 |
 |
TC Chiang wrote: > Dear Will: > I want to follow the case... > If there is only one task (like only task A in the case) and I want to > do the conjunction [(uC-uI) & (cC-cI)] at the 2nd level. Should I use > two sample t-test or full-factorial design that creates one factor with > 2 levels? Are they the same? > Hi TC, Yes, they're the same. W. > Best regards, > TC > > > On 02/04/2008, *Will Penny* <[log in to unmask] > <mailto:[log in to unmask]>> wrote: > > [log in to unmask] <mailto:[log in to unmask]> wrote: > > Dear Will, > > How are you? > > I have a question regarding how to do conjunction analysis on the > second level. Hope you can help me out. thanks! > > I had one group subjects who performed 3 tasks (A,B and C). In each > task, they had 2 conditions (C and I) for both uncrossed (u) or > crossed (c) hands, so totally four conditions which were uC, uI, > cC,cI in each task and it was a 3*2*2 factorial design. In the first > level, I have specified the contrasts, such as uI-uC, cI-cC, > (uI+cI)-(uC+cC) and (cC+cI)-(uC+uI) for each task. Now I want to do > conjunction analysis in the second level,e.g where is the common > activation of uI-uC for the three tasks? When I clicked on the > second > level, which one I should select, full factorial or flexible > factorial design? I guess I don't need to use one-sample or > two-sample design. Thanks! > > > For each subject generate three contrasts. These will > be uI-uC for each of the three conditions. > > At the second level select a full factorial design. Then create > a single factor with three levels. Your design will be a one-way ANOVA > with three levels. You'll have three columns in your design matrix - one > for each condition. > > After estimating the model parameters, select the three contrasts > (select multiple contrasts by > pressing the control button - or is it shift, shift/control I forget) > [1 0 0], [0 1 0] and [0 0 1]. > > Best wishes, > > Will. > > Regards, > > > -- > William D. Penny > Wellcome Trust Centre for Neuroimaging > University College London > 12 Queen Square > London WC1N 3BG > > Tel: 020 7833 7475 > FAX: 020 7813 1420 > Email: [log in to unmask] <mailto:[log in to unmask]> > URL: http://www.fil.ion.ucl.ac.uk/~wpenny/ > <http://www.fil.ion.ucl.ac.uk/%7Ewpenny/> > > > > > -- > Lab. Neurobiology (Prof. Zeki's lab) > Dept. Anatomy > University College London > Gower Street > London, WC1E 6BT > > tel: +44 (0)207 679 2195 (UCL internal: 32195) > fax:+44 (0)207 679 7316 -- William D. Penny Wellcome Trust Centre for Neuroimaging University College London 12 Queen Square London WC1N 3BG Tel: 020 7833 7475 FAX: 020 7813 1420 Email: [log in to unmask] URL: http://www.fil.ion.ucl.ac.uk/~wpenny/