The relevance of multiple‐choice reading test data in studying expository passage comprehension: The saga of a 15 year effort towards an experimental/correlational merger

Abstract

This article argues for the usefulness of merging experimental and correlational approaches in the study of expository prose comprehension. Towards this end I review the studies my colleagues and I have completed in recent years. In the course of reviewing the effects of rhetorical structure, concreteness/abstractness of texts, and other variables, it is shown that the correlational approach yields results similar to the experimental literature. But more importantly, the vast data banks associated with multiple‐choice tests allow us to explore some of the variance that exists across passages and across student ability levels. The types of significant variables studied in explaining especially main idea identification raise issues of how these results might impinge on several prominent theories of prose comprehension. Several suggestions for future research are made.