A test of systems theory to assess the impact of sexual abuse as a child on current risk of physical abuse by mothers

Ontogenic variables, such as the experience of the parent's abuse as a child and current depression or substance abuse, were expected to have a greater impact on the risk of child abuse than microsystem and exosystem variables such as family functioning, domestic violence, income, community safety, and social support. It was expected that social support would have the greatest impact of these broader system variables. It is believed that attachment style mediates the relationship between the ontogenic system level and microsystem/exosystem level variables.;Secondary data were used to examine these systemic impacts. The dataset was obtained from the National Data Archive on Child Abuse and Neglect. The sample consisted of 265 women, the majority of whom were African-American and who had a high school education or GED. The majority of these women were employed, while others were in school or a training program at the time of the interview. Over 50 percent of the sample had never been married.;A multiple regression analysis was conducted to examine the impact of factors from three systems levels on the risk of child abuse. Variables were entered into the multiple regression equation after they had demonstrated significance in bi-variate analyses with Bonferroni corrections. Mother's age was entered first in this equation in order to control for any potential effect of this variable as it has been shown to be significantly related to risk of abuse. Entered in the second step were the ontogenic variables found to be significant: depression and locus of control. In the third and final step, the microsystem and exosystem variables found to have a significant impact on the risk of abuse were entered: the number of social supports available to the mother, the mother's assessment of family functioning and the frequency of domestic violence perpetrated against her.;Locus of control and mother's depression were demonstrated to be the most important predictors in the equation. Domestic violence and social support were the next most important predictors, with family functioning becoming non-significant in this equation. This supports the hypothesis that ontogenic variables have the greater impact in predicting risk of physical abuse.;A path analysis was conducted in order to examine the possible time order and causal nature of these variables. All variables were included in this analysis. The only variables found to impact the risk of physical abuse were depression and locus of control. The only route from experiencing childhood sexual abuse to the risk of committing child physical abuse was through the level of depression currently being experienced by the mother. Mother's locus of control impacted the risk of abuse directly as well as through its impact on depression. Risk of abuse was not influenced directly by the experience of sexual abuse. (Abstract shortened by UMI.)...