I am examining average maximum temperature data for a site in Western Australia that has relocated at least four times. Due to exposure problems, both temperature (y) and the covariate rainfall (x) are measured imprecisely; however, I expect x vs. y relationships to be significant, slopes to be homogeneous and I wish to test the NULL hypothesis that inter-site means are the same.
I'm wondering if there is a RMA procedure that is equivalent to categorical multiple linear regression?
I can of-course use one-way AOV but the fit (or lack of) with rainfall is also important in this case.
Thanking you in anticipation,
Bill Johnston