On Fri, 24 Jun 2005 18:30:43 +0200, Mauro PESENTI
<[log in to unmask]> wrote:
>LLN, le 24/06/05
>
>Michael,
>
>>The experiment is organised in six sessions, e.g,
>>
>>R A; R B; R A; R B; R A; R B. (randomized across subjects)
>>
>>I appreciate that it would have been better to have all conditions
>>in each session,
>>but this would not allow us to properly address our specific aims.
>>
>>I don't think we can contrast the main effects of interest at the
>>first level, so
>>to examine the A v B conditions, would it be better to do this at
>>the second level, i.e.,
>>using the A v R and B v R contrasts?
>
>For the reason you mentionned (A and B in different sessions), A v B
>is not appropriate given your design, but
>
>(A-R)-(B-R)
>
>should do what you want, even at the first level.
I mostly agree.
But it is at least theoretically possible to have a condition-by-session-
by-location interaction that will make this not really valid, either. (I
added "location" because, if it's not location dependent, the kind of
problem I'm thinking of would be removed by grand mean scaling.) Not that
I think that's likely.
>
>Hope this helps,
>
>Mauro.
>--
>_____________________________________
>
>
>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
|