Investigating relationships within the Index of Learning Styles: a data driven approach

Learning styles are incorporated more and more in e-education, mostly in order to provide adaptivity with respect to the learning styles of students. For identifying learning styles, at the present time questionnaires are widely used. While such questionnaires exist for most learning style models, their validity and reliability is an important issue and has to be investigated to guarantee that the questionnaire really assesses what the learning style theory aims at. In this paper, we focus on the Index of Learning Styles (ILS), a 44-item questionnaire to identify learning styles based on FelderSilverman learning style model. The aim of this paper is to analyse data gathered from ILS by a data-driven approach in order to investigate relationships within the learning styles. Results, obtained by Multiple Correspondence Analysis and cross-validated by correlation analysis, show the consistent dependencies between some learning styles and lead then to conclude for scarce validity of the ILS questionnaire. Some latent dimensions present in data, that are unexpected, are discussed. Results are then compared with the ones given by literature concerning validity and reliability of the ILS questionnaire. Both the results and the comparisons show the effectiveness of data-driven methods for patterns extraction even when unexpected dependencies are found and the importance of coherence and consistency of mathematical representation of data with respect to the methods selected for effective, precise and accurate modelling.