Dear statistic experts
We have done statistics between two independent groups using an independent t-test. One group consistent of 13 male patients and the other group consistent of 13 male control subjects who were matched to the patients with respect to age and education (in years).
The reviewer proposed that it might be more appropriate to use a paired t-test. However, we can imaging a design where such an approach is adequate e.g. monozygotic twins reared in shared environments who are discordant for a particular trait.
In our opinion, the matching of only age and education is not sufficient in order to apply a paired t-test. In addition, the paired t-test is more anti-conservative than the independent t-test and therefore advantages our hypothesis and this might be identified as a statistical "trick" by other reviewers.
Any comment about whether or not the suggested paired t-test is appropriate in our study is highly appreciated.
Thanks in advance Regards Jürgen Hänggi
---------------------------------------------------------------------------- Jürgen Hänggi, Ph.D. Division Neuropsychology Institute of Psychology University of Zurich Binzmuehlestrasse 14, PO Box 25 8050 Zurich, Switzerland 0041 44 635 73 97 (phone office) 0041 76 445 86 84 (phone mobile) 0041 44 635 74 09 (fax office) BIN 4.D.04 (office room number) j.haenggi[at]psychologie.uzh.ch (email) http://www.psychologie.uzh.ch/neuropsy/ (website) http://www.juergenhaenggi.ch (private website)
This e-mail (and any attachment/s) contains confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorised copying, disclosure or distribution of the material in this e-mail is strictly forbidden. ----------------------------------------------------------------------------
Hi Jurgen, I don't think it is too unusual to do this kind of thing, ie, a paired test where pairing is based on gender, age, and education, but you know your data and field better than I do. It is not the case that one is more conservative than another (they are the same) assuming that the underlying assumptions are met. The underlying assumptions are more stringent for the paired than for the unpaired. doug
On 07/24/2012 06:22 AM, Jürgen Hänggi wrote:
Dear statistic experts
We have done statistics between two independent groups using an independent t-test. One group consistent of 13 male patients and the other group consistent of 13 male control subjects who were matched to the patients with respect to age and education (in years).
The reviewer proposed that it might be more appropriate to use a paired t-test. However, we can imaging a design where such an approach is adequate e.g. monozygotic twins reared in shared environments who are discordant for a particular trait.
In our opinion, the matching of only age and education is not sufficient in order to apply a paired t-test. In addition, the paired t-test is more anti-conservative than the independent t-test and therefore advantages our hypothesis and this might be identified as a statistical "trick" by other reviewers.
Any comment about whether or not the suggested paired t-test is appropriate in our study is highly appreciated.
Thanks in advance Regards Jürgen Hänggi
Jürgen Hänggi, Ph.D. Division Neuropsychology Institute of Psychology University of Zurich Binzmuehlestrasse 14, PO Box 25 8050 Zurich, Switzerland 0041 44 635 73 97 (phone office) 0041 76 445 86 84 (phone mobile) 0041 44 635 74 09 (fax office) BIN 4.D.04 (office room number) j.haenggi[at]psychologie.uzh.ch (email) http://www.psychologie.uzh.ch/neuropsy/ (website) http://www.juergenhaenggi.ch (private website)
This e-mail (and any attachment/s) contains confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorised copying, disclosure or distribution of the material in this e-mail is strictly forbidden.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Jurgen,
I actually just came across a similar situation in which I tried to convince my PI that a paired t test is more appropriate. In my humble opinion, if you are applying an independent group test, than the matching design is irrelevant in this analysis because you're not taking any advantage that might accompany the matching. On the other hand, a paired t test lends more power if your subjects are sufficiently matched, while being more conservative when the matching is a failure (i.e. if you thick the two groups are actually idependent. In that case you'll get the same t value but a smaller df in a paired t analysis).
That being said, I suggest you consult a statistical advisor for the appropriate solution.
Cherry
On Tue, Jul 24, 2012 at 11:53 AM, Douglas N Greve <greve@nmr.mgh.harvard.edu
wrote:
Hi Jurgen, I don't think it is too unusual to do this kind of thing, ie, a paired test where pairing is based on gender, age, and education, but you know your data and field better than I do. It is not the case that one is more conservative than another (they are the same) assuming that the underlying assumptions are met. The underlying assumptions are more stringent for the paired than for the unpaired. doug
On 07/24/2012 06:22 AM, Jürgen Hänggi wrote:
Dear statistic experts
We have done statistics between two independent groups using an
independent
t-test. One group consistent of 13 male patients and the other group consistent of 13 male control subjects who were matched to the patients
with
respect to age and education (in years).
The reviewer proposed that it might be more appropriate to use a paired t-test. However, we can imaging a design where such an approach is adequate e.g. monozygotic twins reared in shared environments who are discordant for a particular trait.
In our opinion, the matching of only age and education is not sufficient
in
order to apply a paired t-test. In addition, the paired t-test is more anti-conservative than the independent t-test and therefore advantages our hypothesis and this
might be
identified as a statistical "trick" by other reviewers.
Any comment about whether or not the suggested paired t-test is
appropriate
in our study is highly appreciated.
Thanks in advance Regards Jürgen Hänggi
Jürgen Hänggi, Ph.D. Division Neuropsychology Institute of Psychology University of Zurich Binzmuehlestrasse 14, PO Box 25 8050 Zurich, Switzerland 0041 44 635 73 97 (phone office) 0041 76 445 86 84 (phone mobile) 0041 44 635 74 09 (fax office) BIN 4.D.04 (office room number) j.haenggi[at]psychologie.uzh.ch (email) http://www.psychologie.uzh.ch/neuropsy/ (website) http://www.juergenhaenggi.ch (private website)
This e-mail (and any attachment/s) contains confidential and/or
privileged
information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorised copying, disclosure or distribution of the material in this e-mail is strictly forbidden.
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
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.
A paired t-test is a special case of a repeated measures ANOVA with a single within-subject factor and two levels. You should use it when the measurements in one group are correlated with the measurements in the other group and there is a within-subject effect that persists across paired measurements.
Best -Jorge
De: Yizhou Ma ym850@nyu.edu Para: Douglas N Greve greve@nmr.mgh.harvard.edu CC: freesurfer@nmr.mgh.harvard.edu Enviado: Martes 24 de julio de 2012 13:05 Asunto: Re: [Freesurfer] Paired t-test with independent samples
Hi Jurgen,
I actually just came across a similar situation in which I tried to convince my PI that a paired t test is more appropriate. In my humble opinion, if you are applying an independent group test, than the matching design is irrelevant in this analysis because you're not taking any advantage that might accompany the matching. On the other hand, a paired t test lends more power if your subjects are sufficiently matched, while being more conservative when the matching is a failure (i.e. if you thick the two groups are actually idependent. In that case you'll get the same t value but a smaller df in a paired t analysis).
That being said, I suggest you consult a statistical advisor for the appropriate solution.
Cherry
On Tue, Jul 24, 2012 at 11:53 AM, Douglas N Greve greve@nmr.mgh.harvard.edu wrote:
Hi Jurgen, I don't think it is too unusual to do this kind of thing, ie,
a paired test where pairing is based on gender, age, and education, but you know your data and field better than I do. It is not the case that one is more conservative than another (they are the same) assuming that the underlying assumptions are met. The underlying assumptions are more stringent for the paired than for the unpaired. doug
On 07/24/2012 06:22 AM, Jürgen Hänggi wrote:
Dear statistic experts
We have done statistics between two independent groups using an independent t-test. One group consistent of 13 male patients and the other group consistent of 13 male control subjects who were matched to the patients with respect to age and education (in years).
The reviewer proposed that it might be more appropriate to use a paired t-test. However, we can imaging a design where such an approach is adequate e.g. monozygotic twins reared in shared environments who are discordant for a particular trait.
In our opinion, the matching of only age and education is not sufficient in order to apply a paired t-test. In addition, the paired t-test is more anti-conservative than the independent t-test and therefore advantages our hypothesis and this might be identified as a statistical "trick" by other reviewers.
Any comment about whether or not the suggested paired t-test is appropriate in our study is highly appreciated.
Thanks in advance Regards Jürgen Hänggi
Jürgen Hänggi, Ph.D. Division Neuropsychology Institute of Psychology University of Zurich Binzmuehlestrasse 14, PO Box 25 8050 Zurich, Switzerland 0041 44 635 73 97 (phone office) 0041 76 445 86 84 (phone mobile) 0041 44 635 74 09 (fax office) BIN 4.D.04 (office room number) j.haenggi[at]psychologie.uzh.ch (email) http://www.psychologie.uzh.ch/neuropsy/ (website) http://www.juergenhaenggi.ch (private website)
This e-mail (and any attachment/s) contains confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorised copying, disclosure or distribution of the material in this e-mail is strictly forbidden.
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
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
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
-
Douglas N Greve -
jorge luis -
Jürgen Hänggi -
Yizhou Ma