Consumer choice behavior models have been extensively used to analyze individuals’ choice behavior in a variety of fields such as marketing, economics, and transportation. Previous consumer behavior studies have been based on the assumption that consumer choice decisions depend on objective alternative attributes. However, it is not accurate to limit description of consumer subjective behavior to discrimination between good and bad objective alternative attributes. There are two other important factors, subjective taste preference and psychometric constructs, which should be considered as well.

Previously, many researchers have studied objective alternative attributes, subjective taste preference and psychometric constructs separately. Because a framework of integrated models had not been developed, these factors have been incorporated in very few studies. However, choice behavior is more accurately described and forecast by considering these factors, rather than just the one factor (McFadden, 1986).

In this article, a framework of an integrated model is described, composed of objective alternative attributes, subjective taste preference, and psychometric constructs (see Figure 1). This framework is constructed as follows: Subjective taste preference can be treated as unobserved individual heterogeneity in the choice process (Ben-Akiva, Bolduc, & Bradley, 1993); psychometric constructs, such as consumer perceptions and attitudes, can be treated as latent variables    of factor analysis or structural equation models (Muthen, 1984), because such latent variable models have considerable potential ability to combine together choice behavior and psychometric indicators; objective alternative attributes can be dealt with by economic utility in discrete choice models, which are derived from probability forms of chosen alternatives under an assumption of utility-maximizing behavior by the consumer.

The proposed integrated model has considerable potential for use in marketing applications because of its ability to combine market choice data and psychometric data into unified descriptions of behavior and market forecasts. It is evident that consumer behavior is complicated. The development of this model has been an attempt to improve the behavioral representation of previous models by integrating methods that capture unobserved heterogeneity in the choice process, and that explicitly model behavioral factors such as perceptions and attitudes. Although this framework may not be the only method for integrating these factors and more investigation is required, it is an excellent starting point for meaningful exploration of consumer behavior.

References

Ben-Akiva, M., Bolduc, D., & Bradley, M. (1993). Estimation of travel choice models with randomly distributed values of time. Transportation Research Record, 1413, 88-97.

McFadden, D. (1986). The choice theory approach to market research. Marketing Science, 5(4), 275-297.

Muthen, B. (1984). A general structural equation model with dichotomous, ordered categorical, and continuous latent variable indicators. Psychometrika, 49, 115-132.

Ben-Akiva, M., Bolduc, D., & Bradley, M. (1993). Estimation of travel choice models with randomly distributed values of time. Transportation Research Record, 1413, 88-97.

McFadden, D. (1986). The choice theory approach to market research. Marketing Science, 5(4), 275-297.

Muthen, B. (1984). A general structural equation model with dichotomous, ordered categorical, and continuous latent variable indicators. Psychometrika, 49, 115-132.

Appreciation is due to anonymous reviewers.
This study was funded in part by a grant from the National Science Council

ROC (NSC 98-2410- H-244-001).

Yu-Chin Chen, Assistant Professor, Department of Information Management, Kao-Yuan University, No. 1821, Jhongshan Rd., Lujhu Township, Kaohsiung County 821, Taiwan, ROC. Phone: +886-7-6077777 ext 2412; Email: [email protected]