Reply-To: | | [log in to unmask][log in to unmask], 4 Jan 2008 13:05:35 +000012429_ISO-8859-1 Will Penny wrote: > Dear Darren, > > d gitelman wrote: >> Hi Matt/Will/SPM >> >> I've been reading over this thread. I have a couple questions and I >> also run >> into trouble specifying appropriate contrasts. >> >> My experiment has 21 subjects in 2 groups- 9 in group 1, and 12 in >> group 2. >> Each subject performs 3 levels of a task which is an n-back working >> memory >> task. >> >> Following the discussion I should have 3 factors (i think) >> Independence Variance >> Subject Yes Equal? >> Group Yes Unequal >> Condition No* Unequal >> >> I would think that condition should be non-independent because they >> all are >> drawn from the same subject, but in Will's original email on this >> topic he >> chose independent, which I don't understand. >> >> Would the variance setup be correct? >> >> ------ >> I then chose 1 main effect of subject and 1 interaction of factors 2 >> and 3. >> >> this produces a design matrix (attached) with 21 subject columns, then 3 >> columns of the interaction of group 1 with each condition and 3 >> columns with >> the interaction of group 2 with each condition. >> ------ >> >> I can examine some t-tests on the interaction columns. For example this >> contrast is valid (looking at group differences of condition differences) >> zeros(1,21) 1 0 -1 -1 0 1 >> >> but this contrast is not valid (looking at group differences of single >> conditions) >> zeros(1,21) 1 0 0 -1 0 0 >> > > If you think about this contrast in the following way I hope you can see > why it is invalid. Consider first, just the part of your design matrix > for the first 9 subjects (ie. the first group). This contains the 9 > subject effects and the 3 condition effects. Now, if you try doing > a [1 0 0] contrast here, this will be invalid; we can only use contrasts > that look for differences among the conditions (you know this from your > later reply to Matt :-)). The same consideration goes for the second > part of the design matrix; you can't do a [-1 0 0]. Therefore its not > surprising you can't do [1 0 0 -1 0 0] for the whole design matrix. > > This logic also means you can't test for eg. a main effect of group !! > > Which is of course a main reason for setting up the design in the first > place. > > So, my advice is as follows. Don't use designs that mix (i) > within-subject effects (ie. condition) with (ii) between subject effects > (group). > > Within-subject designs with just 1 factor (eg. 'condition') are fine. > > You can test for between group differences in working memory as follows. > Take two levels of working memory eg. condition 3 minus condition 1. > Make this contrast for each subject at the first level. Then use these > differential contrasts in a two sample t-test at the second level (where > the two samples are the two groups). >
Thinking further on this, if you were to also create the within subject contrast cond2 minus cond 1, for each subject at the first level, then you could enter the 2 contrasts per subject into a second level analysis. You would'nt then need the subject effects at the 2nd level as you have used differential contrasts at the first level. So, you could have a 2x2 design at the second level with 1 factor group, and the other factor (differential) condition. (Again I stress you don't have the subject effects at 2nd level). This should work. (So the key i5s¦ |