Predicting Adoption Choices Using Choice Probability Elicitation
(A2021-94683)

AUTHORS

Keyvan Dehmamy, Goethe University Frankfurt; Thomas Otter, Goethe University; Peter Kurz, bms marketing research + strategy; Günter Hitsch, Full Professor of Marketing at University of Chicago; Schum Laura, University of Groningen

ABSTRACT

Choice-based conjoint analysis is a widely used method to estimate consumer preferences for products and services that are not currently available in the market place from survey responses. In a standard conjoint design consumer report the preferred choice among a set of alternatives. This approach assumes that consumers face no uncertainty about their preferred choice even though the product choice does not occur at the time when the survey is taken but at some future point in time. We propose an alternative design that asks the subjects to state the choice probabilities for each possible product choice. This design allows the subjects to express their subjective uncertainty over the possible choices. Using conjoint data, we find that the preference estimates are significantly more precise when obtained using elicited probabilities unless the stated choice probabilities contain much measurement error.