Topic modeling as a technique for studying the institutionalization of service-learning in higher education

Service-learning in higher education has emerged as a significant area of research in recent decades. While this notable growth reflects the academic and scientific community’s interest in the subject, it also introduces new challenges for researchers navigating the field, as the vast universe of information requires more time and knowledge to identify relevant information. In response to these challenges, this document employs a Natural Language Processing (NLP) technique designed to extract data from an extensive corpus of documents and identify underlying or latent themes. The technique, known as topic modeling, was applied to 354 documents from the Scopus database to uncover hidden patterns and themes. The analysis revealed a model of institutionalization of service learning grounded in critical principles of social justice. This model prioritizes local, diverse, and complex realities, emphasizes the perspective and participation of key stakeholders; is legitimized by frameworks of social responsibility and sustainable development; and is directed towards cultivating citizens with critical awareness and social commitment.