Statistical Methods in Meta-Analysis

Diana B. Petitti

in Meta-Analysis, Decision Analysis, and Cost-Effectiveness Analysis

Second edition

Published in print November 1999 | ISBN: 9780195133646
Published online September 2009 | e-ISBN: 9780199863761 | DOI:

Series: Monographs in Epidemiology and Biostatistics

Statistical Methods in Meta-Analysis

Show Summary Details


This chapter addresses only systematic review and meta-analysis. It first introduces the concept heterogeneity then describes how heterogeneity affects the ability to draw valid conclusions from the summary estimate of effect size in study data that are analyzed in a meta-analysis. The difference between the assumptions that underlie the use of a fixed-effects and a random-effects model are explained. The chapter describes the Mantel Haenszel and the Peto methods for calculating a summary measure of effect size in a meta-analysis. These are the two methods used most commonly for information from randomized trials when the analysis is based on the fixed-effects model. General variance-based methods for estimating a summary measure of effect are presented. These methods are often applicable in the meta-analysis of observational data. The formulae used to do statistical tests of heterogeneity are given. Finally, the chapter describes the DerSimonian Laird method for calculating summary estimates of effect based on a random effects model.

Keywords: heterogeneity; Peto; Mantel Haenszel; DerSimonian Laird; fixed effects model; random effects model

Chapter.  7684 words. 

Subjects: Public Health and Epidemiology

Full text: subscription required

How to subscribe Recommend to my Librarian

Buy this work at Oxford University Press »

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.