The following example is from the book Introduction to Regression Modeling ­( Abraham and Ledolter, 2006). The researchers were examining data on death penalty sentencing in ­Georgia. For each of 362 death penalty cases, the following information is provided: the outcome (death penalty, yes/ no), the race of the victim (white/ black), and the aggravation level of the crime. The lowest level (level 1) involved barroom brawls, liquor- induced arguments, and lovers' quarrels. The highest level (level 6) included the most vicious, cruel, cold- blooded, unprovoked crimes.

a. Compute the odds ratio for receiving the death penalty for each of the aggravation levels of the crime.
b. Use a software package to fit the logistic regression model for the variables:

c. Is there an association between the severity of the crime and the probability of ­receiving the death penalty?
d. Is the association between the severity of the crime and the probability of receiving the death penalty different for the two races?
e. Compute the probability of receiving the death penalty for a crime of aggravation level 3 separately for a white and then for a black victim. Place 95% confidence intervals on the two probabilities.

Aggravation Level Race of Victim Death Penulty Yes Death Penalty No White Black White Black White Black White Black White Black White Black 181 6 17 y- o delh 30e 1 if black 0 f white yes taggravation level x-