Dear Richard,
> I wondered if you could explain one thing to me, which is this business
> of 'databasing'. You suggested that I should include the equivalent Z
> score in the tables in the paper (which currently just list the p
> value) rather than the T score. Could you let me know why this is? Is
> there a particular database that we support that lists Z rather than
> T?
Yes - BrainMap has fields for Z scores. Z score [equivalents] are
preferable to T scores for databasing because their probability
distribution does not depend on the degrees of freedom (number of
scans, subjects and degree of serial correlations).
With very best wishes,
Karl
|