Due to the problems arising with the application of traditional Likert scales and to the methodological issues when interpreting the results, the concept of fuzzy scales is increasingly applied in service quality context by adding properties to Likert scales with the aim to model human judgement and thinking more precisely and reliably. Following this initiative, the primary aim of this paper is to introduce a fuzzy number based methodology in order to enhance the reliability of traditional Likert scale based evaluations associated with the measurement and evaluation of service quality, particularly in higher education context. The performance of services also varies as time goes which is considered as well when developing this novel methodology. In order to handle these issues, the linguistic variables expressing customers' (i.e. students') satisfaction are considered to be fuzzy sets with sigmoid-shaped membership functions. In our approach, fuzzy numbers are composed of two sigmoid functions, which are conjoined by Dombi's intersection operator. By utilizing the principles of Dombi's Pliant Arithmetics, descriptive statistical analyses can be easily conducted on the collected data. The intersection of two sigmoid functions can also be considered as a membership function expressing the usually strongly nonlinear relationship between service performance and customer satisfaction. In order to be able to perform that the linguistic variables 'completely dissatisfied', 'dissatisfied', 'rather dissatisfied', 'rather satisfied', 'satisfied', 'completely satisfied', which are widely used expressions on traditional Likert scales, are linked to the service performance level of the organization. As a result, the sigmoid-shaped membership function is able to express the truth of the statement that a given level of service performance belongs to a certain linguistic variable expressing human judgement. This property allows the management of service providers to assess how customer satisfaction relates to the offered level of performance and thereby allows the identification of those aspects, where further improvements of performance attributes could enhance customer satisfaction. The practical benefit of the proposed methodology is demonstrated through a higher education example of a student survey applied to evaluate specific service quality issues in the case of project work courses. The Department of Management and Business Economics at the Budapest University of Technology and Economics has introduced a fuzzy number based survey containing altogether 26 statements focusing on service quality attributes associated with the supervision process of project work courses. Students are provided the opportunity to evaluate not only the supervisor's performance on a fuzzy number based scale, but also the importance of each statement is rated as well. In this study, the application of the proposed methodology is demonstrated for five statements in case of which the addressed quality issues have been given the highest importance scores during the previous semesters. After assigning the linguistic terms to the performance level in each of the examined statements, it can be concluded that the distance between the consecutive levels of satisfaction is not the same as assumed when utilizing a Likert scale based evaluation. In addition to that, the performance level where satisfaction reaches its next level is highly dependent on which service quality attribute is to be evaluated; up to the half of the range of performance, however, respondents are usually `completely dissatisfied' with the perceived performance. These findings demonstrate well how much information is lost or distorted when applying traditional Likert scales to evaluate service quality. Therefore, it is to be demonstrated that fuzzy numbers are able to deal with the uncertainty or ambiguity of human judgement by expressing the variation of performance as time goes on and by being able to characterize the mainly nonlinear relationship between organizational performance levels and degrees of customers' satisfaction even in those cases, where the level of customer satisfaction associated with the experienced level of performance varies when investigating different service quality attributes. The main novelty of the introduced methodology is that fuzzy sets are determined based on customers' (students') expectations, while former studies in the literature have defined previously the parameters of fuzzy sets representing the corresponding linguistic terms.
