ABSTRACT
Researchers who have conducted factor analyses of Schaefer's (1965) Children's Reports of Parental Behavior Inventory (CRPBI) have uniformly applied the Kaiser-Guttman (K-G) rule when estimating the number of factors, and have relied almost exclusively on orthogonal rotations. Because there is a substantial body of evidence noting the tendency of the K-G rule to underestimate the number of factors when the number of scales is moderate to small, many researchers believed that the factor structure of this scale should be reanalyzed in the light of an alternative procedure. In the present study, we analyzed the responses of a combined sample of 222 elementary school and 128 high school students. All of these students completed a short form of the CRPBI. Analyses of various subsamples (i.e., by gender and age) were consistent with analyses of the combined sample in suggesting that, whereas the specific rotation procedure applied was not critical, a five-factor solution may be more appropriate than the three-factor solutions that have often been reported in the literature. These results suggested that there may be two dimensions of control and two dimensions of discipline, in contrast to the singular dimensions previously reported.