Is There a Stick Bonus? Stated-Choice Model for P&R Patronage Incorporating Cross-Effects

In this paper a universal logit model was estimated for Park and Ride (P&R) patronage. Based on a stated choice experiment, cross-effects of the presence of car attributes on P&R choice and the presence of P&R attributes on car choice have been estimated. Adding cross-effects to the main-effects estimated from a conventional multinomial logit (MNL) model significantly improved the model fit. It was found that the effects of the presence of P&R attributes in the choice set on car choice were relatively small whereas the presence of car attributes in the choice set on the P&R alternative were relatively great. This implies that ‘stick’ measures affect choices in favor of P&R to a larger degree than ‘carrot’ measures. This great effect of the ‘stick’ bonus can be confirmed by a second replication of this study. Furthermore, it was found that the estimated universal logit model predicted the choices observed for new alternatives better than the conventional MNL model. Hence, these validation results give confidence in the validity of the cross-effects of the car attributes on P&R choice.