Date of Award

Winter 1979

Project Type


Program or Major


Degree Name

Doctor of Philosophy


This research is a longitudinal study of wife battering using a system's theory approach. Two types of longitudinal analyses are presented. The first is the analysis of retrospective data of histories of violent relationships. The second type of longitudinal analysis is the comparison of the number and the frequency of men's and women's violent acts at two points: in the year prior to entering a shelter for battered women and six months later.

In-depth interviews with 31 battered women who had voluntarily sought help at a shelter provide histories of violence as well as information on their own early family background, that of their spouses, and the history of their relationship. Three life histories are presented in the form of edited transcripts of interviews with the women. In addition 24 of the women were reinterviewed six months after they came to the shelter.

This research is unique in the presentation of longitudinal data, and in the integration of quantitative and qualitative data. Quantitative data are presented on number of violent acts and frequency of violent acts at the two interview points; data are presented on the women's responses to three incidents of violence, the first, the most important and the most recent, and qualitative data are presented in the form of three life histories.

The quantitative and qualitative data form the basis for a six stage model of the wife battering process, and a flow chart which illustrates the system's processes that lead to stabilization and change in the patterns of wife battering.

Important findings include the following: (1) the sample of battered women differed from a random sample on number of marriages, percent cohabiting, exposure to violence as a child, incidence of alcohol abuse, and skewed marital power; (2) men were both more likely to be violent than the women, and more frequently violent; (3) over three incidents of violence the women's reaction changed from denial to seeking intervention, and the processes that led to change were governed by the systems principles of positive and negative feedback; (4) violence of both men and women was reduced between the two interviews, but 50% of the women had been hit in the six month follow-up period; (5) the highest rates of violence in the six month follow-up period were among women who had returned to their husbands and never left again.

Finally, the most significant contribution of the research may be the development of an analytical model to explain wife battering using the concepts of system's theory. This model incorporates dynamic processes of feedback control and the cybernetic processes of system maintenance.