Bridging Disciplinary Boundaries (January 11 - 14, 2007)

Purpose: There is little research examining how paternal demographic and psychosocial characteristics heighten or minimize risk for paternal physical child abuse. We extended the developmental-ecological framework to study father factors that are related to non-punitive parenting and physical abuse proxies. We used father and mother self-reported data from the Fragile Families and Child Wellbeing Study (FFCWS) to investigate these questions, and we assessed how the observed relationships differed as a function of fathers' race/ ethnic status.

Methods: Interviews were conducted when index children were 3 years old on a national birth cohort of 3,356 families across 20 U. S. cities. A comprehensive set of variables were assessed, including father self-report of educational status, household characteristics, and marital status; fathers' earnings and employment status; and psychosocial variables such as parenting stress, involvement with the child, and perceived support from the mother. We controlled for mother-related variables that have been linked to child abuse. Cross-sectional bivariate and negative binomial regressions were conducted to discern how paternal characteristics were related to the CTS-PC subscale measuring non-punitive parenting, as well as CTS-PC subscales measuring psychological and physical aggression directed towards the child and frequency of corporal punishment.

Results: Bivariate results indicated that White fathers were the most likely to use non-punitive discipline, followed by African American and Hispanic fathers; all cell differences were significant. Hispanic fathers directed significantly less psychological and physical aggression towards their children than both white and African American fathers. Among African American fathers, being in a cohabiting relationship was related to less psychological and physical aggression and corporal punishment. For white fathers, higher levels of involvement with the child was related to higher levels of positive parenting and less physical aggression or corporal punishment. There were no significant relationships among earnings variables and paternal use of either positive or negative parenting behaviors; nor did fathers differ in the overall level of involvement with the child as a function of race/ ethnic status.

Implications: Results supported the basic tenets of the developmental-ecological model. There does not seem to be one pathway leading to higher physical child abuse risk, but rather multiple potential factors linked to risk. Consistent with previous research, Hispanic fathers were overall less likely to use coercive parenting strategies. However, some results challenge conventional wisdom. All fathers reported similar levels of involvement with their children. Although some have pointed to the risk for abuse associated with fathers' lowered economic contributions to the family, we found little indication that paternal earnings and employment significantly heightened paternal risk for physical abuse. With the proper control variables paternal economic factors may not be as closely linked to risk as has been previously thought. Furthermore, we found a paradoxical relationship such that cohabiting African American fathers were at lowered abuse risk compared to married African American fathers, an important finding regarding the unique effects of relationship status in African American families that was not replicated with white or Hispanics. Interventions targeting fathers need to be embedded with the relevant cultural framework.