Performance on a Classroom Test: Interrogating Scores, Time and Grades

Abstract

The objective of this paper 1 is to report on an analysis of performance on a classroom test, administered at the beginning and end
of the semester. A descriptive analysis was made for the purpose of reflection: average performance and variation in scores and the time
taken to complete the test at initial and final administrations are worked out. Performance on the test (initial and final scores) is checked for
relationship with final course grades. Rank correlations between scores and time taken to complete the test (twice) are computed. The test
correlates positively with the final course grades of the students: initial test, r = 0.32; final test, r = 0.68. In about the same average time taken
(43 minutes each) for the initial and final test, the average gain in performance was about nine points (8.65). Less variation in scores and
time taken was also observed in the final test. The evidence from rank correlations between scores and time taken tend to suggest that they are
inversely related: bordering on no relationship (R = 0.082) for the initial and entering negative relationship (R = -0.063) for the final. The positive
gain in performance and the relatively moderate correlation of the test with the final course grades is found to be reassuring, while the inverse
relationship between the scores and the time taken calls for a thorough investigation on a larger sample.