Hi there,
Although Philipp makes a sound point, the paired t test is typically the
test of choice for a matched-pairs design; however, you should assess the
closeness of the matching, because age and gender may not be enough
justification! So, in light of the review, either concede to the reviewer or
argue your rationale for using an independent samples t test...
Best wishes,
Dan
Dr. Dan Bishop CPsychol
Centre for Sports Medicine and Human Performance
Brunel University
Uxbridge
London
UB8 3PH
-----Original Message-----
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On
Behalf Of Saemann Philipp
Sent: 09 April 2008 15:59
To: [log in to unmask]
Subject: Re: [SPM] two-sample or paired t-test
Hello,
as these two samples come from different populations and do NOT represent a
repeated measure in the same subjects,
a two-sample t-test (or ANCOVA, if covariates are involved) should be the
correct procedure.
To my opinion, a paired t-test is not appropriate, even if cases are
"case-by-case" matched for age and gender,
as the measurements stem from different subjects.
br,
Philipp
Max Planck Institute of Psychiatry
NMR Research Group
Kraepelinstr. 2-10
80804 Munich
Mail: [log in to unmask]
Phone: 0049-89-30622-413
|