Using fifth through eight-grade students, the Comer School Development Program was evaluated in 10 inner city Chicago schools over 4 years, contrasting them with nine randomly selected no-treatment comparison schools. Comer schools implemented more program details than the controls but were not faithful to all program particulars. Students' ratings of the school's social climate improved soon after the program began. By the last 2 study years, both the students' and teachers' perceptions of the school's academic climate had also improved relative to the control schools. By these last years, Comer schools had gained about 3 percentile points more than the controls in both reading and math and students reported less acting out on a scale whose items are correlated with more serious offending in later life. Students in Comer schools also endorsed more conventional norms about misbehaving and reported greater ability to control their anger. However, the Comer program did not benefit either students' mental health or their participation in activities that adults consider wholesome. Explanations for the achievement and acting out results are offered based on student and staff data about school climate, on insights from an ethnography conducted in the program schools, and on contrast with the evaluation results from Prince George's County, MD, where a different variant of the program failed to achieve any positive outcomes.
