Intimate violence gets female twist.

When violence erupts between spouses or dating partners, men have a reputation as perpetrators and women as victims.

An analysis of data on relationship violence in the general population now finds that, excluding murder and sexual assaults, women prove slightly more likely than men to commit one or more aggressive acts against a partner. Still, the data show that men are more likely than women to inflict injuries that require medical help, says psychologist John Archer of England's University of Central Lancashire.

In contrast to these findings on the general population, studies that primarily focus on battered women in shelters and men in court-ordered treatment programs for batterers and sexual offenders report that males exhibit a far more violent streak than females, Archer finds.

Archer, whose analysis appears in the September PSYCHOLOGICAL BULLETIN along with commentaries by other psychologists, combined data from 82 studies examining sex differences in aggressive acts between heterosexual partners. Many of the studies, which were published between 1979 and 1997, used a self-report questionnaire that asks marriage or dating partners about slapping, pushing, hitting with objects, physical fighting, and other violent behaviors in their relationship.

Among both men and women, many who act violently don't do so in self-defense, Archer contends. A comparable majority of physically aggressive women and men reported initiating attacks on a partner, who may have responded or remained passive.

Most of the studies considered by Archer were conducted in the United States or Europe. He suspects that rates of women's physical aggression in close relationships are particularly high in Western nations. There, widespread discouragement of men from hitting women, as well as women's growing economic power and independence from coercion by their husbands' families, makes open aggression less risky for women than it is elsewhere, he theorizes.

The new analysis shows that "physical aggression by women must be taken seriously if there is a sincere desire to prevent partner abuse," comments psychologist K. Daniel O'Leary of the State University of New York at Stony Brook.

Psychologist Jacquelyn W. White of the University of North Carolina in Greensboro and her colleagues remain skeptical of Archer's findings. They argue that self-reports aren't adequate and may underestimate the amount and severity of violence that men commit in close relationships.