Over recent years there has been growing research into describing wife batterers and their characteristics. However there is little research into wife batterers and those who are violent with their wives and towards strangers. The present study is an exploratory investigation into the relationship between wife battering offenders, stranger violent offenders and those male offenders that are violent in both situations. The aim of the study is to examine the differences between the three above mentioned groups. Ninety inmates from Paparua medium security and Rolleston minimum security prisons were tested on a Social History Questionnaire, the Millon Clinical Multiaxial Inventory(MCMI-1) (Millon, 1984) and The Gudjonsson Blame Attribution Inventory(GBAI) (Gudjonsson, 1987). Subjects were classified into three groups (as indicated ·by Shields et an,1988), that of Wife Battering Only, Generally Violent and Stranger Violent Only. It was hypothesised 1) that there would be significant differences between the Wife Battering Only, Stranger Violent Only and Generally Violent offenders on the MCMI scales. 2) that the Wife Battering Only group would gain higher scores on mental and guilt feeling attribution factors on the GBAI. A null hypothesis was accepted for hypothesis one, however hypothesis two was accepted. The major findings of this study indicate that the Wife Battering Only group and the Generally Violent group were significantly different from each other on a majority of the MCMI scales. Unexpectedly the Stranger Violent Only and Wife Battering Only groups were found to have similar profiles on the MCMI. Margaree's theory of the undercontrolled/overcontrolled was used to describe the Generally Violent group and the Wife Battering and Stranger Violent Only groups.