INSTRUCTIONAL EFFECTS OF TEXT STRUCTURE-BASED READING STRATEGIES ON THE COMPREHENSION OF SCIENTIFIC PROSE

Five different types of text structures were identified for scientific prose. Text structures were defined as organizational patterns used to meaningfully interrelate passage information. These five structures were Generalization (conveys a principle or law), Enumeration (list-like), Sequence (a series of steps or stages), Compare/Contrast (analyzes similarities and differences), and Classification (grouping on similarities and differences). It was hypothesized that training in the use of reading strategies based on these structures would enhance comprehension of scientific text.In Experiment 2 a 10-hour training course for reading strategies were developed for three text structure types (Generalization, Enumeration and Sequence). One group of Junior College chemistry students (Experimental Group) was first pretested on three biology passages representing the three structures above. These subjects then practiced the reading strategies using a chapter in their course-assigned textbook. Posttesting on an alternate set of three biology passages occurred at the end of training. A second group (Control Group) was pre- and posttested on the same biology passages but did not participate in strategy training sessions. Results indicated the Experimental Group improved their ability to answer questions assessing information stated directly in the text (Textbound) by 14% correct. . . . (Author's abstract exceeds stipulated maximum length. Discontinued here with permission of author.) UMIThe goal of Experiment 1 was to examine the effects of instruction on readers' ability to identify the text structure used to organize short test passages. One group of college level subjects (Experimental Group) received a short lecture and instructional booklet describing the five structures. They were then asked to classify a series of 20 short, one to two paragraph passages into one of the five structure types. A second group of subjects (Control Group) was asked to group the same set of passages into five categories on the basis of the author's organization of passage information. Results indicated direct instruction improved the Experimental Group's accuracy in classifying passages (79% correct) relative to the Control Group (28% correct). Further analysis of the Control Group's performance indicated the more text structures they labeled the greater their success in grouping similarly structured passages.