Cross-situational consistency among problem adolescents: an application of the two-factor model.

A study was conducted to measure cross-situational consistency of the orthogonal dimensions of the two-factor model of social-emotional functioning. According to this model, large proportions of variance in social-emotional functioning can be accounted for by two orthogonal, highly replicable dimensions. The setting was Hawthorne-Cedar Knolls, a long-term institution for emotionally disturbed and delinquent youths; the subjects were 206 residents of varied ethnic and socioeconomic backgrounds. Cross-situational correlations between school and cottage settings were found to be .52 for Factor II (Cooperation-Compliance versus Anger-Defiance) and .42 for Factor I (Interest-Participation versus Apathy-Withdrawal). The results confirm the consistency and discriminant validity hypotheses--specifically, correlations between the corresponding factors of different instruments measuring behavior in different settings were large, whereas correlations between noncorresponding factors were very low. The value of using broad, higher order factors in personality research on the trait versus situation issue is discussed.