Experimental analysis has a long tradition in market research. Conjoint analysis and discrete choice models have been frequently used for measuring and analyzing consumer preferences. In the last 40 years, conjoint analysis has been applied with great success, in particular in new product development, competition analysis, pricing research, and benefit segmentation applications. Most of the experimentation in market research involves the design of product profiles on the basis of product attributes specified at certain levels, and requires respondents to repeatedly rate/rank all profiles.

A large strand of methodological research has increased the sophistication of conjoint analysis. In particular, the benefit segmentation purpose of conjoint analysis has further challenged new developments in latent class modeling (e.g., DeSarbo et al., 1992; Kamakura et al., 1994). Furthermore, the multilevel structure of the data has been taken into account in latent class modeling by allowing covariance between different measurements from the some respondent (e.g., Qu et al., 1996). This research discusses different parameterizations of conjoint analysis within a multilevel latent class framework. The conjoint analysis application is focused on the assessment of a professional service by 163 respondents on 12 cards defined on 5 attributes.

References:

Desarbo, W.S., Wedel, M., Vriens, M., Ramaswamy, V. (1992), Latent class metric conjoint analysis, Marketing Letters, 3(3), 273-288.

Kamakura, W. A., Wedel, M., Agrawal, J. (1994), Concomitant variable latent class models for conjoint analysis, International Journal of Research in Marketing, 11(5), 451-464.

Qu, Y., Tan, M., Kutner, M.H. (1996), Random effects models in latent class analysis for evaluating accuracy of diagnostic tests, Biometrics, 52, 797-810.

Keywords: Experimental analysis; Conjoint analysis; Latent class analysis; Multilevel analysis

Biography: José G. Dias is an associate professor at ISCTE - Lisbon University Institute, Lisbon (Portugal). He holds Master Science degrees in Management (ISCTE) and Population Studies (Univ. of Groningen, The Netherlands). He received is PhD degree in Economics (Econometrics) from the University of Groningen, The Netherlands. His main research interests are connected with the application of Statistics and Econometrics in Management (in particular Market Research problems) and Population Studies.