Poster #3 - Data Size Planning for Multifactor ANOVA Designs via Adequately Narrow Confidence Intervals for Partial Eta-Squared

• In Event: 47.099 - MTCC Poster Session 11
  In Poster Session: 47.099-1 - Division D Section 2 Statistical Theory and Quantitative Methodologies

Abstract

Despite their immense appeal, methods of data size planning via accuracy in effect size estimation (AESE) for multifactor ANOVA designs using the widely used effect size measure, partial eta-squared (H2), have not yet been developed. First, we provide a method that determines the required data size so that the expected width of the confidence interval for H2 in multifactor ANOVA designs will be adequately narrow. Then, a second method is developed that determines the required data size so that the observed width of the confidence interval for H2 will be adequately narrow with a preset level of probabilistic certainty. Finally, we provide an R program that enables the practical use of the methods discussed in this article.

Authors

• Reza Norouzian, Texas A&M University - College Station

• Michael Anthony de Miranda, Texas A&M University - College Station