by darryl » Thu May 24, 2012 6:24 am
In theory, yes. Depends exactly what you mean by 'test', but essentially when you create you data file you input a column of site covariate X, a column for site Y, then a third column for the interaction X*Y which is literally just the 2 sets of values multiplied together (easily done in a spreadsheet). You then set up your design matrices accordingly for what model you want to fit to the data.
Cheers
Darryl