I'm conducting a meta-analysis where I have to compare two interventions. One intervention (Intervention A) has around 26 studies, while the other (Intervention B) has >150. The estimates for B are much lower than those for A (the lower the better efficacy in my case). My question are these findings statistically valid? Or is the smaller (favourable) effect for B only due to the large number of studies?
The number of studies in the meta analysis should not affect the overall estimate of the magnitude of the effect but it would affect the precision of the overall estimate.
The aims of meta analysis are not just overall effect estimate . They also should include considerations of homogeneity between studies.
The effect size does not depend on the number of studies.
The number of studies influences precision, not the effect size.
To conclude that A and B are different, you need a subgroup comparison test or meta-regression.
- Badges
- Science topic
- Similar topics
- Computer Communications (Networks)
- Simulators