Home | Business News | Browse by Publication | M | Marketing Science

Structural modeling in marketing: some future possibilities.

Article Excerpt
The growing interest in using structural modeling to test theories of consumer and firm behavior stems from the increased availability of market data at the individual level (both longitudinal and cross-sectional) for a variety of products. The emergence of online consumer panels, where data on behavioral intentions can be collected, and the availability of retail transactional data from online stores can be expected to only accelerate the trend. Hence, the recent review article by Chintagunta et al. (2006) is a timely contribution to the structural modeling, along with other recent contributions by Wittink (2005), Franses (2005a, b), Van Heerde et al. (2005), and Bronnenberg et al. (2005).

Chintagunta et al. (2006) and the other authors mentioned discuss the state of the art in the development and application of structural models in marketing. They discuss the advantages and disadvantages of using structural modeling to test marketplace theories and the cautions to be observed in the validation of structural models. The recommendations they offer suggest the potential for further use of structural models in marketing. The purpose of this commentary is to take some of the recommendations offered by Chintagunta et al. (2006) and use them to discuss some future possibilities.

First, there is the issue of how much behavioral theory should be included in a structural model. At one extreme, no data are needed to develop a theoretical model (Moorthy 1993). In fact, data could be considered a "distraction" or even irrelevant (Shugan 2002a). Theoretical models that do not rely on data are developed from both behavioral (i.e., substantive) assumptions and those made for analytical tractability (Moorthy 1993). With that distinction in mind, theoretical models could be adequate to provide a preliminary "test" of behavioral theories (Shugan 2002b). Likewise, a theoretical supermodel with economic primitives that are invariant to strategy/policy change could be used to initially evaluate alternative (i.e., competing) strategy/policy regimes that are contained within it (Moorthy 1993). The nested structure of...

View this article FREE - Now for a Limited Time, try Goliath Business News
Free for 3 Days!