论文部分内容阅读
不同背景下因调查量表和评价标准的影响而导致的顾客评价的变化分为三类:量表变化、基准变化和结构变化。合理的CSI计算方法需要准确地反映结构变化,不反映量表变化和基准变化。本文采用蒙特卡罗模拟方法比较了两种计算方法在三种变化下的CSI结果。模拟结果表明:通用方法不会反映量表变化,能灵敏地反映基准变化,同时准确地反映结构变化;改进方法不会反映量表变化和基准变化,而是能够准确地反映出结构变化。因此,改进方法优于通用方法。同时,当不同品牌的CSI分值较接近时,CSI的排序不稳定;而当不同品牌CSI分值差异较大时,排序具有稳定性。
Changes in customer reviews due to the impact of survey scales and evaluation criteria in different contexts fall into three categories: scale changes, baseline changes, and structural changes. Reasonable CSI calculation methods need to accurately reflect the structural changes, does not reflect the scale changes and benchmark changes. In this paper, the Monte Carlo simulation method is used to compare the CSI results of the two methods under the three changes. The simulation results show that the common method can not reflect the change of scale, can reflect the baseline change sensitively, and accurately reflect the change of structure at the same time. The improved method can not reflect the change of scale and reference change, but can accurately reflect the change of structure. Therefore, the improvement method is superior to the general method. Meanwhile, when the CSI scores of different brands are closer, the ranking of CSI is unstable; and when the CSI scores of different brands are quite different, the ranking is stable.