The study investigated the effects of income and vehicle ownership on household mobility (measured as trip frequency, miles traveled, and transit use) across states on the basis of the 2017 National Household Travel Survey. Bayesian multilevel (or random-effect) regression models were developed to include state-policy variables and overcome the concern of small sample sizes in some household strata. The analysis results indicated that household income levels were positively associated with vehicle ownership and mobility; however, extremely high-income households were not necessarily more likely than high-income households to own more vehicles. Owning at least one vehicle was the norm for most households, except under two conditions: when the household income was extremely low or when the state transit level of service (LOS) was high. Moreover, states with a high transit LOS exhibited similar household mobility as long as households had similar income levels and vehicle ownership rates; by contrast, household mobility in states with a low transit LOS was relatively varied. Fully equipped (at least one vehicle per driver) and car-light households (having more drivers than vehicles) exhibited a similar trip frequency number; however, the vehicle miles traveled of car-light households were significantly shorter, especially in states with a high transit LOS. These results suggest that the mobility benefits of being fully equipped are more limited than we had anticipated. Transit services can affect household vehicle ownership and mobility only when being provided above a certain LOS.
- MULTILEVEL MODELS
- CAR OWNERSHIP