OBJECTIVES: In past two decades, researchers have proposed to combine fuzzy theory into measurement in various areas. According to their studies, combining fuzzy theory could reduce the properties differences between measurement methods and human cognition, and the results collected by fuzzy scale(FS) were also superior to those by traditional measurement in both validity and reliability. Therefore, the purpose of this study tries to measure the quality of life(QOL) by FS, and examine its own psychometrical properties and comparability with Likert scale(LS). METHODS: WHOQOL-BREF Taiwan version was used to compare the results collected by FS and LS, with a set of data from 404 subjects in repeated experiment design. Cronbach’s alpha and hypothesis test of reliability coefficient were utilized to compare the reliability of WHOQOL-BREF in both two measurement. Confirmatory factor analysis(CFA) was used to examine the construct validity and measurement invariance (ME/I) between LS and FS at domain level.  RESULTS:  The results indicated that Cronbach’s alpha coefficients of FS were significant higher than that of LS in most domains. Moreover, CFA analysis showed that the equal intercept invariance model between FS and LS measurement were supported, all fit index performed well under the factor model at domain level,  CFI, RMSEA and SRMR increased very slightly after imposing the equal intercept constrain proposed by Meredith in 1993.  CONCLUSIONS: According to the above results, FS did improve QOL measurement in reducing the measurement error, and its construct validity was supported by CFA analysis. Besides, the ME/I analysis indicated that FS was still comparable with traditional LS in the lower measurement error of QOL. In future studies, researchers may use previous studies which incorporating fuzzy scale in helping them diagnose and differentiate psychiatric diseases as reference, and look into how the fuzzy scale QOL could be used in other medical fields.