
2

Neither well
nor poorly
3

Poorly
4

Very poorly
5

294. Creates an environment where victims would
feel comfortable reporting sexual harassment or
assault. Longform12e

Very well
1

Well
2

Neither well
nor poorly
3

Poorly
4

Very poorly
5

295. Are you currently in a work environment where female coworkers are
uncommon (less than 25% of your co-workers)? Longform13
a. Yes 1
b. No 2
How much do you agree or disagree with the following statements about the
people in your work group? Select one answer for each statement.
296. There is very little conflict among your coworkers. Longform14a

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

297. Your coworkers put in the
effort required for their
jobs. Longform14b

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

298. The people in your work
group tend to get along.
Longform14c

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

299. The people in your work
group are willing to help
each other. Longform14d

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

166

Sexual Assault and Sexual Harassment in the U.S. Military

300. You are satisfied with
the relationships you have
with your coworkers.
Longform14e

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

During the past 12 months, how often have you experienced any of the following
behaviors, where coworkers or supervisors... Select one answer for each item.
301. Intentionally interfered
with your work performance? Longform15a

Very often
1

Often
2

Sometimes Once or twice
3
4

Never
5

302. Did not provide information or assistance
when you needed it?
Longform15b

Very often
1

Often
2

Sometimes Once or twice
3
4

Never
5

303. Were excessively harsh
in their criticism of your
work performance?
Longform15c

Very often
1

Often
2

Sometimes Once or twice
3
4

Never
5

304. Took credit for work
or ideas that were yours?
Longform15d

Very often
1

Often
2

Sometimes Once or twice
3
4

Never
5

305. Gossiped/talked about
you? Longform15e

Very often
1

Often
2

Sometimes Once or twice
3
4

Never
5

306. Used insults, sarcasm, or
gestures to humiliate you?
Longform15f

Very often
1

Often
2

Sometimes Once or twice
3
4

Never
5

307. Yelled when they
were angry with you?
Longform15g

Very often
1

Often
2

Sometimes Once or twice
3
4

Never
5

308. Swore at you in a hostile
manner? Longform15h

Very often
1

Often
2

Sometimes Once or twice
3
4

Never
5

309. Damaged or stole your
property or military equip- Very often
ment assigned to you?
1
Longform15i

Often
2

Sometimes Once or twice
3
4

Never
5

How much do you agree or disagree with the following statements about the work
you do at your workplace? Select one answer for each statement.
310. Your work provides you
with a sense of pride.
Longform16a

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

311. Your work makes good use
of your skills. Longform16b

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

312. You like the kind of work
you do. Longform16c

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

Survey Instrument

167

313. Your job gives you the
chance to acquire valuable
skills. Longform16d

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

314. You are satisfied with
your job as a whole.
Longform16e

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

315. Suppose that you have to decide whether to stay on active duty. Assuming
you could stay, how likely is it that you would choose to do so? Longform17
a. Very likely 1
b. Likely 2
c. Neither likely nor unlikely 3
d. Unlikely 4
e. Very unlikely 5
316. In general, would you say your health is: Longform18
a. Excellent 1
b. Very good 2
c. Good 3
d. Fair 4
e. Poor 5
317. Sometimes things happen to people that are unusually or especially frightening, horrible, or traumatic. For example, a serious accident or fire, physical or
sexual assault or abuse, earthquake or flood, war, seeing someone be killed or
seriously injured, or having a loved one die through homicide or suicide. Have
you ever experienced this kind of event? Please count any event in your entire
life. Longform19
a. Yes 1
b. No 2
[If Longform19=2 (No) skip to Longform20]
In the past month, have you…
318. Had nightmares about the event(s) or thought
about the event(s) when you did not want to?
Longform19a

Yes 1

No 2

319. Tried hard not to think about the event(s) or went
out of your way to avoid situations that reminded
you of the event(s)? Longform19b

Yes 1

No 2

320. Been constantly on guard, watchful, or easily
startled? Longform19c

Yes 1

No 2

168

Sexual Assault and Sexual Harassment in the U.S. Military
321. Felt numb or detached from people, activities, or
your surroundings? Longform19d

Yes 1

No 2

322. Felt guilty or unable to stop blaming yourself
or others for the event(s) or any problems the
event(s) may have caused? Longform19e

Yes 1

No 2

Over the last two weeks, how often have you been bothered by any of the following problems? Select one answer for each item.
323. Little interest or pleasure in doing things
Longform20a

Not at all
0

More than Nearly every
Several days
half the days
day
1
2
3

324. Feeling down, depressed, or hopeless
Longform20b

Not at all
0

More than Nearly every
Several days
half the days
day
1
2
3

325. Trouble falling or staying asleep, or
sleeping too much Longform20c

Not at all
0

More than Nearly every
Several days
half the days
day
1
2
3

326. Feeling tired or having little energy
Longform20d

Not at all
0

More than Nearly every
Several days
half the days
day
1
2
3

327. Poor appetite or overeating
Longform20e

Not at all
0

More than Nearly every
Several days
half the days
day
1
2
3

328. Feeling bad about yourself — or that
you are a failure or have let yourself or
your family down Longform20f

Not at all
0

More than Nearly every
Several days
half the days
day
1
2
3

329. Trouble concentrating on things, such
as reading the newspaper or watching
television Longform20g

Not at all
0

More than Nearly every
Several days
half the days
day
1
2
3

330. Moving or speaking so slowly that other
people could have noticed. Or the opposite —being so fidgety or restless that
you have been moving around a lot
more than usual Longform20h

Not at all
0

More than Nearly every
Several days
half the days
day
1
2
3

[If service=USCG, ask]
331. How much of a problem do you believe sexual assault is within the USCG?
USCG2
a. No problem at all 1
b. A little problem 2
c. A moderate problem 3
d. A large problem 4
e. A major problem 5

Survey Instrument

169

DoD provides two types of sexual assault reports.
• Restricted reports allow people to get information, collect evidence, and receive
medical treatment and counseling without starting an official investigation of the
assault.
• Unrestricted reports start an official investigation in addition to allowing the services available in restricted reporting.
How satisfied have you been with the availability of information on... Select one
answer for each item.
Very
dissatisfied
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very satisfied
5

Very
333. How to file an unrestricted
dissatisfied
report? Longform21b
1

Neither
Dissatisfied satisfied nor
2
dissatisfied
3

Satisfied
4

Very satisfied
5

332. How to file a restricted
report? Longform21a

334. Have you had any military training during the past 12 months on topics
related to sexual assault? Longform22
a. Yes 1
b. No 2
[If Longform22=2 (No) skip to Longform24]
My Service’s sexual assault training... Select one answer for each item.
335. Provides a good understanding of what actions
are considered sexual
assault. Longform23a

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

336. Teaches that the consumption of alcohol
may increase the likelihood of sexual assault.
Longform23b

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

337. Teaches how to avoid
situations that might
increase the risk of being
a victim of sexual assault.
Longform23c

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

170

Sexual Assault and Sexual Harassment in the U.S. Military

338. Teaches how to intervene when you witness
a situation involving a
fellow Service member
(bystander intervention).
Longform23d

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

339. Teaches how to obtain
medical care following a sexual assault.
Longform23e

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

340. Explains the role of
the chain of command in
handling sexual assaults.
Longform23f

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

341. Explains the reporting options available if
a sexual assault occurs.
Longform23g

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

342. Identifies the points
of contact for reporting sexual assault (e.g.,
SARC, Victim Advocate).
Longform23h

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

343. Explains how sexual
assault is a mission
readiness problem.
Longform23i

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

344. Explains the resources
available to victims
(e.g., Safe Helpline).
Longform23j

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

[If service=USCG, ask]
Rate your knowledge of the following items:
345. How to contact your
Sexual Assault
Extremely
Knowledgeable
Response
Knowledgeable
4
Coordina5
tor (SARC)
USCG1a

Extremely
Neutral Unknowledgeable
Unknowledgeable
3
2
1

346. How to
contact your
Extremely
Knowledgeable
Victim Advo- Knowledgeable
4
cate (VA)
5
USCG1b

Extremely
Neutral Unknowledgeable
Unknowledgeable
3
2
1

Survey Instrument

171

347. How to
access mediExtremely
Knowledgeable
cal services
Knowledgeable
4
following a
5
sexual assault
USCG1c

Extremely
Neutral Unknowledgeable
Unknowledgeable
3
2
1

348. Sexual
Extremely
Knowledgeable
assault policy
Knowledgeable
4
information
5
USCG1d

Extremely
Neutral Unknowledgeable
Unknowledgeable
3
2
1

349. Sexual harassExtremely
ment policy
Knowledgeable
Knowledgeable
information
4
5
USCG1e

Extremely
Neutral Unknowledgeable
Unknowledgeable
3
2
1

[If service=USCG, ask]
Rate the effectiveness of the following communication forms in increasing your
knowledge on sexual assault:
350. All Hands Blogs on
SAPR USCG4a

Extremely
Effective
5

Effective
4

Neutral
3

Ineffective
2

Extremely
Did Not See
Ineffective
0
1

351. ALCOAST on SAPR
USCG4b

Extremely
Effective
5

Effective
4

Neutral
3

Ineffective
2

Extremely
Did Not See
Ineffective
0
1

352. SAPR Program
Website USCG4c

Extremely
Effective
5

Effective
4

Neutral
3

Ineffective
2

Extremely
Did Not See
Ineffective
0
1

353. Area SAPR Program Website
USCG4d

Extremely
Effective
5

Effective
4

Neutral
3

Ineffective
2

Extremely
Did Not See
Ineffective
0
1

354. Sexual Assault
Awareness Month
(SAAM) Event
USCG4e

Extremely
Effective
5

Effective
4

Neutral
3

Ineffective
2

Extremely
Did Not See
Ineffective
0
1

355. Mandated Training
(GMT) USCG4f

Extremely
Effective
5

Effective
4

Neutral
3

Ineffective
2

Extremely
Did Not See
Ineffective
0
1

356. Sexual Assault Pre- Extremely
vention Workshop Effective
(SAPW) USCG4g
5

Effective
4

Neutral
3

Ineffective
2

Extremely
Did Not See
Ineffective
0
1

Extremely
Effective
5

Effective
4

Neutral
3

Ineffective
2

Extremely
Did Not See
Ineffective
0
1

357. SAPR Strategic Plan
USCG4h

[If service=USCG, ask]

172

Sexual Assault and Sexual Harassment in the U.S. Military

In regard to sexual assault resources, rate the degree to which you understand the
role of the following positions:
358. Sexual Assault Response Coordinator (SARC) USCG5a

Do not
Understand Understand
understand
their role
their role
their role
somewhat 2 very well 3
1

359. Victim Advocate (VA) USCG5b

Do not
understand
their role
1

Understand
their role
somewhat
2

Understand
their role
very well
3

Do not
understand
360. Special Victim Counsel (SVC) USCG5c
their role
1

Understand
their role
somewhat
2

Understand
their role
very well
3

361. Chaplain USCG5d

Do not
understand
their role
1

Understand
their role
somewhat
2

Understand
their role
very well
3

362. Health care provider USCG5e

Do not
understand
their role
1

Understand
their role
somewhat
2

Understand
their role
very well
3

363. Have you had any military training during the past 12 months on topics
related to sexual harassment? Longform24
a. Yes 1
b. No 2
How much do you agree with the following? Select one answer for each item.
364. When you are in a social
setting, it is your duty
to stop a fellow Service
member from doing something potentially harmful
to themselves or others.
Longform25a

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

365. If you are sexually
assaulted, you can trust
the military system to
protect your privacy.
Longform25b

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

366. If you are sexually
assaulted, you can trust
the military system to
ensure your safety following the incident.
Longform25c

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

Strongly
agree
5

Survey Instrument
367. If you are sexually
assaulted, you can trust
the military system to
treat you with dignity and
respect. Longform25d

Strongly
disagree
1

Disagree
2

Neither agree
nor disagree
3

Agree
4

173

Strongly
agree
5

368. In your opinion, has sexual harassment in our nation become more or less
of a problem over the last 2 years? Longform26
a. Less of a problem today 1
b. About the same as 2 years ago 2
c. More of a problem today 3
d. Do not know? 97
369. In your opinion, has sexual harassment in the military become more or
less of a problem over the last 2 years? Longform27
a. Less of a problem today 1
b. About the same as 2 years ago 2
c. More of a problem today 3
d. Do not know? 97
370. In your opinion, has sexual assault in our nation become more or less of a
problem over the last 2 years? Longform28
a. Less of a problem today 1
b. About the same as 2 years ago 2
c. More of a problem today 3
d. Do not know? 97
371. In your opinion, has sexual assault in the military become more or less of
a problem over the last 2 years? Longform29
a. Less of a problem today 1
b. About the same as 2 years ago 2
c. More of a problem today 3
d. Do not know? 97
[ALL MARINE AND AIR FORCE specific questions go at the end.]
[If service=USMC, ask]
[MARINE SPECIFIC QUESTIONS START]
372. If you needed to talk about a sexual assault, which USMC point of contact
would you be most comfortable talking to? Select one USMC1
a. SARC 1
b. UVA/VA 2

174

Sexual Assault and Sexual Harassment in the U.S. Military

c.
d.
e.
f.
g.
h.
i.

SAFE Helpline or other hotline counselor 3
Chaplain or religious leader 4
A supervisor or someone in your chain of command 5
An officer or NCO outside your chain of command 6
Victims’ Legal Counsel or other military counsel 7
Medical professional 8
Law enforcement professional 9

373. Do you know who the Uniformed Victim Advocate (UVA) in your unit
is? USMC2
a. Yes 1
b. No 2
374. Would you be willing to report a sexual assault to your UVA? USMC3
a. Yes 1
b. No 2
375. Which of the following topics would you like to see included in SAPR
training? Select all that apply
a. Healthy relationships USMC4a
b. Professional ethics USMC4b
c. Respect USMC4c
d. Sexual harassment prevention USMC4d
e. Command climate USMC4e
f. Personal boundaries USMC4f
[MARINE SPECIFIC QUESTIONS END]
[AIR FORCE SPECIFIC QUESTIONS START]
[If service=USAF, ask]
376. Do you know anyone in the Air Force who has intervened in a situation to
prevent a sexual assault since [X Date]? USAF1
a. Yes 1
b. No 2
c. Not Sure 3
[AIR FORCE SPECIFIC QUESTIONS END]
[ARMY AND NAVY DO NOT HAVE ANY EXTRA ITEMS]
[If longform respondent, present:]

Survey Instrument

Thank you for completing the survey. You may have found that the
questions did not completely cover your experiences. Nonetheless, the
answers you provided are very important to this study.
Sometimes answering questions like this can be upsetting. If you feel
you need support or would like to talk to someone, you can call:
DoD Safe Helpline number (877-995-5247)
Military Crisis Line (1-800-273-8255)
RAINN (1-800-656-HOPE)
A SAFE helpline counselor can also explain how to report a sexual
assault and how to find out the current status of a sexual assault report.

175

Prior WGRA Form

The RAND study team developed the prior WGRA form to be very similar to the
2012 WGRA, but included fewer questions to improve response/completion rates.
Items in the prior WGRA form are numbered with the same item numbers as in the
2012 WGRA. Where the item numbers are not sequential (skipping numbers) it means
that the corresponding 2012 WGRA item is not included in the 2014 survey. The last
item in the prior WGRA form was created by the RAND study team and is numbered
SAFU25 to match the same item in the RMWS.

177

178

Sexual Assault and Sexual Harassment in the U.S. Military

Administrative Variables:
[PROGRAMMING NOTE: [X Date] = Date survey was taken -12 months. The first
time a respondent takes the survey the [X Date] value will be computed and stored in
the database. Upon re-entry, this date will be loaded from the database using the date
computed when the respondent took the survey the first time].

Survey Instrument

[Landing Page START]

2014 RAND Military Workplace Study

Please sign in

Continue
Technical Assistance: support@randmilitarystudy.org, 1-855-365-5914
[Landing Page END]

179

180

Sexual Assault and Sexual Harassment in the U.S. Military

[Consent START]
Welcome to the 2014 RAND Military Workplace Study

Dear [Soldier/Sailor/Airman/Marine/Coast Guardsman]:

RCS # DD-P&R(QD)1947
Expires: 7/25/2015

Survey Instrument

181

182

Sexual Assault and Sexual Harassment in the U.S. Military

Who do you contact if you have questions or concerns about the survey?
• Questions about the overall study or RAND: Contact the RAND team by
email: WGRS2014@rand.org or go to the RAND website link: www.WGRS2014.
rand.org.
• Questions about computer, technical or survey problems: Contact the Westat
Survey Helpdesk toll free number: 1-855-365-5914 (OCONUS please call collect: 240-453-2620) or by email: support@randmilitarystudy.org.
• Questions about your rights as a participant in this study: Contact the
RAND Human Subjects Protection Committee: 310-393-0411, ext. 6369 in
Santa Monica, California.
• Questions about the licensing of the survey: Information about DoD surveys
can be found at http://www.dtic.mil/whs/directives/corres/intinfocollections/iic_
search.html; this survey’s RCS # is DD-P&R(QD)1947 and the expiration date
is 7/25/2015.
Some of the questions in this survey will be personal. For your privacy, you may want
to take this survey where other people won’t see your screen. It is okay to forward the
survey to a personal email address or you may complete it on a smartphone.
Press the Continue button if you agree to do the survey.
Print Informed
Consent
Continue
You can print a copy of this Informed Consent Statement by pressing the following
button:
[Consent END]

Survey Instrument

183

2014 RAND Military Workplace Study – Prior
Thank you for agreeing to participate in this important study.
Please answer each question thoughtfully and truthfully. This will allow us to provide
an accurate picture of the different experiences of today’s military members. If you
prefer not to answer a specific question for any reason, just leave it blank.
Some of the questions in the survey will be personal. For your privacy, you may want
to take this survey where other people won’t see your screen.
[Programming Note: Use Gender Variable from the sample file if Gender is not selected
by Respondent]
2. Are you...? PF2
Male 1
Female 2
3. Are you Spanish/Hispanic/Latino? PF3
No, not Spanish/Hispanic/Latino 1
Yes, Mexican, Mexican-American, Chicano, Puerto Rican, Cuban, or other
Spanish/Hispanic/Latino 2
4. What is your race? Select one or more races to indicate what you consider yourself
to be.
White PF4_1
Black or African American PF4_2
American Indian or Alaska Native PF4_3
Asian (e.g., Asian Indian, Chinese, Filipino, Japanese, Korean, or Vietnamese) PF4_4
Native Hawaiian or other Pacific Islander (e.g., Samoan, Guamanian, or
Chamorro) PF4_5
6. In the past 12 months, have you been deployed to a combat zone or to an
area where you drew imminent danger pay or hostile fire pay? PF6
Yes 1
No 2
7. To what extent do/would you feel safe during deployments from being sexually assaulted on your base/installation/ship? PF7
Very safe 1
Safe 2
Neither safe nor unsafe 3

184

Sexual Assault and Sexual Harassment in the U.S. Military

Unsafe 4
Very unsafe 5
8. To what extent do/would you feel safe from being sexually assaulted on
your home base/installation/ship? PF8
Very safe 1
Safe 2
Neither safe nor unsafe 3
Unsafe 4
Very unsafe 5
9. Are you currently in a work environment where members of your gender
are uncommon? PF9
Yes 1
No 2
10. What is the gender of your immediate supervisor? PF10
Male 1
Female 2
11. How much do you agree or disagree with the following statements about
your supervisor? Select one answer for each statement. PF11

You trust your supervisor.
PF11_1

Strongly
agree
1

Agree
2

Neither agree
nor disagree
3

Disagree
4

Strongly
disagree
5

Your supervisor ensures
that all assigned personnel
are treated fairly. PF11_2

Strongly
agree
1

Agree
2

Neither agree
nor disagree
3

Disagree
4

Strongly
disagree
5

There is very little conflict
between your supervisor and the people who
report to him/her. PF11_3

Strongly
agree
1

Agree
2

Neither agree
nor disagree
3

Disagree
4

Strongly
disagree
5

Your supervisor evaluates
your work performance
fairly. PF11_4

Strongly
agree
1

Agree
2

Neither agree
nor disagree
3

Disagree
4

Strongly
disagree
5

Your supervisor assigns
work fairly in your work
group. PF11_5

Strongly
agree
1

Agree
2

Neither agree
nor disagree
3

Disagree
4

Strongly
disagree
5

You are satisfied with the
direction/supervision you
receive. PF11_6

Strongly
agree
1

Agree
2

Neither agree
nor disagree
3

Disagree
4

Strongly
disagree
5

17. Suppose that you have to decide whether to stay on active duty. Assuming
you could stay, how likely is it that you would choose to do so? PF17
Very likely 1

Survey Instrument

185

Likely 2
Neither likely nor unlikely 3
Unlikely 4
Very unlikely 5
18. Overall, how satisfied are you with the military way of life? PF18
Very satisfied 1
Satisfied 2
Neither satisfied nor dissatisfied 3
Dissatisfied 4
Very dissatisfied 5
19. How often during the past 12 months have you experienced any of the following behaviors where coworkers or supervisors... Select one answer for
each item.

Intentionally interfered
with your work performance? PF19_1

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Did not provide information or assistance when
you needed it? PF19_2

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Were excessively harsh in
their criticism of your work
performance? PF19_3

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Took credit for work or
ideas that were yours?
PF19_4

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Gossiped/talked about
you? PF19_5

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Used insults, sarcasm, or
gestures to humiliate you?
PF19_6

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Yelled when they were
angry with you? PF19_7

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Swore at you in a hostile
manner? PF19_8

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Damaged or stole your
property or equipment?
PF19_9

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

186

Sexual Assault and Sexual Harassment in the U.S. Military

27. During the past 12 months, did any of the following happen to you? If it
did, do you believe your gender was a factor? Select one answer for each
item.

You were rated lower than you deserved on your
last military evaluation. PF27_1

No, or does
not apply
1

Yes, but your Yes, and your
gender was gender was a
NOT a factor
factor
2
3

Your last military evaluation contained unjustified
negative comments. PF27_2

No, or does
not apply
1

Yes, but your Yes, and your
gender was gender was a
NOT a factor
factor
2
3

You were held to a higher performance standard
than others in your military job. PF27_3

No, or does
not apply
1

Yes, but your Yes, and your
gender was gender was a
NOT a factor
factor
2
3

You did not get a military award or decoration
given to others in similar circumstances. PF27_4

No, or does
not apply
1

Yes, but your Yes, and your
gender was gender was a
NOT a factor
factor
2
3

Your current military assignment has not made
use of your job skills. PF27_5

No, or does
not apply
1

Yes, but your Yes, and your
gender was gender was a
NOT a factor
factor
2
3

Your current assignment is not good for your
career if you continue in the military. PF27_6

No, or does
not apply
1

Yes, but your Yes, and your
gender was gender was a
NOT a factor
factor
2
3

You did not receive day-to-day, short-term tasks
in your military job that would have helped you
prepare for advancement. PF27_7

No, or does
not apply
1

Yes, but your Yes, and your
gender was gender was a
NOT a factor
factor
2
3

You did not have a professional relationship with
someone who advised (mentored) you on military
career development or advancement. PF27_8

No, or does
not apply
1

Yes, but your Yes, and your
gender was gender was a
NOT a factor
factor
2
3

You did not learn until it was too late of opportu- No, or does
nities that would have helped your military career. not apply
PF27_9
1

Yes, but your Yes, and your
gender was gender was a
NOT a factor
factor
2
3

You were unable to get straight answers about
your military promotion possibilities. PF27_10

No, or does
not apply
1

Yes, but your Yes, and your
gender was gender was a
NOT a factor
factor
2
3

You were excluded from social events important
to military career development and being kept
informed. PF27_11

No, or does
not apply
1

Yes, but your Yes, and your
gender was gender was a
NOT a factor
factor
2
3

Survey Instrument

187

You did not get a military job assignment that you
wanted and for which you were qualified. PF27_12

No, or does
not apply
1

Yes, but your Yes, and your
gender was gender was a
NOT a factor
factor
2
3

Have you had any other adverse personnel actions
in the past 12 months? PF27_13

No, or does
not apply
1

Yes, but your Yes, and your
gender was gender was a
NOT a factor
factor
2
3

[Ask if PF27_12 = 3]

28. You answered “Yes, and your gender was a factor” to “You did not get a
military job assignment that you wanted and for which you were qualified” above. Was this assignment legally open to women? PF28
Yes 1
No 2
[Ask if any PF27_1- PF27_13 = 3 OR any PF27_1 PF27_13 = 2]
29. Do you consider ANY of the behaviors which you marked as happening
to you in the previous question to have been... Select one answer for each
item.

Sex discrimination?
PF29_1

None 1

Some 2

All 3

Racial/ethnic discrimination? PF29_2

None 1

Some 2

All 3

Age discrimination?
PF29_3

None 1

Some 2

All 3

Religious discrimination? PF29_4

None 1

Some 2

All 3

None 1

Some 2

All 3

Other? PF29_5

In this question you are asked about sex/gender-related talk and/or behavior that was
unwanted, uninvited, and in which you did not participate willingly.
30. How often during the past 12 months have you been in situations involving
• Military Personnel (Active Duty or National Guard/Reserve)
–– on- or off-duty
–– on- or off-installation or ship; and/or
• DoD/Service Civilian Employees and/or Contractors
–– in your workplace or on your installation/ship

188

Sexual Assault and Sexual Harassment in the U.S. Military

where one or more of these individuals (of either gender)... Select one
answer for each item.

Repeatedly told sexual
stories or jokes that were
offensive to you? PF30_1

Never
1

Once or twice Sometimes
2
3

Often 4
Often
4

Very often
5

Referred to people of your
gender in insulting or
offensive terms? PF30_2

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Made unwelcome
attempts to draw you
into a discussion of sexual
matters (e.g., attempted
to discuss or comment on
your sex life)? PF30_3

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Treated you “differently”
because of your gender
(e.g., mistreated, slighted,
or ignored you)? PF30_4

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Made offensive remarks
about your appearance,
body, or sexual activities?
PF30_5

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Made gestures or used
body language of a sexual
nature that embarrassed
or offended you? PF30_6

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Made offensive sexist
remarks (e.g., suggesting that people of your
gender are not suited for
the kind of work you do)?
PF30_7

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Made unwanted attempts
to establish a romantic
sexual relationship with
you despite your efforts to
discourage it? PF30_8

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Put you down or was condescending to you because
of your gender? PF30_9

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Continued to ask you for
dates, drinks, dinner, etc.,
even though you said
“No”? PF30_10

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Made you feel like you
were being bribed with
some sort of reward
or special treatment to
engage in sexual behavior? PF30_11

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Survey Instrument

189

Made you feel threatened
with some sort of retaliation for not being sexually cooperative (e.g., by
mentioning an upcoming
review)? PF30_12

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Touched you in a way that
made you feel uncomfortable? PF30_13

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Intentionally cornered you
or leaned over you in a
sexual way? PF30_14

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Treated you badly for
refusing to have sex?
PF30_15

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Implied faster promotions
or better treatment if you
were sexually cooperative?
PF30_16

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Attempted to have sex
with you without your
consent or against your
will, but was not successful? PF30_17

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Had sex with you without
your consent or against
your will? PF30_18

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

Other unwanted genderrelated behavior? PF30_19

Never
1

Once or twice Sometimes
2
3

Often
4

Very often
5

[Ask if any PF30_1-PF30_16 = 2 OR Any PF30_1-PF30_16 = 3 OR Any PF30_1-PF30_16 = 4 OR Any
PF30_1-PF30_16 = 5]

31. How many of these behaviors that you selected as happening to you, do
you consider to have been sexual harassment? PF31
None were sexual harassment 1
Some were sexual harassment; some were not sexual harassment 2
All were sexual harassment 3
32. In the past 12 months, have you experienced any of the following intentional sexual contacts that were against your will or occurred when you
did not or could not consent where someone... PF32
• Sexually touched you (e.g., intentional touching of genitalia, breasts, or
buttocks) or made you sexually touch them?
• Attempted to make you have sexual intercourse, but was not successful?
• Made you have sexual intercourse?

190

Sexual Assault and Sexual Harassment in the U.S. Military

• Attempted to make you perform or receive oral sex, anal sex, or penetration by a finger or object, but was not successful?
• Made you perform or receive oral sex, anal sex, or penetration by a finger
or object?
Yes 1
No 2
[If PF32=2, then skip to the end of survey]
[Ask if PF32 = 1]
33. In the past 12 months, how many separate incidents of sexual touching,
attempted or completed intercourse, oral or anal sex, or penetration by a
finger or object did you experience? PF33
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 or more 9
Think about the situation(s) you experienced in the past 12 months that involved the
behaviors in the previous question. Tell us about the one event that had the greatest
effect on you.
[Ask if PF32 = 1]
34. What did the person(s) do during the situation? Select one answer for each
behavior. PF34

Sexually touched you (e.g., intentional touching of Did not do
genitalia, breasts, or buttocks) or made you sexuthis 1
ally touch them PF34_1

Did this 2

Attempted to make you have sexual intercourse,
but was not successful PF34_2

Did not do
this 1

Did this 2

Made you have sexual intercourse PF34_3

Did not do
this 1

Did this 2

Attempted to make you perform or receive oral
Did not do
sex, anal sex, or penetration by a finger or object,
this 1
but was not successful PF34_4

Did this 2

Survey Instrument

Made you perform or receive oral sex, anal sex, or Did not do
penetration by a finger or object PF34_5
this 1

Did this 2

[Ask if PF32 = 1]
35. Did the situation occur... Select “Yes” or “No” for each item. PF35

At a military installation? PF35_1

Yes 1

No 2

During your work day/duty hours? PF35_2

Yes 1

No 2

While you were on TDY/TAD, at sea, or during
field exercises/alerts? PF35_3

Yes 1

No 2

While you were deployed to a combat zone or to
an area where you drew imminent danger pay or
hostile fire pay? PF35_4

Yes 1

No 2

During the delayed entry program? PF35_5

Yes 1

No 2

During recruit training/basic training? PF35_6

Yes 1

No 2

During any type of military combat training?
PF35_7

Yes 1

No 2

During Officer Candidate or Training School/Basic
or Advanced Officer Course? PF35_8

Yes 1

No 2

During military occupational specialty school/technical training/advanced individual training/professional military education? PF35_9

Yes 1

No 2

[Ask if PF32 = 1]
36. How many offender(s) were involved? Select one. PF36
One person 1
More than one person 2
Not sure 3
[Ask if PF32 = 1]
37. What was/were the gender(s) of the offender(s)? Select one. PF37
Male only 1
Female only 2
Both male and female 3
Not sure 4
[Ask if PF32 = 1]

191

192

Sexual Assault and Sexual Harassment in the U.S. Military

38. Was the offender(s)… Select “Yes” or “No” for each item.

Someone in your chain of command? PF38_1

Yes 1

No 2

Other military person(s) of higher rank/grade who
was not in your chain of command? PF38_2

Yes 1

No 2

Your military coworker(s)? PF38_3

Yes 1

No 2

Your military subordinate(s)? PF38_4

Yes 1

No 2

Other military person(s)? PF38_5

Yes 1

No 2

DoD/Service civilian employee(s)? PF38_6

Yes 1

No 2

DoD/Service civilian contractor(s)? PF38_7

Yes 1

No 2

Your spouse/significant other? PF38_8

Yes 1

No 2

Person(s) in the local community? PF38_9

Yes 1

No 2

Unknown person(s)? PF38_10

Yes 1

No 2

[Ask if PF32 = 1]
39. Did the offender(s) use drugs to knock you out (e.g., date rape drugs, sedatives)? PF39
Yes 1
No 2
Not sure 97
[Ask if PF32 = 1]
40. Had either you or the offender(s) been drinking alcohol before the incident? PF40
Yes 1
No 2
[Ask if PF32 = 1]
41. Had either you or the offender(s) been using drugs before the incident?
PF41
Yes 1
No 2
[Ask if PF32 = 1]

Survey Instrument

193

42. Did the offender(s)... Select “Yes” or “No” for each item.

Threaten to ruin your reputation if you did not
consent? PF42_1

Yes 1

No 2

Threaten to physically harm you if you did not
consent? PF42_2

Yes 1

No 2

Use some degree of physical force (e.g., holding
you down)? PF42_3

Yes 1

No 2

[Ask if PF32 = 1]
43. Did the offender(s)... Select “Yes” or “No” for each item.

Sexually harass you before the situation? PF43_1

Yes 1

No 2

Stalk you before the situation? PF43_2

Yes 1

No 2

Sexually harass you after the situation? PF43_3

Yes 1

No 2

Stalk you after the situation? PF43_4

Yes 1

No 2

[Ask if PF32 = 1]
44. As a result of this situation, to what extent did... Select one answer for each
item.

You consider requesting
a transfer? PF44_1

Not at all
1

Small
extent
2

Moderate
extent
3

Large
extent
4

Very large
extent
5

You think about getting out of your Service?
PF44_2

Not at all
1

Small
extent
2

Moderate
extent
3

Large
extent
4

Very large
extent
5

Your work performance
decrease? PF44_3

Not at all
1

Small
extent
2

Moderate
extent
3

Large
extent
4

Very large
extent
5

[Ask if PF32 = 1]
45. Did you report this situation to a civilian authority or organization? PF45
Yes 1
No 2
DoD provides two types of reporting of sexual assault. Unrestricted reporting is for
victims who want medical treatment, counseling, and an official investigation of the
assault. Restricted reporting is for victims who want information and to receive medical treatment and counseling without prompting an official investigation of the assault.

194

Sexual Assault and Sexual Harassment in the U.S. Military

[Ask if PF32 = 1]
46. Did you report this situation to an installation/Service/DoD authority or
organization? PF46
Yes 1
No 2
DoD provides two types of reporting of sexual assault. Unrestricted reporting is for
victims who want medical treatment, counseling, and an official investigation of the
assault. Restricted reporting is for victims who want information and to receive medical treatment and counseling without prompting an official investigation of the assault.
[Ask if PF32 = 1 AND PF46= 1]
47. Did you make... Select one. PF47
Only a restricted report? 1
Only an unrestricted report? 2
A restricted report that was converted to an unrestricted report? 3
[Ask if PF32 = 1 AND PF46 = 1 AND (PF47 = 2 OR 3)]
48. How satisfied have you been with your treatment by the... Select one answer
for each item.

Sexual Assault Very satisfied
Victims’ Advo1
cate assigned
to you? PF48_1

Satisfied
2

Neither
Dissatisfied
satisfied nor
4
dissatisfied
3

Very
dissatisfied
5

Does not
apply
6

Sexual Assault Very satisfied
Response Coor1
dinator (SARC)
handling your
report? PF48_2

Satisfied
2

Neither
Dissatisfied
satisfied nor
4
dissatisfied
3

Very
dissatisfied
5

Does not
apply
6

Commander
Very satisfied
handling your
1
report? PF48_3

Satisfied
2

Neither
Dissatisfied
satisfied nor
4
dissatisfied
3

Very
dissatisfied
5

Does not
apply
6

Criminal
Very satisfied
investigator(s)
1
handling your
report? PF48_4

Satisfied
2

Neither
Dissatisfied
satisfied nor
4
dissatisfied
3

Very
dissatisfied
5

Does not
apply
6

Satisfied
2

Neither
Dissatisfied
satisfied nor
4
dissatisfied
3

Very
dissatisfied
5

Does not
apply
6

Trial Defense Very satisfied
Office person1
nel? PF48_5

Survey Instrument

195

Legal Office
Very satisfied
personnel
1
(prosecution)?
PF48_6

Satisfied
2

Neither
Dissatisfied
satisfied nor
4
dissatisfied
3

Very
dissatisfied
5

Does not
apply
6

Legal assisVery satisfied
tance (not
1
prosecution)?
PF48_7

Satisfied
2

Neither
Dissatisfied
satisfied nor
4
dissatisfied
3

Very
dissatisfied
5

Does not
apply
6

Medical per- Very satisfied
sonnel? PF48_8
1

Satisfied
2

Neither
Dissatisfied
satisfied nor
4
dissatisfied
3

Very
dissatisfied
5

Does not
apply
6

Very satisfied
1

Satisfied
2

Neither
Dissatisfied
satisfied nor
4
dissatisfied
3

Very Very
dissatisfied
5

Does not
apply
6

Safe Helpline Very satisfied
staff? PF48_10
1

Satisfied
2

Neither
Dissatisfied
satisfied nor
4
dissatisfied
3

Very
dissatisfied
5

Does not
apply
6

Chaplain?
PF48_9

[Ask if PF32 = 1 AND PF46 = 1 AND (PF47 = 2 OR 3) AND (PF48_4 = 4 OR 5)]
52. Was the criminal investigator(s) handling your report... PF52
Military? 1
Civilian? 2
Both? 3
[Ask if PF32 = 1 AND PF46 = 1 AND (PF47 = 2 OR 3) AND (PF48_5 = 4 OR 5)]
53. Was the Trial Defense Office personnel... PF53
Military? 1
Civilian? 2
Both? 3
[Ask if PF32 = 1 AND PF46 = 1 AND (PF47 = 2 OR 3) AND (PF48_6 = 4 OR 5)]
54. Was the Legal Office personnel (prosecution)... PF54
Military? 1
Civilian? 2
Both? 3
[Ask if PF32 = 1 AND PF46 = 1 AND (PF47 = 2 OR 3 AND (PF48_7 = 4 OR 5)]

196

Sexual Assault and Sexual Harassment in the U.S. Military

55. Was the legal assistance (not prosecution)... PF55
Military? 1
Civilian? 2
Both? 3
[Ask if PF32 = 1 AND PF46 = 1 AND (PF47 = 2 OR 3 AND (PF48_8 = 4 OR 5)]
56. Was the medical personnel... PF56
Military? 1
Civilian? 2
Both? 3
[Ask if PF32 = 1 AND PF46 = 1]
59. As a result of this situation, did you... Select “Yes,” “No,” or “Don’t know”
for each item.

Experience any professional retaliation (e.g., loss
of privileges, denied promotion/training, transferred to less favorable job)? PF59_1

Yes 1

No 2

Don’t
know 97

Experience any social retaliation (e.g., ignored
by coworkers, being blamed for the situation)?
PF59_2

Yes 1

No 2

Don’t
know 97

Experience any administrative actions (e.g.,
placed on a medical hold, placed on a legal
hold, transferred to a different assignment)?
PF59_3

Yes 1

No 2

Don’t
know 97

Experience any punishments for infractions/violations, such as underage drinking or fraternization? PF59_4

Yes 1

No 2

Don’t
know 97

[Ask if PF32 = 1 AND PF46 = 1]
60. How satisfied have you been with... Select one answer for each item.

The quality of sexual
assault advocacy services
you received?
PF60_1

Very
satisfied
1

Satisfied
2

Neither
satisfied
nor
dissatisfied
3

Dissatisfied
4

Very
dissatisfied
5

Does not
apply
6

The quality
of counseling services
you received?
PF60_2

Very
satisfied
1

Satisfied
2

Neither
satisfied
nor
dissatisfied
3

Dissatisfied
4

Very
dissatisfied
5

Does not
apply
6

Survey Instrument

197

The quality of
medical care
you received?
PF60_3

Very
satisfied
1

Satisfied
2

Neither
satisfied
nor
dissatisfied
3

Dissatisfied
4

Very
dissatisfied
5

Does not
apply
6

The amount
of time investigation
process took/
is taking?
PF60_4

Very
satisfied
1

Satisfied
2

Neither
satisfied
nor
dissatisfied
3

Dissatisfied
4

Very
dissatisfied
5

Does not
apply
6

How well you
were/are kept
informed
about the
progress of
your case?
PF60_5

Very
satisfied
1

Satisfied
2

Neither
satisfied
nor
dissatisfied
3

Dissatisfied
4

Very
dissatisfied
5

Does not
apply
6

The Safe
Helpline
service you
received?
PF60_6

Very
satisfied
1

Satisfied
2

Neither
satisfied
nor
dissatisfied
3

Dissatisfied
4

Very
dissatisfied
5

Does not
apply
6

The reporting
process overall? PF60_7

Very
satisfied
1

Satisfied
2

Neither
satisfied
nor
dissatisfied
3

Dissatisfied
4

Very
dissatisfied
5

Does not
apply
6

[Ask if PF32 = 1 AND PF46 = 1]
68. When you reported the situation, were you offered... Select “Yes” or “No”
for each item.

Sexual assault advocacy services (e.g., referrals or
offers to accompany/transport you to appointments)? PF68_1

Yes 1

No 2

Counseling services? PF68_2

Yes 1

No 2

Medical or forensic services? PF68_3

Yes 1

No 2

Legal services? PF68_4

Yes 1

No 2

Chaplain services? PF68_5

Yes 1

No 2

[Ask if PF32 = 1 AND PF46 = 1]
69. What were your reasons for reporting the situation to any installation/
Service/DoD authority or organizations? Select “Yes” or “No” for each
item.

198

Sexual Assault and Sexual Harassment in the U.S. Military

Prevent the offender from continuing in the military PF69_1

Yes 1

No 2

Stop the offender from hurting you again PF69_2

Yes 1

No 2

Stop the offender from hurting others PF69_3

Yes 1

No 2

Seek justice PF69_4

Yes 1

No 2

It was the right thing to do PF69_5

Yes 1

No 2

Seek help dealing with an emotional incident
PF69_6

Yes 1

No 2

Punish the offender PF69_7

Yes 1

No 2

Discourage other potential offenders PF69_8

Yes 1

No 2

Identify a fellow military member who is acting
inappropriately PF69_9

Yes 1

No 2

Seek closure on the incident PF69_10

Yes 1

No 2

Seek medical assistance PF69_11

Yes 1

No 2

Seek mental health assistance PF69_12

Yes 1

No 2

Stop rumors by coming forward PF69_13

Yes 1

No 2

Yes 1

No 2

Other PF69_14

[Ask if PF32 = 1 AND PF46 = 1]
70. How long after the situation occurred did you report it? Select one. PF70
Within 24 hours 1
Within 2-3 days 2
Within 4-7 days 3
Within 8-14 days 4
Within 15-30 days 5
More than 30 days 6
[Ask if PF32 = 1 AND PF46 = 1 AND (PF70 = 2 OR 3 OR 4 OR 5 OR 6]
71. Why did you delay reporting the situation? Select “Yes” or “No” for each
item.

Did not realize at first that the situation was a
crime PF71_1

Yes 1

No 2

Had to figure out how to report PF71_2

Yes 1

No 2

Wanted to think about the situation before deciding to report PF71_3

Yes 1

No 2

Survey Instrument

Wanted to seek advice first from a friend or family
member PF71_4

Yes 1

No 2

Wanted to seek advice/counseling from a professional (e.g., medical personnel, chaplain, mental
health counselor, Safe Helpline) before deciding to
report PF71_5

Yes 1

No 2

Waited until you felt safe from the offender
PF71_6

Yes 1

No 2

Waited until you could reach a specific authority
(e.g., your chaplain, your doctor, your commander)
PF71_7

Yes 1

No 2

Decided to report after receiving training or a
briefing on sexual assault PF71_8

Yes 1

No 2

Researched sexual assault before deciding to
report PF71_9

Yes 1

No 2

Was in a location where you could not contact an
authority PF71_10

Yes 1

No 2

199

[Ask if PF32 = 1 AND PF46 = 2]
72. What were your reasons for not reporting the situation to any of the installation/Service/DoD authorities or organizations? Select “Yes” or “No” for
each statement.

You thought it was not important enough to
report. PF72_1

Yes 1

No 2

You did not know how to report. PF72_2

Yes 1

No 2

You felt uncomfortable making a report. PF72_3

Yes 1

No 2

You did not think anything would be done.
PF72_4

Yes 1

No 2

You heard about negative experiences other victims went through who reported their situation.
PF72_5

Yes 1

No 2

You thought you would not be believed. PF72_6

Yes 1

No 2

You thought reporting would take too much time
and effort. PF72_7

Yes 1

No 2

You were afraid of retaliation/reprisals from the
person(s) who did it or from their friends. PF72_8

Yes 1

No 2

You thought your performance evaluation or
chance for promotion would suffer. PF72_9

Yes 1

No 2

You thought you would be labeled a troublemaker. PF72_10

Yes 1

No 2

200

Sexual Assault and Sexual Harassment in the U.S. Military

You did not want anyone to know. PF72_11

Yes 1

No 2

You did not think your report would be kept confidential. PF72_12

Yes 1

No 2

You feared you or others would be punished for
infractions/violations, such as underage drinking
or fraternization. PF72_13

Yes 1

No 2

You were afraid of being assaulted again by the
offender. PF72_14

Yes 1

No 2

You thought you might lose your security clearance/personnel reliability certification. PF72_15

Yes 1

No 2

[Ask if PF32 = 1 AND (PF46 = 1 OR 2)]
73. In retrospect, would you make the same decision about reporting if you
could do it over? PF73
Yes 1
No 2
[Ask if PF32 = 1]
SAFU25. It can be difficult to remember the exact date when events occurred. In this
study, it is important to know which events happened in the last 12 months, and which
events happened earlier.
Thinking about when the event occurred, how certain are you that it occurred in
the last 12 months? If the event occurred over a long time, think about whether it ever
happened after [X Date]. SAFU25
1) Definitely occurred AFTER [X Date]. 1
2) Not sure if it occurred BEFORE OR AFTER [X Date]. 2
3) Definitely occurred BEFORE [X Date]. 3

Survey Instrument

201

In early 2014, the Department of Defense Sexual Assault Prevention and
Response Office asked the RAND National Defense Research Institute to conduct
an independent assessment of sexual assault, sexual harassment, and gender
discrimination in the military—an assessment last conducted in 2012 by the
department itself with its Workplace and Gender Relations Survey of Active Duty
Personnel. This volume documents the methodology used in the resulting RAND
Military Workplace Study, which invited close to 560,000 service members to
participate in a survey fielded in August and September of 2014. It describes the
survey methods, how the new questionnaire was designed, and how sampling,
recruitment, and analytic weighting were pursued. It also includes the entire
survey instrument.

N AT I ONAL DEFENSE R ESEAR C H IN S TITUTE
$60.00

www.rand.org

ISBN-10 0-8330-8860-2
ISBN-13 978-0-8330-8860-4
56000

9 780833 088604
RR- 870/1- OSD

File Type	application/pdf
File Title	Sexual Assault and Sexual Harassment in the U.S. Military: Volume 1. Design of the 2014 RAND Military Workplace Study
Subject	This volume documents the methodology used in the 2014 RAND Military Workplace Study, an independent assessment of sexual assaul
Author	Edited by Andrew R. Morral, Kristie L. Gore, Terry L. Schell
File Modified	2014-12-04
File Created	2014-12-03