A computational approach to evaluating curricular alignment to the united nations sustainable development goals

The United Nations (UN) considers universities to be key actors in the pursuit of Sustainable Development Goals (SDGs). Yet, efforts evaluate embeddedness SDGs university curricula tend rely on manual analyses curriculum documents for keywords contained sustainability lexica, with little consideration diverse contexts such keywords. efficacy these efforts, relying expert co-elicitation both subject-matter and sustainability, suffers from drawbacks associated keyword searches, as limited coverage concepts, difficulty extracting intended meaning potential greenwashing through “keyword stuffing.” This paper presents a computational technique, derived natural language processing (NLP), which develops lexicon root (RKs) relative importance by adapting Term Frequency–Inverse Document Frequency (TF-IDF) method corpus documents. Identifying RKs module/course descriptors offers basis evaluating 5,773 modules university's using classification criteria provided Association Enhancement Sustainability Higher Education's (AASHE). Applying this our analysis found 286 (5%) “sustainability focused” further 769 (13%) inclusive,” appear address 1, 17, 3, 7, 15. Whilst technique does not exploit machine learning methods applied large amounts trained data, it is, nevertheless, systemic evolutive. It, therefore, an appropriate trade-off, faculty analytics skills can apply. By supplementing existing approaches curriculum, developed contribution benchmarking curricular alignment SDGs, facilitating pursue meaningful enhancement, whilst complying reporting requirements. is useful first-pass any portfolio. Further testing validation offer avenue future design-science research.