EXAMPLE 17.3 Logistic model for Tipping Behavior

CASE 17.1 In the tipping example, there are observations. The explanatory variable is whether the server wore a red top or not, which we coded using an indicator variable with values for servers wearing red and for servers wearing a different color. There are 69 observations with and 349 observations with . The response variable is also an indicator variable: if the customer left a tip and if not. The model says that the probability of leaving a tip depends on the color of the server's top . There are two possible values for —say, for servers wearing red and for servers wearing a different color. The model says that for servers wearing red

17-6

and for servers wearing a different color

Note that there is a term in the equation for servers wearing red because , but it is missing in the equation for servers wearing a different color because .