This paper explores a number of methodological issues related to the econometric analysis of stated preference data in the context of estimating the value of automobile travel time. Estimates of parameters and the willingness to pay (WTP) to save time are obtained using conventional ordered probit and rank-ordered logit models and an innovation called mixed logit. We find that the average WTP is low and does not exhibit much variation among motorists. Although our findings using data on respondents' rankings of alternatives are robust, we find that caution should be used in estimating stated preferences based on respondents' ratings.