Dear forum friends, I have a question to ask: When conducting a meta-analysis and selecting studies, if the parameters (beta coefficients and their standard errors) are estimated using diverse methods across studies—such as GMM, 2SLS, OLS, FE, and RE—can they all be reasonably converted into partial correlation coefficients (r) to synthesize a comparable effect size?
-
Login or Register
- Log in with
