Description/Abstract:
ABSTRACT
THE INFLUENCE OF GLOBAL TEXT FEATURES ON THE THEMATIC PROCESSING OF EXPOSITORY TEXT
Diane Carol Jarin
Using more complex expository text passages than usually studied with children, this study examined the interactive effect of presence or absence of thematic initial paragraph sentences and a thematic linking sentence on comprehension. Subjects in the main experiment were 112 4th, 5th and 6th grade children who tested as excellent readers on reading comprehension measures. For two different passage contents, four forms were constructed. Form 1 contained both a thematic sentence for each paragraph and a thematic passage linking sentence at the end of paragraph one; Form 2 contained only a thematic sentence for each paragraph; Form 3 contained only a thematic linking sentence; and Form 4 contained neither type of thematic sentence. When thematic sentences were not present, detail sentences appeared as substitutes.
Subjects were divided randomly into four groups, each of which read one Form of both passages. After silently reading each passage, subjects wrote a short summary of the passage information. Twenty additional subjects were asked questions pertaining to their comprehension processes, immediately after summary writing.
Written summaries were rated on a Thematic Recall Scale reflecting the accuracy of their thematic content. Inter-rater and Equivalent Form ratings were highly reliable and the two contents produced no significant differences.
A three-way analysis of variance with repeated measures on the within-subjects factor was performed on the scores. Highly significant Thematic Recall Scale score differences and an interaction factor indicated that Form 1 scores were higher than scores for the other Forms, which did not differ significantly. The significant interaction effect appeared primarily responsible for the significant main effects of each type of thematic sentence.
The joint presence of both thematic sentences (Form 1) seemed needed by these good readers to produce organized comprehension as tested by written recall. Subjects who read other than Form 1 passages tended to write isolated details and extraneous information. The results and the interviews indicated that joint (rather than separate) presence of both thematic sentence types facilitated written summary construction. Knowledge of techniques for identifying thematic sentences did not ensure accurate responses. Implications for teaching reading comprehension and text structure identification are discussed.