Hi,
I'm trying to run a cortical thickness group comparison (HD v Controls), controlling for site, gender (1 or 2) and age. Without age as a covariate the results are as expected but once age is added on all significant results disappear. I think this is a problem
with the method rather than because all results are dependent on age. My fsgd and contrast files for the glm are below. I would really appreciate your input as to what might be going wrong to effect the results so strongly. Am I doing something wrong?
GroupDescriptorFile 1
Title HDvControl
Class HD-london-1
Class HD-leiden-1
Class HD-paris-1
Class HD-ulm-1
Class HD-london-2
Class HD-leiden-2
Class HD-paris-2
Class HD-ulm-2
Class C-london-1
Class C-leiden-1
Class C-paris-1
Class C-ulm-1
Class C-london-2
Class C-leiden-2
Class C-paris-2
Class C-ulm-2
Variables Age
Input 30009 HD-paris-2 26
Input 30014 C-ulm-2 51
Input 30016 C-london-1 48
Input 30024 C-leiden-1 38
Input 30029 HD-london-2 66
Input 30030 HD-london-1 54
Input 30037 HD-london-2 42
Input 30041 C-london-1 43
Input 30042 HD-london-2 40
Input 30043 HD-london-2 41
Input 30047 HD-london-2 66
Input 30051 C-london-2 45
Input 30054 HD-london-2 55
Input 30056 C-london-2 66
Input 30060 C-london-1 50
Input 30068 C-london-1 51
Input 30071 HD-london-1 43
Input 30072 HD-london-2 44
Input 30074 C-london-2 58
Input 30075 HD-london-2 48
Input 30078 C-london-2 53
Input 30079 HD-london-1 58
Input 30082 C-london-2 62
Input 30089 HD-london-2 54
Input 30093 HD-leiden-1 44
Input 30098 HD-leiden-2 45
Input 30100 HD-leiden-2 42
Input 30102 C-leiden-2 48
Input 30107 HD-leiden-2 61
Input 30112 HD-leiden-2 50
Input 30114 C-leiden-1 48
Input 30115 HD-leiden-2 47
Input 30117 HD-leiden-2 49
Input 30118 C-leiden-1 51
Input 30119 HD-leiden-2 59
Input 30124 C-leiden-1 57
Input 30125 HD-leiden-2 39
Input 30136 C-leiden-2 51
Input 30138 C-leiden-2 56
Input 30139 HD-leiden-2 33
Input 30142 HD-leiden-2 32
Input 30148 HD-leiden-1 48
Input 30154 C-leiden-1 47
Input 30155 HD-leiden-2 46
Input 30160 HD-leiden-2 38
Input 30162 HD-leiden-2 52
Input 30163 C-leiden-2 42
Input 30168 C-ulm-2 58
Input 30177 C-london-2 49
Input 30185 C-ulm-2 34
Input 30189 HD-ulm-1 50
Input 30191 C-ulm-1 44
Input 30193 HD-ulm-1 46
Input 30195 C-ulm-2 50
Input 30198 C-ulm-1 56
Input 30199 HD-ulm-1 53
Input 30208 HD-ulm-1 59
Input 30209 C-ulm-2 44
Input 30210 HD-ulm-1 30
Input 30214 HD-ulm-2 57
Input 30215 HD-ulm-1 34
Input 30220 HD-ulm-2 54
Input 30222 HD-ulm-1 64
Input 30223 C-ulm-1 44
Input 30229 HD-ulm-1 40
Input 30235 HD-ulm-1 48
Input 30240 HD-ulm-1 36
Input 30241 C-paris-2 65
Input 30247 HD-paris-1 24
Input 30248 HD-paris-1 44
Input 30250 HD-paris-2 62
Input 30254 HD-paris-1 37
Input 30256 HD-paris-1 23
Input 30259 HD-paris-1 64
Input 30265 C-paris-2 50
Input 30269 C-paris-2 58
Input 30273 C-paris-2 49
Input 30275 C-paris-1 59
Input 30280 HD-london-2 42
Input 30281 HD-london-2 65
Input 30288 HD-london-2 51
Input 30311 HD-paris-1 40
Input 30312 HD-paris-2 51
Input 30314 HD-paris-2 46
Input 30315 HD-leiden-2 67
Input 30318 C-paris-2 64
Input 30320 C-paris-1 49
Input 30321 HD-paris-1 61
Input 30326 C-paris-2 45
Input 30334 HD-london-1 49
Input 30347 C-paris-1 62
Input 30349 HD-paris-2 48
Input 30352 C-paris-2 66
Input 30357 C-ulm-1 28
Input 30361 HD-paris-2 63
Input 30363 HD-ulm-1 51
Input 30365 C-ulm-2 46
Input 30369 HD-ulm-2 45
Input 30378 C-leiden-1 53
Input 30385 HD-leiden-2 61
Input 30392 HD-ulm-2 51
Contrast:
1 1 1 1 1 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Thanks,
Elin