Revealing Vocational Training on Achieving UN’s Sustainable Development Goals: Analysis Through Machine Learning

Vocational training plays a crucial role in supporting the achievement of the United Nations Sustainable Development Goals (SDGs), outlined explicitly in SDG Targets 4.3, 4.4, and 4.5. However, there is a lack of comprehensive studies examining the teaching of knowledge in state-level vocational training programs to support the attainment of SDGs. The primary objective of this study is to investigate the connection between SDG education and vocational training. To achieve this, we analyzed the curricula of i) four vocational training courses and ii) three applied technological and applied studies courses adopted by the government of New South Wales in Australia. The classification was based on a public training dataset from OSDG and subject descriptions via logistic regression (LR) and a generative pre-trained transformer (GPT) model. The findings from the subject-level analysis demonstrate the effectiveness of the adopted approach. Across all curricula, SDG 9 is the most prominently incorporated SDG. However, policymakers should be aware of the limited SDG representation related to social equality in vocational training. To evaluate the classification's performance, the authors have also manually classified each module of a course. While there is substantial agreement between human reviewers, the agreement between human reviewers, LR and GPT approach is only fair, indicating less consistency in the SDG classifications between human, LR, and GPT assessments.