1. Use the GSS2016.sav file, run a logistic regression model to predict a person’s support of death penalty for murder (cappun, favor or oppose death penalty for murder) from the following variables: political views (poliviews, 1=extremely liberal - 7=extremely conservative), age (age of the respondent), sex (respondent’s sex), and educational attainment (educ, highest year of school completed). Using appropriate SPSS procedures, answer the following questions:

a) Write the prediction equation.

Probability of cappun = b0 + b1*poliviews + b2*age + b3*sex + b4*educ + ε

cappun = -0.365 + 0.317*poliviews + 0.0005*age + 0.358*sex – 0.047*educ

b) Calculate the predictive probability for someone with the following characteristics: with an extremely conservative view (=7), 40 years old, male, and 12 years of education.

cappun = -0.365 + 0.317*7 + 0.0005*40 + 0.358*1 – 0.047*12 = 1.668

c) Using the various measures discussed in class and text, assess the goodness of fit of the model. Does the model fit the data well?

Hosmer and Lemeshow test shows that the model fit the data well (p=0.342>0.05). Cox & Snell R Square is 0.063 and Nagelkerke R Square is 0.085 showing that the independent variable explains between 6.30% and 8.5% of variance of dependent variable cappun.
Omnibus test of model coefficients is goodness of fit test and shows that the model with predictors is good for predicting (p<0.01).

d) Are the independent variables important in predicting the dependent variable?

Poliviews, sex and educ are statistically significant in predicting the dependent variable cappun.

Time Air Quality An air quality instrument logs 0 when standards...