One framework, eleven research areas
The statistical framework of evidence synthesis is shared across disciplines, but almost everything around it differs: the study designs that dominate, the outcomes and effect measures, the appraisal tools, the reporting conventions and even whether pooling is customary. A meta-analysis of randomized trials in cardiology has little in common, in practice, with a meta-analysis of correlations between organizational constructs or of elasticities in economics.
Each research-area page explains what evidence synthesis looks like in that discipline, which methods suit it, which do not, and what an expert reviewer in the field will look for. Medicine and health sciences are the flagship area and have the most developed set of specialty pages.
The research areas
Medicine and health sciences
Trials, cohort studies, diagnostic accuracy and prognostic research; the flagship area, with fifteen specialties.
Life and biological sciences
Effect-size synthesis in experimental and observational biology, including preclinical evidence.
Psychology and behavioral sciences
Standardized effect sizes and correlations, moderators, replicability and publication bias.
Management and business
Correlation-based synthesis across organizations, construct measurement and context moderators.
Economics and econometrics
Meta-regression analysis of estimates such as elasticities, with attention to publication selection.
Education
Learning-outcome effect sizes, clustered designs and intervention moderators.
Social sciences
Policy evaluation, qualitative synthesis, realist review and systematic review.
Environment and sustainability
Ecological effect sizes, systematic maps and environmental evidence.
Agriculture and food sciences
Field-trial synthesis, ratio-based effect sizes and yield outcomes.
Sports and exercise sciences
Small-sample trials, standardized effects and training-intervention moderators.
Engineering and technology
Where meta-analysis is established and where systematic mapping fits better.
What differs between areas
- Study designs. Randomized trials and cohort studies dominate health research. Surveys and experiments dominate psychology and management, field trials dominate agriculture, and quasi-experimental estimates dominate economics.
- Effect measures. Risk and hazard ratios are typical in health, standardized mean differences and correlations in the behavioral sciences, ratio-based effects in ecology, and elasticities in economics.
- Appraisal and reporting. Health reviews use RoB 2, ROBINS-I, QUADAS-2 and GRADE with PRISMA reporting. Other fields use different tools or none, and the reporting norms differ.
- Customary practice. In some disciplines meta-analysis is routine. In others a systematic or scoping review without pooling is the more appropriate design, and the page for the field says so.
If your field is not listed, or sits between two areas, describe your project. The areas are a guide to what we know best and not a limit on what can be discussed.
Get a quoteTell us your field, study type and target journal.
Frequently asked questions
Is meta-analysis used in every discipline?
No. It is established in clinical medicine, psychology and education, used in parts of the social, management and ecological sciences, and uncommon in some others. Where pooling is rare, a systematic or scoping review without pooling is often the better design.
Why are the specialties of medicine given more detail?
Medicine is where the methods of evidence synthesis are most developed and most standardized, and the flagship area has a page for each major specialty. Other areas are being developed in the same way.
My field is not on the list. Can you still help?
Possibly. Describe your question and study type, and the first reply will say whether evidence synthesis suits it and whether we have the right expertise.