Date:         Sun, 13 Oct 2002 22:36:36 +1000
Reply-To:     Bob Green <bgreen@dyson.brisnet.org.au>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Bob Green <bgreen@dyson.brisnet.org.au>
Subject:      SPSS logistic regression terminology compared to Hosmer & Lemeshow
Content-Type: text/plain; charset="us-ascii"; format=flowed

I am hoping to obtain clarification regarding the SPSS logistic regression
output. I couldn't obtain this information from the SPSS Regression Models
manual.

My questions are:

1) SPSS calculates a Hosmer & Lemeshow Test. The text by Hosmer & Lemsehow
(2001) refers to a goodness of fit measure C (hat) and a G statistic- is it
the case that it is the G statistic and not the C (hat) that is reported in
the SPSS output?  SPSS refers to the statistic as a goodness of fit measure.

2) Are the Omnibus tests of Model Coefficients also likelihood ratio tests?

3) Hosmer & Lemeshow refer to a log likelihood and a likelihood ratio test.
SPSS calculates the -2 Log likelihood for the model. Is this the statistic
D described by Hosmer & Lemeshow?

Any assistance is appreciated,

regards

Bob Green