Heterogeneity is concerned with the differences we observe between the results of the included studies. This presentation shows how to explore these differences and how this will inform our understanding of the effects we’re observing, and how we should interpret them. It includes a discussion of the fixed-effect and random-effects meta-analysis models, assessing heterogenetiy using the chi-squared test or the I2 statistic, and exploring heterogeneity using subgroup analysis.
The presentation has been translated into several languages other than English. Select an available language from the list at the top right of this page.
- Describe three general types of heterogeneity across studies
- Identify the main sources of heterogeneity
- Choose fixed-effect or random effects meta-analysis based on your assumptions about heterogeneity
- Identify heterogeneity by visual inspection of the forest plots or by examining relevant statistics
- Explore your results to explain heterogeneity
- Recognise the difference between subgroup analysis and sensitivity analysis
Exploring heterogeneity - slidecast
Compiled by Miranda Cumpston.
Based on materials by Georgia Salanti, Julian Higgins, Steff Lewis, the Cochrane Statistical Methods Group, the Australasian Cochrane Centre and the Dutch Cochrane Centre.