Violence against women is a major social problem in the United States. National surveys estimate that at least 2 million women each year are battered by an intimate partner, and crime data from the Federal Bureau of Investigation (FBI) record about 1,500 murders of women by husbands or boyfriends each year. Overall, the Bureau of Justice Statistics reports that women sustained about 3.8 million assaults and 500,000 rapes a year in 1992 and 1993: more than 75 percent of these violent acts were committed by someone known to the victim, and 29 percent of them were committed by an intimate—a husband, ex-husband, boyfriend, or ex-boyfriend. Studies estimate that between 13 and 25 percent of all U.S. women will experience rape in their lifetimes. These figures are believed to be underestimates.

The consequences of this violence against women may be long-lasting. Both rape and intimate partner violence (battering, which is frequently accompanied by psychological abuse) are associated with a host of both short- and long-term problems, including physical injury and illness, psychological symptoms, and, in extreme cases, death. And the conse-

quences go far beyond the individual female victims, affecting their children, families, and friends, as well as society at large.

Research on violence against women is a relatively young and fragmented field. At this early yet critical time in the developing field, Congress (in the Violence Against Women Act of 1994) directed the National Research Council to develop a research agenda to increase the understanding and control of violence against women, including rape and domestic violence. Congress specified that the agenda focus primarily on preventive, educational, social, and legal strategies, including consideration of the needs of underserved populations.

In convening the Panel on Research on Violence Against Women, the National Research Council specifically charged the panel with the following tasks: synthesize the relevant research literature and develop a framework for clarifying what is known about the nature and scope of violence against women; supplement the research review with lessons learned by field professionals and service providers; and identify promising areas of research to improve knowledge of the scope of the problem and interventions for dealing with it.

After reviewing the literature on battering, rape, and sexual assault, the panel concludes that significant gaps exist in understanding of the extent and causes of violence against women and the impact and effectiveness of preventive and treatment interventions. In order to begin filling those gaps, the panel recommends a research agenda to facilitate development in four major areas: preventing violence against women, improving research methods, building knowledge about violence against women, and developing the research infrastructure.

The panel concludes that in order to significantly reduce the amount of violence against women in the United States

the focus must be on prevention. The development of effective preventive interventions will require a better understanding of the causes of violent behavior against women, as well as rigorous evaluations of preventive intervention programs. The panel, therefore, recommends:

• longitudinal research to study the developmental trajectory of violent behavior against women and whether and how it differs from the development of other violent behaviors;
• the inclusion of questions about violent behavior against women in research on the causes of other violent behavior;
• the examination of risk factors, such as poverty, childhood victimization, and brain injury, for sexual and intimate partner violence in studies of at-risk children and adolescents;
• rigorous evaluation of both short- and long-term effects of programs designed to prevent sexual and intimate partner violence, including school-based education programs, media campaigns, and legal changes intended to deter violence against women; and
• the inclusion of intimate partner and sexual violence outcomes in evaluations of nonviolent conflict resolution programs and other programs designed to prevent delinquency, substance abuse, teenage pregnancy, gang involvement, and general violence.

Researchers working on violence against women come from a wide spectrum of disciplines, each of which has its own terms and perspectives. Many studies in the field suffer from methodological weaknesses, including small sample sizes, lack of control groups, and weak instrumentation.

The panel recommends several key topics for improving research on violence against women:

• clear definition by researchers and practitioners of the terms used in their work;
• the development and testing of scales and other tools of measurement to make operational the key and most used definitions;
• improvement in the reliability and validity of research instruments with guidance from subpopulations with whom the instruments will be used, for example, people of color or specific ethnic groups;
• clarification of theory and the outcomes expected from the intervention in evaluation research;
• the use of randomized, controlled outcome studies to identify the program and community features that account for effectiveness (or lack thereof) of legal and social service treatment interventions with various groups of offenders;
• both qualitative and quantitative research to recognize the confluence of the broad social and cultural context in which women experience violence, as well as individual factors, with attention to such factors as race, ethnicity, socioeconomic status, age, and sexual orientation in shaping the context and experience of violence in women's lives.

There are many gaps in understanding of violence against women. There is relatively little information about violence against women of color, disabled women, lesbians, immigrant women, and institutionalized women. Research on violence against women has advanced along categorical lines (i.e., sexual violence, physical violence) rather than on women's experiences, which are believed to include multiple forms of violence. Although there appear to be some similarities and some differences between generally violent behavior and violence directed at women, the extent of the similarities and differences remains unknown.

The panel recommends the following areas as the most important next research steps:

• the development of both national and community-level survey studies using the most valid instrumentation and questioning techniques to measure incidence and prevalence of violence against women (both victimization and perpetration), with particular attention to surveys of racial and ethnic minorities, and other underrepresented population subgroups;
• the inclusion of questions pertaining to violence against women in national and community surveys of topics such as women's mental or physical health or social or economic well-being;
• identification and secondary analysis of existing data sets with respect to violence against women;
• research on the consequences of violence against women that includes intergenerational consequences, effects of race and socioeconomic status on consequences, and costs to society, including lost productivity and the use of the criminal justice, health, and social service systems;
• studies that describe current services for victims of violence and evaluate their effectiveness;
• studies to investigate the factors associated with victims' service-seeking behavior, including delaying seeking of services or not seeking services at all, in order to identify barriers to service seeking and alternative approaches and settings for service provision; and
• studies that examine discretionary processes in the criminal and civil justice systems, including legal research on the theory and implementation of new laws and reforms, police and prosecutorial decision making, jury decision making, and judicial decision making.

Responsibility for research on violence against women is spread across a number of federal agencies, with most funding coming from the Centers for Disease Control and Prevention and the National Institutes of Health of the U.S. Department of Health and Human Services, and the National Institute of

Justice of the U.S. Department of Justice. A number of other agencies, including the National Science Foundation and the U.S. Department of Education, have programs that could contribute to the development of research on violence against women. The panel concludes that research on violence against women will be strengthened by a research infrastructure that supports interdisciplinary efforts and helps to integrate those efforts into service programs and institutional policies, especially in the area of preventive intervention.

The panel recommends two key actions for improving research capacity and strengthening ties between researchers and practitioners:

• development of a coordinated research strategy by government agencies; and
• establishment of a minimum of three to four research centers, within academic or other appropriate settings, to support the development of studies and training programs focused on violence against women, to provide mechanisms for collaboration between researchers and practitioners, and to provide technical assistance for integrating research into service provision.

The problem of violence against women in the United States will not be solved in the short term or without concentrated attention. Well-organized research will be critical to the long-term goal of preventing and ameliorating the effects of violence against women.