Drawing construction as a strategy for learning from text

Classroom use of student-generated drawings has been encouraged for a number of purposes (e.g., R. Hubbard & K. Ernst, 1996). The present study examined the use of drawing as a learning strategy for 5th- and 6th-grade students reading science text. Three experimental drawing conditions and a reading control tested the hypothesis that drawing is effective only when students are supported during the construction process. Drawing (draw) participants constructed drawings only, whereas illustration comparison participants compared drawings with a provided illustration. Prompted illustration comparison (PIC) participants answered prompting questions to guide this comparison process. Dependent measures included a free-recall and recognition posttest, drawing accuracy, on-line self-monitoring behaviors, and time on task. PIC participants constructed the most accurate drawings and also scored significantly higher on the free-recall posttest. No differences were found on recognition posttest items. Although all drawing conditions spent significantly more time on task, these participants also engaged in significantly more self-monitoring events than did reading control participants. PIC participants also engaged in more events than did draw participants.