Effectiveness of a Graduate Measurement Course / Pamela M. Broadston and Robert L. Kennedy.

To determine the effectiveness of a graduate educational assessment course in measurement processes, this study compared test scores from the beginning and from the end of the course. The three sections that were the focus of the study were offered in 2002 and 2003 by the same instructor. The course was offered as a traditional face-to-face class using PowerPoint presentations during all of the lectures. All sections incorporated several quizzes, hands-on activities, and a 50-question multiple choice final examination. There were 44 participants for whom there was complete information, 35 females and 9 males. The assumptions for the dependent t-test and the Wilcoxon test could not be met, so a quantile (sign) test was run to compare the pretest and posttest scores. The assumption that the measurement scale be at least ordinal was met since the data comprised frequency counts, but random selection was not possible since students are not randomly assigned to classes. However, the students did not exhibit any obviously exclusive characteristics. The test indicated that the null hypothesis of no statistically significant difference between the pretest and posttest scores could be rejected at the p=0/.00024 level (13 higher and 0 lower). It is concluded, then, that there were differences relative to the pretest and posttest scores, suggesting that the class was effective for learning the measurement topics introduced. The syllabus is attached. (Contains 2 figures, 2 tables, and 32 references.) (SLD)

Microfiche. [Washington D.C.]: ERIC Clearinghouse microfiches : positive.