Development of a Model for Adaptive Training via Computer-Assisted Instruction Utilizing Regression Analysis Techniques.
FLORIDA STATE UNIV TALLAHASSEE COMPUTER-ASSISTED INSTRUCTION CENTER
Pagination or Media Count:
The intent of the study was to investigate the efficacy of a dynamic decision model for an ongoing learning situation presented via computer-assisted instruction CAI. In addition to correctness on learning frames within a concept, criterion frames testing the concepts, and correctness of end of unit quiz questions, the study was concerned with variables such as latencies for each of the above measures, and the subjects confidence of his response on the criterion and quiz questions. The first step involved the investigation, through correlational analysis, of the relationship of these variables with performance on the final examination for the two hour course on concepts of Boolean algebra. Based on the results of correlation and regression analysis with the final exam as the dependent measure, the relevant variables were incorporated into the decision model. The adaptive decision model was shown to be effective in identifying those trainees who needed remedial instruction. However, the use of the adaptive model did not significantly improve performance on a course presenting basic concepts of Boolean algebra when compared with a group of trainees that did not have the benefit of the adaptive model. Author
- Humanities and History