Statistical Science

Selection Models and the File Drawer Problem

Satish Iyengar and Joel B. Greenhouse

Full-text: Open access

Abstract

Meta-analysis consists of quantitative methods for combining evidence from different studies about a particular issue. A frequent criticism of meta-analysis is that it may be based on a biased sample of all studies that were done. In this paper, we use selection models, or weighted distributions, to deal with one source of bias, namely, the failure to report studies that do not yield statistically significant results. We apply selection models to two approaches that have been suggested for correcting the bias. The fail-safe sample size approach calculates the minimum number of unpublished studies showing nonsignificant results that must have been carried out in order to overturn the conclusion reached from the published studies. The maximum likelihood approach uses a weighted distribution to model the selection bias in the generation of the data and estimates various parameters of interest. We suggest the use of families of weight functions to model plausible biasing mechanisms to study the sensitivity of inferences about effect sizes. By using an example, we show that the maximum likelihood approach has several advantages over the fail-safe sample size approach.

Article information

Source
Statist. Sci. Volume 3, Number 1 (1988), 109-117.

Dates
First available in Project Euclid: 19 April 2007

Permanent link to this document
https://projecteuclid.org/euclid.ss/1177013012

Digital Object Identifier
doi:10.1214/ss/1177013012

JSTOR
links.jstor.org

Keywords
Meta-analysis file drawer problem selection bias weighted distributions maximum likelihood estimation

Citation

Iyengar, Satish; Greenhouse, Joel B. Selection Models and the File Drawer Problem. Statist. Sci. 3 (1988), no. 1, 109--117. doi:10.1214/ss/1177013012. https://projecteuclid.org/euclid.ss/1177013012


Export citation

See also

  • See Comment: Larry V. Hedges. [Selection Models and the File Drawer Problem]: Comment. Statist. Sci., Volume 3, Number 1 (1988), 118--120.
  • See Comment: Robert Rosenthal, Donald B. Rubin. [Selection Models and the File Drawer Problem]: Comment: Assumptions and Procedures in the File Drawer Problem. Statist. Sci., Volume 3, Number 1 (1988), 120--125.
  • See Comment: Nan Laird, G. P. Patil, C. Taillie. [Selection Models and the File Drawer Problem]: Comment. Statist. Sci., Volume 3, Number 1 (1988), 126--128.
  • See Comment: M. J. Bayarri. [Selection Models and the File Drawer Problem]: Comment. Statist. Sci., Volume 3, Number 1 (1988), 128--131.
  • See Comment: C. Radhakrishna Rao. [Selection Models and the File Drawer Problem]: Comment. Statist. Sci., Volume 3, Number 1 (1988), 131--131.
  • See Comment: William DuMouchel. [Selection Models and the File Drawer Problem]: Comment. Statist. Sci., Volume 3, Number 1 (1988), 132--133.
  • See Comment: Satish Iyengar, Joel B. Greenhouse. [Selection Models and the File Drawer Problem]: Rejoinder. Statist. Sci., Volume 3, Number 1 (1988), 133--135.