Diagnosis of Interactive Conjoint Analysis Based on Social Comparison

Diagnosis of Interactive Conjoint Analysis Based on Social Comparison

Dennis Castel (Department of Computer Science and Intelligent System, Osaka Prefecture University, Osaka, Japan), Fei Wei (Department of Computer Science and Intelligent System, Osaka Prefecture University, Osaka, Japan), Ryosuke Saga (Department of Computer Science and Intelligent System, Osaka Prefecture University, Osaka, Japan) and Hiroshi Tsuji (Department of Computer Science and Intelligent System, Osaka Prefecture University, Osaka, Japan)
Copyright: © 2014 |Pages: 16
DOI: 10.4018/ijkss.2014040102
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

To help non-professional background respondents to get awareness from complicated calculations of conjoint analysis, this paper presents overview and functions of preference diagnosis, transforming results of conjoint analysis from tacit knowledge to explicit knowledge. This process also leads to the creation of new knowledge. In addition, this paper also discusses a design of diagnosis model by positioning on individual, groups and clusters. A simple example is also discussed. The proposal includes two types of strategy: by-individual feature, a diagnosis generated in the result of conjoint analysis for individual and by-social feature, a diagnosis based on the comparison with analytical result of other respondents.
Article Preview

1. Introduction

Marketing researchers have made great use of the conjoint analysis to estimate the impact of selected product (or service) characteristics on consumer preferences (Gustafsson, 2010; Sheng, 2008). The conjoint analysis allows to find the best combination of attributes and levels for new product, and to point out what customers are truly willing for product features. It became popular because the conjoint analysis is a low-price and flexible analysis method (Iyengar, 2012; Kotri, 2006; Lohrke, 2010).

In the case of complex combination of product, for example when the number of attribute for product is over five, the respondents could be confused; the precision of evaluation may drop significantly and the analysis results could not be consistent (Orme, 2010). Therefore, a diagnosis based on respondents’ results should be necessary in order to gives them a chance to re-consider their evaluation, allowing them to accept or correct for their responses.

Moreover, for the marketing researcher, they need a feedback from respondents to control the validity and reliability of conjoint analysis. Based on different evaluated product, conjoint analysis should be applied not only as a market simulation method, but also as a tool based on SECI model which can help respondents to lean and create knowledge (Felfernig, 2008; Gourlay, 2004).

At first, we will explain some terms of the evaluation process and review the traditional conjoint analysis. Then, a proposed interactive conjoint analysis by web-based questionnaire system will be described. There are two remarkable features in interactive conjoint analysis. Firstly, in comparison to the traditional conjoint analysis, it is easier to evaluate the respondents’ preferences for product or service. Secondly, it can reduce the probability of inconsistent results of analysis. To implement diagnosis system based on interactive conjoint analysis, we develop a knowledge-based system named CASIMIR. It will be introduce in the last chapter.

To transform the acquired information to knowledge based on Data Information Knowledge Wisdom chain (Buchanan, 1998) and optimize conjoint analysis results, we suggest two types of dynamic strategy: by-individual feature, a diagnosis generated in the result of conjoint analysis for a single respondent, and by-social feature, a diagnosis which is based on the comparison with analytical result with other respondents result.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 8: 4 Issues (2017): 3 Released, 1 Forthcoming
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing