Assessing assault self-reports by batterer program participants and their partners

Self-reports on domestic violence inventories remain the basis of court and clinical decision-making and program outcome evaluations, but little resea rch has investigated the reliability and validity of these self-reports wit h clinical populations. We investigated the most widely used self-report in ventory, the Conflict Tactics Scale, using a multisite database of men admi tted to batterer programs and their female partners (n = 840). Concurrent v alidity was assessed by comparing batterer and victim reports to police inc ident reports at program intake (n = 145). Victims (29%) were more likely t han batterers (19%) to report no assault when the police reports indicated an assault Batterers were, however, more likely to minimize the severity of assaults than their victims. Reliability was assessed by comparing agreeme nt between men and women at intake and at 12-month follow-up (n = 558). Ove rall man-woman agreement was higher at follow-up (74%) than at intake (64%) . However, occurrence agreement declined substantially (from 61 to 17%), an d male underreport and male denial markedly increased. Based on the men's a nd women's descriptions of the assaults, the women who underreport appear t o do so primarily to preserve the relationship and men do so in outright de nial. The findings imply the need to collect both men's and woman's reports at intake and contradict the notion that agreement increases as a result o f the batterer's sensitization to violence in a program.