The study involved both a longitudinal component in which subjects were tested during classification and again after assignment to correctional units and a cross-sectional component consisting of a survey of subjects at four diverse correctional units. Interviews, questionnaires, and case record analyses were the data collection methods. Findings indicate that reception unit residents were without a full day's schedule of meaningful activities. Placement decisions and time goals were set by the classification committee in the absence of formal criteria; the committee was most often influenced in its decisionmaking by the nature of the resident's offense. The classification process was not regulated by procedures designed to ensure fairness or integrity. Residents participating in a program offering a comprehensive work, school, and guided group interaction treatment approach made impressive gains in self-esteem. No differences were observed between residents in such programs at secure and open facilities in this area. Though a majority of the residents perceived guided group interaction positively, some were alienated from their group interaction positively by subjects in open units allowing freedom of movement and community involvement. Boys who perceived social climate more favorably also had higher self-esteem. A fuller schedule should be offered at the reception unit, formal guidance should be provided to the classification committee, therapeutic alternatives should be developed for guided group interaction, the right to refuse treatment should be respected, and programs at nonsecure units should be supplemented by other enriching activities. Tabular data, charts, and 13 references are included. (Author abstract modified)