Can we do a meta-analysis on a phenomenon for eg. behavioural spillover without being domain specific? I would appreciate if anyone can share some recent examples of such studies where meta is done on a particular phenomenon.
I'm not sure how systematic reviews with meta-analyses (SRMAs) are conducted in your field. In medicine, SRMAs are performed with respect to a deliberately circumscribed research question or objective involving, at the very least, characteristics of patients, interventions, comparator (if applicable) and outcomes. For me, this implies that SRMAs are domain-specific (and perhaps more in-depth than in the level of domain) by nature. The meta-analytic procedures you intend to apply are only as good as the rationale and assumptions, explicitly or implicitly, you are going to utilize them for.
In theory, you could do a meta-analysis on pretty much anything. The concern, though, is that the effect sizes that you will aggregate in a meta-analysis that is not well-defined or adequately constrained won't be very meaningful. This is not a problem to be solved by the methods (meta-analysis) but a reflection to have about what hypothesis you want to test or research question you want to answer.
David Moreau - The hypothesis is to examine existence of a phenomenon, since mixed findings exists in the literature. However, the phenomenon has been tested in various contexts, majorly in pro-environment behaviours. I am confused do I need to include studies testing the said phenomenon only in pro-environmental context or I can also include studies in other contexts like health, organisation behaviour, sustainable consumption etc. If you could through some light or refer some studies, it would be helpful. Thanks in advance
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis