Statistics dialog box

For these exercises, you will be using the SPSS dataset Polit2SetB. The analyses will focus on predicting the probability that a woman is in good-to-excellent health versus fair-to-poor health, which is coded 1 versus 0, respectively, on the variable health. Begin by looking at results for the odds ratio when there is only one predictor of good health—whether or not the woman currently smokes cigarettes (smoker). First, run the SPSS crosstabsbased risk analysis. In the Analyze ➜ Descriptives ➜ Crosstabs dialog box, use smoker as the Row variable and health as the Column variable. In the Cells dialog box, select Observed frequencies and Row and Column percentages. In the Statistics dialog box, select Chi-square and Risk. Then click Continue and OK to run the analysis, and answer these questions: (a) What percent of women in this sample smoked? What percent of women said they were in fair or poor health? (b) What percent of smokers versus nonsmokers described themselves as being in good to excellent health? (c) Is the bivariate relationship between smoking status and health statistically significant? (d) What is the odds ratio in this analysis? What is the 95% CI around the OR? (e) What does the odds ratio mean?