Accuracy and Precision of Generalized Omega-Squared: An Empirical Investigation of Research Design and Nonnormality

• In Event: 33.058 - Effect Sizes, Sample Sizes, and Power

Abstract

Generalized omega-squared has been proposed as a variance-explained effect size that is comparable across a variety of ANOVA designs, including any number of between- and within-subjects factors. A simulation study was conducted to investigate the accuracy and precision of this effect size across designs, as well as its sensitivity to non-normal population distributions. Factors investigated in the simulation included sample size, population distribution, population eta-squared value, and ANOVA designs structure (number of factors and number of factor levels). Results indicated substantial bias in effect size estimates with small samples in the within-subject designs compared to bias in the between-subject designs. With small samples, the magnitude of bias differed across research design factors, but these differences decreased with larger samples.

Authors

• Anh P Kellermann, University of South Florida

• Patricia Rodriguez de Gil, University of South Florida

• Thanh Vinh Pham, University of South Florida

• Jeffrey D. Kromrey, University of South Florida

• Jeanine Romano, American Board of Pathology

• Yi-Hsin Chen, University of South Florida