0

Using emmeans, I have already coded for the difference and significance in means between: White Christian (WC) Men and Black Christian (BC) men, and then White Muslim (WM) men and Black Muslim (BM) men, for a certain stereotype Dimension. Here is my code for it:

pairwise_emm2 <- emmeans(fitdata3, pairwise ~ Religion * Race | Dimension, type = "response") pairwise_emm2$contrasts
I now want to code for the differences between these differences. Meaning I want to know if the WC-BC difference is greater than the WM-BM difference, and if it is significant. Can someone please help me figure out the code for that? Thank you!

Phil
  • 7,287
  • 3
  • 36
  • 66
Ifeoma Okoli
  • 1
  • This usually would involve bootstrapping where you estimate your outcome metric e.g. 1000 times based on different sub samples of your data. So I suggest you you could search a bit in that direction (bootstrapping in R). – deschen Feb 23 '21 at 20:19
  • This might be a better fit for [stats.se] since you seem to be seeking statistical advice for how to model this particular parameter. Otherwise you should provide a [reproducible example](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) with sample input and desired output here if this is really just a programming question. – MrFlick Feb 23 '21 at 20:58

0 Answers0