ERIC Number: ED445081
Record Type: Non-Journal
Publication Date: 2000-Jan-29
Reference Count: N/A
Understanding and Interpreting Effect Size Measures in General Linear Model Analyses.
Vasquez, Linda M.; Gangstead, Sandra Kay; Henson, Robin K.
Best practice is now understood by many to include reporting effect sizes when submitting manuscripts for publication since some (and potentially more) journals in the education field require, or strongly encourage, authors to report magnitude of effect measures with their statistical interpretation discussion. Therefore, it is important that researchers familiarize themselves with various effect size measures and how to interpret them. This paper reviews many of the effect size statistics available, including both corrected and uncorrected measures such as eta squared, omega squared, epsilon squared, the Wherry formula, the Herzburg formulas (both fixed and random models), and the Lord formula. These statistics are explained and demonstrated with heuristic data. Several recommendations are presented to facilitate appropriate interpretation of effect sizes, including the consideration of estimate bias, fixed or random effects design use, and whether analyses were conducted using univariate or multivariate statistical approaches. One table shows actual calculations using general linear model analyses. (Contains 42 references.) (SLD)
Publication Type: Information Analyses; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A