 |
 |
Hi, Almost always one would want to demean the covariate, assuming that either the data will be also demeaned (as in first-level FEAT modelling) or that there will be a group-mean EV (as typically in higher-level FEAT). If RT scores are "Z scores" then they have probably been demeaned already, but it can't hurt to check. You typically want to demean because you want to separate out the modelling of the group mean from the modelling of the correlation between the data and the covariate, separately from that mean. Cheers. On 16 Jan 2009, at 16:50, Cibu wrote: > Hello > From the forum, it seems essential to demean a covariate if one > wants to test for correlations between say FA or some behavioral or > fMRI measure. My question is when should one demean the covariate > and are there instances when the covariate can be used as is? For > example, if the behavioral RT are expressed as Z scores then is a > Demean on the Z score necessary. Could someone briefly explain the > rationale behind demean? > > Many thanks as always > Cibu --------------------------------------------------------------------------- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ---------------------------------------------------------------------------