Text Mining and Variants of Correspondence Analysis for analysing written Italian of University students

This paper aims at investigating the textual similarities/dissimilarities of the written Italian used by university students of different departments of North, Centre and South Italy. The students’ text data are part of a large survey dataset, collected from the University of Bologna aiming at analysing an alleged decline in Italian language and at highlighting peculiar linguistic features of the Italian language used by university students. The text data comes from a sample of 2159 participants belonging to different departments of Italian universities. Here we focus on studying the association between the written production of semiformal texts of Italian students and the University geographical area (North, Centre and South of Italy) through a non-symmetrical variant of simple correspondence analysis.