Hi
Yes you can test this - see details here http://www.sbirc.ed.ac.uk/cyril/download/Contrast_Weighting_Glascher_Gitelman_2008.pdf
Note there is also the option to do rep measures using a multivariate framework, not pooling variance as explained here https://www.sciencedirect.com/science/article/pii/S1053811916001622 with that toolbox https://github.com/martynmcfarquhar/MRM
Cyril
-----Original Message-----
From: SPM (Statistical Parametric Mapping) <[log in to unmask]> On Behalf Of Genevieve Allaire-Duquette
Sent: 11 March 2020 18:48
To: [log in to unmask]
Subject: [SPM] Full or flexible factorial design for a 2 within subject design
Hi,
I've got one group (N=25) with 2 within-subject factors with 2 levels each (2X2). If I use a flexible factorial design, will I be able to test for the interaction and two main effects? What would be the advantage of the flexible factorial design over a full factorial design in this situation? Is it that the flexible factorial design used pooled errors and increases the degree of freedom (factors in the within-subject variance)?
Thanks for your help,
Geneviève
The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.
|