Predicting multiple-text integration

Abstract This study investigated how single-text reading comprehension and individual differences contribute to multiple-text integration. One hundred fourteen students read sets of multiple texts on two distinct topics and completed two multiple-text integration tasks for each topic: an essay and a sentence verification task. The tasks were evaluated across three integration levels: selecting main ideas, forming generalizations, and generating deep-level inferences. The researchers conducted a series of linear regressions and hierarchical linear models separately for each integration task and integration level. The findings revealed that reading comprehension measures differentially influenced the two tasks and three integration processing levels. Additionally, individual differences — including reading strategies, interest, and vocabulary — were found to predict multiple-text integration performance. Overall, the results suggest that reading comprehension skills and individual differences uniquely predict various aspects of intertextual integration, highlighting the complex nature of how readers process and synthesize information across multiple texts.