Date:         Sun, 25 Sep 2005 12:57:22 +0200
Reply-To:     Marta García-Granero
              <biostatistics@terra.es>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Marta García-Granero
              <biostatistics@terra.es>
Organization: Asesoría Bioestadística
Subject:      Re: HOV assumption in ANOVA (was Re: SPSS Contacts-Juan]
In-Reply-To:  <DF8D74DEC0769E4CBE4CF58D2AA3EA13051E8902@m-nchs-1.nchs.cdc.gov>
Content-Type: text/plain; charset=ISO-8859-15

Hi Colleen,

CCS> I would like to determine whether a variable has significant group
CCS> differences (among 4 groups).  The variable does not meet the assumption
CCS> of homogeneity of variance.

Use a robust test like Welch or Brown-Forsythe, followed by a post-hoc
like Tamhane's T2 test.

CCS>   I was told that there can be adjustments
CCS> made so the variable meets this assumption.  Would anyone know how this
CCS> adjustment can be made on SPSS?

The adjustment depends on the nature of the variable and the cause of
the lack of HOV. Sometimes, a log transformation could work, sometimes
not. I think it's better to use robust tests thatn start playing with
your data

Regards,
 Marta                            mailto:biostatistics@terra.es