Development and application of performance improvement verification model: a case study of an e-learning system

Many researchers have employed fuzzy linguistic scales to measure customer fuzzy perceptions and increase the reliability of their research. However, complex response methods may affect the precision of responses and the willingness of respondents. Thus, based on the concepts of distance and probability, we propose a simple response method for fuzzy linguistic scales to address these issues. In addition to a performance evaluation matrix to quickly identify items in need of improvement, we included order statistics and nonparametric statistics in a performance improvement verification model that solves the problems posed by sample size differences in pre- and post-improvement. This model is also applicable when the population does not follow a normal distribution. Finally, we used a case study with a CALL system to demonstrate the application of the proposed model, which fulfils the spirit of continuing improvement in total quality management.