Windows on comprehension: Reading comprehension processes as revealed by two think-aloud procedures

To address methodological questions regarding use of the think-aloud (TA) procedure and theoretical questions regarding the roles of prior knowledge and strategy use in reading comprehension, 24 college students each read 3 passages in 3 different presentation modes (marked TA, unmarked TA, and control) and answered essay comprehension questions. There was no effect of presentation mode on essay scores. TA comments were coded into 4 categories, 2 of which were significantly correlated with comprehension scores for marked but not unmarked passages. The authors conclude that the marked procedure elicited more veridical protocols. A second coding and analysis of the marked protocols showed that students who scored high on the comprehension test were more likely to have made many TA comments reflecting a "knowledge-transforming" approach to the text.