[Skip to Navigation]
December 1996

Studies of Ventricular Enlargement-Reply

Author Affiliations

Department of Psychiatry School of Medicine Case Western Reserve University 10900 Euclid Ave Cleveland, OH 44106 Helio Elkis, MD, PhD São Paulo, Brazil

Arch Gen Psychiatry. 1996;53(12):1166-1167. doi:10.1001/archpsyc.1996.01830120106018

In Reply  The critique by Persaud of our meta-analysis' is based on the following 2 contentions: (1) There is a generally accepted quantitative criterion for the fail-safe N (Nfs) and (2) the quantitative criterion of Rosenthal2—a criterion that would require us to achieve an Nfs equal to 65—is a reasonable criterion to apply to our meta-analysis. In our opinion, both contentions are refutable.One point that needs to be made before we proceed is that several types of meta-analysis exist. Rosenthal discusses Nfs in the context of a "combined significance" metaanalysis, whereas we conducted an "effect size" metaanalysis.3 (See the book by Hedges and Olkin3 for a critique of the combined significance approach and a defense of the effect size approach.) Separate formulas for the calculation of Nfs are required for these 2 types of meta-analysis. This is not debatable. What is debatable is

Add or change institution