Abstract
Ninety-eight average 7th- and 8th-grade readers were assigned to one of four groups based on their own proficiency predictions (high vs. low) under two task demands (reading for essay vs. multiple-choice exams). Subjects read one of two folk tales, took the appropriate exam, retrospectively reported strategies used, and freely recalled the folk tale read. Results indicated that when scores on the essay and free recall measures were adjusted for prior reading achievement, the self-perceived high proficiency group performed significantly better than the self-perceived low proficiency group. There was also some evidence to suggest that students' perceptions of proficiency affected their choice of strategic activity. Implications of these findings for educational practice and research methodology are discussed.