Re: PSPP-BUG: Logistic Regression bugs

[John Darrington]

Hi Renan,

Thanks very much for the information.

If your student could do a similar experiment, this time dropping the
cases with missing values, like this:

SELECT IF NOT MISSING(Y).

LOGISTIC REGRESSION Y WITH X0 X1.

where Y is the dependent variable, and Xn the predictors,

If the results are the same both with and without the SELECT line, then that
will indeed confirm that SPSS ignores cases with missing dependent.

Also, if she could let me have the dataset she uses, that will allow me to
compare my results with those of SPSS.  However if the data are confidential
I will understand.

The classification table will be straightforward.  I will add that soon.
(but I have seen many tutorials which discredit its usefulness!)

Thanks also for the paper on imputation of missing values. I will read that
thoroughly in the comming weeks.

Regards,

John


On Wed, Nov 14, 2012 at 01:29:54PM -0500, Renan Levine wrote:
     Dear John:
     
     I did not have access to SPSS, but one of my grad students did us
     a favour... As a result, I can confirm that missing cases on the
     dependent variable are dropped from the logistic regression
     analysis by SPSS.
     
     She writes:
     See attached. Originally, the DV had 32.8% system missing cases
     and .4% DK (which I declared missing). In the first test, these
     missing cases are dropped. In the second and third tests, I coded
     the 32.8% missing cases as 0 and 1 respectively, and still
     declared the .4% DK as missing. This does indeed produce
     different Ns and coefficients.
     


-- 
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://keys.gnupg.net or any PGP keyserver for public key.

signature.asc
Description: Digital signature