The Optimal Design of Choice Experiments That Incorporate Ties
Stephen Bush
School of Mathematical Sciences, University of Technology, Sydney, Broadway, NSW, Australia

In this presentation we consider the design and analysis of stated choice experiments that give the respondent the opportunity to indicate that two or more alternatives are equally attractive. We consider an extension to the model introduced by Davidson (1970), for paired comparisons, to allow for an arbitrary number of alternatives in each choice set. We then discuss the connection between the optimal designs for models that do not incorporate ties and the optimal designs for models in which ties are allowed.

In a choice experiment, we present a series of choice sets to the respondent sequentially. Each choice set consists of m options, each of which describes a product or state, which we generically call an item. Each item is described by a set of attributes, the features that we are interested in measuring. Respondents are asked to select the most preferred item in each choice set. We then use the multinomial logit model to determine the importance of each attribute.

Choice experiments are widely used in transportation, marketing, health and environmental research to measure consumer preferences. From these consumer preferences, we can calculate willingness to pay for an improved product or state, and hence make policy decisions based on these preferences.

Keywords: Choice Experiment; Experimental Design; Multinomial Logit Model; Davidson Ties Model

Biography: Stephen Bush is a lecturer in Statistics at the University of Technology, Sydney, where he has been a member of faculty for two years. His main area of research interest is in the efficient design of experiments for a variety of nonlinear models. He is also interested in the design and analysis of experiments that measure preference behaviour in individuals.