Meta-Analysis

I am involved in statistical research. I use computer-based data simulation (the Monte Carlo method) to compare statistical methods or determine the reliability of a statistical method under unusual conditions. Actually, all of my work has looked at comparisons between different methods of meta-analysis (a technique for assimilating effect sizes from different studies addressing the same basic research question). I have used Monte-Carlo methods to compare different fixed- and random-effects methods of meta-analysis (Hedges and colleagues' method and Hunter and Schmidt's method) in a variety of situations such as the number of studies included in the meta-analysis, the relative sample-size of those studies, the simulation parameters for the superpopulation and the amount of variability in effect sizes (see Field, 2005, 2003, 2001 ).