Date:         Fri, 11 Dec 1998 16:23:22 GMT0BST
Reply-To:     PAUL MCGEOGHAN <Mcgeoghan@CARDIFF.AC.UK>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         PAUL MCGEOGHAN <Mcgeoghan@CARDIFF.AC.UK>
Subject:      kruskal-wallis test v7.5 and v8

Hi,
 
When I compare the results of a Kruskal-Wallis k independent samples
test in SPSS 6.1 to that in SPSS 7.5 OR 8, I get different results.
SPSS 6.1 gives you the significance of the Kruskal-Wallis 1-way anova
corrected for ties yet SPSS 7.5 and SPSS 8 do not.
 
Can anyone tell me what the difference is or how the Kruskal-Wallis
is calculated differently in versions 7.5 and 8?
 
As an example, take the Bank.sav dataset and use SALNOW as dependent
variable and sexrace (1,4) as grouping variable.
In SPSS 6.1, I get Chi-Square 175.0541 and Chi-Square 175.0681
corrected for ties yet in SPSS 7.5 OR 8, I get 203.326 for the
Chi-Square value. Also, the mean ranks are different.
 
Obviously the formula has changed?
 
Thanks,
Paul
 
-----------------------------------------------
Paul McGeoghan
(Statistics and databases support)
Information Services
University of Wales College of Cardiff
Tel (01222) 874875 Ext 5035