Hello Douglas,
I did as you suggested and I got the results, but I still have difficulty understanding how exactly the choice of Design Matrix Type influences the analysis, so I have several questions:
1. When I look at thickness difference between two groups and choose age as nuisance parameter, doesn't it mean that calculation take into account the difference in slopes already (as I correct for age)? So therefore, when I also choose DODS or DOSS I do kind of double correction (matrix type + nuisance), don't I? 2. When I choose e.g. DODS and do the analysis, can I see the slopes (or the values of slopes) for two groups somehow? In which file this output is? 3. Can you recommend me anything to read regarding mathematics of how the analysis is done with either matrix type? 4. Just in case you know: how is this step implemented in SPM? It also uses glm, but it doesn't ask for the DODS vs DOSS choice.
Thank you, I appreciate you time and help, Tanja.
On Wed, Aug 17, 2011 at 6:19 PM, Tetiana Dadakova tetiana.d@gmail.com wrote:
Thank you, Douglas.
On Wed, Aug 17, 2011 at 5:48 PM, Douglas N Greve greve@nmr.mgh.harvard.edu wrote:
Hi Tanja, you should use DODS and test for a difference between the age slopes of the groups (ie, an interaction between group and age). If there are no significant interactions, you should then use DOSS to test for the difference between thicknesses. doug
Tetiana Dadakova wrote:
Dear all,
I have two groups of subjects, and I want to see a difference in thickness between them controlling for age. I couldn't find any information in the literature on how the subject's brains in my groups develop with age. Therefore I can't tell whether thickness as a function of age for both groups has the same slope or different.
My question is if in this case I should use DODS?
Thank you, Tanja. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
For question #3; #4.
#3 Any good stastistical textbook will be useful. The two that come to mind are: (a) Meyers and Wells. Research Design and Statistical A nalysis (b) Keppel and Wickens. Design and Analysis: A Researcher's Handbook (4th Edition)
#4
SPM uses the General Linear Model for all of its second level models. The DODS and DOSS are terms used in the freesurfer package and not in other packages; however, the designs can easily be generated in SPM.
DODS -- Design matrix with different slopes per group Select two-sample t-test (or flexible factorial) Specify your two groups of scans Specify the covariate values, names, and select interaction with factor 1
You can ask if the slopes are different
DOSS -- Design matrix with same slopes across all groups Select two-sample t-test (or flexible factorial) Specify your two groups of scans Specify the covariate values, names, and select interaction none.
The slopes are modelled as a single regressor, so you can't ask about the difference in slopes between groups.
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 Mon, Aug 22, 2011 at 8:46 AM, Tetiana Dadakova tetiana.d@gmail.comwrote:
Hello Douglas,
I did as you suggested and I got the results, but I still have difficulty understanding how exactly the choice of Design Matrix Type influences the analysis, so I have several questions:
- When I look at thickness difference between two groups and choose
age as nuisance parameter, doesn't it mean that calculation take into account the difference in slopes already (as I correct for age)? So therefore, when I also choose DODS or DOSS I do kind of double correction (matrix type + nuisance), don't I? 2. When I choose e.g. DODS and do the analysis, can I see the slopes (or the values of slopes) for two groups somehow? In which file this output is? 3. Can you recommend me anything to read regarding mathematics of how the analysis is done with either matrix type? 4. Just in case you know: how is this step implemented in SPM? It also uses glm, but it doesn't ask for the DODS vs DOSS choice.
Thank you, I appreciate you time and help, Tanja.
On Wed, Aug 17, 2011 at 6:19 PM, Tetiana Dadakova tetiana.d@gmail.com wrote:
Thank you, Douglas.
On Wed, Aug 17, 2011 at 5:48 PM, Douglas N Greve greve@nmr.mgh.harvard.edu wrote:
Hi Tanja, you should use DODS and test for a difference between the age slopes of the groups (ie, an interaction between group and age). If
there
are no significant interactions, you should then use DOSS to test for
the
difference between thicknesses. doug
Tetiana Dadakova wrote:
Dear all,
I have two groups of subjects, and I want to see a difference in thickness between them controlling for age. I couldn't find any information in the literature on how the subject's brains in my groups develop with age. Therefore I can't tell whether thickness as a function of age for both groups has the same slope or different.
My question is if in this case I should use DODS?
Thank you, Tanja. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
The information in this e-mail is intended only for the person to whom
it is
addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you
in
error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Tetiana Dadakova wrote:
Hello Douglas,
I did as you suggested and I got the results, but I still have difficulty understanding how exactly the choice of Design Matrix Type influences the analysis, so I have several questions:
- When I look at thickness difference between two groups and choose
age as nuisance parameter, doesn't it mean that calculation take into account the difference in slopes already (as I correct for age)? So therefore, when I also choose DODS or DOSS I do kind of double correction (matrix type + nuisance), don't I?
If there is a difference between the slopes, then it is very hard to interpret a difference between the thicknesses (some statisticians say that you should not even try). The DOSS will correct for age assuming that the age slopes are the same.
- When I choose e.g. DODS and do the analysis, can I see the slopes
(or the values of slopes) for two groups somehow? In which file this output is?
If you have two groups and DODS, then the slopes will be the 3rd and 4th frame in the beta.mgz file.
- Can you recommend me anything to read regarding mathematics of how
the analysis is done with either matrix type?
Donald Mclaren gave some good refs.
- Just in case you know: how is this step implemented in SPM? It also
uses glm, but it doesn't ask for the DODS vs DOSS choice.
As Donald points out, DODS and DOSS are FS-specific names, but you can implement them in SPM. Sorry, I don't have enough SPM experience to tell you how. doug
Thank you, I appreciate you time and help, Tanja.
On Wed, Aug 17, 2011 at 6:19 PM, Tetiana Dadakova tetiana.d@gmail.com wrote:
Thank you, Douglas.
On Wed, Aug 17, 2011 at 5:48 PM, Douglas N Greve greve@nmr.mgh.harvard.edu wrote:
Hi Tanja, you should use DODS and test for a difference between the age slopes of the groups (ie, an interaction between group and age). If there are no significant interactions, you should then use DOSS to test for the difference between thicknesses. doug
Tetiana Dadakova wrote:
Dear all,
I have two groups of subjects, and I want to see a difference in thickness between them controlling for age. I couldn't find any information in the literature on how the subject's brains in my groups develop with age. Therefore I can't tell whether thickness as a function of age for both groups has the same slope or different.
My question is if in this case I should use DODS?
Thank you, Tanja. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
freesurfer@nmr.mgh.harvard.edu