About AMA

Executive Summary
The complexity of many products is almost constantly increasing. The purchase of a new car or the booking of a vacation package is usually the outcome of a comprehensive decision-making process that takes into account many aspects of the available product alternatives. To understand and predict these purchase decisions, preference measurement approaches are needed that accommodate large numbers of attributes without overburdening the respondents with respect to cognitive load and interviewing length.

This article stimulates and expands on the substantial discussion on preference measurement approaches for complex products by proposing a modified version of the analytic hierarchy process, which is based on the collection of paired comparisons of attributes and attribute levels and thus is called paired comparison-based preference measurement (PCPM). Because of the compositional character of PCPM, the requirements of stimulus design can be kept at a minimum, which has been emphasized as a significant benefit of self-explicated approaches. However, in contrast to the latter, PCPM’s paired comparisons facilitate the identification and handling of measurement errors.

It is universally accepted that no approach dominates in all situations or in all scopes of application. Each of the popular methods has its strengths and weaknesses. Thus, the authors theoretically and empirically compare the new approach to ACA (adaptive conjoint analysis) and CASEMAP (computer-assisted self-explication of multi-attributed preferences), which are two of the most established preference measurement techniques for complex, high-involvement products. The analyses focus on the theoretical assumptions of the three approaches, particularly pertaining to the underlying utility model, levels of measurement, anchor points, and treatment of measurement errors. Using two between-subject studies (summer vacation packages and cell phones), the authors investigate the respective effects on interviewing length, individual hit rates, and aggregate choice share predictions.

In both studies, PCPM yields significantly higher predictive accuracy in view of the individual hit rates. Moreover, PCPM is better or at least on par with ACA and CASEMAP with respect to aggregate choice share predictions. The available results also show that PCPM significantly reduces the survey length compared with the benchmark approaches.

In-depth analyses reveal that two aspects in particular of PCPM foster the beneficial results: PCPM elicits more distinctive attribute importances, and it levels out substantial inconsistencies in the respondents’ responses. Overall, both studies provide empirical support for the predictive validity and managerial usefulness of PCPM in understanding preferences for complex products.

Biography
Sören Scholz is Lecturer of Marketing in the Department of Business Administration and Economics at Bielefeld University, Germany. He received a master’s degree in Business Administration from the same university. His recent research interests include environmental scanning, the analysis of consumer preferences, and managerial decision making.

Martin Meissner is Lecturer of Marketing in the Department of Business Administration and Economics at Bielefeld University, Germany. He obtained a master’s degree in Business Administration from Bielefeld University. His research focuses on consumer decision behavior.

Dr. Reinhold Decker is Full Professor of Marketing in the Department of Business Administration and Economics at Bielefeld University, Germany. He holds a Master’s degree in Industrial Engineering and a PhD in Marketing from the University of Karlsruhe. His research focuses on model-based decision support in marketing, data mining, and online preference measurement and has been published in various scholarly journals and conference proceedings. He was a visiting professor at the Moscow Academy of Economics, the New University of Lisbon, and the University of Vienna, and he currently serves on the editorial board of Review of Managerial Science.

Member Comments (0):

To rate or comment on articles, you must be a logged in AMA member. Click here to join