In this paper we demonstrate a modeling approach that can be used to better understand the use of safety belts in single- and multi-occupant vehicles, and the effect that vehicle, roadway and occupant characteristics have on usage rates. Using data from a roadside observational survey of safety-belt use in Indiana, a mixed (random parameters) logit model is estimated. Potentially interrelated choices of safety-belt use by drivers and front-seat passengers are examined. The approach we use also allows for the possibility that estimated model parameters can vary randomly across vehicle occupants to account for unobserved effects potentially relating to roadway characteristics, vehicle attributes, and driver behavior. Estimation findings indicate that the choices of safety-belt use involve a complex interaction of factors and that the effect of these factors can vary significantly across the population. Our results show that the mixed logit model can provide a much fuller understanding of the interaction of the numerous variables which correlate with safety-belt use than traditional discrete-outcome modeling approaches.