Meta-analysis

Background resources on systematic reviews and meta-analysis — start here if you are new to the topic.

Six individual studies, each with its own effect estimate and confidence interval, pooled into a single combined estimate that is more precise than any one study alone

What is a systematic review / meta-analysis?

An introduction to systematic reviews and meta-analysis, and how the two terms relate.


Meta-analysis steps

The steps involved in conducting a meta-analysis, from defining the research area to publishing the results.


Methods and formulas

Effect sizes, heterogeneity, moderators, and publication bias — with the underlying formulas.


Glossary

Definitions of the most important terms used in meta-analysis.


References

Books and articles on meta-analysis, organized by topic.


In-depth topics

Companion material for the steps above — multilevel models and reporting guidelines.

Multilevel models

Handling dependent effect sizes when a study contributes more than one outcome, timepoint, or subscale.


PRISMA flow diagram and reporting guideline

How to read the PRISMA flow diagram, what the checklist covers, and where it fits among other reporting guidelines.


Specialized topics

Extensions of the standard process above, relevant only for specific kinds of data or research questions.

Network meta-analysis

Comparing three or more treatments at once by combining direct and indirect evidence, and the transitivity assumption it relies on.


Meta-analysis of single-case experimental designs

Overlap-based effect sizes (PND, NAP, Tau-U) for single-subject A-B designs, where standard effect sizes don't apply.