U.S. flag

An official website of the United States government, Department of Justice.

Nonphysical Abuse Among Batterer Program Participants

NCJ Number
Journal of Family Violence Volume: 17 Issue: 4 Dated: December 2002 Pages: 293-314
Date Published
December 2002
22 pages

This study used a multisite sample of batterer program participants (n=840) with a longitudinal 15-month follow-up to identify risk markers for the nonphysical abuse of women.


Nonphysical abuse generally refers to controlling behaviors, verbal abuse, and threats; it is often identified as "psychological abuse," because its intent and effect is often to diminish another person's self-esteem and mental well-being. Participants for this study were drawn from four "well-established" programs for batterers in Pittsburgh, Houston, Dallas, and Denver. They were selected to represent a range of program format in terms of duration and services. A 15-month follow-up included baseline interviews at program intake and telephone interviews every 3 months for 15 months after intake. Outcome variables for the study were three categories of nonphysical abuse: verbal abuse, controlling behaviors, and threats. Outcome variables were measured by follow-up reports from the female victims. The selection of independent variables was based on the prevailing predictive model used in recent research to identify risk markers for physical and nonphysical abuse. Independent variables pertained to demographics, abuse in one's family-of-origin, heavy drinking, previous antisocial behavior, prior nonphysical abuse, and program attendance. The study found that the only substantial and consistent risk markers for nonphysical abuse were nonphysical and physical abuse prior to program intake and dropping out of the program. The programs apparently helped to reduce nonphysical abuse. Other significant risk markers varied across the categories of nonphysical abuse, suggesting that the categories do not form a continuum or hierarchy of abuse. Recurrent abuse, moreover, continues to be more difficult to predict in clinical samples than in the general population. 2 tables, 1 figure, and 36 references

Date Published: December 1, 2002