IPA分析法的修正及其在游客满意度研究的应用

The Modified Importance-performance Analysis Method and its Application in Tourist Satisfaction Research

传统的重要性绩效(IPA)分析法以受访者自述满意度和重要性评价来衡量各评价要素对提升客户满意度的意义。多数学者指出,这种分析方法存在两个弊端:一是重要性评价受满意程度的影响,二是单个评价要素的满意与否不一定能带来整体满意度相对称的变化,因此,单个评价要素的重要性也应当有所不同。所以,采用自述重要性的分析方式并不能反映客户的真实感受,分析方法需要改进。该研究以一山地景区的游客满意度为例,使用对数转化和偏相关系数法从满意度评价计算出引申重要性评价,去除满意度的影响,从而能较为客观地反映出游客对评价指标重要性感知。该研究对两种IPA分析方法进行了实证比较研究,并指出了结果存在差异性的原因,为完善使用IPA方法提供了有价值的理论阐述和实践检验。

Consumer satisfaction analysis is an important approach to examine business performances and operations. Many satisfaction analysis methods aim at identifying critical service performance factors that can effectively improve customer satisfaction levels and loyalty. Due to its simple and effective technique, importance- performance analysis （IPA） has been widely used in customer satisfaction research to prioritize influential factors and guide quality-based marketing strategies. The IPA method has two dimensions to attribute importance and performance （or satisfaction as in satisfaction research） , and generally uses an interviewee self-reported approach to acquire performance and importance data. While the traditional approach is convenient in data collection, and seemingly objective to reveal consumers＇ real perceptions towards the service, it neglects the assumptions of the IPA method. To correctly display the interaction between performance and importance, the IPA assumptions for conventional IPA method holds that attribute importance and attribute performance variables are independent, and the latter has a linear relationship with overall performance. However, these assumptions may largely be erroneous in the real world as using a self-reported approach for both dimensions may lead to paradoxical results. An individual＇ s importance perception is inevitably affected by his or her evaluation towards the same attribute performance, thus the relationship between customer attribute performance and overall performance is asymmetrical, and the relationship between attribute importance and attribute performance is causal. Consequently, the self- reported approach for importance evaluation cannot meet the assumptions, and the findings make the application of conventional IPA questionable. To eliminate performance＇ s effect on importance, this study adopts a modified IPA method that uses partial correlation coefficient as implicitly derived-importance. The partial correlation is generated in two steps. First, natural logarithmic transformation is applied to attribute performance to capture more sensitivity in the variables. Second, partial correlation coefficients are calculated between logarithmic attribute performance variables and the overall performance variable. As it the exogenous effects of alienates other variables＇ on the relationship between the overall satisfaction and the specified attribute performance, partial correlation coefficient reflects only the net association between the specified attribute performance and overall performance. Thus it can be a more accurate way to reflect the true importance of the same attribute. To empirically test the differences between the two IPA methods in tourist satisfaction research, this study conducts a tourist satisfaction survey using self- expressed question design at Jigongshan Mountain scenic area in the Spring of 2012. Data analysis results show a significant relationship between attribute satisfaction and performance, which verifies the conventional IPA＇s inadequacy in satisfying the method assumptions. When comparing the results from two IPA methods, the modified method reveals significant differences in the distribution of variables in the four vectors, which suggests that the referential information acquired from conventional IPA can cause decision makers to take incorrect actions when attempting to improve customer satisfaction. Comparing to the conventional method, the modified IPA approach effectively eliminates the latent subjective effects from individuals＇ satisfaction evaluation and also reduces the onus on interviewees. The modified IPA adds to the knowledge body of satisfaction research and is an applicable tool for practical management work.