Hi,
Thanks for the reply. After ANOVA is realized that i still need to find
individual contrasts between the 4 conditions. for that purpose i would need
to do an extension of the tripled group difference right? so just to
summarize, all 12 subjects were scanned under 5 different conditions(i had
mentioned 4 conditions before but i have an additional condition that i did
not apply in the ANOVA). To get a pentuple paired difference would the
following be the correct model specification
(2 subjects for simplicity)
1 1 1 1 1 0
1 1 1 1 0 1
-1 0 0 0 1 0
-1 0 0 0 0 1
0 -1 0 0 1 0
0 -1 0 0 0 1
0 0 -1 0 1 0
0 0 -1 0 0 1
0 0 0 -1 1 0
0 0 0 -1 0 1
A=a+b+c+d
B=-a
C=-b
D=-c
E=-d
A-B = 2a+b+c+d [2 1 1 1 0..]
B-A = [-2 -1 -1 -1 0..]
and so on..
is this correct? In contrast to this type of test , could we do a separate
2nd level paired t-test analysis for finding differences between conditions?
I know that it is inconvenient and that it leads to a multiple comparisons
issue but how would the pentuple paired-difference test avoid the latter.
-vish
|