Text

When absence begets inference in conjoint analysis

---

An enduring issue in the development and use of conjoint analysis is consumer response to missing attributes in partial profiles. Academic research has failed to produce a consistent answer to this issue, and practitioners are more likely to view it as a nuisance factor. Bradlow, Hu, and Ho (2004) argue that inference making is common, follows a specified process, and when properly incorporated into the analysis, can enhance the predictive validity of the technique. The authors applaud the attempt to raise consciousness about the behavioral underpinnings of conjoint analysis, and they are intrigued by the specific model that Bradlow, Hu, and Ho advance. At the same time, however, the likelihood and substance of inference making is likely to vary as a function of the respondent's understanding of the task, the costs and benefits of engaging different imputation rules, prior beliefs about attribute relationships, and the way the beliefs are modeled. From a pragmatic perspective, the tractability and accuracy of proposed model will vary with its ability to represent attributes across a range of realistic levels. Nonetheless, progress toward a solution to the motivating problem requires formal attempts to address it. Bradlow, Hu and Ho offer a model that may stimulate additional research with respect to its assumptions and relative value compared with that of other, yet-to-be proposed alternatives. The authors identify additional behavioral topics that may enhance the psychological plausibility of conjoint analysis..Printed Journal

---

Ketersediaan

Call Number	Location	Available
	PSB lt.dasar - Pascasarjana	1

Penerbit	American Marketing Association.,
Edisi	-
Subjek	Models Market research Conjoint analysis
ISBN/ISSN	222437
Klasifikasi	-
Deskripsi Fisik	-
Info Detail Spesifik	-
Other Version/Related	Tidak tersedia versi lain
Lampiran Berkas	Tidak Ada Data