Robust conjoint analysis

Beginning with data collection, Robust Conjoint Analysis demonstrates quite a few comparative advantages over other conjoint methodologies. A computer-aided graphical survey collects information through a user-friendly interface. Compared to a numerical or quantitative approach in data collection, the graphic interactive approach reflects an easier assessment of the trade-offs among choices. The ability of dynamically updating information that Robust Conjoint Analysis possesses during the data collection procedure eases limitations placed on the cognitive burdens of survey respondents. Our experience has shown that in many instances, as data collection progresses, a quick scan and acceptance of an interaction graph is all that is required of the respondent in the data collection procedure. Furthermore, when the computer-aided data collection procedure is not feasible, an alternative table format of data collection procedure can be used in traditional survey fashion. Another significant advantage of Robust Conjoint Analysis is that, unlike other models which pre-include or pre-exclude interactions among attributes, the Robust Conjoint Analysis procedure is able to detect whether or which interactions are to be included in the analysis model automatically. Similar to what many hybrid models attempt to achieve, two extensions of Robust Conjoint Analysis offer options for introducing the benefits of the self-explicated method and the full-profile method. Unlike hybrid models that combine the two methods in methodology formulation, these two extensions are not essential in the model formulation but rather serve the functions of reducing respondent efforts in data collection and increasing model accuracy. The empirical comparison study we include in this research shows that Robust Conjoint Analysis provides preference models which predict respondents\u27 choice behaviors more accurately than preference models of self-explication and full profile models. Outcomes of the validity comparison analysis show that predictions made by Robust Conjoint Analysis are more closely matched to the actual results both in the first choice prediction and preference ranking among choices. Robust Conjoint Analysis provides a method for overcoming many obstacles or assumptions encountered by other conjoint analysis methods. We expect the introduction of Robust Conjoint Analysis to contribute to the body of research on conjoint methods by offering an alternative approach, which also allows enhancement via the inclusion of other conjoint analysis methods