Multiple Imputation for Confidence Interval Estimations for Standardized Linear Contrasts of Means: The One-Way Fixed-Effects Between-Subjects Univariate Case

• In Event: Poster Session 3
  In Poster Session: Division D Section 2 Quantitative Methods and Statistical Theory

Abstract

The authors define effect size (ES) as a standardized linear contrast of means for one-way, fixed-effects between-subjects univariate designs. Within this context, this study aims to investigate the performance of multiple imputation (MI) for constructing Bonett’s confidence interval for ES under conditions manipulated by six factors: (1) number of independent levels, (2) population variances, (3) population ESs, (4) sample sizes, (5) missing rates, and (6) the number of imputations. These conditions will be specified in the MCMC algorithm of SAS MI procedure, and simulation results will further our understanding about Bonett’s method in the presence of missing data. This study is confined to (1) one-way, fixed-effects between-subjects univariate designs, and (2) a single contrast of means.

Authors

• Li-Ting Chen, Indiana University

• Chao-Ying J. Peng, Indiana University