Where academic tradition
meets the exciting future

Visual Conjoint Analysis (VCA): A Topology of Preferences in Multi-Attribute Decision Making

Peter Sarlin, Shahrokh Nikou, József Mezei, Harry Bouwman, Visual Conjoint Analysis (VCA): A Topology of Preferences in Multi-Attribute Decision Making. Quality and Quantity , 1–21, 2014.

http://dx.doi.org/10.1007/s11135-014-9992-z

Abstract:

This paper proposes an approach denoted visual conjoint analysis (VCA). Conjoint analysis is commonly used in marketing to understand consumers’ decision criteria, particularly why consumers prefer and select certain products and their variations. Yet, little efforts have been made to provide visual means for exploring and visualizing preferences and utilities of consumers. In this paper, we propose an approach that enables identifying a low-dimensional topology of consumer profiles and their demographic characteristics. Through a two-step approach, VCA makes use of techniques for (i) data reduction and (ii) dimension reduction in combination with conjoint analysis. It provides a two-dimensional representation (dimension reduction) of a small number of respondent segments (data reduction). This provides means for two key tasks: (i) identifying the topology of multivariate respondent profiles in a lower dimension, focusing on neighborhood relations, and (ii) visual representations of information describing the respondent profiles, as well as the combination of the two tasks. The approach is applied to a real-world case of consumers’ preferences of mobile platform ecosystems.

BibTeX entry:

@ARTICLE{jSaNiMeBo14a,
  title = {Visual Conjoint Analysis (VCA): A Topology of Preferences in Multi-Attribute Decision Making},
  author = {Sarlin, Peter and Nikou, Shahrokh and Mezei, József and Bouwman, Harry},
  journal = {Quality and Quantity},
  publisher = {Springer},
  pages = {1–21},
  year = {2014},
  keywords = {Conjoint analysis, Cluster analysis, Data reduction, Dimension reduction, Visual conjoint analysis},
}

Belongs to TUCS Research Unit(s): Institute for Advanced Management Systems Research (IAMSR)

Publication Forum rating of this publication: level 1

Edit publication