Graduate Introductory Statistics : In Class vs. On Line / Robert L. Kennedy and Corliss Jean McCallister.

This study compared the effectiveness of traditional teaching, electronic mail (e-mail), and combination approaches for teaching graduate introductory statistics classes. The e-mail courses that are the focus of this study were offered in the fall terms of academic years, 1997 through 2000 by the same instructor. All sections incorporated the use of a computer for data analysis. There were 41 participants in the traditional classes, 20 in the electronic-only classes, and 28 using both methods. In all classes, the majority of students were white females. Twenty-item multiple-choice pre-tests and post-tests on basic statistical topics were given. An analysis of covariance (ANCOVA) was run using post-test scores as the response variable and pre-test scores as the covariate. The ANCOVA technique involves features of both the analysis of variance and regression, so assumptions for both were tested. Normality and homoscedasticity across all groups were verified. Homogeneity of regression was observed in scatter plots of pre-test scores versus post-test scores and their trend lines by treatment and control groups. The test indicated that the null hypothesis of no statistically significant difference among the traditional, electronic, and combined classes scores could not be rejected at the 0.05 level. The effect size (f=0.16) was small. The paper concludes that offering the course by any of these three approaches seemed neither to help nor hinder the performance of students to the extent measured by the multiple-choice tests. The course syllabus is attached. (Contains 29 references.) (SLD)

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