Close Home Forum Sign up / Log in

MANOVA Guru: Please Help!

J

I am comparing two groups on three levels and felt a MANOVA analysis would effectively analyze the data. One group has 40 participants and the other 180 participants due to the nature of the groups. This is unavoidable. Will the in-equivalent groups jeopardize the integrity of the study, or is there another test that would be better to evaluate the data?

T

Hello

Mixed effects models (otherwise known as multilevel models) or just simple regression would better analyze the data, as it can deal with unequal groups. You would treat group as a predictor with two levels (i.e., group 1, group 2).

But what do you mean by "on three levels"? Three dependent variables/outcomes (e.g., score 1, score 2, score 3), or one dependent variable/outcome with three levels (e.g., ill, healthy, deceased)? This will affect your choice of analysis.

Tudor

49682