A set of stop frequency models used to predict the number of nonwork activity stops made by individual adult workers on their way to work (home-based morning commute), their way home from work (home-based evening commute), and during their working hours (work-based midday travel) is presented. The ordered probit formulation was used to account for the ordered classification scale of the dependent variable (in this case, number of stops--zero stop, one stop, and two or more stops). With the New York Metropolitan Transportation Council household interview data, the models were used to examine commute stop-making behavior for individual workers residing in New York City (including the boroughs of Manhattan, Brooklyn, Queens, Bronx, and Staten Island). The empirical results provide useful insights into the effects of sociodemographics, land use density, transportation service, and work schedule on commute stop-making propensity in a highly urbanized environment. The findings also provide exploratory methodological evidence that could lead to an approach for predicting the change in commute stop-making propensity as a result of the changes in sociodemographic conditions and transportation-land use scenarios.
Abstract