This study investigates the relationship between selected student characteristics and an indicated preference for self-directed study. Additional relationships between academic achievement, self concept, cognitive style, and teacher ratings are also examined. Subjects for the study were 58 students from grades five through eight in a large county school system. Scores from tests of achievement, mental ability, self concept, reflection/impulsivity, and field dependence/independence were used for the analyses. In addition, students were asked their preference for self-directed study in math, English, both subjects, or neither subject. Teachers were also asked to rate each student's ability to profit from this instructional mode. Results indicated significant correlations between field dependence/independence and math and English achievement. No significant correlations were found between reflection/impulsivity and achievement; between cognitive style and self concept; or between preference for self directed study and teacher ratings, cognitive style, achievement, self concept, IQ, race, sex, age, or grade level. The paper suggests that students should not be assigned to self-directed study on the basis of these variables, but that student choice should be taken into consideration. The results suggest that instruction can be improved through the consideration of student needs and student preferences. The need for further research is emphasized. (BD)