Definition
Meta-analysis is the statistical combination of results from two or more separate studies- generally as part of a systematic review project - with the following aims:
Caution: Meta-analyses can mislead researchers, "particularly if specific study designs, within-study biases, variation across studies, and reporting biases are not carefully considered" (Deeks JJ, et al. Chapter 10: Cochrane Handbook for Systematic Reviews of Interventions). |
Start with a research question that you believe is best answered by:
1. conducting a systematic review AND
2. using statistical methods to summarize the results of the studies you included in your review (University of Edinburgh. Systematic Reviews and meta-analyses: a step-by-step guide).
For a comprehensive look at the steps involved in performing a meta-analysis see Chapter 12.2.1 Principles of meta-analysis from the Cochrane Handbook for Systematic Reviews of Interventions.
For a handy web-based effect size calculator see the Practical Meta-Analysis Effect Size Calculator, on the Campbell Collaboration website, developed by David B. Wilson, Ph.D., George Mason University.