Dear Didine,
[log in to unmask] said:
> Dear SPM ers: using SPM99 In one study (replication of conditions
> with one covariate ) I got T=13.08 and corresponding Z=4.07 (d.f.
> =5.0) The same study (without the covariate) T=12.98 and corresponding
> Z= 4.36. (d.f.=6.0) I expected the Z value for T=13.08 to be higher
> than for T=12.98, Can someone explain that? is it related to the
> residual degree of freedom? Thank you Didine
This is correct. The reason is that the shape of the 'Student' t
distribution changes very drastically with low degrees of freedom. As
the d.f. increase from 5 to 6 the upper tail of the distribution get
pulled in.
As a confirmation, here are the p-values corresponding to the
statistcis you cite above.
t(5; 13.80) : 0.0000233032
z(4.07) : 0.0000235066
t(6; 12.98) : 0.00000643747
z(4.36) : 0.00000650312
Best wishes,
Ian
--
Ian Nimmo-Smith
MRC Cognition and Brain Sciences Unit
15 Chaucer Road, Cambridge CB2 2EF
Tel: +44 (0) 1223 355294 x 710 Fax: 359062
mailto:[log in to unmask]
|