Note: The original seminar was held in Italian. Here you can find a translation of the abstract.
The present work is part of the “Plan for Science Degrees” (Piano Lauree Scientifiche, PLS), aiming at presenting and discussing the potential of the analysis of complex data to high school seniors and university freshmen. In particular, the focus has been on the analysis of textual data, a field that has seen an exponential development in recent years, also from the spread and relevance assumed by social media and social networks. In this paper we introduce text mining methods, as presented during the seminar, and the main results of an experiment carried out during the meeting, in which audience was an integral part of the study. Initially, we showed how to perform a statistical survey with which it was possible to obtain the responses of individual students, then we processed the results. A concluding discussion was carried out on the validity of the instruments and the interpretation of the results, which is a non-trivial issue task the complex nature of the data. The aim of the seminar was to show the methodological implications and the potentiality of statistical analysis on textual data, a theme widely dealt with in the degree courses of the Federico II: Statistics for Business and Society (“Statistica per l’Impresa e la Società ”, bachelor degree) and Statistics for Decision Making (“Scienze Statistiche per le Decisioni”, master degree) at the Department of Political Sciences.