Background Ensuring that educational outcomes translate into workforce readiness remains a major challenge, particularly in Technical and Vocational Education and Training (TVET). Increasing evidence suggests that vocational skill gaps originate from uneven foundational competencies developed during primary education. At the same time, the growing use of Artificial Intelligence (AI) in educational assessment raises concerns regarding transparency and interpretability. This study develops and empirically examines an Explainable Artificial Intelligence (XAI)–based competency assessment framework that bridges primary education outcomes and vocational skill readiness in a transparent and pedagogically meaningful manner Method A quantitative explanatory design was employed using competency data from 612 learners. Primary education competencies—literacy, numeracy, problem-solving, self-regulation, collaboration, motivation, and creativity—were modeled to predict vocational skill readiness. Multiple machine learning models were evaluated using 10-fold cross-validation. A Random Forest Regressor was selected based on superior predictive performance. Explainability was achieved using SHAP (Shapley Additive Explanations) to generate global and local interpretations of model outputs. Results The Random Forest model explained 76% of the variance in vocational skill readiness (R 2 = 0.76), outperforming baseline models. Global SHAP analysis identified problem-solving and self-regulation as the strongest predictors, followed by literacy and collaboration. Local SHAP decomposition revealed compensatory competency patterns, indicating that strengths in transversal skills can offset weaknesses in other domains. These explainable insights enable early identification of skill gaps and differentiated intervention strategies. Conclusion Vocational readiness is a longitudinal construct shaped by foundational competencies developed in primary education. Integrating XAI with competency-based assessment enhances transparency, supports ethical AI use, and enables cross-level curriculum alignment, contributing to sustainable workforce development.
Jatmoko et al. (Wed,) studied this question.