Candidate: Jason T. Otto

Date: Thursday, March 11, 2004 2:00 pm - 4:00 pm
2536 Wood

Abstract: The objective of this dissertation was to produce effective computer-based programmed instruction modules to serve as supplemental training for an ongoing college seminar in behavior analysis. Computer-based programmed instruction supplemented a checklist of a strategy for diagramming behavioral contingencies in the first study and supplemented difficult textbook material in the other studies. In all, the instruction involved 31 concepts, rules, or objectives. Microsoft® PowerPoint® and Macromedia FlashTM were the authoring tools used to develop these supplemental modules. The modules involved multiple-choice-branching programming, which students completed as homework assignments that were delivered with a compact disk (Studies 1-4) and the World Wide Web (Study 5). In general, the goal was to measure the benefit of adding computer-based programmed instruction to current materials with which students were having difficulty. In Study 1 comparing paper-based with computer-based programmed instruction, students first took a pretest, then completed either a chapter from a paper-based workbook or a similar computer-based programmed instruction module, and finally took a posttest. In Studies 2-5, students first read a textbook chapter (Malott & Trojan Suárez, 2004), then took a test, or provided an original example of the concept being trained; then after completing computer-based programmed instruction, students took another test or provided another original example. Among the five studies, all but one showed statistically significant improvements following computer-based programmed instruction.
Study 1, involving a strategy for diagramming behavioral contingencies, showed large, statistically significant pretest-posttest improvement both when students completed paper-based programmed instruction by itself and the computer-based programmed instruction by itself. Study 2, involving behavioral-contingency diagrams of sick social cycles, showed no statistically significant improvement between students' original examples after reading the textbook and subsequent original examples after completing computer-based programmed instruction. Studies 3 through 5, involving stimulus equivalence, generalization gradients, and discrete trial/free-operant procedures respectively, showed statistically significant improvements after completing the relevant computer-based programmed instruction. Social validity in the form of student evaluations indicated the computer-based programmed instruction was highly preferred compared to the paper-based programmed instruction workbook used throughout the seminar, primarily because the computer provided feedback on the correctness of answers.

Related Topics

Main List of Archives:
Dissertation Defenses

Current Dissertation Defenses