And for the comparison to be appropriate, you further have to make sure that you are using the exact same main and interaction terms in SPSS as are being used in QDEC. The mere presence of interaction terms in a model can have big effects on the significance (or lack thereof) of the main effects. I believe that QDEC will by default include the interaction terms, while I suspect that perhaps these were not included in SPSS. Also, you need to match on "DODS" vs. "DOSS" as well.

cheers, -Mike H.

Depends on some of the details. Is spss demeaning the age? If so, are you passing qdec demeaned ages? It can also be that it shows up better in the ROI analysis (ROI analyses are often more powerful than map-based).

doug

Dankner, Nathan (NIH/NIMH) [F] wrote:

...

Hi all,

I've covaried age in a group difference analyses on cortical thickness data both in qdec for vertex-based and in spss for the gyral-based data. For the vertex-based data, covarying age has almost no effect, and if anything intensifies the group differences. However, when run in qdec, covarying age in this dataset wipes out almost any group differences. Is covarying in qdec to be trusted? It seems strange to me that these two methods would result in such extreme differences in results. Thanks,

Nathan

---

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

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer