So I am running an experiment where I am interested in the proportion of time spent doing certain behaviors under different treatments.

I know MANOVA can be used for this. I have 3 behavior times (Dependent variables) and two treatments (independent variables). I am also curious if chi-squared analysis can also be used? I would have a 3x2 table with and in each cell the proportion of time spent doing behavior X. I would then test to see if the proportions are equivalent.

Is one test better than the other? Is Chi-squared not even an option?

Thank you.