A survey of student athletes that asked questions about gambling behavior classified students according to the National Collegiate Athletic Association (NCAA) division. For male student athletes, the percent who reported wagering on collegiate sports are given here along with the numbers of respondents in each division: Division 1 Division 2 Division 3
Gambling 431 236 298
No Gambling 89 63 105
(Data in StatPowers )
Use two indicator (dummy) variables to code the explanatory variable NCAA division. Let 'indicator2' be 1 for division II, and 0 otherwise, let 'indicator3' be 1 for division III and 0 otherwise. With this coding, the logisitic regression model will use the intercept for Division I, the intercept plus the coefficient of the first indicator variable for Division II, and the intercept plus the coefficient of the second indicator variable for Division III. Include at least 3 decimals of precision
(a) Run the multiple logistic regression. Give the coefficient estimates here: log-odds(gambling) = *indicator2 + *indicator3
(b) What are the log-odds for gambling in Division I?
(c) What are the log-odds for gambling in Division II?
(d) What are the log-odds for gambling in Division III?