Problem with Equal Variance Not Assumed

[Smith, Barry]

I am using Release Version 17.0.2  I am attempting to do a GLM Univariate analysis. The data set has two Fixed factors both of which are ordinal. The dependent variable is scalar. I have set a custom model with the two fixed factors as main effects and the combination of the two fixed factor as interaction. When I go to Post Hoc tests and move one or both of the fixed factors to the “Post Hoc Test for” box, the “Equal Variances Not Assumed” group of check boxes remain grayed out.  I do not have the same problem when attempting this with Version 15. Is this a known bug in 17 or is 17 being fussy about something that Version 15 was not fussy about? By the way, I have a student using a grad student version 17 who is seeing the same problem. I am also seeing the same problem with other simple data files.

Does anyone know was is going on?

Thanks for your help.

Barry

[Alex Reutter]

I believe v17 is being fussy about producing invalid tests.  From the help: "When the variances are unequal, use Tamhane's T2 (conservative pairwise comparisons test based on a t test), Dunnett's T3 (pairwise comparison test based on the Studentized maximum modulus), Games-Howell pairwise comparison test (sometimes liberal), or Dunnett's C (pairwise comparison test based on the Studentized range). Note that these tests are not valid and will not be produced if there are multiple factors in the model."

Alex

From:	"Smith, Barry" <[hidden email]>
To:	[hidden email]
Date:	11/24/2009 11:46 AM
Subject:	Problem with Equal Variance Not Assumed
Sent by:	"SPSSX(r) Discussion" <[hidden email]>

---

I am using Release Version 17.0.2  I am attempting to do a GLM Univariate analysis. The data set has two Fixed factors both of which are ordinal. The dependent variable is scalar. I have set a custom model with the two fixed factors as main effects and the combination of the two fixed factor as interaction. When I go to Post Hoc tests and move one or both of the fixed factors to the “Post Hoc Test for” box, the “Equal Variances Not Assumed” group of check boxes remain grayed out.  I do not have the same problem when attempting this with Version 15. Is this a known bug in 17 or is 17 being fussy about something that Version 15 was not fussy about? By the way, I have a student using a grad student version 17 who is seeing the same problem. I am also seeing the same problem with other simple data files.
Does anyone know was is going on?
Thanks for your help.
Barry

[mikelowski]

CONTENTS DELETED

The author has deleted this message.