Predicting Joint Choice Using Individual Data

  • Authors:
  • Anocha Aribarg;Neeraj Arora;Moon Young Kang

  • Affiliations:
  • Stephen M. Ross School of Business, University of Michigan, Ann Arbor, Michigan 48109;Wisconsin School of Business, University of Wisconsin--Madison, Madison, Wisconsin 53706;Wisconsin School of Business, University of Wisconsin--Madison, Madison, Wisconsin 53706

  • Venue:
  • Marketing Science
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Choice decisions in the marketplace are often made by a collection of individuals or a group. Examples include purchase decisions involving families and organizations. A particularly unique aspect of a joint choice is that the group's preference is very likely to diverge from preferences of the individuals that constitute the group. For a marketing researcher, the biggest hurdle in measuring group preference is that it is often infeasible or cost prohibitive to collect data at the group level. Our objective in this research is to propose a novel methodology to estimate joint preference without the need to collect joint data from the group members. Our methodology makes use of both stated and inferred preference measures, and merges experimental design, statistical modeling, and utility aggregation theories to capture the psychological processes of preference revision and concession that lead to the joint preference. Results based on a study involving a cell phone purchase for 214 parent-teen dyads demonstrate predictive validity of our proposed method.