I have been poked for using cohen's d as an effect size statistic in a meta-analysis of pharmacogenetic studies. Apparently is has been argued that (McGough et al., 2009) cohen's d is not appropriate as "given the large stimulant treatment effects [...], effects of medication dose are likely to outweigh any genetic effects."
Further it is argued that: "the proper question is not whether different genotypes should be considered as different treatments but how much variance in treatment outcome is explained by genetic as opposed to medication effects. Cohen f 2 provides a signal-noise ratio, indicating how much variability is attributable to the model’s fixed effects and thus provides a dimensionless measure of the magnitude of gene and gene-dose signals."
This confuses me as I have found no meta-analysis using cohen's f and even wikipedia writes that cohen's f "estimates for the sample rather than the population and is biased (overestimates effect size for the ANOVA)."
McGough, J.J., McCracken, J.T., Loo, S.K., Manganiello, M., Leung, M.C., Tietjens, J.R., Trinh, T., Baweja, S., Suddath, R., Smalley, S.L., Hellemann, G., Sugar, C.A., 2009. A Candidate Gene Analysis of Methylphenidate Response in Attention-Deficit/Hyperactivity Disorder. J Am Acad Child Adolesc Psychiatry. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2888980/?tool=pubmed
Is there a general recommendation which is the appropriate effect size measure for a meta-analysis of pharmacogenetic studies?
