Ciao Eugenio, thank you very much for your prompt and kind response.Yes, I'm looking average over the entire support of the path distribution and their differences among 2 pathological and 1 control groups.I haven't a strong hypothesis.When I performed MANOVA, Tukey or Bonferroni post hoc find differences but p for Wilk's lambda is not significant.If I understand your suggestion, I should performed a simple t-test for each tract.Bur, for Bonferroni's correction, I should consider: A. 0.05/18 tracts X 3 groups (I suspected that it should be too hard)orB. 0.05/18 tractsor more simplyC. 0.05/3 groups Thanks,
Stefano
----Messaggio originale---- Da: iglesias@nmr.mgh.harvard.edu Data: 24-mar-2014 22.11 A: stdp82@virgilio.it, "Freesurfer support list"freesurfer@nmr.mgh.harvard.edu Ogg: Re: [Freesurfer] TRACULA statistics concern
Ciao Stefano, can you be a bit more specific about what you're exactly trying to to? Am I correctly assuming that you are looking at differences between your groups? Are you looking at average FA/MD/RD/AD over whole tracts, or at values along them? Do you have any strong hypothesis on which tract will give you differences? In any case, people often look at the tracts of interest one by one (which a simple test, e.g., t-test) and then apply Bonferroni correction. Kind regards, /Eugenio
On Mon, 2014-03-24 at 21:39 +0100, stdp82@virgilio.it wrote:
Hi list,
I'm performing TRACULA on three groups for all 18 tracts and for FA, MD, RD, AD.
Which is the best way to perform statistic on TRACULA outcomes? I think that MANCOVA is too hard and it delete interesting and consolidate (in literature) results.
Thanks,
Stefano _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
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.
Hi again, if you perform the same analysis over and over for each of the 18 tracts, then the Bonferroni corrected p-value would be 0.05/18 Cheers, /Eugenio
On Mon, 2014-03-24 at 22:47 +0100, stdp82@virgilio.it wrote:
Ciao Eugenio,
thank you very much for your prompt and kind response. Yes, I'm looking average over the entire support of the path distribution and their differences among 2 pathological and 1 control groups. I haven't a strong hypothesis. When I performed MANOVA, Tukey or Bonferroni post hoc find differences but p for Wilk's lambda is not significant. If I understand your suggestion, I should performed a simple t-test for each tract. Bur, for Bonferroni's correction, I should consider:
A. 0.05/18 tracts X 3 groups (I suspected that it should be too hard) or B. 0.05/18 tracts or more simply C. 0.05/3 groups
Thanks,
Stefano
----Messaggio originale---- Da: iglesias@nmr.mgh.harvard.edu Data: 24-mar-2014 22.11 A: stdp82@virgilio.it, "Freesurfer support list"freesurfer@nmr.mgh.harvard.edu Ogg: Re: [Freesurfer] TRACULA statistics concern
Ciao Stefano, can you be a bit more specific about what you're exactly trying to to? Am I correctly assuming that you are looking at differences between your groups? Are you looking at average FA/MD/RD/AD over whole tracts, or at values along them? Do you have any strong hypothesis on which tract will give you differences? In any case, people often look at the tracts of interest one by one (which a simple test, e.g., t-test) and then apply Bonferroni correction. Kind regards, /Eugenio
On Mon, 2014-03-24 at 21:39 +0100, stdp82@virgilio.it wrote:
Hi list,
I'm performing TRACULA on three groups for all 18 tracts and for FA, MD, RD, AD.
Which is the best way to perform statistic on TRACULA outcomes? I think that MANCOVA is too hard and it delete interesting and consolidate (in literature) results.
Thanks,
Stefano _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
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.
Unless you're comparing groups in a pairwise manner, i.e., A vs B, B vs C, and A vs C. In that case, you would have 18 x 3 comparisons and the threshold for significance would be 0.05 / (18 x 3) Cheers, /Eugenio
On Mon, 2014-03-24 at 17:50 -0400, Juan Eugenio Iglesias wrote:
Hi again, if you perform the same analysis over and over for each of the 18 tracts, then the Bonferroni corrected p-value would be 0.05/18 Cheers, /Eugenio
On Mon, 2014-03-24 at 22:47 +0100, stdp82@virgilio.it wrote:
Ciao Eugenio,
thank you very much for your prompt and kind response. Yes, I'm looking average over the entire support of the path distribution and their differences among 2 pathological and 1 control groups. I haven't a strong hypothesis. When I performed MANOVA, Tukey or Bonferroni post hoc find differences but p for Wilk's lambda is not significant. If I understand your suggestion, I should performed a simple t-test for each tract. Bur, for Bonferroni's correction, I should consider:
A. 0.05/18 tracts X 3 groups (I suspected that it should be too hard) or B. 0.05/18 tracts or more simply C. 0.05/3 groups
Thanks,
Stefano
----Messaggio originale---- Da: iglesias@nmr.mgh.harvard.edu Data: 24-mar-2014 22.11 A: stdp82@virgilio.it, "Freesurfer support list"freesurfer@nmr.mgh.harvard.edu Ogg: Re: [Freesurfer] TRACULA statistics concern
Ciao Stefano, can you be a bit more specific about what you're exactly trying to to? Am I correctly assuming that you are looking at differences between your groups? Are you looking at average FA/MD/RD/AD over whole tracts, or at values along them? Do you have any strong hypothesis on which tract will give you differences? In any case, people often look at the tracts of interest one by one (which a simple test, e.g., t-test) and then apply Bonferroni correction. Kind regards, /Eugenio
On Mon, 2014-03-24 at 21:39 +0100, stdp82@virgilio.it wrote:
Hi list,
I'm performing TRACULA on three groups for all 18 tracts and for FA, MD, RD, AD.
Which is the best way to perform statistic on TRACULA outcomes? I think that MANCOVA is too hard and it delete interesting and consolidate (in literature) results.
Thanks,
Stefano _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
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
freesurfer@nmr.mgh.harvard.edu
-
Juan Eugenio Iglesias -
stdp82＠virgilio.it