Dear Colleagues,
Can you suggest some references and also the acceptable number of articles to conduct a Meta-Analysis?
Dear Adnan;
Meta-analysis can be done with 2 studies (Least acceptable number), in this case they can be called "Pooled analysis". But, the more studies you have, the higher validity and reliability your meta obtains.
Kind Regards
Dear Adnan;
Meta-analysis can be done with 2 studies (Least acceptable number), in this case they can be called "Pooled analysis". But, the more studies you have, the higher validity and reliability your meta obtains.
Kind Regards
That is right. I have conducted an analysis with 3 on FFP effectiveness in OP poisoning. Some outcomes were analyzed with two.
As Abushoul and Mirfazaelian have said, the minimum number of studies would be 2.
However, this should not be your goal. The purpose of a meta-analysis is to provide a numerical estimate to an effect in the literature. Therefore, the analysis should contain every study available. I would recommend that you do a thorough review of the literature and include every study you can find, published or unpublished (remember to use selection criteria).
I would only set a maximum number of studies if you want just a sample of studies, when a huge number exists. However, I would advise that you do not do this if it can be avoided, as the meta-analysis would not be comprehensive if you did.
The "acceptable number of articles" is therefore every article available on the phenomenon investigated.
- Badges
- Science topic
- Similar topics
- Composition Studies
- Articles