우세분석을 이용한 고객만족도 영향변수의 상대적 중요도 측정

Abstract

In order to improve customer satisfaction, it is essential to measure the predictor variables that actually influence customer satisfaction. Whenever multiple regression analysis is used to determine the relative importance of the predictor variables. Typically, the relative importance of predictors is assessed by simply comparing their standardized regression coefficients. However, when the predictors are correlated, which is generally the case in most the causal study in primary research, it is recognized that regression coefficients cannot be used to unambiguously explain variance shared by two or more predictors. The dominance analysis, originally pioneered by Budescu(1993), has been recognized as a better approach and used by many researchers in recent years to measure the relative importance of predictor variables. The purpose of this article is to call greater attention to the dominance analysis and suggest an application of it in customer satisfaction survey. In this article, we conducted a review of the literature and applied it in public transportation services. The results(the relative importance of predictors) obtained by the dominance analysis are compared with results obtained by multiple regression analysis.