Comparison of two outcome groups in a school-based mental health project.

Two samples of school-maladjusted children responding most and least favorably to a helping program with nonprofessional child-aides were compared on demographic, referral, and program experience variables. Outcome measures other than teacher ratings used to select the samples validated the inferred improvement differences. Subjects with poor outcomes were significantly more likely to reside in the city, older, from lower socioeconomic backgrounds, and initially more maladjusted than those with good outcomes, but there were no group differences in age, repeat in grade, or program experience variables. Implications of these data for future program directions and modifications of aide-training procedures are discussed.