Do subtypes of maritally violent men continue to differ over time?

In previous batterer typology studies, only 1 study gathered longitudinal data and no research examined whether subtypes continue to differ from one another over time. The present study did so. We predicted that, at 1.5- and 3-year follow-ups, the subtypes identified at Time 1 (A. Holtzworth-Munroe, J. C. Meehan. K. Herron, U. Rehman, G. L. Stuart, 2000; family only, low level antisocial, borderline/dysphoric, and generally violent/antisocial) would continue to differ in level of husband violence and on other relevant variables (e.g., generality of violence, psychopathology, jealousy, impulsivity, attitudes toward violence and women). Although many group differences emerged in the predicted direction, not all reached statistical significance, perhaps because of small sample sizes. Implications of the findings (e.g.. not all marital violence escalates; possible overlap of the borderline/dysphoric and generally violent/antisocial subgroups) are discussed, as are methodological issues (e.g., need for more assessments over time, the instability of violent relationships).