Computer-enriched instruction (CEI) is better for preview material instead of review material: An example of a biostatistics chapter, the central limit theorem

This study examines the timing using computer-enriched instruction (CEI), before or after a traditional lecture to determine cross-over effect, period effect, and learning effect arising from sequencing of instruction. A 2x2 cross-over design was used with CEI to teach central limit theorem (CLT). Two sequences of graduate students in nursing participated in this study. Sequence A was given the CEI and handout first, followed by a traditional lecture and handout. Sequence B was given a lecture and handout first, followed by the CEI and handout. A pre-test and posttests (posttest1 after the first presentation and posttest2 after the second presentation) were given to measure instructional outcomes. All tests were 10 multiple choice questions with four options and only one option was the correct answer. No significant difference in scores was seen in posttest1 between students in sequence A (6.1+/-2.5) and B (6.5+/-2.1) (p=0.52 by unpaired t-test; p=0.80 after adjusted for pre-test score by ANCOVA). No significant carryover effect was seen (p=0.66). A significant period effect was observed, in which 7.7 for period 2 was greater than 6.3 for period 1, the difference of 1.4 with 95% CI between 0.8 and 2.0, p<0.0001. When the course material was taught twice, traditional lecture had a higher average score (7.3) than CEI (6.6) and the difference between two teaching methods was 0.7 with 95% CI between 0.1 and 1.3, p=0.02). Sequence A's increase (2.1) represented a preview effect of CEI and sequence B's increase (0.8) represented a review effect of CEI. The study time needed for the first presentation was the same for the two teaching methods (55min); the study time needed for the first presentation was always longer than that of the second presentation, regardless of which teaching method was being used. When comparing the subjective rating on two teaching methods between two sequences, sequence A students rated traditional lecture significantly better than sequence B students did in regards to satisfaction (p=0.0003), clarity (p=0.002), understanding (p=0.02). About two-thirds of sequence A students preferred CEI followed by the lecture, and 43.8% of the sequence B students preferred the lecture followed by CEI. In conclusion, when CLT was taught once, CEI and traditional lecture were equivalent in terms of the test score (posttest1) and teaching time. When CLT was taught twice, traditional lecture had a higher average test score than CEI. Preview effect of CEI in CLT was suggested because of higher difference in test score in sequence A than sequence B, and better subjective rating of traditional lecture in sequence A than in sequence B.