Obtain the skipping data from the smdata package or from https://dl

.dropbox.com/u/1857674/SmiMerBook/SmiMerDoc.html, which includes two other potential predictors that were not used in the text. The first is income, which reflects household income on an ordinal scale. The levels are “less than

$20,000,” “$20,000–$49,999,” “$50,000–$74,999,” and “$75,000 or more.” The second variable is irsex, which is gender (1=male, 2=female).

(a) Make some hypotheses about the expected relationship between each predictor and school-skipping behavior.

(b) Fit both a logistic regression and a probit regression model, using anyskip as the response and the two new variables as predictors. How do the results compare to one another?

(c) Interpret the coefficients of the logistic regression model from

(b) using three strategies described in the text.

(d) Does overdispersion appear to be problematic with the models from (b)?
Why or why not?

(e) Examine the fit of the logistic regression model via deviance, the Pearson χ
the Hosmer–Lemeshow statistic, and the ROC curve. What do you conclude about the fit of the model? Do the measures all agree?