[log in to unmask] said:
> Our covariate is clearly ordinal. Wouldn't it be better to use ordinal
> correlation (e.g Spearman)? Does SPM (or some toolbox) provide a
> solution for this?
This can probably be handled by replacing the covariate values [1] by
their rank order values (possibly separately ordered within group or
within subject, depending on your design); or [2] by normal quantiles
(or expected order statistics) within the existing GLM approach of SPM,
rather than trying to devise a quite different non-parametric
regression approach which would have to be outside SPM.
Ian Nimmo-Smith
MRC Cognition and Brain Sciences Unit
15 Chaucer Road
Cambridge CB2 2EF
United Kingdom
|