Robust variance estimation in meta‐regression with dependent effect size estimates

Research Synthesis Methods - Tập 1 Số 1 - Trang 39-65 - 2010
Larry V. Hedges1, Elizabeth Tipton1, Matthew C. Johnson1
1Northwestern University, Statistics, Evanson, IL, U.S.A.

Tóm tắt

Abstract

Conventional meta‐analytic techniques rely on the assumption that effect size estimates from different studies are independent and have sampling distributions with known conditional variances. The independence assumption is violated when studies produce several estimates based on the same individuals or there are clusters of studies that are not independent (such as those carried out by the same investigator or laboratory). This paper provides an estimator of the covariance matrix of meta‐regression coefficients that are applicable when there are clusters of internally correlated estimates. It makes no assumptions about the specific form of the sampling distributions of the effect sizes, nor does it require knowledge of the covariance structure of the dependent estimates. Moreover, this paper demonstrates that the meta‐regression coefficients are consistent and asymptotically normally distributed and that the robust variance estimator is valid even when the covariates are random. The theory is asymptotic in the number of studies, but simulations suggest that the theory may yield accurate results with as few as 20–40 studies. Copyright © 2010 John Wiley & Sons, Ltd.

Từ khóa


Tài liệu tham khảo

10.1002/9780470743386

Cooper HC, 2009, The Handbook of Research Synthesis and Meta‐analysis

Lipsey ML, 2001, Practical Meta‐analysis

Sutton AJ, 2000, Methods for Meta‐analysis in Medical Research

10.1007/s004420050381

Hedges LV, 1985, Statistical Methods for Meta‐analysis

Olkin I, 1994, The Handbook of Research Synthesis, 339

Olkin I, 2009, The Handbook of Research Synthesis and Meta‐analysis, 357

10.1037/0033-2909.103.1.111

10.1037/1082-989X.1.3.227

10.1002/sim.4780122405

Hedges LV, 2007, The Handbook of Statistics, 26, 919

10.1016/B978-012691360-6/50018-5

10.2307/1164634

10.1002/9780471722199

10.1002/sim.791

10.1002/sim.2913

10.1111/j.1467-985X.2008.00593.x

10.1890/0012-9658(1999)080[1142:SIIEMA]2.0.CO;2

KonstantopoulosS.Variance components estimation in meta‐analysis. Unpublished doctoral dissertation University of Chicago 2003.

10.3102/1076998607309080

10.1093/biomet/76.3.622

Raudenbush SW, 2009, The Handbook of Research Synthesis and Meta‐Analysis, 295

10.1016/0304-4076(85)90158-7

10.3102/00346543071003393

Eichler F, 1967, Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, 59

Huber P, 1967, Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, 221

10.2307/1912934

10.1002/9780470316436