There are four different but complementary approaches to conducting meta-analysis research. First, you should identify all articles related to the particular topic you want to investigate. Here, you can make use of pertinent data bases and search engines available on the issue(e.g., Cochrane collaboration, EMBASE, etc.). Second, After finding a large number of articles, you should decide on certain inclusion or eligibility criteria for choosing the studies which most appropriately satisfy your cause. Third, you should abstract the relevant data from each study. Here, you should use several reviewers to avoid possible errors in the interpretation of the targeted data. Finally, in the analysis stage, you should exp[lain the degrees of between and within group variances (i.e., homogeneity vs. heterogeneity) in the existing data reported by the reviewers by employing proper statistical techniques. Good luck with your research.
I answer according to my field of competence: the bibliographic retrieval. It is necessary to make a rigorous review of the scientific literature, published or not.Construct a search strategy with, on the one hand, relevant keywords appropriate to your subject and on the other hand on the bibliographic tools used (for example, the use of MeSH Terms to query MEDLINE).Depending on your theme, different databases should be used (PubMed/MEDLINE, Scopus, Web of Science, Embase, Cochrane, etc...)You should not forget the "gray" literature, the relevant but unpublished documents (reports, theses, dissertations, etc.) and the abstracts of congresses.You can already at this stage establish criteria for strictly bibliographic selection (or exclusion), such as titles of journals, the presence of structured abstracts, notoriety of research teams, access to documents, etc.,Work in a rigorous way, with several persons, researchers, and with a good software of bibliographic management ...