68
distinguishing between fathers of clinic-referred and non-referred children. Although norms have not been established for the PLOC-SF, the results of this study suggest that fathers of children referred for behavior problems tended to have a more external locus of control than fathers of comparison children. In other words, they perceived themselves as having less control over their child's behavior.
A statistically significant difference between groups was also found on the PPVTR (Dunn & Dunn, 1981), the measure used to screen for delay in receptive vocabulary. Children in the clinic-referred group received a mean score of 92.6 (aD = 14.8) and the children in the comparison group received a mean score of 102.9 (SD = 17.3). Although scores for both groups fall within the average range, the lower score for the clinic-referred children is consistent with the literature reporting an association between language delays and behavior problems in young children (Cantwell, Baker, & Mattison, 1979; Cohen, Davine, & Meloche-Kelly, 1989; Richman, Stevenson, & Graham, 1982). Behavioral Differences Between Clinic-referred and Non-referred Families
Discriminative validity identifies differences between groups of individuals whose dissimilarities are established by other criteria. In this study, significant differences were found on four of the seven DPICS II summary variables. First, child compliance was significantly different between the two groups. The clinic-referred children on average across the three situations were compliant with 58% of paternal total commands compared to the non-referred group who were compliant with 70% of paternal total commands. It is important to note that the compliance ratios described here include the child's compliance only to those direct and indirect commands for which they had an opportunity to comply,