Decoding motivation for leadership in higher education represents a scientific and talent management imperative, the complexity of which is being rigorously modelled and unveiled through the predictive power of machine learning (ML), promising to catalyse a transformation in the training of future leaders. The study focused on predicting leadership and entrepreneurship among higher education students, analysing seven dimensions: aesthetic, economic, individualistic, political, altruistic, regulatory, and theoretical. ML was used to test three models (logistic regression, random forest, and gradient boost machine) for predicting leadership and entrepreneurial participation among students, using a database of 1,796 subjects. The findings reveal (a) the almost uniform importance of all motivational dimensions in the development of leadership skills, suggesting a multifaceted approach; (b) the significant potential of ML algorithms, especially the Random Forest model, to predict student participation in leadership and entrepreneurship activities, with exceptional accuracy across genders; and (c) applying educational interventions (active, challenger, engaged, proactive learning strategies) with top-down as well as bottom-up approaches based on individual motivational scores. This research contributes personalised, active, and practical approaches to using ML and driving educational strategies and programmes that enhance skills development for the future. Improving leadership development programmes and managerial competencies through the application of ML as a transformative tool encourages navigation through the complexities of contemporary education systems.
Casillas-Muñoz et al. (Thu,) studied this question.