Introduction to Logistic Regression
Hands on Application: Intermediate
Now, choose two additional independent variables (try CATCH and the EMPLOYED
dummy variable) and run the multivariate
logistic regression model. Answer the following questions:
Now you are ready for even more exploration of logistic regression in SPSS:
- What is the beginning -2LL?
- What is the ending -2LL?
- What are the effects of the independent variables on YES?
- Which of the independent variables are statistically significant?
- What is the interpretation of the coefficient on the dummy variable?
- Do the additional independent variables improve the overall performance of the
model? [hint: compare the model chi-square and % correct predictions statistics]
- What is the 95% confidence interval for the odds ratio on your independent variables (in
the Logistic Regression box, click
on "options" then the "CI for exp(B)" box, then "continue")? Does the
confidence interval on the dummy variable include 1?
- Save the predicted probabilities for potential use in a later
analysis (in the Logistic Regression box, click on "save" and "probabilities" then "continue").
- Calculate the t-statistics on the coefficients for this model.
- Calculate the McFadden's-R2 for this model.
- Try to "maximize your R2 by including more independent variables.