I am conducting a meta-analysis to explore the factors contributing to between-study heterogeneity through meta-regression.

However, I have encountered a problem in my analysis. I have the following components:

  • A proportional covariate representing the male-to-all percentage, which is non-normally distributed.
  • The outcome variable, which is the Freeman-Tukey double arcsine transformed proportion. It may or may not follow a normal distribution.

Should I assess normality for these components and consider transforming them to achieve normality, if required?

Any advice or suggestions would be greatly appreciated. Thank you!

More Alireza Sadeghi's questions See All
Similar questions and discussions