Editor 's Note: This Research in Brief is based on two studies -1) Jacquelyn W. White and Paige Hall Smith of the University of North Carolina -Greensboro conducted a study of victimization and perpetration of violence among college students. This study, "Developmental Antecedents of Violence Against Women " ((NCJ 187775), was supported by grant number 98 -WT -VX -0010 from the National Institute of Justice.
2) Jane A. Siegel of Rutgers University -Camden and Linda M. Williams of the Stone Center at Wellesley College conducted an analysis of interview data - collected previously -from women with  contemporaneously documented histories of child sexual abuse and from a matched comparison group of women with no documented abuse. This study, "Risk Factors for Violent Victimization of Women: A Prospective Study " ((NCJ 189161), was supported by grant number 98 -WT -VX -0028 from the National Institute of Justice. These unpublished reports are available from the National Criminal Justice Reference Service. The White and Smith report is online at www.ncjrs.org/ pdffiles1/nij/grants/187775.pdf;  the Siegel and Williams report is online at www.ncjrs.org/pdffiles1/nij/grants/189161.pdf.

Studies investigating the root causes of violence against women devote more attention to perpetrators' behavior than to identifying risk factors among victims, in part to avoid the appearance of "blaming the victim." Responsibility for violence against women rests with the assailants, but identifying potential risk factors is essential to developing preventive strategies.

The National Institute of Justice (NIJ) funded two studies to identify factors that could predict which women were most likely to become victims of sexual or physical violence (see "Editor's Note ").

The first, a study of college students (see "College Student Study: Methodology "), found that women who were physically or sexually abused (or witnessed violence) in childhood but not in adolescence were at no greater risk for physical or sexual victimization in college. Those who experienced both childhood and adolescent victimization were at greater risk to be victimized in college.

The second, a study of victims of child sexual abuse (see "Childhood Sexual Abuse Study: Methodology "), found that childhood sexual abuse was a risk factor only when combined with sexual abuse during adolescence. The similarity of these findings from two studies with different samples and methodologies reinforces their significance and should help practitioners and policy-makers identify and prioritize children and adolescents most in need of appropriate interventions.

College Student Study: Findings about victims

Victimization. By the end of four years of college, 88 percent of women had experienced at least one incident of physical or sexual victimization in their lifetimes, and 64 per ­cent had experienced both. Almost 78 percent experienced at least one incident of physical victimization in their lifetimes, and 79 percent experienced at least one incident of sexual victimization.

Physical victimization was assessed using Straus ' ((1979) Conflict Tactics Scales, which measures self-reported experiences of such behaviors as being hit, shoved, or pushed. Sexual victimization was measured using the Koss et al. (1987) Sexual Experiences Survey, which includes a range of victimization experiences from coerced sexual contact to rape.

Revictimization. Previous studies suggest that young women, and college students in particular, are at increased risk of becoming victims of sexual abuse (Koss et al., 1987). This study found that women who were physically or sexually abused or who witnessed domestic violence in childhood (before age 14) were at greater risk for physical and/or sexual victimization in high school.

Women who were victimized in high school were at much greater risk for physical and/or sexual victimization in college. After controlling for victimization in high school, however, those who were abused or witnessed violence in childhood were not at greater risk for college victimization.

Women who were both sexually and physically abused during high school were most likely to suffer sexual abuse in college, followed by women who were abused sexually but not physically. Those who experienced physical abuse but no sexual abuse in high school were no more likely to suffer sexual abuse in college than women who had never been abused. Women who experienced co-occurrence (both physical and sexual abuse) during high school were most likely to become victims of physical violence by their romantic partners, followed by women who had experienced physical violence in high school but no sexual abuse. Women who had experienced sexual abuse in high school but no physical violence were no likelier to become victims of physical violence in college than those who had never been abused.

Mental health. Women who experienced co-occurrence in high school and the first year of college reported the highest levels of psychological distress (a composite measure consisting of anxiety, depression, loss of control, and lack of emotional ties). Women who had not been abused reported the lowest levels, and other women reported intermediate levels.

Physical health. Women who experienced co-occurrence during high school and the first year of college reported more doctor visits than women with no assault experiences. Women who had experienced only physical assault, however, rated their overall general health lower than all other women.

Problem behaviors. Women who experienced co-occurrence during high school were likeliest to report suicidal thoughts during high school, although suicidal thoughts in college were not related to co-occurrence in high school. An increased number of sex partners (defined as men with whom women had sexual inter ­course, consensual or not) was associated with all types of victimization. Women who experienced co-occurrence in high school and those who had been sexually victimized but not physically assaulted had the most sex partners, followed by those who had only been physically assaulted. By the end of the fourth year in college, women who had experienced co-occurrence were most likely to have engaged in unprotected sex. Alcohol use was highest for women who experienced co-occurrence in high school and the first year of college; it was lowest among those who had not been victimized.

College Student Study: Findings about perpetrators

Co-occurrence of assault. It is common for a man to commit both sexual and physical assault, although not necessarily against the same woman. The percentage of men who did so declined from nine percent in high school to five percent in the first year of college to two percent in the fourth year of college. The relationship between sexual and physical assault is significant: A young man who committed one type of assault was five times more likely to commit the other during high school (the odds remained roughly the same over the college years).

Family violence. Young men who both sexually and physically assaulted women experienced more family violence and witnessed significantly more family violence than other men.

Childhood sexual abuse. A significant relationship was found between childhood sexual victimization and perpetration of sexual and physical assault in high school. Men who committed physical and sexual assault or sexual assault alone in high school reported more childhood sexual abuse than men who had no history of assault or who committed only physical assault.

Factors associated with co-occurrence. Men who sexually and physically assaulted women were significantly more likely to abuse drugs and alcohol, have many sexual partners, engage in delinquent behavior, and approve of male use of violence against women than men who committed neither form of assault. For some of these factors, however, there were no differences between those who committed both forms of assault and those who committed only one.

Predictors of perpetration. Men who committed sexual assault during high school were four times more likely to commit sexual assault during the first year of college than men with no history of sexual assault. The results of this study, however, show a decline in sexual aggression over time.

Child Sexual Abuse Study: Findings

Risk of victimization. The relationship between child ­hood and adult victimization is complex. Sexual abuse in childhood alone or adolescence alone was not found to be associated with a higher risk of adult physical or sexual victimization. Only women who had been sexually abused as both children and adolescents were found to be at increased risk of victimization as adults, with rates of both sexual and physical adult victimization significantly higher than those for all women. Three-quarters of these women were sexually assaulted when adults, and more than 8 in 10 (84 percent) experienced severe domestic violence. (Severe violence was defined as beating, punching, or hitting with something that could hurt; choking; burning or scalding on purpose; and having serious injuries or needing to see a doctor as a result of a fight with a partner.) This is not to say, however, that few of the others were victimized as adults. Rates of reported sexual victimization and domestic violence were high: 28 percent of women who experienced no childhood or adolescent sexual victimization reported being sexually assaulted as adults, and 60 percent reported experiencing at least minor violence perpetrated by an intimate partner.

Situational risk factors. Having multiple sexual partners significantly increased the risk of adult sexual abuse. Alcohol abuse was found to be statistically significant in predicting increased risk of adult sexual victimization but not domestic violence.

"Double victims." Women who were double victims -victimized in both childhood and adolescence -differed from the other women inter ­viewed in ways that could have contributed to their later behavior. They were significantly more likely to have reported being beaten by their mothers. They had a significantly higher number of living situations (distinct household compositions) and childhood caregivers between birth and age 18.

Double victims were more likely to have had a boyfriend and to have started drinking
alcohol at an earlier age. They were also likelier to have engaged in at least two delinquent behaviors: running away from home and prostitution before the age of 18.
Running away was one of two factors that significantly increased the odds of becoming a double victim once the other risk factors were taken into account together; the other was having a mother who had a criminal history. Nearly half (48 percent) of the double victims reported that their mothers had been arrested when the victims were children or teenagers -a rate seven times greater than that reported for the mothers of women not abused as children. Some of these mothers may have been incarcerated, which would account for the significantly greater number of living situations and childhood caregivers the women report ­ed having and their higher scores on a measure of parental neglect.

Aggressive behavior and domestic violence. Women who engaged in aggressive behavior themselves (i.e., the use of force -hitting, kicking, or punching -or a weapon against another person, including a domestic partner) were at increased risk of being severely abused by their partners. Aggression was categorized as minor or severe depending on the type of force/weapon used and the degree of injury inflicted. One must be cautious, however, before concluding that female aggressive behavior is a risk factor for victimization; much of this violence was reported to be in self-defense. For example, 38 percent of those who used force against their partner said that they were never the first to do so, and 40 per ­cent said that they used force primarily or exclusively to protect themselves from imminent harm. This is especially true for victims of severe domestic violence: Only 27 percent of the victims interviewed reported initiating violence most or all of the time; 44 percent said that they never initiated violence. More than half reported using force primarily or exclusively to protect themselves, and only five percent said that their own use of force was never caused by the need to protect themselves.

Limitations of the studies

Although both studies present useful findings on groups that are important targets for intervention, each has several limitations. The analyses of "double victims " in the child sexual abuse study were exploratory and need to be replicated. These findings are more tenuous than the study's other results. Another limitation was that the analyses of childhood and adolescent victimization in the college student study were retrospective, and such analyses may be affected by the faulty memory of some participants. Although the child sexual abuse study was designed so that it would not have to rely on  retrospective accounts of childhood victimization, some control group members had to be reclassified as victims based on retrospective accounts of being sexually abused before age 13.

In addition, although the child sexual abuse study provided information about a population traditionally neglected in research on violence against women, the fact that the sample is disproportionately poor, urban, and black and has very high victimization rates raises questions about whether it is sufficiently representative to draw meaningful general conclusions from the study's findings. The same reservation applies to the college student study. Taken together, however, the findings of the two studies, substantiate each other.

Devising prevention strategies

The results of both studies, despite their limitations, indicate the need for directing interventions toward boys and girls who witness or experience violence. Boys who are victims of violence are especially in need of intervention, because they are likely to grow up and mistreat their partners. Early intervention is likely to go along way toward preventing further violence later in life.

The research suggests that professionals who have contact with children and adolescents should be educated about the importance of taking gender-based violence seriously. Prevention and intervention strategies can help potential victims and perpetrators overcome social influences that tend to support or condone violence against women by including gender-based violence prevention activities in other programs that target adolescent boys and girls, such as substance abuse and pregnancy prevention programs. Prevention and intervention strategies should support girls and women who report abuse and promote psychological healing and social resolution. But programs for adolescent girls -especially those who have experienced childhood victimization -should, in addition, incorporate activities that could prevent violence against women.

References:

Koss, M.P., Gidycz, C., and Wisniewski, N. (1987). The scope of rape: Incidence and prevalence of sexual aggression and victimization in a national sample of higher
education students. Journal of Consulting and ClinicalPsychology, 55,162 -170.

Straus, M.A. (1979). Measuring intrafamily conflict and violence: The Conflict Tactics Scales. Journal of Marriage and the Family, 41,75 -88.