The purpose of this study was to test the effectiveness of high and low question generation as part of an instructional reading comprehension strategy called Reciprocal Teaching. A post test-only control group design was employed with aptitude factor tests administered prior to the study. In this design, the dependent variables were delayed written free recall and question generation for science and social studies expository texts; covariates were field dependence/field independence, induction, and verbal comprehension. gender and race/ethnicity also served as independent variables. One hundred eleven middle school students from five 7th grade classes were randomly assigned to four treatment groups in a 2x2 design. All groups differed in the level of questions to be generated from the passages: (a) low level questions; (b) mixed high and low level questions; (c) high level questions and (d) control. Subjects in all groups were trained for five consecutive days to use the four metacognitive activities of Reciprocal Teaching (predicting, questioning, summarizing and clarifying) while reading expository text. Scripts were developed for all groups to ensure uniformity of instruction. After training, subjects participated in two equivalent post test sessions. Analysis of covariance was performed with significant full model main effects for high level questions on total score for the question generation science text measure (F=3.80, pThe purpose of this study was to test the effectiveness of high and low question generation as part of an instructional reading comprehension strategy called Reciprocal Teaching.
|Title||:||The Testing of an Instructional Strategy for Improving Reading Comprehension of Expository Text in Science and Content Area Reading|
|Author||:||William Michael Brown|