Constructing Confidence Intervals for Effect Sizes in ANOVA Designs

Abstract

A confidence interval for effect sizes provides a range of plausible population effect sizes (ES) that are consistent with data. This article defines an ES as a standardized linear contrast of means. The noncentral method, Bonett's method, and the bias-corrected and accelerated bootstrap method are illustrated for constructing the confidence interval for… (More)

1 Figure or Table

Topics

  • Presentations referencing similar topics