Dear Maria,
I agree with Donald that including these is reasonable and will likely
reduce the error (and help your model fit better). I would add that if
the covariates significantly differed across group, the interpretation
would be trickier, but it sounds like you are safe.
A couple of articles which may be useful in this discussion:
Barnes J, Ridgway GR, Bartlett J, Henley SMD, Lehmann M, Hobbs N,
Clarkson MJ, MacManus DG, Ourselin S, Fox NC (2010) Head size, age and
gender adjustment in MRI studies: a necessary nuisance? NeuroImage
53:1244-1255.
Miller GA, Chapman JP (2001) Misunderstanding analysis of covariance.
Journal of Abnormal Psychology 110:40-48.
and although you didn't mention total gray matter, this issue is often
raised as well, so I also mention:
Peelle JE, Cusack R, Henson RNA (2012) Adjusting for global effects in
voxel-based morphometry: Gray matter decline in normal aging.
NeuroImage 60:1503-1516.
Hope this helps!
Best regards,
Jonathan
--
Jonathan Peelle, PhD
Department of Otolaryngology
Washington University in St. Louis
Office: (314) 362-9044
http://peellelab.org || http://jonathanpeelle.net
On Tue, Nov 13, 2012 at 8:37 AM, MCLAREN, Donald
<[log in to unmask]> wrote:
> Any time that you have a covariate with a large range and that
> covariate is associated with your DV, it should be included. It will
> reduce the residual error of the model and increase the significance.
>
> The key question is whether the covariate has the same relationship
> with the DV in each group. If the DV/covariate relationship is not
> significantly different between the groups, then you only need a
> single column for each covariate. If the relationship is different,
> then you need to model each covariate as N columns for N groups. See
> mumford.fmripower.org/mean_centering/ for more details.
>
> In brief, mean centering across everyone leads to covariate-adjusted
> means (e.g. what are the group means if each group had the same
> covariate value); mean centering within each group leads to the same
> group means, but you are controlling for the covariates; not mean
> centering leads to the group terms being the y-intercepts.
>
> Hope this helps.
>
> Best Regards, Donald McLaren
> =================
> D.G. McLaren, Ph.D.
> Research Fellow, Department of Neurology, Massachusetts General Hospital and
> Harvard Medical School
> Postdoctoral Research Fellow, GRECC, Bedford VA
> Website: http://www.martinos.org/~mclaren
> Office: (773) 406-2464
> =====================
> This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED
> HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is
> intended only for the use of the individual or entity named above. If the
> reader of the e-mail is not the intended recipient or the employee or agent
> responsible for delivering it to the intended recipient, you are hereby
> notified that you are in possession of confidential and privileged
> information. Any unauthorized use, disclosure, copying or the taking of any
> action in reliance on the contents of this information is strictly
> prohibited and may be unlawful. If you have received this e-mail
> unintentionally, please immediately notify the sender via telephone at (773)
> 406-2464 or email.
>
>
> On Tue, Nov 13, 2012 at 8:45 AM, Maria Serra <[log in to unmask]> wrote:
>> Dear SPM experts,
>>
>> we have performed a VBM analysis with 4 groups (3 with different grade of
>> pathology and 1 healthy control group). We carried out an ANOVA between 4
>> groups with no covariates given that there were no significant differences
>> between groups regarding variables which could affect gray matter volume
>> (TIV, Age, Gender, Schooling..).
>>
>> Even though, reviewers have suggested us to include TIV and Age as
>> covariates, would it be correct? in case of including them, would they act
>> as a confounding agent, introducing noise to the analysis?
>>
>> Thank you in advance for the help.
>>
>>
>> --
>> Maria Serra
>>
>> PIC (Port d'Informació Científica)
>> Campus UAB, Edifici D
>> E-08193 Bellaterra, Barcelona
>> Telf. +34 93 586 8232
>> www.pic.es
>>
>> __
>>
>> Please consider the environment before printing this e-mail
>>
>>
>>
|