Open Conference Systems, STATISTICS AND DATA SCIENCE: NEW CHALLENGES, NEW GENERATIONS

Font Size: 
New fuzzy composite indicators for dyslexia
Isabella Morlini, Maristella Scorza

Last modified: 2017-05-22

Abstract


Composite indicators should ideally measure a multidimensional concept that cannot be captured by a single variable. In this paper, we suggest a method based on fuzzy set theory for the construction of synthetic indexes of dyslexia, using the set of metric variables measured by means of reading tests. These indexes allow us to include in the diagnosis of dyslexia different aspects of the reading performance, like the speed and the accuracy (the number of misspelled words).  A few criteria for assigning values to the membership function are discussed, as well as criteria for defining the weights of the variables. An application regarding the diagnosis of dyslexia in primary school in Italy is presented. In this application, the fuzzy approach is compared with the crisp approach actually used in detecting dyslexic children. The application shows that the proposed indices work well in identify the level of impairment of the students and the results are in agreement with the percentages of dyslexic identified with the traditional diagnostic criterion but give more insights and go beyond the rigid unrealistic partition between “dyslexic” and “not dyslexic” student. Indeed, the limit between a “bad” and a “pathological” performance in psychometric tests is somehow fuzzy