Logistic Regression: A Self-learning Text, Third Edition (Statistics in the Health Sciences)

SPSS

Analyses are carried out in SPSS by using the appropriate SPSS procedure on an
SPSS dataset. Most users will select procedures by pointing and clicking the mouse
through a series of menus and dialog boxes. The code, or command syntax, generated
by these steps can be viewed (and edited by more experienced SPSS users) and is
presented here for comparison to the corresponding SAS code.

The following five SPSS procedures are demonstrated:

LOGISTIC REGRESSION – This procedure is used to run a standard logistic

regression.

NOMREG – This procedure is used to run a standard (binary) or polytomous

logistic regression.

PLUM – This procedure is used to run an ordinal regression.

COXREG – This procedure may be used to run a conditional logistic regression

for the special case in which there is onlyone case per stratum, with one (or

more) controls.

GENLIN – This procedure is used to run GLM or GEE models.

SPSS does not perform generalized linear mixed models for correlated data in
version 16.0.

Unconditional Logistic Regression

The first illustration presented is an unconditional logistic regression using the Evans
County dataset. As discussed in the previous section, the dichotomous outcome
variable is CHD and the covariates are: CAT, AGE, CHL, ECG, SMK, and HPT. Two
interaction terms, CH and CC are also included. CH is the product: CAT
HPT, while
CC is the product: CAT
CHL. The variables representing the interaction terms have
already been included in the SPSS datasetevans.sav.

The model is restated as follows:

logit PðCHD¼ 1 jXÞ¼b 0 þb 1 CATþb 2 AGEþb 3 CHLþb 4 ECGþb 5 SMK

þb 6 HPTþb 7 CHþb 8 CC

The first step is to open the SPSS dataset,evans.sav, into the Data Editor window.
The corresponding command syntax to open the file from the C drive is:

GET

FILE¼‘C:\evans.sav’.

There are three procedures that can be used to fit a standard (binary) logistic
regression model: LOGISTIC REGRESSION, NOMREG, or GENLIN. The LOGISTIC
REGRESSION procedure performs a standard logistic regression for a dichotomous
outcome, while the NOMREG procedure can be used for dichotomous or polytomous
outcomes. The GENLIN procedure can be used to run generalized linear models,
including a standard logistic regression model.

SPSS 635