This paper provides a general overview of a major study of student absenteeism as well as a narrower look at the problems associated with measuring student absences. The study data came from questionnaires completed by students, teachers, and administrators at six high schools in two urban districts in the western United States; interviews with teachers and administrators; and district records. After describing the sample, the paper reviews problems of underreporting absences, difficulties in assessing the significance of absence figures, and problems in defining "absence." A measure of absence based on class periods rather than on full or half days is proposed. The paper next discloses the magnitude of the absence problem. Students' reasons for attending or not attending classes are examined, and the effects of class subject, size, difficulty, teacher, and grading system on absence rates are reviewed. Differences between frequently and infrequently absent students are noted. The differences between teachers with high student absence rates and those with low rates are explored in terms of classroom practices, attendance monitoring procedures, and beliefs concerning absence control. The paper concludes with eight specific recommendations to schools for increasing control over absences. An appendix lists the individual reports based on the study. (PGD)