A0241
Title: Effect size comparison for populations with an application in psychology
Authors: Sudeep Bapat - Indian Institute of Technology Bombay (India) [presenting]
Bhargab Chattopadhyay - Indian Institute of Technology Jodhpur (India)
Abstract: Effect size estimates are widely reported in various behavioral studies. In precise estimation or power analysis studies, sample size planning revolves around the standard error (or variance) of the effect size. These studies are generally carried out under sampling-budget constraints. Hence, the optimum allocation of resources to populations with different inherent population variances is important, as this affects the effect size variance. Sequential analysis is used to determine the theoretically optimal sample size, which maximizes either the power of the test for a testing problem or the precision of estimation by minimizing the variance under budget constraints. In the process, sequential theory is used to arrive at optimum sample sizes of the corresponding populations to achieve minimum variance. The developed sequential method is a distribution-free method and does not need knowledge of population parameters. Mathematical justification of the characteristics enjoyed by this method is laid out along with simulation studies. Wide applicability is demonstrated in the effect size comparison context.
