Good day everyone. I hope you are all staying healthy during these rough times.
I reach out to you, because I would like to know if its possible to do a meta-analysis of some papers that have done a Structural Equation Modeling, based on a proposed model/framework of theirs. The latent variable is similar in all papers, but the observable variables may vary from study to study.
In case its possible to do such thing, what would be the minimum data I would require in order to do so?
Any help regarding this topic or hints on where to look is highly appreciated.
thanks in advance.
Dear Jose,
You can use the following Peterson and Brown (2005)'s formula to convert the standardized beta coefficient that is reported in papers that used SEM to correlations and use them in your meta-analysis. You can find the formula in this paper: Peterson, R. A., & Brown, S. P. (2005). On the use of beta coefficients in meta-analysis. The Journal of Applied Psychology, 90(1), 175–181. doi:10.1037/0021-9010.90.1.175 .
Just be careful that his method has been criticized by some recent studies.
Good luck with your research
Thanks Ahmad Aljarah , I will check it out and keep note of your warning.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis