This study examined the variability in certain academic behaviors of remedial college freshmen and investigated the predictive relationship of four measures of these students' academic attitudes and habits to those behaviors. The academic behaviors included students' selection of an anticipated instructional pace; attendance at optional lecture/discussion sessions; election of additional work in preparation for unit tests; actual rate of progress ion through the course; accuracy of anticipated pace selection; number of test trials needed to pass each unit test; and attribution of failure for unit tests failed. The 101 freshmen in this study were heterogeneous in regard to race, sex, and past academic performance. They were assigned to seven sections of a 15-week PSI remedial reading course, based on Nelson-Denny Reading test scores between the 11th and the 28th percentiles. During the first week of the semester, students were administered Rotter's I-E Scale, the Survey of Study Habits and Attitudes, and the State-Trait Anxiety Inventory. The second week, the students were oriented to the course structure by completing the first instructional unit as a class. The students then proceeded through the remaining seven units independently, at their own pace.