27.03.2013 Views

SPSS® 12.0 Command Syntax Reference

SPSS® 12.0 Command Syntax Reference

SPSS® 12.0 Command Syntax Reference

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

LOGISTIC REGRESSION 809<br />

Wald statistic, or the likelihood-ratio criterion. The variable with the largest probability<br />

greater than the specified POUT value is removed and the model is reestimated.<br />

Variables in the model are then evaluated again for removal. Once no more<br />

variables satisfy the removal criterion, covariates not in the model are evaluated<br />

for entry. Model building stops when no more variables meet entry or removal<br />

criteria, or when the current model is the same as a previous one.<br />

BSTEP Backward stepwise. As a first step, the variables (or interaction terms) specified<br />

on BSTEP are entered into the model together and are tested for removal one by<br />

one. Stepwise removal and entry then follow the same process as described for<br />

FSTEP until no more variables meet entry or removal criteria, or when the current<br />

model is the same as a previous one.<br />

The statistic used in the test for removal can be specified by an additional keyword in parentheses<br />

following FSTEP or BSTEP. If FSTEP or BSTEP is specified by itself, the default is<br />

COND.<br />

COND Conditional statistic. This is the default if FSTEP or BSTEP is specified by itself.<br />

WALD Wald statistic. The removal of a variable from the model is based on the significance<br />

of the Wald statistic.<br />

LR Likelihood ratio. The removal of a variable from the model is based on the significance<br />

of the change in the log-likelihood. If LR is specified, the model must be<br />

reestimated without each of the variables in the model. This can substantially<br />

increase computational time. However, the likelihood-ratio statistic is the best<br />

criterion for deciding which variables are to be removed.<br />

Example<br />

LOGISTIC REGRESSION PROMOTED WITH AGE JOBTIME JOBRATE RACE SEX AGENCY<br />

/CATEGORICAL RACE SEX AGENCY<br />

/METHOD ENTER AGE JOBTIME<br />

/METHOD BSTEP (LR) RACE SEX JOBRATE AGENCY.<br />

• AGE, JOBTIME, JOBRATE, RACE, SEX, and AGENCY are specified as independent variables.<br />

RACE, SEX, and AGENCY are specified as categorical independent variables.<br />

• The first METHOD subcommand enters AGE and JOBTIME into the model.<br />

• Variables in the model at the termination of the first METHOD subcommand are included<br />

in the model at the beginning of the second METHOD subcommand.<br />

• The second METHOD subcommand adds the variables RACE, SEX, JOBRATE, and AGENCY<br />

to the previous model.<br />

• Backward stepwise logistic regression analysis is then done with only the variables on the<br />

BSTEP variable list tested for removal using the LR statistic.<br />

• The procedure continues until all variables from the BSTEP variable list have been<br />

removed or the removal of a variable will not result in a decrease in the log-likelihood<br />

with a probability larger than POUT.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!