dear Stijn
as I told, i'm a beginner users. What i stated is the product of days of mailing list mining. I reported results and observations from several users. Some of them are partly unclear for me.
your first doubt is discussed in these two posts:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0906&L=FSL&D=0&P=346845
https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind1102&L=FSL&D=0&P=264039
point 4 contrasts are explained in the first link i wrote.
since [0 0 1 0] should express the positive correlation between age and group1 means
the [0 0 1 -1] contrast should assess differences between two groups correlation with age.
this issue is discussed here
https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind1005&L=FSL&D=0&P=89430
it adds that, to test for different age slopes between the groups, an F-test on either [0 0 1 -1] and [0 0 -1 1] should be performed.
hope that helps
cheers
Alberto
|