UC Davis

This paper investigates the choices to engage in various activities while commuting. Using data collected through a survey of 2149 Northern California commuters, we develop binary logistic regression models of the decision to engage in each of eight types of activities on the commute: rest/sleep, use a laptop, use a smartphone, listen to audio, read printed materials, edit papers/ printed materials, read electronic documents, and edit electronic documents. Explanatory variables include socio-demographic characteristics and attitudinal and time use factors. We model the engagement in each type of activity while traveling by each of five different modes (bicycle, commuter rail, transit, shared ride, and drive alone) – 33 models in all (excluding impractical combinations). The results illuminate the individual-specific traits affecting commuters’ propensities to engage in activities while traveling. Those propensities exhibit large differences across modes. Longer commutes result in higher propensities to engage in almost all modeled activities for commuter rail and ridesharing. Age, gender, income, trip distance, education level, attitudes and preferences towards the adoption of technology, familial obligations, expectations about time use, and attitudes towards multitasking all affect the propensity to engage in activities while commuting.