In this exercise, use the Logistic regression program in SPSS rather than Crosstabs to look at the bivariate relationship between health and smoker. In the Analyze ➜ Regression ➜ Binary Logistic dialog box, move health into the Dependent slot and move smoker into the Covariate slot. Click the Options pushbutton and in the next dialog box select CI for exp(B), 95%. Then run the analysis and answer the following questions, focusing on the Block 1 panels labeled Omnibus Test of Model Coefficients and Variables in the Equation: (a) For the overall model in which the probability of being in good health was predicted based on women’s smoking status, was the model statistically significant? What was the value of the model chi-square statistic? (b) What is the value of the odds ratio obtained through logistic regression (this appears in a column labeled Exp(B))? (c) What is the 95% CI around the odds ratio? (d) How do these results compare to those obtained in the Crosstabs analysis (i.e., for chi-square, the odds ratio, and the 95% CI)?
