This simulation study modified the repeated measures mean difference effect size, d=RM , for scenarios with unequal pre- and post-test score variances. Relative parameter and SE bias were calculated for dRM ≠ versus dRM = . Results consistently favored d≠RM over d=RM with worse positive parameter and negative SE bias identified for d=RM for increasingly heterogeneous variance conditions.
Smith, Lindsey J. Wolff and Beretvas, S. Natasha
"Estimation of the Standardized Mean Difference for Repeated Measures Designs,"
Journal of Modern Applied Statistical Methods:
2, Article 27.
Available at: http://digitalcommons.wayne.edu/jmasm/vol8/iss2/27