Marco Vriens

Commercial use of conjoint analysis in Europe: Results and critical reflections

Abstract We report the incidence of conjoint analysis applications by European market research suppliers. Based on responses to a survey, we document about 1,000 commercial projects over a five-year period and show breakdowns by product category, study purpose, and other characteristics such as study design, data collection, and data analysis. The results are compared with information collected for an earlier time period in the United States. We also discuss recent conjoint research results, identify issues that warrant further study, and suggest how the current practice of conjoint analysis can be improved.

Latent Class Metric Conjoint Analysis

A latent class methodology for conjoint analysis is proposed, which simultaneously estimates market segment membership and part-worth utilities for each derived market segment using mixtures of multivariate conditional normal distributions. An E-M algorithm to estimate the parameters of these mixtures is briefly discussed. Finally, an application of the methodology to a commercial study (pretest) examining the design of a remote automobile entry device is presented.

Verbal versus Realistic Pictorial Representations in Conjoint Analysis with Design Attributes

Abstract The current generation of high-powered graphics software offers an effective means for presenting product designs. Armed with the right tools for generating photorealistic representations of alternative designs, product development teams can obtain useful consumer input about product design attributes. However, generating computer-based models carries greater costs than producing verbal representations (written, key-word descriptions). 1 If a verbal representation can effectively communicate the relevant design and styling attributes, can product developers justify the costs associated with generating a computer-based model? Marco Vriens, Gerard H. Loosschilder, Edward Rosbergen, and Dick R. Wittink highlight a fundamental question in the choice between verbal and pictorial representations 2 : Does the type of representation used affect the nature and the quality of the results that product developers obtain? Specifically, does the type of representation used in a study affect the information that the study provides about market segmentation and the relative importance of different design attributes? And does the choice of representation type affect a study’s reliability and predictive accuracy? To address these questions, the authors conducted a study with a European subsidiary of a Japanese manufacturer of car stereo equipment. The study involves the selection of product designs from those made available by the manufacturer. Respondents were asked to evaluate both verbal representations and photorealistic pictorial representations of proposed car stereo designs. Half the respondents evaluated the verbal representations first, while the other half rated the pictorial representations first. In this study, the pictorial representations produced higher relative importance ratings for two of the three design attributes, as well as somewhat greater heterogeneity (that is, segmentation) among respondents. However, the verbal representations produced greater predictive accuracy, especially for respondents who rated the verbal descriptions after they had evaluated the pictorial representations. These results suggest that the pictorial representations improved the respondents’ understanding of the design attributes, while the verbal representations seem to facilitate judgment.

Measuring perceived service quality using integrated conjoint experiments

Proposes the use of integrated conjoint experiments to measure perceived service quality. It also demonstrates the process of modelling the hierarchical relations between operationally defined service attributes, strategically relevant service dimensions, and overall preference for banks or banking products. The proposed method, which is based on hierarchical information integration theory, avoids some of the limitations and problems of SERVQUAL and traditional conjoint analysis. The approach is demonstrated with an application to retail banks involving four service dimensions and 28 attributes. Conclusions are drawn about which dimensions and attribute changes will yield the strongest improvements in a bank’s utility and competitive position. The paper ends with a discussion of topics for further research.

Solving marketing problems with conjoint analysis

Conjoint analysis has become a popular research tool after its introduction in marketing in 1971 by Green and Rao. Both the academic as well as the commercial community have shown a strong interest in the conjoint methodology. In the academic literature attention is either directed at methodological issues or is aimed at application issues. A considerable amount of academic research is now available, both in the field of marketing as well as in other disciplines. The methodological developments have been reviewed extensively, as well as the commercial use of conjoint analysis. In the academic literature no systematic evaluation of the developments in the area of applications has been given. In this article we aim to fill this gap in the literature. After a brief discussion of the basic concepts of conjoint analysis we discuss the range of marketing problems which can be addressed by conjoint analysis. Issues to which attention is addressed include among others: (1) strategic issues, (2) segmentation issue...

Ratings-Based versus Choice-Based Latent Class Conjoint Models

Traditional ratings- or rankings-based conjoint analysis has been very popular in commercial practice. Recently, the choice-based conjoint approach has become an attractive alternative for measuring preference structures. However, little is known about the extent to which both approaches produce similar results or about how they compare in terms of predictive accuracy. This paper presents a conceptual and empirical comparison of ratings-based and choice-based conjoint approaches. The authors add to previous empirical investigations by comparing both approaches with respect to differences in relative attribute importances and predictive accuracy while controlling for task order. In particular, the authors compare segment-level (latent class) models. The results show substantial differences in the segment-level relative attribute importances. These results are consistent with previous research and the prominence hypothesis. At the segment and aggregate level the choice-based approach clearly outperforms the ratings-based approach.

Labor-market behavior of long-term unemployed: A multidisciplinary approach

In the beginning of the eighties unemployment increased dramatically in the Netherlands. Since 1984 short-term unemployment has decreased, while at the same time long-term unemployment has continued to rise, absolutely as well as relatively. The issue of long-term unemployment is current and no solution seems to be in sight. Scientific research in the Netherlands on unemployment has put emphasis for a long time on the labor-market behavior of the unemployed. Labor-market behavior has in the 70s and 80s been investigated by researchers from various disciplines, starting from different notions of labor market behavior, applying various models and different methodologies. The disciplinary approach is therefore dominant, not only in the Netherlands but in other countries as well. In this article we have tried to elaborate a multidisciplinary model, in which economic, sociological and psychological approaches with respect to labor market behavior have been integrated into one conceptual framework. In order to do so, we have developed a non-recursive model of labor-market behavior. This model was estimated and tested applying a simultaneous estimation procedure (as implemented in the Linear Structural Relations, LISREL-program). Results show that the multidisciplinary perspective is not only theoretically appealing, but also has some empirical advantages.

Segmentation Procedures for Conjoint Models: A Monte Carlo Study of Performance

Benefit segmentation procedures in conjoint analysis group consumers according to the relative utilities attached to the various attributes. Traditionaly a two-stage procedure is used for benefit segmentation, in which conjoint models are estimated at the individual level, and estimated utilities are clustered to obtain segments. This two-stage procedure has a number of drawbacks, which were discussed by e.g. Wedel and Kistemaker (1989). One of the most important flaws is its failure to optimize some measure of the accuracy with which preferences are predicted within segments. The methods proposed by Hagerty (1985), Kamakura (1988), DeSarbo, Oliver and Rangaswamy (1989), Wedel and Kistemaker (1989) and Wedel and Steenkamp (1989) were developed to overcome this flaw and use simultaneous estimation and segmentation. Comparisons of the methods to date have been limited to a few studies on empirical data, while the performance of the methods may vary under different conditions. In this study we investigate the relative performance of these methods and the two-stage procedure, under a variety of conditions, in a Monte Carlo study. In the synthetic data sets the following five factors are systematically varied: 1) the number of subjects, 2) the number of degrees of freedom for estimation at the individual level, 3) the number of segments, 4) the error in the preferences and 5) the distribution of partworths within segments.