Advisor

• Dr. Paul Resta

Degree

• Doctoral

Abstract

The purpose of this experimental study was to determine if learner-generated summaries with feedback in a computer-based hypermedia instructional program resulted in higher performance in the immediate and delayed recall of information than multiple-choice questions with feedback in a computer-based hypermedia instructional program. At the beginning of the semester, students were randomly assigned to either treatment one or treatment two: (1) hypermedia study guide incorporating multiple-choice questions with interactive feedback, and (2) hypermedia study guide incorporating learner-generated summaries with interactive feedback. Students remained in their assigned treatment throughout the entire semester. The instructional materials were programmed in Linkway Live. The instrumentation used in the study consisted of an interactive computer-mediated objective test developed for the study. Ten of the items measured recognition (multiple-choice items in which there were four alternatives for each item), and the remaining ten items were constructed to measure cued-recall (short answer). A directional t-test on both immediate and delayed recall mean scores of the two treatment groups revealed a significant difference. The critical t value was 1.67. $P(vert Tvert > 1.67) = 0.05.$ The general conclusion is that the use of learner-generated summaries in computer-based hypermedia instructional programs may enhance both immediate and delayed recall of information as measured by fill-in-the-blank items. Furthermore, the significant difference in performance between the two treatment groups indicates that the addition of constructed responses in computer-based hypermedia instructional programs may facilitate learning of complex knowledge structures and extend the limits of learning on computer-based programs beyond the typical immediate recall of information common in most computer-based instructional environments.