The study examined whether there were differences in the quality of classroom time spent by 79 mainstreamed learning disabled fifth grade students and 141 regular fifth grade students; and also whether differences in qualtiy of time related to student, instructional, and classroom variables. Trained observers coded learner moves, content area, and difficulty level on 5 minute cycles during math instruction with six students in each of 37 classes. Among findings were the following: over one third of the designated math period was devoted to nonmath activities; mainstreamed students tended to spend less time engaged in math activities than regular students in the same class; the key time variable distinguishing between mainstreamed and regular students was that regular students spent more time on tasks where they experienced success; instructors' interactions with students, either individually or in groups, influenced the amount of time engaged in math activities; and the introduction of handicapped students into regular classrooms was related to decreases in the engaged time of both mainstreamed and regular students. Over half the document consists of an appendix containing the complete observation training manual developed for the study and provides explanations of the observation categories and procedures for coding observations. (DB)