Hi, I am testing whether the proportions are different among three samples. For example, in sample 1, there are n1 participants with feature A, and n2 with feature B; in sample 2, there are n3 participants with feature A, and n4 with feature B; in sample 3, there are n5 participants with feature A, and n6 with feature B. I want to test whether the proportions of A (each participant has features either A or B, and no other features) are the same among the three samples.
Normally, chi-square test can be used to do the test. However, there is one assumption which should be met before using this chi-square test for this purpose--the three samples are independent. However, somehow, in my study, the three samples are not independent. I am wondering if anyone knows any other tests that I can use as an alternative to chi-square test when this assumption is violated? Many thanks
- Badges
- Science topic