Rather than examining stress at the level of individuals, as is usually done, this study looked at stress at the institutional level or from an "administrative perspective." The purpose was to measure changes in levels of stress in schools over the academic year. Schools sampled included three public high schools in Salinas, California. The majority in two of the schools was white and in the third school the majority was Hispanic. Daily measures of stress were collected from routine administrative records, such as student contacts with school and community services or staff. Stress indicators included several categories: dietary (cafeteria customer counts), medical, academic, facilities use, and behavioral. Findings indicated that patterns of stress in schools are highly variable over the course of the academic year; are systematic, not random, in school environments; and may vary in predictable ways between schools. There were differences in the patterns between the white schools and the Hispanic school. More should be done to investigate time-ordered patterns of stress in educational settings and to control for this source of variation in planning, evaluation, and research. (Author/JM)