The influence of first year behaviour in the progressions of university students

Advanced clustering techniques are used on educational data concerning various cohorts of university students. First, K-means analysis is used to classify students according to the results of the self assessment test and the first year performance. Then, the analysis concentrates on the subset of the data involving the cohorts of students for which the behavior during the first, second and third year of University is known. The results of the second and third year are analyzed and the students are re-assigned to the clusters obtained during the analysis of the first year. In this way, for each student we are able to obtain the sequence of traversed clusters during three years, based on the results achieved during the first. For the data set under analysis, this analysis highlights three groups of students strongly affected by the results of the first year: high achieving students who start high and maintain their performance over the time, medium-high achieving students throughout the entire course of study and, low achieving students unable to improve their performance who often abandon their studies. This kind of study can be used by the involved laurea degree to detect critical issues and undertake improvement strategies