LEARNING SET FORMATION IN PROGRAMMED INSTRUCTION
Abstract
Two different orders of three units of programed instruction were administered to groups of students matched on (a) intelligence or (b) relevant achievement tests. Comparisons were made between groups that were (a) high or (b) average on each matching variable. The hypotheses being tested were that after varied amounts of prior practice in programed instruction, (a) learning set formation would not be demonstrated by the high intelligence and high achievement groups, and (b) learning set formation would be demonstrated by the average intelligence and average achievement groups. Only partial support was obtained for each hypothesis. The data indicated the following: (a) In a programed sequence, error rate is a more appropriate measure than achievement for observing learning set formation. (b) Learning set formation is observable in programed instruction for all learners regardless of individual differences. Since, reduced error rate was the indication of learning set formation, the phenomenon can be measured only in programs involving a moderately high error rate. (c) Since error rate differed for some of the experimental groups while achievement remained the same, the results were interpreted to mean that a moderately high error rate program which offers opportunity for correction of response errors may be as effective in producing learning as a low error rate program which confirms correct responses.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1964
- Accession Number
- AD0609802
Entities
People
- James H. Reynolds
- John S. Abma
- Robert Glaser
Organizations
- University of Pittsburgh