Models for the Design of Instructional Text

Abstract

IN ORDER to identify guidelines for writing textbooks that better facilitate students' learning, a number of researchers have analyzed the revision strategies of experts and the effects of their revisions. The present study partially replicates and extends a study originally conducted by Graves et al. (1988) comparing a segment from a high school history text and revisions of that text by text linguists, composition teachers, and Time-Life writers. In revising the text, the Time-Life writers emphasized making the text more interesting, whereas the other two groups of experts focused on clarifying the structure and providing cohesion. In the current study, the performance of 11th-grade students on free-recall and short-answer tests as well as their ratings of interest and ease of learning all indicate that the revision by the composition teachers was most effective. This result confirms prior work by Britton, Van Dusen, Gulgoz, & Glynn (1989), but fails to confirm the original study by Graves et al., in which superior recall was found for the revision by the Time-Life writers. The authors further analyze the information recalled by students in relation to the hierarchical structure of the text in an attempt to understand the experts' specific revision techniques. The authors suggest that presenting textbook writers with a series of text-independent guidelines may be less effective than alternative strategies such as protocol-aided revision, in which the text is revised based on comments made by readers as they try to understand it.