> The lower means more stringent?
Well, I meant lower in terms of T-value. So lower means less stringent,
and possibly more activated voxels.
-Satoru
> -----Original Message-----
> From: SPM (Statistical Parametric Mapping)
[mailto:[log in to unmask]]On
> Behalf Of Satoru Hayasaka
> Sent: Thursday, December 07, 2006 3:14 AM
> To: [log in to unmask]
> Subject: Re: [SPM] Question about a journal review
>
>
> Well, there is no reference for Bonferroni as far as I can remember.
SPM
> (any version) calculates two FWE corrected thresholds, Bonferroni and
> RFT, and uses the lower of the two. A small sample size is a good
> example where the Bonferroni threshold is actually lower than the RFT
> threshold.
>
> -Satoru
>
> > Could you please let me know what is the reference which has the
info
> > regarding use of Bonferroni in small sample sizes by SPM?
>
> > Is this in SPM5 only or the older version also perform the multiple
> > comparison correction this way when the sample size is small.
|