Key points are not available for this paper at this time.
A causal-comparative analysis was done of the test performance of undergraduate students in five different sections of educational psychology taught over a 3-year period. A pretest-posttest design was used to test the hypothesis that a testing program employing fewer retests than original tests produced significant gains in mean highest test/retest scores. Original tests and retests were of comparable mean difficulty, discrimination, and cognitive level. Correlations between the unretested means and highest test/retest means for retesters in the five sections were statistically significant. Various retesting programs were not found to be significantly different in producing gains in mean test scores among retesters. Subgroup analysis revealed that retesting was used most frequently by students at the lower end of the marking spectrum. Further subgroup analysis revealed no significant difference in the frequency of retesting among sophomores, juniors, and seniors in the five sections. Students’ response to the use of retests was overwhelmingly positive.
Ward Mitchell Cates (Mon,) studied this question.