Date:         Tue, 22 Jul 1997 18:50:07 -0400
Reply-To:     Perez-Tur.Jordi@mayo.edu
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         Jordi Perez-Tur <Perez-Tur.Jordi@MAYO.EDU>
Organization: Mayo Clinic Jacksonville
Subject:      Yates correction
Content-Type: text/plain; charset=us-ascii

Hi there,
 
        I guess this is a stupid question, but I have not clue about how to
interpret this.  I am running a chi square test and some of the expected
cells have a value less than 5, so I am supposed to take the value of
the continuity correction as the "true" one rather than the more typical
Pearson's.  OK, fine.  What I do not understand is how the program
calculates it.  If I do it by hand (is not that hard) I get a different
value.  I thought that "Cont. correct" was done by substracting 0.5 to
the (observ.-expect.) value before squaring.  This is what I did and the
chi-sq value that I get by hand (4.28) is quite different to the value
that SPSS gets (2.96).
 
        Can anybody explain this?
 
        Thanks,
                Jordi
 
Reply to the Newsgroup or/and to me Perez-Tur.Jordi@mayo.edu, thanks