The reliability of the fuzzy scale for assessing the responses of a person in an opinion questionnaire (which is usually treated as a Likert-type questionnaire) is analyzed.Thus, a comparison between a classical Likert scale-based questionnaire and its fuzzy rating scale-based version in terms of their reliability is established.For this purpose, some questions of the Grasha-Riechmann student learning style scales have been answered by a group of students using both scales.Considering that the Likert version of this questionnaire is reliable, the aim is to check the reliability of its fuzzy version by comparing the corresponding Cronbach's alpha coefficients.