Or, just to throw out a different terminology that you might encounter,
the tests are being performed using "Type III" sums-of-squares.
cheers,
-MH
On Wed, 2010-12-22 at 11:36 -0500, MCLAREN, Donald wrote:
> To specifically answer the last part of your question, the statistical
> test applied is testing whether or not its significant after
> controlling for the null-space of the contrast (e.g. where the
> contrast is 0). So if you test the first, it controls for the 2nd and
> 3rd. If you test the 2nd, it controls for the 1st and 3rd.
>
>
> Best Regards, Donald McLaren
> =================
> D.G. McLaren, Ph.D.
> Postdoctoral Research Fellow, GRECC, Bedford VA
> Research Fellow, Department of Neurology, Massachusetts General
> Hospital and Harvard Medical School
> 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 Wed, Dec 22, 2010 at 11:33 AM, MCLAREN, Donald
> <[log in to unmask]> wrote:
> Stasa,
>
> (1) multivariate regression is a regression analysis with
> multiple dependent variables, not multiple independent
> variables. whne you have multiple independent variables, the
> model is either a multiple regression or ANCOVA.
>
> (2) Since you don't have any grouping variables, your example
> is a multiple regression. If you added groups, then you'd have
> a ANCOVA model.
>
> (3) The order of the independent variables does not matter.
> Just make sure you know the order you entered so that you know
> what you are testing. The statistical tests that you run are
> the same as adding the term last in a stepwise regression
> model. The equation for the estimation of the betas is: pinv
> (X)*Y. Where X is your independent variable matrix and Y is
> your dependent variable.
>
> Hope this helps.
>
> Best Regards, Donald McLaren
> =================
> D.G. McLaren, Ph.D.
> Postdoctoral Research Fellow, GRECC, Bedford VA
> Research Fellow, Department of Neurology, Massachusetts
> General Hospital and Harvard Medical School
> 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 Wed, Dec 22, 2010 at 10:43 AM, Stasa Tadic <[log in to unmask]>
> wrote:
> Hi all,
> I would like to analyze relation between two
> variables/measures (fMRI data as dependent variable,
> white matter hyperintensities counts as independent
> variable) while controling for other variables like
> age or cognitive function (or both). I am using SPM 5.
> If I use multivariate regression analysis in SPM5,
> does the way of ordering the variables assume that the
> first will be inferred and others (second or third )
> will be fixed/controled for...
> Thanks,
> Stasa
>
>
>
|