Strength-of-Preference Indicators and an Ordered-Response Model for Ordinarily Dichotomous, Discrete Choice Data

Recent environmental regulations raise the importance of accurate, survey-based methods to estimate environmental values, while interviewing costs stimulate research to improve dichotomous choice survey methods. This paper proposes using strength-of-preference indicators along with dichotomous contingent response data. We model strength-of-preference indicators as quasi-cardinal measures using an ordered-response model. We find that strength-of-preference indicators provide substantial efficiency gains while increasing respondents' effort minimally, with evidence that interpersonal comparison of preference ratings induces no bias. The ordered response model of utility and willingness to pay is superior, even in predicting the binary choice results.