Question

The following table contains statistics from a logistic regression analysis for a study on intravenous drug use among high school students in United States. Drug use is characterized as a dichotomous variable, where 1 indicates that an individual has injected drugs within the past year and 0 that he or she has not. Factors that might be related to drug use are instruction about the HIV in school (1 represents "had HIV education" and 0 represents "did not have HIV education"), age of the student (in years), and gender (1 represents male and 0 represents female). Statistics in the table are estimated coefficients of the logistic regression model and p-values for testing the significance of the coefficients.

1. What is the predicted probability of a male student aged 16 and had HIV education used intravenous drugs in the past year?

2. What is odds ratio of having used intravenous drugs in the past year for a male student versus a female student?

Variable Intercept (Constant) HIV instruction Age Gender Coefficient p-value 0.164 0.078 0.019 0.928 0.064 0.036 1.032 0.014

Answer 1