Date of Award


Document Type


Degree Name

Master of Arts (MA)




In recent years, there have been numerous approaches to the study of wife abuse, e.g., sociological, psychoanalytic, and environmental. An approach recently suggested by Walker (1977-78, 1978, 1979a, 1980) incorporates the notion of learned helplessness. Her conceptualization of the psychological characteristics of abused women suggests that such women have a traditional attitude toward the female sex-role, a predominantly external locus of control, and low self-esteem. The present investigation was an attempt to validate Walker's model.

Responses to the Attitudes toward Women Scale (AWS), the Rotter Internal-External locus of control scale (RIE), and the Texas Social Behavior Inventory (TSBI) were gathered from three groups: 18 women from a battering situation, 16 non-battered women in psychotherapy, and 20 women who were neither in therapy nor battered. In addition, demographic information and information regarding the subjects' childhood experiences with family violence were collected from each of the three groups, and treated as covariates. A second part of the investigation attempted to relate the test profile exhibited by the battered women to specific characteristics of their battering experiences. A questionnaire concerned with information such as the length of time in the abusive relationship, the type and severity of abuse, and the type and effectiveness of the help sought following termination of the abusive relationship, was thus administered to the group of abused women, factor analyzed, responses converted to factor scores for each subject, and related to their responses on the AWS, RIE, and TSBI.

With respect to attitudes .toward the female sex-role, no evidence was found to support the notion that abused women differ from non-battered women in therapy, nor from women neither in therapy nor battered. A more external locus of control appeared attributable to women in therapy than to the other tv?o categories of women, with abused women only slightly more external in their locus of control than control group women. On the measure of self-esteem, there were initial indications that abused women had lower self-esteem than did women in the control group, but the introduction of statistical controls for concurrent differences between these two groups on demographic and exposure to family violence as a child variables reduced that difference to a non-significant level. Attempts to relate scores on the AWS, RIE, and TSBI to elements of the battering experience revealed that sex-role attitudes were largely unrelated to subjects’ factor scores. For abused women, aspects of the abusive situation surrounding the severity of the abuse, judicial involvement, and the immediacy and scope of the abuse were found to be predictive of their locus of control, while those aspects related to judicial involvement, the involvement of neighbors, and public knowledge of the abusive relationship were significantly related to their level of self-esteem.

The above findings were discussed in terms of their implications for Walker's learned helplessness model of the psychological characteristics of abused women, and in relation to the need for further study in the area which incorporates the principles of standardized measurement and allows for comparisons between groups of abused women and other selected groups of women.