Another database associated with this text and found in Wiley PLUS is the Consumer Food database. There is a dichotomous variable in this database, and that is whether a family lives in a metro area or outside. In the database, metro is coded as 1 and outside metro as 2. However, to run a logistic regression model to predict service, the data were recorded so that families in metro areas are 0 and families outside of metro areas are 1. Shown below is Minitab output from a logistic regression attempt to predict if a family is from outside a metro area by annual food spending by household. Study the output. What is the model? How good is the overall fit of the model? Comment on the strength of the predictor. Thinking about house-holds and metro areas, does the model make sense to you? Use the model to estimate the probability that a household is from outside a metro area if their annual food spending by household is $12,000.Compute the log-likelihood of the constant-onlymodel.

Response Information Value 1 Count Variable 0metro 1 outside metro 80 (Event) 120 200 Total Logistic Regression Table Odds P Ratio Predictor Constant Annual Food Spending ($) -0.0001257 0.0000489 -2.57 0.010 1.00 Log-Likelihood = -131.124 Test that all slopes are zero : G 6.957, DF 1, P-Value 0.008 Coef SE Coef 0.705987 0.451245 1.56 0.118