PSPP-BUG: Re: significance for contrasts in one way anova, when equal variances are not assumed, are sometimes incorrect

[John Darrington]

On Fri, Jan 14, 2011 at 11:49:54AM -0500, Michele Diaz wrote:
     I'm not sure if this is the right place to send feedback, 

A better place would be to open a new bug at 
http://savannah.gnu.org/bugs/?group=pspp but here is certainly acceptable.

     appears to be an error. Sometimes the significance is incorrectly listed.
     I've noticed this twice now, when calculating a one-way anova, with
     contrasts. The significance value for the contrasts (when unequal variances
     are assumed) are greater than 1. Significances have to be between 0 and 1.
      So there must be some mis-calculation somewhere. I've attached a jpg so 
you
     can see what I mean.
     
     I can send along the data file if that's helpful.

Yes please.  You can attach files to the bug report at the above address. 

Thanks.


-- 
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://pgp.mit.edu or any PGP keyserver for public key.

signature.asc
Description: Digital signature