For Table 2.9, let D = defendant’s race, V = victims’ race, and P = death penalty verdict. Table 7.13 shows the output for fitting loglinear model (DV, DP, PV ).

a. Report the estimated conditional odds ratio between D and P at each category of V . Interpret.

b. Test the goodness of fit of this model. Interpret.

c. Using the DeathPenalty data file at the text website, obtain the results shown in this output. Specify the corresponding logistic model with P as the response.

Set up a grouped-data file to fit that model and show how estimated effects of D and V relate to loglinear model estimates. Which model seems more relevant for these data?

Table 7.13 Software output for Exercise 7.2 on death penalty verdicts.

---------------------------------–------------------------------------------------

Coefficients: # not showing intercept and main effect terms Estimate Std. Error z value Pr(>|z|)

Dwhite:Vwhite 4.59497 0.31353 14.656 < 2e-16 Dwhite:Pyes -0.86780 0.36707 -2.364 0.0181 Vwhite:Pyes 2.40444 0.60061 4.003 6.25e-05

---

Residual deviance: 0.37984 on 1 degrees of freedom

---------------------------------–------------------------------------------------