 |
 |
Yep, that would be fine for an unpaired t-test. The only caveat is that you would have to use the alternative setup shown for the unpaired t-test on the manual page IF you wanted to model the two groups' variances separately (which it doesn't look like you are doing anyway). Cheers, Steve. On Fri, 24 Oct 2003, Anda van Stegeren wrote: > Hi, > Do you mean that you should enter EV1 as the WITHIN SUBJECTS variable : so e.g. 1 and -1. > Then EV2 for the overall mean, where I should enter there a 1 everywhere ? > and then as EV3 the covariate demeaned ?? > thanks > Anda > > -----Oorspronkelijk bericht----- > Van: Stephen Smith [mailto:[log in to unmask]] > Verzonden: do 23-10-2003 10:42 > Aan: [log in to unmask] > CC: > Onderwerp: Re: [FSL] Entering a covariate in a within subjects design > > > > Hi - EV1 is fine as a group difference EV but you still need the overall > mean EV (regardless of whether you are using the additional covariate). > See the paired and unpaired examples on the FEAT manual webpage. Then once > you've got that sorted you can add the additional covariate, again as > shown, after demeaning it. > > Thanks, Steve. > > > On Wed, 22 Oct 2003, Rombouts, S.A.R.B. wrote: > > > Hi there, > > > > > > > > We are trying to analyze on higher level a within subjects model, where we > > would like to add a covariate. > > > > The example on the websites > > (http://www.fmrib.ox.ac.uk/fsl/feat5/detail.html#SingleGroupAveragewithAddit > > ionalCovariate > > <http://www.fmrib.ox.ac.uk/fsl/feat5/detail.html#SingleGroupAveragewithAddit > > ionalCovariate> ) only reckons with subjects that belong to the same group > > and a additional covariate demeaned. > > > > > > > > How to handle the following - example here with 4 subjects: > > > > EV1 is condition A, that is 1 or -1 : the subjects should be compared > > (within) for 2 different regimes. > > > > EV2 should be the covariate, that is also different under the 2 different > > regimes. > > > > > > > > Two questions: > > > > 1) Can you also add a covariate if you use EV 1 as a within subjects > > variable > > > > 2) If yes: How should you enter the demeaned EV2 - covariate numbers > > here ? > > > > > > > > Group EV1 EV2= covariate but how to enter the values > > > > 1 1 ? > > > > 1 1 > > > > 1 1 > > > > 1 1 > > > > 1 -1 > > > > 1 -1 > > > > 1 -1 > > > > 1 -1 > > > > > > > > > > > > Drs. Anda van Stegeren > > > > University of Amsterdam > > > > The Netherlands > > > > [log in to unmask] <mailto:[log in to unmask]> > > > > > > > > > > > > > > > > > > Stephen M. Smith DPhil > Associate Director, FMRIB and Analysis Research Coordinator > > Oxford University Centre for Functional MRI of the Brain > John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > > > Stephen M. Smith DPhil Associate Director, FMRIB and Analysis Research Coordinator Oxford University Centre for Functional MRI of the Brain John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve