ERIC Number: ED339743
Record Type: Non-Journal
Publication Date: 1991-Nov
Reference Count: N/A
A Comparison of the Glass Meta-Analytic Technique with the Hunter-Schmidt Meta-Analytic Technique on Three Studies from the Education Literature.
Hough, Susan L.; Hall, Bruce W.
The meta-analytic techniques of G. V. Glass (1976) and J. E. Hunter and F. L. Schmidt (1977) were compared through their application to three meta-analytic studies from education literature. The following hypotheses were explored: (1) the overall mean effect size would be larger in a Hunter-Schmidt meta-analysis (HSMA) than in a Glass meta-analysis (GMA) due to correction for measurement error when compared on the same set of experimental data; (2) the overall mean effect size calculated using the pooled within-group standard deviation in HSMA would not differ significantly from that in a GMA that uses the control group standard deviation; (3) most of the variation between study effect sizes would be due to sampling error according to sampling error correction formulas from the HSMA method; and (4) no moderator variables would be found because most of the variation between study effect sizes is due to sampling error. A correlated t-test was used to compare the overall mean effect sizes that were calculated using GMA and HSMA. Pearson correlations and analyses of variances were run on the study data. Three meta-analytic studies were selected and statistical data from each of the individual studies were collated. Results support Hypotheses 1 and 2, but reject Hypotheses 3 and 4. It is argued that the HS correction formulas are technically more accurate, but that the Glass method is adequate in portraying effect size and more easily calculated. Three tables present data from the meta-analyses. A 21-item list of references is included. (SLD)
Publication Type: Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A