0

I am conducting a meta-analysis of individual studies that report probabilities and their standard error as an effect size. I plan to convert the probabilities and standard errors to the natural logarithm of the odds, however, I am not sure how to convert the standard errors of the probabilities into SEs of the natural logarithm of the odds.

Is there a formula to convert these? Better yet, is there a way to do this in R-studio?

Daniel
  • 1
  • mayne this helps: https://stats.stackexchange.com/questions/158481/how-to-convert-the-standard-error-of-the-log-odds-ratio-to-the-odds-ratio-standa – user380392 Feb 20 '23 at 11:24

1 Answers1

0

Best to convert SE to n given that SE²=pq/n Then run a meta-analysis of np (cases) and n using MetaXL using the Freeman-Tukey transformed proportion. Also see these papers: FTT paper and JECH paper

Suhail
  • 3