This is the accessible text file for GAO report number GAO-07-148R entitled 'Prevalence of Domestic Violence, Sexual Assault, Dating Violence, and Stalking' which was released on November 13, 2006. This text file was formatted by the U.S. Government Accountability Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. November 13, 2006: Congressional Committees: Subject: Prevalence of Domestic Violence, Sexual Assault, Dating Violence, and Stalking: In hearings conducted between 1990 and 1994, Congress noted that violence against women was a problem of national scope and that the majority of crimes associated with domestic violence, sexual assault, and stalking were perpetrated against women. These hearings culminated in the enactment of the Violence Against Women Act (VAWA) in 1994 to address these issues on a national level.[Footnote 1] VAWA established grant programs within the Departments of Justice (DOJ) and Health and Human Services (HHS) for state, local, and Indian tribal governments and communities. These grants have various purposes, such as providing services to victims and training for law enforcement officers and prosecutors. The 2006 reauthorization of VAWA expanded existing grant programs and added new programs addressing, among other things, young victims, the housing and economic needs of victims, and the health care system's response to domestic violence, dating violence, sexual assault, and stalking.[Footnote 2] The total fiscal year 2006 appropriation level for violence against women programs is about $560 million--approximately $382 million for programs administered by DOJ and about $178 million for programs administered by HHS.[Footnote 3] Although criminal justice, health, and domestic violence experts believe that valid and reliable estimates have the potential to be of use to policy makers, service providers, and researchers, there are concerns that current crime statistics do not provide a full assessment of the problem. The Violence Against Women and DOJ Reauthorization Act of 2005, enacted January 5, 2006, requires GAO to conduct a study and report on data indicating the prevalence of domestic violence, dating violence, sexual assault, and stalking among men, women, youth, and children, as well as services available to these victims.[Footnote 4] We developed two objectives to respond to this mandate. 1. To what extent do national data collection efforts report prevalence of men, women, youth, and children who are victims of domestic violence, sexual assault, dating violence, and stalking? 2. What support services (e.g., counseling, medical, legal, etc.) are available to victims of these categories of crime and what are the number and characteristics of victims receiving these services by type of service? This report addresses the first objective. Our work on the second objective is ongoing. To assess the extent to which national data collection efforts report prevalence of victims of domestic violence, sexual assault, dating violence, and stalking, we obtained information from and interviewed officials at DOJ's Office of Justice Programs' Bureau of Justice Statistics, National Institute of Justice, and Office of Juvenile Justice and Delinquency Prevention, Office on Violence Against Women, and Federal Bureau of Investigation's Criminal Justice Information Services Division. We also obtained information and interviewed officials at HHS's National Institutes of Health and Centers for Disease Control and Prevention's National Center for Injury Prevention and Control--Division of Violence Prevention. Further, we gathered information from research and advocacy organizations related to the crimes under study, including the National Academy of Sciences' National Research Council; Respecting Accuracy in Domestic Abuse Reporting; Men's Health Network; Rape, Abuse, and Incest National Network; National Center for Victims of Crime; National Domestic Violence Hotline; and Stalking Resource Center. We reviewed pertinent federal laws and conducted literature searches, focusing on reporting systems and surveys from which results were issued or reported since 2001.[Footnote 5] However, we did not independently evaluate the methodology used in any of these studies and we are not making any assessments regarding their overall merit. In October 2006, we briefed your offices on the results of our work or provided a copy of our briefing slides to your staff. This report conveys the information provided during those discussions. We conducted our work from April through October 2006 in accordance with generally accepted government auditing standards. Results: Since 2001, the amount of national research that has been conducted on the prevalence of domestic violence and sexual assault has been limited, and even less research has been conducted on dating violence and stalking. No single, comprehensive effort currently exists that provides nationwide statistics on the prevalence of these four categories of crime among men, women, youth, and children. Rather, various national efforts address certain subsets of these crime categories among some segments of the population. For example, the Centers for Disease Control and Prevention's (CDC) National Violent Death Reporting System, which collects incident-based data from multiple sources, such as coroner/medical examiner reports, gathers information on violent deaths, including those resulting from domestic violence and sexual assaults. This system and the other national data collection efforts were not intended to provide comprehensive estimates on the prevalence associated with domestic violence, sexual assault, dating violence, and stalking. Some of these national data collection efforts focus largely on incidence--the number of separate times a crime is committed against individuals during a specific time period-- rather than prevalence--the unique number of individuals who were victimized during a specific time period.[Footnote 6] Obtaining both incidence and prevalence data is important for determining services to provide to victims of crimes. In addition, HHS noted that both types of data are important for determining the impact of violence and strategies to prevent it from occurring. Table 1 in the attached briefing slides (see enc. II, p. 29) shows the 11 national efforts we identified that have reported data since 2001 on certain aspects of domestic violence, sexual assault, dating violence, and stalking. More detailed information about these efforts is contained at the end of enclosure II. The national data collection efforts we reviewed cannot provide a basis for combining their results to compute valid and reliable nationwide prevalence estimates because the efforts use varying definitions. For example, CDC's Youth Risk Behavior Surveillance System's definition of dating violence included the intentional physical harm inflicted upon a survey respondent by a boyfriend or girlfriend. In contrast, the Victimization of Children and Youth Survey's definition did not address whether the physical harm was intentional. Officials from the National Institute of Justice (NIJ) and Bureau of Justice Statistics (BJS) acknowledge that estimating prevalence in the absence of widely accepted and used uniform definitions is a challenge. Certain agencies have taken steps to build consistency in some of their collection efforts. For example, CDC, in collaboration with the Office of Justice Programs (OJP) and others, established uniform definitions for certain forms of domestic violence in 1999 and for sexual assault in 2002, with the intent of promoting and improving consistency among the research community. CDC and OJP encourage but do not require grantees to use these definitions as part of their research efforts and cannot always use these definitions in their own work. Although CDC and OJP acknowledge that using standard definitions of these offenses may be advantageous, they believe there are circumstances that preclude such use. For example, CDC officials said that some of the current data collection efforts, such as the Youth Risk Behavior Surveillance System, were developed before CDC established the uniform definitions. They also said it would be difficult to alter the definitions used in these efforts because the efforts continue to be used to provide comparable data to measure trends over time in the United States. However, in 2004, the National Academy of Sciences' National Research Council reported that top priority should be given to developing clear definitions and cautioned that without consistency in the use of terms across studies, accurate prevalence estimates will remain elusive. Further, the national data collection efforts we reviewed cannot provide a basis for combining their results to compute valid and reliable nationwide prevalence estimates because the efforts have varying scopes in terms of the incidents and categories of victims that are included. For example, the National Child Abuse and Neglect Data System includes only reported sexual assaults against children, not unreported incidents. Furthermore, as noted earlier, CDC's Youth Risk Behavior Surveillance System definition of dating violence includes intentional physical harm inflicted upon a survey respondent, but excludes youth who are not in grades 9-12 and those who do not attend school. In contrast, the Victimization of Children and Youth Survey was addressed to youth ages 12 and older, or those who were at least in the sixth grade. Although perfect data may never exist because of the sensitivity of these crimes and the likelihood that not all occurrences will be disclosed, initiatives are under way to provide additional information related to the prevalence of these issues. For example: ² Domestic violence, sexual assault, and stalking initiatives: CDC, NIJ, and the Department of Defense are collaborating on a National Intimate Partner and Sexual Violence Survey to address certain forms of domestic violence, sexual assault, and stalking. CDC reported that it expects the survey to be fully administered by 2008. Although the survey will gather information regarding experiences that occurred during an individual's life span, it will not be administered to people under age 18. ² Other stalking initiatives: The Office on Violence Against Women and BJS told us that they collaborated to conduct the National Crime Victimization Survey--Stalking Supplement and expect to report results in the summer of 2007. According to BJS, this survey will obtain information about the identity of the stalker, nature of the stalking incidents, consequences to the victim, and actions the victim took about the incident, including whether it was reported to the police. However, while this supplement will gather data on stalking incidents involving adult victims, it will not collect information on stalking associated with youth ages 12-17. ² Other domestic violence and sexual assault initiatives: CDC began collecting data through a telephone survey on intimate partner violence and sexual violence as part of its ongoing Behavioral Risk Factor Surveillance System. In 2005, CDC administered the intimate partner violence module to approximately 77,000 people in 16 states and administered the sexual violence module to about 115,000 people in 26 states. In addition to these efforts, under an Office of Juvenile Justice and Delinquency Prevention (OJJDP) grant, the University of New Hampshire is planning to conduct the National Study on Children's Exposure to Violence. DOJ officials told us the data will be collected from September 2007 to June 2008, and OJJDP officials said that the study will assess variations in incidence and lifetime prevalence of children's exposure to a broad array of violence and abuse. Furthermore, NIJ recently sponsored two seminars aimed at identifying key issues related to measuring the prevalence of dating violence, domestic violence, and sexual assault against women, improving interagency coordination on these issues, and highlighting the results of the latest efforts on domestic violence and sexual assault. If these efforts are completed as planned, CDC and DOJ will make progress in collecting information needed to determine the extent to which men, women, youth, and children are victims of domestic violence, sexual assault, dating violence, and stalking. However, some information gaps will remain, particularly in the areas of dating violence among victims age 12 and older and stalking among victims under age 18. To cost-effectively address information gaps, it is important to consider additional costs that would be incurred in collecting new or different data as well as the usefulness of such data. It is equally important to consider the benefits resulting from the use of these data (different allocations of resources) and the availability of funds to collect such data (a cost-benefit analysis). According to DOJ officials, a cost-benefit analysis should precede any future large- scale effort aimed at national prevalence estimates. Conclusions: Current national data collection efforts cover portions of domestic violence, sexual assault, dating violence, and stalking among different segments of the population. Because the efforts use different definitions and vary in scope, they cannot be combined and leveraged to determine the nationwide prevalence of these categories of crime. The absence of comprehensive nationwide prevalence information somewhat limits the ability to make informed policy and resource allocation decisions about the statutory requirements and programs created to help address these four categories of crime and victims. Although policymakers may never have perfect data, DOJ and HHS have collaborated to obtain more uniformity across research efforts and have initiatives in the early stages that could be used to collect information regarding the prevalence of certain crimes for some segments of the population. However, some information gaps will remain. DOJ and HHS have not yet determined to what extent they can cost-effectively revise current efforts or design new initiatives so as to collect more consistent data that can be combined to better build nationwide estimates. To decide on the cost-effectiveness of obtaining better data, policymakers would need to consider the marginal costs to collect more or different data, the utility of obtaining better data, benefits to be derived from the use of better data (different allocations of resources), and availability of funds to gather better data. Recommendations for Executive Action: To provide Congress and agency decision makers with more comprehensive information on the prevalence of domestic violence, sexual assault, dating violence, and stalking to assist them in making policy decisions on grants and other issues associated with these four categories of crime, we are recommending that the Attorney General and the Secretary of Health and Human Services collaboratively take the following four actions: ² determine the extent to which initiatives being planned or under way can be designed or modified to address existing information gaps; ² identify and evaluate alternatives for addressing any remaining gaps; ² incorporate such alternatives deemed cost-effective in future budget requests; and: ² to the extent possible, require the use of common definitions when conducting or providing grants for federal research to leverage individual collection efforts so that the results of such efforts can be readily combined to achieve nationwide prevalence estimates. Agency Comments: We provided a draft copy of this report with the attached briefing slides to HHS and DOJ for comment. HHS provided formal written comments on a draft of this report on October 25, 2006, which are presented in enclosure III. In commenting on the draft report, HHS concurred with the recommendations and stated that it will continue to expand its collaborations with DOJ to improve data collection and monitoring of violence. HHS also provided technical comments, which we have incorporated as appropriate. DOJ declined to provide formal written comments on a draft of this report. However, DOJ provided technical comments on the draft briefing slides, which we incorporated as appropriate. In its technical comments, DOJ expressed concern regarding the potential costs associated with implementing our proposed recommendations and suggested that a cost-benefit analysis be conducted. We agree that performing a cost-benefit analysis is a critical step, as acknowledged by our recommendation that DOJ and HHS incorporate alternatives for addressing information gaps deemed cost-effective in future budget requests. DOJ officials also expressed concern that our work was primarily focused on issues associated with prevalence data. As discussed in this report and the attached briefing slides, we believe obtaining information on both prevalence and incidence data is important for determining services to provide to victims of crime. However, we did not conduct a detailed analysis of incidence data because doing so was outside the scope of our review. We are sending copies to the Senate Committee on Homeland Security and Governmental Affairs, the House Committee on Government Reform, the Secretary of Health and Human Services, the Attorney General, the Director of the Office of Management and Budget, and other interested parties. We will also make copies available to others on request. In addition, the report will be available on GAO's Web site at [Hyperlink, http://www.gao.gov]. If your office or staff have any questions concerning this report, please contact me at (202) 512-8777 or by e-mail at Larencee@gao.gov. Other GAO contacts and key contributors to this report are listed in enclosure IV. Signed by: Eileen Regan Larence, Director: Homeland Security and Justice Issues: Enclosures (4): Enclosure I: List of Committees: The Honorable Arlen Specter: Chairman: The Honorable Patrick J. Leahy: Ranking Minority Member: Committee on the Judiciary: United States Senate: The Honorable Michael B. Enzi: Chairman: The Honorable Edward M. Kennedy: Ranking Minority Member: Committee on Health, Education, Labor, and Pensions: United States Senate: The Honorable F. James Sensenbrenner, Jr. Chairman: The Honorable John Conyers, Jr. Ranking Minority Member: Committee on the Judiciary: House of Representatives: The Honorable Joe Barton: Chairman: The Honorable John D. Dingell: Ranking Minority Member: Committee on Energy and Commerce: House of Representatives: Enclosure II: Briefing to Congressional Committees: Prevalence of Domestic Violence, Sexual Assault, Dating Violence, and Stalking: Briefing to Congressional Committees: Introduction: In hearings conducted between 1990 and 1994, Congress noted that violence against women was a problem of national scope and that the majority of crimes associated with domestic violence, sexual assault, and stalking were perpetrated against women. These hearings culminated in the enactment of the Violence Against Women Act (VAWA) in 1994 to address these issues on a national level.[Footnote 7] VAWA established grant programs within the Departments of Justice (DOJ) and Health and Human Services (HHS) for state, local, and Indian tribal governments and communities. These grants have various purposes, such as providing services to victims and training for law enforcement officers and prosecutors. The total fiscal year 2006 appropriation level for violence against women programs is about $560 million- approximately $382 million for programs administered by DOJ and about $178 million for programs administered by HHS.[Footnote 8] Although the Bureau of Justice Statistics (BJS) reported a decline in family violence[Footnote 9] between 1993 and 2002, some service providers and advocacy groups do not believe current crime statistics provide a full assessment of the problem. * For example, National Domestic Violence Hotline officials cautioned that statistical reports must be used carefully because they do not account for all cases of violence, such as those that are unreported.[Footnote 10] Criminal justice, health, and domestic violence experts believe that valid and reliable estimates have the potential to be of use to policymakers, service providers, researchers, and others in determining the success of programs to combat domestic violence, sexual assault, dating violence, and stalking and the need for changes or additions to these programs. * The Violence Against Women and Department of Justice Reauthorization Act of 2005, enacted January 5, 2006, requires GAO to conduct a study and report on data indicating the prevalence of domestic violence, dating violence, sexual assault, and stalking among men, women, youth, and children, as well as the services available to these victims.[Footnote 11] Objectives: In responding to this mandate, we outlined two objectives. We assessed: 1. to what extent national data collection efforts report prevalence of men, women, youth, and children who are victims of domestic violence, sexual assault, dating violence, and stalking and: 2. the support services (e.g., counseling, medical, legal, etc.) available to victims of these categories of crime and the numbers and characteristics of victims receiving these services by type of service. This briefing addresses the first objective. Our work on the second objective is ongoing. Scope and Methodology: To address the first objective, we obtained information from and interviewed officials of the following key federal entities because they are involved in ongoing efforts to (1) collect and maintain information or (2) conduct or fund research to address certain aspects of domestic violence, sexual assault, dating violence, and stalking: * DOJ: - Office of Justice Programs' BJS, National Institute of Justice (NIJ), Office of Juvenile Justice and Delinquency Prevention (OJJDP): - Office on Violence Against Women (OVW): - Federal Bureau of Investigation's (FBI) Criminal Justice Information Services Division: * HHS: - National Institutes of Health: - Centers for Disease Control and Prevention's (CDC) National Center for Injury Prevention and Control-Division of Violence Prevention: Through research efforts, we also identified the following research or advocacy organizations and contacted them to obtain information related to the prevalence of these categories of crime: * National Academy of Sciences' National Research Council: * Respecting Accuracy in Domestic Abuse Reporting: * Men's Health Network: * Rape, Abuse, and Incest National Network: * National Center for Victims of Crime: * National Domestic Violence Hotline: * Stalking Resource Center: We reviewed pertinent federal laws related to domestic violence, sexual assault, dating violence and stalking. We conducted literature searches of DOJ publications, HHS publications, prior GAO report Congressional Research Service reports., and the Internet. As assault we reviewed information on pertinet national health and crime reporting systems and surveys that are ongoing, episodic and onetimeffrts from federal and non-federal sources. To obtain recent information, we focused on reporting systems and surveys from which results were issued or re ported since 2001.[Footnote 12] We did not independently evaluate the methodology used in any of these studies. We attended an NIJ workshop that gathered representatives from various program and research funding agencies to discuss the results of recent research and data collection efforts. We conducted our work from April through October 2006 in accordance with generally accepted government auditing standards. Results in Brief: No single, comprehensive effort currently exists that provides nationwide statistics on the prevalence of these four categories of crime for men, women, youth, and children. Designing a single effort would be costly given the resources required to collect such data and may be duplicative of some existing efforts. Available national data collection efforts contain information on various subsets of domestic violence, sexual assault, dating violence, and stalking among certain segments of the population and were not intended to provide comprehensive estimates. Some of these efforts collect data on incidence rather than prevalence. Both are important for appropriately determining resources needed to provide services to victims of crimes. * Incidence refers to the number of separate times a crime is committed against individuals during a specific time period. * Prevalence is the unique number of individuals who were victimized during a specific time period. We cannot combine the results of the various collection efforts to estimate the prevalence of these four categories of crime nationwide among all segments of the population because the efforts: * use different definitions to measure the various offenses, and: * have varying scopes such as including different categories of victims and not always including estimates of unreported incidents. Several initiatives are under way that could help address some information gaps if they are completed as planned, but other gaps will remain, such as in the area of dating violence. To provide Congress with more comprehensive information on the prevalence of domestic violence, sexual assault, dating violence, and stalking to assist in carrying out its legislative and oversight agenda, we are recommending that DOJ and HHS: * determine the extent to which initiatives being planned or under way can be designed to address existing gaps in information on domestic violence, sexual assault, dating violence, and stalking, * identify and evaluate alternatives to address any remaining gaps, * incorporate alternatives deemed cost-effective in future budget requests, and: * to the extent possible, require the use of common definitions when conducting or providing grants for federal research to leverage individual collection efforts so that the results of such efforts can be readily combined to achieve nationwide prevalence estimates. Background: Domestic violence, sexual assault, dating violence, and stalking are complex subjects. They can include many forms of violence, such as verbal threats, physical assaults, murder, and rape and a wide range of victims (e.g., spouses, intimate partners, children, and other family members). Some of these forms of violence have not always been considered crimes. The offenses involved in these forms of violence, like most violent crimes, generally are prosecuted at the state and local levels. VAWA created a number of grant programs to address domestic violence, sexual assault, and stalking as well as authorized additional funding for domestic violence shelters.[Footnote 13] In 2000, during the reauthorization of VAWA, language was added to the law to provide greater emphasis on dating violence in efforts to address violence against women. The 2006 reauthorization of VAWA expanded existing grant programs and added new programs addressing, among other things, young victims, the housing and economic needs of victims, and the health care system's response to domestic violence, dating violence, sexual assault, and stalking.[Footnote 14] Limited National Data Are Available to Estimate Prevalence: Since 2001, the amount of national research that has been conducted on the prevalence of domestic violence and sexual assault has been limited, and even less research has been conducted on dating violence and stalking. No single, comprehensive data collection effort has been conducted to determine the prevalence of these four categories of crime among men, women, youth, and children. Rather, various national efforts address certain subsets of these categories of crime among some segments of the population and were not intended to provide comprehensive estimates. Designing a single, comprehensive data collection effort to address these four categories of crime among all segments of the population independent of existing efforts would be costly, given the resources required to collect such data. Furthermore, it would be inefficient to duplicate some existing efforts that already collect data for certain aspects of these categories of crime. Some of these efforts focus largely on incidence rather than prevalence. * Incidence refers to the number of separate times a crime is committed against individuals during a specific time period. * Prevalence is the unique number of individuals who were victimized during a specific time period. The following hypothetical statements illustrate incidence and prevalence: * 4,110 separate occurrences of domestic violence against women were reported during 1938 (incidence), whereas: * 2,500 women were victims of these 4,110 occurrences (prevalence). Obtaining both incidence and prevalence data is important for determining services to provide to victims of crimes. In addition, HHS noted that both types of data are important for determining the impact of violence and strategies to prevent it from occurring. The National Violence Against Women Prevention Research Center reported that to effectively measure rape, it is important to determine how many rape cases have occurred (incidence) and how many women have ever been raped (prevalence) to determine the level of services that the state will need to provide to victims.[Footnote 15] As reflected in table 1, we identified 11 national efforts that have reported data on certain aspects of these categories of crime. These efforts provide a mixture of prevalence and incidence data. Additional information on these efforts is provided in appendix I. Table 1: National Data Collection Efforts with Results Reported or Issued since 2001 Relevant to Domestic Violence, Sexual Assault, Dating Violence, and Stalking. Data collection effort: Summary Uniform Crime Reporting Program; categories of crime: Sexual Assault; Prevalence data: No; Incidence data: Yes. Data collection effort: National Incident-based Reporting System; categories of crime: Domestic violence, Sexual assault, stalking; Prevalence data: No; Incidence data: Yes. Data collection effort: national crime Victimization Survey; categories of crime: Domestic violence, sexual assault; Prevalence data: Potentially[A]; Incidence data: Yes. Data collection effort: National Violent Death Reporting System[B]; categories of crime: Domestic violence, sexual assault; Prevalence data: Yes; Incidence data: Yes. Data collection effort: Youth Risk Behavior Surveillance System; categories of crime: Sexual assault, dating violence; Prevalence data: Yes; Incidence data: no. Data collection effort: National Child Abuse and Neglect Data System; categories of crime: Sexual Assault; Prevalence data: Yes; Incidence data: No. Data collection effort: National Electronic Injury Surveillance System- All injury program; categories of crime: Domestic violence, sexual assault; Prevalence data: No; Incidence data: Yes. Data collection effort: Extent, nature, and Consequences of Rape Victimization: Findings from the National Violence Against Women Survey; categories of crime: Sexual Assault; Prevalence data: Yes; Incidence data: Yes. Data collection effort: The Harris Poll #50; categories of crime: Domestic violence, sexual assault; Prevalence data: yes; Incidence data: No. Data collection effort: The Victimization of Children and Youth; categories of crime: Sexual assault, dating violence; Prevalence data: Yes; Incidence data: Yes. Data collection effort: Injury Control and Risk Survey-2; categories of crime: Stalking; Prevalence data: Yes; Incidence data: Yes. [A] Although BJS collects data that could be used to determine prevalence related to domestic violence and sexual assault, BJS has declined to produce prevalence estimates because of the difficulties and complexities associated with the task. [B] This system contains data only on fatalities, thus, an individual is entered into the system one time upon death. Therefore, the prevalence and incidence rates for data in this system would be the same. Source: GAO's analysis of national data collection efforts. [End of table] Varying Definitions and Scope Make It Difficult to Combine Results for Nationwide Estimates: In 2004, the National Academy of Sciences' National Research Council reported that currently available information on prevalence was inadequate because it had been derived from efforts with varying definitions and scope. The national data collection efforts we reviewed cannot provide a basis for combining their results to compute valid and reliable nationwide prevalence estimates because the efforts use varying definitions and have varying scopes. National data collection efforts currently available use different definitions to measure these various forms of violence. * For example, CDC's Youth Risk Behavior Surveillance System's definition of dating violence included the intentional physical harm inflicted upon a survey respondent by a boyfriend or girlfriend. In contrast, the Victimization of Children and Youth Survey's definition did not address whether the physical harm was intentional. NIJ and BJS officials acknowledge that estimating prevalence in the absence of widely accepted and used uniform definitions is a challenge. Agencies have taken steps to build consistency into some of their collection efforts. * For example, CDC, in collaboration with OJP and others, established uniform definitions for certain forms of domestic violence in 1999 and for sexual assault in 2002, with the intent of promoting and improving consistency among the research community. CDC and OJP encourage but do not require grantees to use these definitions as part of their research efforts and report they cannot always use these definitions in their own work when these efforts began before the definitions were developed. Although CDC and OJP acknowledge that using standard definitions of these offenses may be advantageous, they believe there are circumstances that preclude such use. * For example, CDC officials said that some of the current data collection efforts, such as the Youth Risk Behavior Surveillance System, were developed before CDC established the uniform definitions. They also said it would be difficult to alter the definitions used in these efforts because the efforts continue to be utilized to provide comparable data to measure trends over time in the United States. In 2004, the National Academy of Sciences' National Research Council reported that top priority should be given to developing clear definitions and cautioned that without consistency in the use of terms across studies, accurate prevalence estimates will remain elusive. Varying Scopes Make It Difficult to Combine Results for Nationwide Estimates: National data collection efforts also have varying scopes in terms of the incidents and categories of victims and that are included. * For example, the National Child Abuse and Neglect Data System includes only reported sexual assaults against children, not unreported incidents. * Furthermore, as noted earlier, CDC's Youth Risk Behavior Surveillance System's definition of dating violence includes intentional physical harm inflicted upon a survey respondent in grades 9-12. In contrast, the question on dating violence included in the Victimization of Children and Youth Survey was addressed to youth a ages 12 and older, or those who were at least in the sixth grade, but its definition of dating violence did not address if the physical harm was intentional. Without comparable information, including both reported and unreported incidents, it is not possible to combine prevalence estimates from national data collection efforts, and these efforts likely underestimate the prevalence of these categories of crime. Recent Initiatives May Address Some Information Gaps: Perfect data may never exist because of the sensitive nature of these issues and the likelihood that all occurrences related to domestic violence, sexual assault, dating violence, and stalking will not be disclosed. However, initiatives are under way to provide additional information related to the prevalence of these issues. If these efforts are completed as planned, CDC and DOJ will make progress in collecting information needed to determine the extent to which men, women, youth, and children are victims of these four categories of crime. However, some information gaps will remain, particularly in the areas of dating violence among victims ages 12 and older and stalking among victims under age 18. To cost-effectively address information gaps, it is important to consider additional costs that would be incurred in collecting new or different data as well as the usefulness of such data. It is equally important to consider the benefits resulting from the use of these data (different allocations of resources) and the availability of funds to collect such data (a cost-benefit analysis). According to DOJ officials, no cost-benefit analysis has been performed, and such an analysis should precede any future large-scale effort aimed at national prevalence estimates. Domestic violence, sexual assault, and stalking initiatives: CDC, NIJ, and the Department of Defense are collaborating on a National Intimate Partner and Sexual Violence Survey to address certain forms of domestic violence, sexual assault, and stalking. CDC reported that it expects the survey to be fully administered by 2008. Although the survey will gather information regarding experiences that occurred during an individual's life span, it will not be administered to victims under age 18. Other stalking initiatives: OVW and BJS told us that the collaborated to conduct the National Crime Victimization Survey-Stalking Supplement and expect to report results in the summer of 2007. * According to BJS, this survey will obtain information about the identity of the stalker, nature of the stalking incidents, consequences to the victim, and actions the victim took about the incident including whether it was reported to the police. * However, while this supplement will gather data on stalking incidents involving adult victims, it will not collect information on stalking associated with Youth ages 12-17. Other domestic violence and sexual assault initiatives: CDC began collecting data through a telephone survey on intimate partner violence and sexual violence as part of its ongoing Behavioral Risk Factor Surveillance System. In 2005, CDC administered the intimate partner violence module to approximately 77,000 people in 16 states and administered the sexual violence module to about 115,000 people in 26 states. Under an OJJDP grant, the University of New Hampshire is planning to conduct the National Study on Children's Exposure to Violence. DOJ officials told us the data will be collected from September 2007 to June 2008. * OJJDP officials said that the study will assess variations in incidence and lifetime prevalence of children's exposure to a broad array of violence and abuse. NIJ recently sponsored two seminars aimed at identifying key issues related to measuring the prevalence of dating violence, domestic violence, and sexual assault against women, improving interagency coordination on these issues, and highlighting the results of the latest efforts on domestic violence and sexual assault. * In July 2006, NIJ sponsored an interagency seminar to discuss challenges associated with measuring the prevalence of dating violence. * In August 2006, NIJ sponsored a Violence Against Women seminar for program and research funding agencies to present the results of recent prevalence studies and to discuss unresolved questions that may guide future research efforts. Conclusions: Current national data collection efforts cover portions of these four categories of crime among different segments of the victim population. Because the efforts use different definitions and vary in scope, they cannot be combined and leveraged to determine the nationwide prevalence of these categories of crime. The absence of comprehensive nationwide prevalence information somewhat limits the ability to make informed policy and resource allocation decisions about the statutory requirements and programs created to help address these four categories of crime and victims. Although policymakers may never have perfect data, DOJ and HHS have collaborated to obtain more uniformity across research efforts and have initiatives in the early stages that could be used to collect information regarding the prevalence of certain crimes for some segments of the population. However, some information gaps will remain. DOJ and HHS must determine to what extent they can cost-effectively revise current efforts or design new initiatives so as to collect more consistent data that can be combined to better build nationwide estimates. Policymakers need to decide whether it is worth the cost to obtain better data. This decision should be based on judgments about the: * marginal costs to collect more or different data, * utility of obtaining better data, * benefits to be derived from the use of better data (different allocations of resources), and: * availability of funds to gather better data. Recommendations for Executive Action: To provide Congress and agency decision makers with more comprehensive information on the prevalence of domestic violence, sexual assault, dating violence, and stalking to assist them in making policy decisions on grants and other issues associated with these four categories of crime, we are recommending that the Attorney General and the Secretary of Health and Human Services collaboratively take the following four actions: * determine the extent to which initiatives being planned or under way can be designed or modified to address existing information gaps, * identify and evaluate alternatives for addressing any remaining gaps, * incorporate such alternatives deemed cost-effective in future budget requests, and: * to the extent possible, require the use of common definitions when conducting or providing grants for federal research to leverage individual collection efforts so that the results of such efforts can be readily combined to achieve nationwide prevalence estimates. Agency Comments and Our Evaluation: We provided draft copies of these briefing slides to HHS and DOJ for comment. HHS concurred with the recommendations and stated that it will continue to expand its collaborations with DOJ to improve data collection and monitoring of violence. HHS and DOJ provided technical comments, which we have incorporated as appropriate. In its technical comments, DOJ expressed concern regarding the potential costs associated with implementing our proposed recommendations and suggested that a cost-benefit analysis be conducted. We agree that performing a cost-benefit analysis is a critical step, as acknowledged by our proposed recommendation that DOJ and HHS incorporate alternatives for addressing information gaps deemed cost-effective in future budget requests. DOJ officials also expressed concern that our briefing was primarily focused on issues associated with prevalence data. As discussed in the briefing, we believe obtaining information on both prevalence and incidence data is important for determining services to provide to victims of crime. Appendix I: National Data Collection Efforts with Results Reported or Issued since 2001: We identified 11 national data collection efforts that address various aspects of domestic violence, sexual assault, dating violence, and stalking from which results were reported or issued since 2001. The national data collection efforts are discussed in table 2 and include information on (1) the agency or sponsor responsible for conducting the effort; (2) whether domestic violence, sexual assault, dating violence, or stalking was included in the scope of the effort; (3) the frequency in which the effort is conducted; and (4) the data limitations associated with the determination of reliable prevalence estimates related to domestic violence, sexual assault, dating violence, and stalking on a national basis. These efforts provide a mixture of prevalence and incidence data. Table 1: National Data Collection Efforts with Results Reported or Issued since 2001 Relevant to Domestic Violence, Sexual Assault, Dating Violence, and Stalking: Data collection efforts: Summary Uniform Crime Reporting Program (UCR); [Hyperlink, http://www.fbi.gov/ucr/ucr.htm], [Hyperlink, http://www.fbi.gov/ucr/05cius/]; Agency/sponsor: Federal Bureau of Investigation; Description: Summary UCR is a DOJ statistical program designed to measure the magnitude, nature, and impact of certain crimes in the United States. More than 17,000 law enforcement agencies nationwide (about 94 percent of the total population in 2005) voluntarily report crime data brought to their attention to the Federal Bureau of Investigation, and findings are published in a detailed annual report; Categories of crime: Sexual assault; Data Collection Frequency: Ongoing; Data limitations associated with the determination of reliable nationwide prevalence estimates: * Does not include crimes that are not reported to the police; * This effort does not include the full range of sexual assaults (i.e., male rapes, forcible groping); * Does not include data on prevalence. Data collection efforts: National Incident-Based Reporting System (NIBRS); [Hyperlink, http://www.fbi.gov/ucr/ucr.htm], [Hyperlink, http://www.fbi.gov/filelink.html?file=/ucr/cius_03/pdf/03sec5.pdf]; See PDF file page 5 (report page 341) under center column heading labeled "DATA"; Agency/ sponsor: Federal Bureau of Investigation; Description: NIBRS is an incident-based reporting system designed to collect more detailed information than is reported under the traditional Summary UCR program. According to DOJ, to date about 33 percent of the 17,000 law enforcement agencies nationwide report crime information to the FBI using NIBRS; Categories of crime: Domestic violence, sexual assault, stalking; Data Collection Frequency: Ongoing; Data limitations associated with the determination of reliable nationwide prevalence estimates: * Does not include crimes that are not reported to the police; * Currently, 30 states as well as the District of Columbia are NIBRS certified. Of these, 10 have participation from all law enforcement agencies in their state; * Does not include data on prevalence. Data collection efforts: National Crime Victimization Survey (NCVS); [Hyperlink, http://www.ojp.usdoj.gov/bjs/cvict.htm], [Hyperlink, http://www.ojp.usdoj.gov/bjs/abstract/cv05.htm]; Agency/sponsor: Bureau of Justice Statistics; Description: NCVS is a DOJ statistical program designed to measure the magnitude, nature, and impact of certain crimes in the United States, including crimes reported and not reported to the police. Twice a year, the U.S. Census Bureau interviews household members ages 12 and over in a nationally representative sample of approximately 42,000 households (about 75,000 people); Categories of crime: Domestic violence, sexual assault; Data Collection Frequency: Ongoing; Data limitations associated with the determination of reliable nationwide prevalence estimates: * Does not include individuals under age 12; * Does not include individuals not living in a household; * Does not include homicides; * Although BJS collects data that could be used to determine prevalence related to domestic violence and sexual assault, BJS has declined to produce prevalence estimates because of the difficulties and complexities associated with the task. Data collection efforts: National Violent Death Reporting System (NVDRS); [Hyperlink, http://www.cdc.gov/ncipc/profiles/nvdrs/facts.htm], [hyperlink, http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5415a1.htm]; Agency/sponsor: Centers for Disease Control and Prevention; Description: NVDRS collects incident-based data from multiple sources, such as death certificates, coroner/medical examiner reports, and police reports. Information is collected about the relationship between victims and suspects as well as circumstances preceding the death, such as whether intimate partner violence was involved; Categories of crime: Domestic; Violence, sexual assault; Data Collection Frequency: Ongoing; Data limitations associated with the determination of reliable nationwide prevalence estimates: * Does not include incidents other than violent deaths; * NVDRS is used in 17 states. Data collection efforts: Youth Risk Behavior Surveillance System (YRBSS); [Hyperlink, http://www.cdc.gov/healthyyouth/yrbs/overview.htm], [Hyperlink, http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5519a3.htm]; Agency/sponsor: Centers for Disease Control and Prevention; Description: YRBSS collects data through a nationally representative school-based survey of students in grades 9-12 that monitors priority health risk behaviors that contribute to the leading causes of death, disability, and social problems among youth and adults in the United States. Students are asked to complete a self-administered questionnaire covering a variety of health risk behaviors and topics; Categories of crime: Sexual assault, dating; violence; Data Collection Frequency: Biennial; Data limitations associated with the determination of reliable nationwide prevalence estimates: * Does not include individuals who do not attend school; * Excludes students in grades other than 9-12; * Respondents are asked only one question on physical dating violence and only one question on sexual assault. Data collection efforts: National Child Abuse and Neglect Data System (NCANDS); [Hyperlink, http://www.acf.hhs.gov/programs/cb/systems/], [Hyperlink, http://www.acf.hhs.gov/programs/cb/stats_research/index.htm#can]; Agency/ sponsor: Children's Bureau, Administration on Children, Youth and Families, Administration for Children and Families; Description: NCANDS is a voluntary system that currently collects annual case-level child abuse and neglect data from almost all states (48 states and the District of Columbia for fiscal year 2005) as well as key aggregated child abuse and neglect statistics from all states' child protective services agencies; Categories of crime: Sexual; assault; Data Collection Frequency: Ongoing; Data limitations associated with the determination of reliable nationwide prevalence estimates: * Excludes data on abused children not reported to child protective service agencies. Data collection efforts: National Electronic Injury Surveillance System- All Injury Program (NEISS-AIP); [Hyperlink, http://www.cpsc.gov/cpscpub/pubs/3002.html], [Hyperlink, http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5121a3.htm]; Agency/sponsor: Centers for Disease Control and Prevention and the Consumer Product Safety Commission; Description: NEISS-AIP collects data about all types and external causes of nonfatal injuries and poisonings treated in emergency departments in a subset of a nationally representative sample of 100 U.S. hospitals. The system collects data about the relationship of the perpetrator to the victim (e.g., spouse, parent), as well as the context of the crime (e.g., sexual assault); Categories of crime: Domestic violence, sexual assault; Data Collection Frequency: Ongoing; Data limitations associated with the determination of reliable nationwide prevalence estimates: * Does not include injuries that are not reported to an emergency room; * Does not include data on prevalence. Data collection efforts: Extent, Nature, and Consequences of Rape Victimization: Findings from the National Violence Against Women Survey (January 2006); [Hyperlink, http://www.ojp.usdoj.gov/nij/pubs- sum/210346.htm]; Agency/sponsor: Centers for Disease Control and Prevention and National Institute of Justice; Description: This report was prepared using data collected in a telephone survey administered to a sample of 8,000 women and 8,005 men. Although this report was issued in 2006, the data were originally collected from November 1995 to May 1996 to obtain information about lifetime experiences with various types of violence; Categories of crime: Sexual assault; Data Collection Frequency: Onetime; Data limitations associated with the determination of reliable nationwide prevalence estimates: * Does not include individuals that do not live in a residential household; * Does not include individuals under age 18; * Data are over 10 years old; * Does not include individuals without a telephone. Data collection efforts: The Harris Poll[®] #50 (June 2006); [hyperlink, http://www.harrisinteractive.com/harris_poll/index.asp?PID=677]; Agency/ sponsor: Harris Interactive; Description: An online survey of 2,377 adult respondents ages 18 and older was conducted to gauge the magnitude of domestic violence in the United States; Categories of crime: Domestic violence, sexual assault; Data Collection Frequency: Onetime; Data limitations associated with the determination of reliable nationwide prevalence estimates: * Does not include individuals under age 18; * Does not include individuals without a computer or Internet access; * Non-probability sample. Data collection efforts: The Victimization of Children and Youth: A Comprehensive, National Survey (February 2005); [Hyperlink, http://www.unh.edu/ccrc/pdf/CV74.pdf]; Agency/sponsor: University of New Hampshire and University of North Carolina at Chapel Hill; Description: This survey examined a large spectrum of violence, crime, and victimization experiences in a nationally representative sample of about 2,000 children and youth ages 2 to 17 years in the contiguous United States. Telephone interviews were conducted with 1,000 children age 10 to 17 years and the caregivers for 1,030 children age 2 to 9 years; Categories of crime: Sexual assault, dating violence; Data Collection Frequency: Onetime; Data limitations associated with the determination of reliable nationwide prevalence estimates: * Does not include individuals over age 17; * Does not include individuals without a telephone. Data collection efforts: Injury Control and Risk Survey-2 (2006); [Hyperlink, http://www.ajpmonline.net/article/PIIS074937970600167X/abstract]; Agency/sponsor: Centers for Disease Control and Prevention; Description: The first survey, conducted in 1994, involved a nationally representative telephone survey of 5,238 individuals aged 18 and older, and the results were reported in 1999. The second national telephone survey, conducted from 2001 to 2003, collected data from a nationally representative sample of 9,684 respondents (4,877 women and 4,807 men); Categories of crime: Stalking; Data Collection Frequency: Episodic; Data limitations associated with the determination of reliable nationwide prevalence estimates: * Does not include individuals under age 18; * Does not include individuals without a telephone. Source: GAO analysis of national data collection efforts. [End of table] Enclosure III: Comments from the Department of Health and Human Services: Office of the Assistant Secretary for Legislation: Department Of Health & Human Services: Washington, D.C. 20201: OCT 25 2006: Eileen R Larence: Director, Homeland Security And Justice Issues: U.S. Government Accountability Office: Washington, DC 20548: Dear Ms. Larence: The Department of Health and Human Services has reviewed the U.S. Government Accountability Office's (GAO) draft report entitled, "Prevalence of Domestic Violence, Sexual Assault, Dating Violence, and Stalking"(GAO 07-148R). HHS concurs with the draft Recommendations for Executive Action, and will continue and expand its collaborations with the Department of Justice to improve data collection and monitoring of violence. However, in addition to requiring the use of common definitions, it may be beneficial to include measurement. Although definitions may be similar, if different operationalizations and measures are used, the result could be that survey questions will be different and, if so, prevalence estimates could not be combined across those studies. The Department provided several technical comments directly to your staff. The Department appreciates the opportunity to comment on this draft report before its publication. Sincerely, Signed by: Vincent J. Ventimilgia: Assistant Secretary for Legislation: Enclosure IV: GAO Contact and Staff Acknowledgments: GAO Contact:Eileen R. Larence, (202) 512-8777: Acknowledgments: In addition to the contact named above, Debra B. Sebastian, Assistant Director; David P. Alexander; Frances Cook; Katherine M. Davis; Melissa Hermes; Varflay C. Kesselly; Deborah Ortega; and Clarence Tull made key contributions to this report. (440497): FOOTNOTES [1] Pub. L. No. 103-322, §§ 40001-703, 108 Stat. 1796, 1902-55 (1994). [2] Pub. L. No. 109-162, tits. I-VII, 119 Stat. 2960, 2972-3053 (2006). [3] These amounts reflect the across-the-board rescission reductions of 1 percent for fiscal year 2006 discretionary appropriations. [4] Pub. L. No. 109-162, § 119, 119 Stat. 2960, 2989-90 (2006). [5] Some of these systems and surveys obtain information about incidents not reported to police or other authorities. [6] The following hypothetical statement illustrates incidence and prevalence: 4,110 separate occurrences of domestic violence against women were reported during 1938 (incidence); whereas 2,500 women were victims of these 4,100 occurrences (prevalence). [7] Pub. L. No. 103-322, §§ 40001-703, 108 Stat. 1796, 1902-55 (1994). [8] These amounts reflect the across the board rescission reductions of 1 percent for fiscal year 2006 discretionary appropriations. [9] The Bureau of Justice Statistics defines "family violence" as all types of violent crime committed by an offender 3 who is related to the victim either biologically or legally, through marriage or adoption. [10] While some data collection efforts, such as the National Crime Victimization Survey, obtain information from survey participants about crimes not reported to the police, some survey participants may not disclose all incidents of crime during the survey. [11] Pub. L. No. 109-162, § 119, 119 Stat. 2960, 2989-90 (2006). [12] Some of these systems and surveys obtain information about incidents not reported to police or other authorities. [13] Pub. L. No. 103-322, §§ 40121, 40151-52, 40231, 40241, 40295, 40602, 108 Stat. 1796,1910-16,1920-21, 1932-35, 1940-42, 1951. [14] Pub. L. No. 109-162, tits. I-VII, 119 Stat. 2960, 2972-3053 (2006). [15] National Violence Against Women Prevention Research Center, "Rape in South Carolina: A Report to the State," 20 April 9, 2003.