74
score on the PSI. This finding is difficult to interpret. It may be a chance finding, or perhaps, fathers who are experiencing more personal distress actually tend to give fewer commands and are less engaged with their children in general than fathers who are experiencing less distress. In support of this interpretation is the finding that fathers of children with behavior problems who report less marital satisfaction also report less parenting involvement (Stormshak, Speltz, DeKlyen, & Greenberg, 1997).
Limitations and Future Directions
Several limitations of this study warrant discussion. First, it has been found that fathers are more reactive to observation than mothers, and interact differently with children based on the setting of the assessment (home vs. laboratory; Cassidy, Parke, Butkovsky, & Braungart, 1992). The results of this study, therefore, may not generalize to other settings. Second, the majority of this sample was comprised of Caucasian boys from two-parent families. Further study of referred and non-referred minorities is necessary to have adequate information on children from diverse backgrounds and their fathers. It will also be important for future research to compare the interactions of children from families with different constellations (e.g., single parent, divorced parents with joint custody, step-families). Although girls were included in this study, they were not sufficient in number to make statistical comparisons between clinic-referred and nonreferred girls or between clinic-referred boys and clinic-referred girls. It has been suggested that boys and girls at this age differ in their interactions with parents of different gender. Future research with larger samples of girls would provide the opportunity to examine this issue.