The file drawer formula in meta-analysis estimates the number of null difference effect sizes (ESs) from unpublished studies in researcher file drawers that are necessary to reduce the mean ES to a magnitude no longer considered practical or meaningful. Unlike the file drawer concept, this paper proposes a novel use of Orwin's 1983 formula that produces an estimate of the potential bias in the mean ES due to missing ESs from studies that were located and retrieved, but rejected because of insufficient data for ES calculation.
Get full access to this article
View all access options for this article.
References
1.
CohenJ. (1969) Statistical power analysis for the behavioral sciences.New York: Academic Press.
2.
HunterJ. E., & SchmidtF. L. (1990) Methods of meta-analysis: correcting error and bias in research findings.Newbury Park, CA: Sage.
3.
KlineR. B. (2004) Beyond significance testing: reforming data analysis methods in behavioral research.Washington, DC: American Psychological Association.
4.
OrwinR. G. (1983) A fail-safe N for effect size in meta-analysis. Journal of Educational Statistics, 8, 157–159.
5.
RosenthalR. (1979) The “file drawer problem” and tolerance for null results. Psychological Bulletin, 85, 638–641.
6.
SibleyB. A., & EtnierJ. L. (2003) The relation between physical activity and cognition in children: a meta-analysis. Pediatric Exercise Science, 15, 243–256.
7.
YoungK. M.NorthernJ. J.ListerK. M.DrummondJ. A., & O'BrienW. H. (2007) A meta-analysis of family-behavioral weight-loss treatments in children. Clinical Psychology Review, 27, 240–249.
