Who Leads, What Matters? Machine Learning and the Complexity of University Performance

Abstract

The role of leadership in public institutions, particularly universities, is often linked to goal-setting and decision-making processes that impact efficiency. In Spain, public university rectors are directly elected by academics, staff, and students, offering a unique context for studying leadership influence. This study uses a unique database to analyze Spanish public universities across five categories: academic and research performance, social objectives, internationalization, university characteristics, and rector profiles. Using a K-Means unsupervised machine learning algorithm, we identify five distinct clusters of Spanish public universities, each characterised by a specific combination of institutional performance indicators and management characteristics.