A0205
Title: A meta-analysis based hierarchical variance model for powering one and two-sample t-tests
Authors: Xinlei Wang - University of Texas at Arlington (United States) [presenting]
Jackson Barth - Southern Methodist University (United States)
Abstract: Sample size determination (SSD) is essential in statistical inference and hypothesis testing, as it directly affects the accuracy and power of the analysis. An SSD methodology is proposed for one and two-sample t-tests that ensure clinical relevance using a pre-determined unstandardized effect size. The novel approach leverages Bayesian meta-analysis to account for the uncertainty surrounding the variance, a common issue in SSD. By incorporating prior knowledge from related studies via a Bayesian gamma-inverse gamma model, an informative posterior predictive distribution is obtained for the variance that leads to better decisions about sample size. An empirical Bayes approach is proposed for efficient posterior sampling, which is further combined with a discretized simulation approach to facilitate computation. Simulations and empirical studies demonstrate that the methodology outperforms other aggregate approaches (simple average, weighted average, median) in variance estimation for SSD, especially in meta-analyses with large disparities in sample size and moderate variance. Thus, it offers a robust and practical solution for sample size determination in t-tests.
