Escala Likert difusa: um estudo sobre diferentes abordagens

Abstract

To analyze human attitudes is one of the key-points in Psychometric Scales. Being the most known and used, the Likert Scale is commonly presented as a five-point or four-point rating scale ranging from “strongly disagree” to “strongly agree”. The importance of the scale is based on the necessity to collect the ideas of certain groups for the decision making. Therefore, psychometric scales include the uncertainty that works on the subjectivity of individual answers. The Likert Scale brings some weakness to this evaluation process by losing information about the uncertainty of the answers provided. So, an approach of the Likert Scale incorporating the Fuzzy Sets Theory - considering the uncertainty and/or the ambiguity in the answer of the interviewed - make the discussion relevant. Towards that, in this paper are shown some literature strategies that seek to improve the Classic Likert Scale by the Fuzzy Sets Theory. In order to confirm if the Fuzzy Likert Scales are more satisfying than Classic Likert Scale, tests are fulfilled in the previous considerations seeing hypothetical and real cases, evaluating metrics such as: mean, standard deviation and agreement among the answers. So, bringing reliability to the areas that use the method in order to be more assertive on its analizes and considering best what the group studied indeed tried to show/inform.