Meta-analysis steps

This page walks through the steps of conducting a meta-analysis. For an introduction to what a systematic review and a meta-analysis are, see What is a systematic review / meta-analysis?

Defining the research area

Clarifying the research question the meta-analytic review addresses.

Defining inclusion and exclusion criteria

Which studies are eligible for the meta-analysis, based on study and publication characteristics.

Searching the literature

Search strategies for retrieving the relevant literature on the topic of the meta-analysis.

Selecting primary studies

Removing duplicates and checking each retrieved reference against the inclusion criteria.

Coding primary studies

Extracting the data needed for the analysis from each primary study, including an assessment of study quality.

Computing effect sizes

Effect size, variance, standard error, and confidence interval for each study, and how to report them.

Aggregating effect sizes

Combining effect sizes into an overall estimate with the fixed-effect or random-effects model.

Assessing heterogeneity

Measuring how much study effect sizes disagree with each other, using the Q statistic and I².

Testing moderators

Testing whether categorical or continuous factors affect the magnitude of the effect sizes.

Evaluating publication bias

Methods for assessing whether published studies differ systematically from unpublished ones.

Publishing a meta-analysis

Reporting standards and guidelines to refer to when writing up a meta-analysis.