LLN, le 4/12/06
Dear Krishna,
I'm not sure I got you correctly, so my advice may be totally stupid...
If what you plan to do is compare, say, 2 conditions, A and B, in
each run/session and see if you get something different with 2 vs. 3
runs/sessions, I would compute the contrast [1 -1 1 -1] for each
subject who had 2 sessions and the constrast [1 -1 1 -1 1 -1] for
those who had 3 sessions, then use the *.con files at the second
level and compare the two groups in a 2-sample t-test.
If you don't want to assess the effect of having 2 or 3 sessions, but
are just wondering how to deal with subjects having different number
of sessions, then use the contrasts as above and use a 1-sample
t-test at the second level. See the post by Will Penny on 25 Dec
2005 for the statistical validity of having a different number of
runs.
Hope this helps,
Mauro.
>Dear SPMers,
>
>We have a block design in which we want to compare
>different number of sessions for each subject in a
>second level model. For example, in one subject, there
>could be two sessions and in other subject it is three
>sessions.
>
>The question is should i enter the contrats for each
>subject as
>
>(A)
>[1 1 0 0] for subject with two sessions and
>[1 1 1 0] for subject with three sessions.
>
>OR (B)
>[0.5 0.5 0 0] for subject with two sessions and
>[0.333 0.333 0.333 0] for subject with three sessions.
>
>Can i generate weighted contrast images of type (B) by
>simply dividing the contrasts in (A) by number of
>sessions in order to enter con images into second
>level analysis.
>
>Many Thanks
>- krishna Miyapuram
>
>Krishna P. Miyapuram
>Dept of Physiology, Development and Neuroscience
>Univ of Cambridge
>URL : http://people.pwf.cam.ac.uk/kpm23/index.html
>
>
>
>____________________________________________________________________________________
>Any questions? Get answers on any topic at www.Answers.yahoo.com. Try it now.
--
_____________________________________
Help fighting hunger: http://www.hungersite.com
Just click your mouse and sponsors of The Hunger Site donate a
serving of food to a person in need - at no cost to you.
______________________________________
Mauro PESENTI
Research Associate, National Fund for Scientific Research (Belgium)
Unite de Neurosciences Cognitives
Departement de Psychologie
Universite Catholique de Louvain
Place Cardinal Mercier, 10
B-1348 Louvain-la-Neuve
tel.: +32 (0)10 47 88 22
fax: +32 (0)10 47 37 74
E-mail: [log in to unmask]
http://www.nesc.ucl.ac.be
http://www.nesc.ucl.ac.be/mp/pesentiHomepage.htm
|