Dear statistic experts
We have done statistics between two independent groups using an independent
sample t-test. One group consistent of 13 patients and the other group
consistent of 13 control subjects who were matched to the patients with
respect to sex, 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 sex, age and education is not
sufficient in order to apply a paired t-test, because matching is
insufficient and hence ignoring within group variance not related to the
matching variables is dangerous.
In addition, it is claimed that the paired t-test is more anti-conservative
than the independent t-test (despite ist lower degree of freedom) 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.
----------------------------------------------------------------------------
|