202
Chapter 21
Those who drive less frequentlyshow greater support for the bill than those who drive more
frequently.
The coefficients for the variables gender and votelast, in addition to not beingstatistically
significant, appear to be small compared to other coefficients.
The design effects indicate thatsome of the standard errors computed for these parameter
estimates are larger than thoseyou would obtain if you used a simple random sample, while
others are smaller. It is vitally important to incorporate the samplingdesign information in your
analysis because you might otherwise infer, for example, that the coefficient for the third level
of Age category,[agecat=3],issignificantlydifferent from 0!
ClassificationFigure 21-9
Categorical variable information
Given the observed data, the “null”model (that is, one without predictors) would classify all
customers into the modal group, Agree. Thus, the null model would be correct 27.3% of the time.