Integrating Conjoint and Maximum Difference Scaling Data
提出一个模型，整合联合分析和最大差异缩放数据，以识别所有产品特征的成分效用，并用外部商品作为共同参考水平，结合满意度数据发现市场机会，以大型电视调查数据为例。
Customer preferences for product features play an important role in designing successful goods and services. Preferences for features are typically obtained by utilizing a model of choice where the utility for all but one level of an attribute is estimable. That is, the traditional discrete choice model can provide information on the change in utility between attribute-levels, but cannot separately estimate the utility associated with all levels of an attribute. In this paper, we propose a model that integrates conjoint and Maximum Difference scaling data to identify part-worth utilities for all product features, using the outside good as a common reference level, instead of the usual practice of having a reference level for each product attribute. The preference data are also integrated with satisfaction data to identify market opportunities for new and existing products. We illustrate our model with data from a survey measuring customer satisfaction and preferences for large-screen TVs. This paper was accepted by Raphael Thomadsen, marketing. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2023.02560 .