Hi Heather,
According to the literature, FA changes with age. Thus, you should find age and gender match controls. You age mean is so different between patient and control group. It is not an ideal situation to begin with. I would prefer age- and gender- match controls. Thus, you need to regress your age effect out (classic glm explaination). Sometime it can still be hard to tease all the age effect out.
Let's see a simple example.
Controls: 8 5 6 7 9 Patients: 11 9 11 14 15
mean age for controls: 7 mean age for patients: 12
significantly different.
We want to know the difference between the two groups regarding measure A. For instance: measure A usually increases with age for controls.
Measure A for controls: 30 15 20 25 35 for patients: 50 35 50 55 60
Significantly different without controlling for age. Well, are the differences caused by group effect (patients v.s. controls) or purely age itself? We can not know if we only use a simple two sample t-test.
Well, let's put age as a covariate. Bingo, group difference is not significant anymore. From this, you can see use of age as a covariate is necessary in your case. Detect the real differences between your conditions.
Happy Holiday!
-Yingying
Hi Heather :
If there is a relationship between age and and your dependent variable (e.g. FA) then you should definitely include age as a covariate even if age is matched between the groups. This is because age is explaining a lot of the variance associated with your dependent variable. The best way to confirm this is to simulate this situation with a non-imaging dataset from everyday life (e.g. children's test scores and SES). Then you can run the GLM model with and without the covariate to convince yourself. It happens that TBSS is non-parametric (whereas GLM is parametric) but I think the priniciple still applies.
Raj
Duke University
Thanks, Yingying.
The two groups are controls and patients and we are interested in diffusion metrics associated with the condition and not due to developmental changes (age).
What does demeaning age within groups (for controls and patients separately) tell us and what would demeaning all subjects (patients and controls) with the grand mean tell us?
Many thanks,
--Heather
On Tue, Dec 20, 2011 at 12:00 PM, Yingying Wang
<[log in to unmask]> wrote:
Hi Heather,
According to your message, you might want to do the two group separately. Also it depends on your hypothesis.
For example: two group: controls and patients
What's your interest?
1) Use age as covarite (nuisance variable) if you're not interested in developmental changes.
Best, Yingying
>>> Heather Collins <
[log in to unmask]> 12/20/2011 9:50 AM >>>
Dear FSL experts,
I am running TBSS on 2 groups of subjects. The mean ages are not different between the groups. However, when any diffusion metric is plotted over age, the groups show different slopes: one increases with age whereas the other decreases with age. (It looks like an X)
I have combed through the archives and cannot find information about this kind of situation.
1) Should age be used as a covarite in this situation?
2) What does it mean for age to be demeaned within each group versus demeaning both groups using the grand mean? Which should I use in this situation?
Thank you in advance for your time and help!
Cheers,
--Heather
