Bias in meta-analyses using Hedges’ d

ECU Author/Contributor (non-ECU co-authors, if there are any, appear on document)

James R. Bence (Creator)

Elizabeth A. Hamman (Creator)

Craig W. Osenberg (Creator)

Paula Pappalardo (Creator)

Scott D. Peacor (Creator)

Institution

East Carolina University (ECU )

Web Site: http://www.ecu.edu/lib/

Abstract: The type of metric and weighting method used in meta-analysis can create bias and alter coverage of confidence intervals when the estimated effect size and its weight are correlated. Here, we investigate\r\nbias associated with the common metric, Hedges’ d, under conditions common in ecological meta-analyses.\r\nWe simulated data from experiments, computed effect sizes and their variances, and performed meta-analyses applying three weighting schemes (inverse variance, sample size, and unweighted) for varying levels\r\nof effect size, within-study replication, number of studies in the meta-analysis, and among-study variance.\r\nUnweighted analyses, and those using weights based on sample size, were close to unbiased and yielded\r\ncoverages close to the nominal level of 0.95. In contrast, the inverse-variance weighting scheme led to bias\r\nand low coverage, especially for meta-analyses based on studies with low replication. This bias arose\r\nbecause of a correlation between the estimated effect and its weight when using the inverse-variance method.\r\nIn many cases, the sample size weighting scheme was most efficient, and, when not, the differences in efficiency among the three methods were relatively minor. Thus, if using Hedges’ d, we recomm

Additional Information

Publication

Other

Language: English

Date: 2023

Subjects

bias\; coverage\; effect size\; Hedges’ d\; meta-analysis\; sample size\; weights

Email this document to