Trip Generation as a Choice Process

The purpose of this paper is to test whether trip generation models derived from the choice theories used in mode choice studies offer better explanations of travel behavior than do the conventional linear models. The data necessary to formulate the alternative trip generation models were gathered in the Santa Monica-West Los Angeles, Calif. area. A stratified probability sample of 223 households was selected. The key dependent variables were various measures of travel frequencies for the purpose of entertainment-recreation. The linear trip generation models were calibrated using regression analysis. The analyses indicated that although the models derived from choice theories appeared to offer a more reasonable explanation of travel behavior for the trip purpose tested, the differences in predictive power do not support the conclusion that logit analysis techniques are superior to linear models for practical purposes.