• METHODESEditerSupprimer

    The Influence of Design Dimensions on Stated Choices in an Environmental Context2015

    MEYERHOFF J., OEHLMANN M., WELLER P.Journaux et Revues (scientifiques)

    analyse conjointe / dichotomique, consentement à payer, incertitudes / biais, préférences déclarées

    Environmental and Resource Economics
    July 2015, Volume 61, Issue 3, pp 385–407

    Discrete choice experiments are increasingly used in the context of environmental valuation. However, there is still little known about the influence of the complexity of the choice task on model outcomes. In this paper we investigate task complexity in terms of the design dimensionality of the choice experiment by systematically varying the number of choice sets, alternatives, attributes, and levels as well as the level range. We largely follow a Design of Designs approach originally introduced in transportation. First, we analyse the influence of the design dimensionality on participants’ dropout behaviour finding that the probability to drop-out of the survey is influenced by socio-demographic characteristics and increases with the number of choice sets, attributes as well as with designs having five alternatives. Second, we investigate the impact of the design dimensions on stated choices by estimating a multinomial logit model, and heteroskedastic logit models. Results show that the error term variance is influenced by socio-demographic characteristics as well as by all design dimensions. Moreover, we find that accounting for the impact of the design dimension on the error variance does not significantly change willingness to pay estimates.

  • Retour