A dynamic approach of AHPSort II to classify stem master’s courses in Italy.

This paper proposes a dynamic AHPSort II for performance evaluation in public admin-istration by introducing a new methodological approach for determining the profiles use-ful for alternatives sorting. In particular, the paper analyzes the performance of master's graduates [1] in STEM (Science, Technology, Engineering, Mathematics) disciplines in Italy. Data are from the Almalaurea surveys on the employment status of graduates, re-lated to the period 2018-2022, and refer to graduates three years after graduation. The analysis considers nine indicators that account for academic performance, employment, and satisfaction for the course of study and for the current work. STEM disciplines in-clude 41 master's degree classes that correspond to over 600 courses. In the first step, information on individual master’s courses is analyzed and synthesized in order to identify any similarities among STEM disciplines and/or associations among the indicators examined. The results obtained by applying the Principal Component Anal-ysis show that degree courses with similar characteristics define groups mostly coinciding with the 4 STEM categories. In the second step, in order to sort the 41 master's degree classes taking into account the 9 indicators, we use AHPSortII [2], a multicriteria method that, among those proposed in the literature, allows handling a large number of alterna-tives and criteria. The procedure can be repeated or easily automated. In applying this method, contrary to what is provided in the literature [3] [4], we propose to determine the central (or limiting) profiles using thresholds defined as a function of tertiles; this pre-vents arbitrariness that represents the main weakness of the AHPSortII method. The method is applied with reference to each year of the period 2018-2022, then proposing for the first time a dynamic approach.