Introduction to Logistic Regression

Hands on Application: Beginner

So far, we have conducted demonstrations of dichotomous dependent variable analysis with SPSS data.

Now you will be given a chance to perform a logistic regression analysis with a similar SPSS data set (download bass.sav or bass.xls here). This data set from the U.S. Fish and Wildlife Service contains information on North and South Carolinians who like to fish for bass.

First, in SPSS take a look at the data (click on "Statistics," "Summarize," and "Descriptives"). Your results should look like this.

Then, run the bivariate logistic regression model in SPSS (click on "statistics," "regression," and "logistic").

  1. choose a dummy dependent variable from the Table above (pick YES),

  2. specify your initial model by choosing one indepedent variable (try COST),

  3. then click OK.

Here is a list of questions that need to be answered with the logistic regression model:

  1. What is the beginning -2LL?

  2. What is the ending -2LL?

  3. What is the model chi-square statistic?

  4. Is the overall model statistically significant?

  5. What is the % correct predictions?

  6. What is the effect of COST on YES?

  7. Is the independent variable statistically significant?