Unbalanced ANCOVA

smed

New Member
#1
Hi all,

I have data in this format:
2 independent variables: 1 factor with 3 levels (3 different locations)
1 continuous IV (depths)
Dependent variable: Area of coral

1) I want to test for the effect of both depth and site and look for an interaction between the two, so I am thinking I should use ANCOVA, or should I group the depths into 15,30,60 and 90 (as they are clumped like that-see attachment) and do factorial ANOVA?
2) However, my data is unbalanced. I have 9 repeats for site a, 11 for b and 12 for c. I could make it balanced by using 9 repeats for all sites, but I would lose data. Can I analyse the unbalanced data, and if so, how? I've heard you can maybe use something called type 3 SS but I don't really understand how that applies to ANCOVA.
3) Finally, I have different numbers of repeats at each depth (see attachment), does this matter, or only if I'm doing factorial ANOVA?:confused:
Sorry for all the questions, any help would be much appreciated!
 
Last edited: