Five populations were sampled: adjudicated delinquents in the State's two camps; youths in the Lookout Mountain School for boys, excluding those who had been paroled from the camps but were not now in the school; youths paroled from the camps, but returned to the school by a court action (recidivists); and youths paroled from either camp who had not committed new offenses for 1 year and who lived in the Denver metropolitan area. The nonrecidivist population was aged between 17 and 23, and the others between 15 and 18. The sample included 82 camp youths, 119 school youths, 36 recidivists, and 89 nonrecidivists. The ratio of youths to counselors was 8 to 1 in the camps, and 15 to 1 in the school. The study instrument used to assess interaction was a revised version of a Reusch, Block, and Bennet test (1953) consisting of 50 statements which described interaction dimensions, and a 7-part, modified Likert-type response scale for each statement. The data was factor-analyzed and tested through variances procedures. Results showed that the factors of teacher-role interaction, contrariety, and noncompetitiveness were common to the samples. Insignificant differences were found between recidivist and nonrecidivist groups in their perceptions of interaction patterns. Tables and a bibliography of approximately 30 references are provided. Appendixes containing the interaction tests, the personal information form, additional test data, and photos of the correctional facilities are included.