Effect Size Indexes for Dichotomized Outcomes Under Variance Heterogeneity: An Empirical Investigation of Accuracy and Precision

• Activity Number: 423
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Social Statistics Section
• Author(s): Jeffrey Kromrey*+ and Bethany Bell
• Companies: University of South Florida and University of South Carolina
• Keywords: Effect Sizes ; simulation ; statistical bias ; dichotomy

Abstract

The accuracy and precision of seven effect size indices proposed for estimating a population standardized mean difference from a 2 X 2 sample table of dichotomized variables were investigated using Monte Carlo methods. Normally distributed, continuous data were generated for two groups and the continuous variable was dichotomized at specified cut points. Factors manipulated in the simulation study's completely crossed design included overall sample size (n1 + n2 = 30, 60, 120, 240), ratio of sample sizes in the two groups (1:1, 1:2, 1:4), population effect size (0, .2, .5, .8), continuous score cut point for dichotomization (.10, .25, .40, .50, .70), and population variance ratio (1:1, 1:2, 1:4). For each sample generated, each of the seven proposed effect size indices was calculated. Both the statistical bias and the RMSE were computed across the set of 10,000 replications of each condi