0
$\begingroup$

I performed generalized linear mixed models to test the effects of a treatment on two different groups (I performed a model for each group). Now I would like to test which of the groups is most affected by the treatment, that is to compare the magnitude of the response. I know how to test for difference in means, but now how to test for a difference in a magnitude. Could someone give me a clue?

$\endgroup$
0
$\begingroup$

Are you using group as a predictor in your model? If so, the coefficient tells you the difference in magnitude. You could bootstrap CIs for the coefficient to get your answer. A little bit more information on what/how you are doing the analysis would be helpful.

$\endgroup$
  • $\begingroup$ Thank you very much. I thnink your answer give me some hint about what I should do. I performed two separate models, one for each group and got a coefficient for each group. But I thinl I should put both groups within the same model, using a dummy variable and test for a significant interactions among this doomy variable and my treatment. I am not proficient with bootsrtapping. Would you have an example? $\endgroup$ – user34771 May 7 '14 at 20:10
  • $\begingroup$ It would be helpful if you described how you are doing this. $\endgroup$ – Lost in transcription May 8 '14 at 16:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.