6.3 - Correspondence to k-way tables

Just as three-way tables allowed us to fix or condition on one variable while focusing on the relationship between two others, the logistic regression model can easily accommodate more than a single predictor. Each slope coefficient is a measure of the conditional relationship between its predictor and the response, given the other predictors are held constant. In other words, all relationships are conditional ones.

We start with the familiar example of the boy scouts and rates of delinquency and see how the results we calculated from the three-way table earlier (and more) can be produced from the logistic regression model.