Pursuing higher level education has become an increasingly more widespread ambition amongst young students. Moreover, many of them are not solely looking to continue their education but are also searching for opportunities to further their overall knowledge and professional skills. For this reason many university students leave home to study in other cities and Countries. Therefore, the necessities to find suitable accommodation services, a community with strong social bonds and opportunities to learn and develop new skills are strongly intertwined.

Consequently, dormitories and halls of residence are an option. In Italy, in addition to these accommodation options, there are also “Collegi di Merito” (merit-based residences), recidences recognized and accredited by the Italian Ministry of University and Reserach, which are a particularly attractive option, since they offer all of these benefits and services. Merit-based residences provide accommodation and formative services and demand that the students meet the merit requirements (ie: a certain GPA; attending cultural and training activities).

Collegio di Milano is a merit-based inter-university campus, initiative of Fondazione Collegio delle Università Milanesi, that hosts Italian and international students enrolled in the universities of Milan and admitted upon selection. In this context of an ever growing demand and the necessity to select those who are better aligned with the context and the requirements, it becomes of fundamental importance to properly optimize the selection process, making it as efficient and functional as possible.

The research aim to analyze the profiles of students residing at Collegio di Milano to understand what indicators point to a good fit with the specific context and then provide guidance on how to make the selection process more effective. The presentation will discuss the emerging profiles, highlighting potential innovative strategies on how to efficiently intercept candidates who seems to be less aligned with the residence’s requirements.

A TwoStep Cluster analysis was performed on a sample of 134 students who resided in the Collegio di Milano from 2020 to 2023 on the basis of their performance in the selection process (eg. solft skills profiles) and their ability of meeting the requirements, which is an essential part of living in Collegio di Milano. Afterwards, ANOVAs were carried out in order to underscore potential differences between groups in terms of performance in the other phases of the selection process.

The analysis showcases four different profiles: two are aligned with the requirements but have significantly different test results, the other two do not meet the requirements, but performed well in the test. Further analysis have underscored how only the first two groups are significantly different from each other.

This demonstrates how potentially unfit candidates are difficult to spot, as they are not particularly different from aligned candidates. However, distinct patterns in the soft skills self-evaluation score might actually be an indicator to understand the tuning of the candidate to the context of Collegio di Milano. Moreover, other competences, which are not currently being evaluated, could potentially be useful as well in identifying the correct profiles. Integrating this information with AI tools could highly optimize the first phase of the selection process by recognizing potentially fit and unfit candidates.

Keywords: University students, hall of residence, selection, assessment, Artificial Intelligence, evaluation.