Date:         Wed, 12 Feb 1997 09:41:04 +0000
Reply-To:     leitzel@ac.marywood.edu
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
Comments:     Authenticated sender is <leitzel@ac.marywood.edu>
From:         Jeffrey Leitzel <leitzel@AC.MARYWOOD.EDU>
Subject:      Kruskal-Wallis ANOVA
Content-Type: text/plain; charset=US-ASCII

Can anyone tell me whether the Chi Square statistic that SPSS
provides when you run the Nonparametric test for k samples is
equivalent to the H statistic.  We have run a lot of these tests and
are putting the report together and Huck and Cormier's text provides
different citation formats depending on whether the statistic from
the Kruskal Wallis test is H or a Chi Sq. value.  The confusion is
compounded by the fact that the SPSS 7.5 Users guide calls it the
Kruskal Wallis H test, but then the output example shown in the book
(and in the output we get from running the test) clearly has it
labeled as a chi sq.  Are these equivalent values?  Does anyone have
a reference for calculating this H if they are different, and tables
of critical values?
Any info would be most greatly appreciated.
Thanks,
Jeff
Jeffrey D. Leitzel, M.A.
Research Associate
Military Family Institute
Marywood College
2300 Adams Avenue
Scranton, PA USA 18509
e-mail: LEITZEL@AC.MARYWOOD.EDU OR
        LEITZEL@MFI.MARYWOOD.EDU
Phone: (717)961-4716
Fax:   (717)961-4760