0
$\begingroup$

I do not even know if such a test exists, although, ideally, it does. I want to run a comparison of proportions test as seen here, that adjusts for a 3 part categorical variable (either "A","B", or "C") as well. The link is just an example - what it basically does is run a chi-square test (I think) based on a ratio and sample sizes and determines the statistical significance of the difference between ratios in groups [comparison of proportions, as it is called]. What would be the proper statistical test to look into running to determine the statistical significance in the comparison of proportions that also takes into account that categorical variable?

Thanks. I hope this isn't a terrible/obvious question.

$\endgroup$
1
  • $\begingroup$ Please add whatever information is necessary to understand & answer your question into the question itself. People aren't going to want to have to navigate elsewhere & read through something just so they can answer your question for you. Moreover, we want these threads to remain intelligible even after the link goes dead. A small dataset to illustrate your situation would also be helpful. $\endgroup$ Jun 30 '17 at 19:57
1
$\begingroup$

It sounds like what you want to do is stratify your sample based on your categorical variable and then compute a Mantel-Haenszel Odds Ratio.

$\endgroup$

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.