KMAN Publication Inc. (KMANPUB) Journal of Adolescent and Youth Psychological Studies (JAYPS) 2981-2526 7 Serial Number 45 2026 05 10 Explainable AI Analysis of Grit, Academic Hope, and Persistence in Iranian EFL Students Explainable AI Analysis of Grit, Academic Hope, and Persistence in Iranian EFL Students 1 11 10.61838/kman.jayps.5074 EN Journal Article 2025 10 23 <table> <tbody> <tr> <td> <p><strong>Objective:</strong> The objective of this study was to examine the predictive roles and relative importance of grit and academic hope in explaining academic persistence among Iranian EFL students using explainable artificial intelligence techniques.</p> <p><strong>Methods and Materials: </strong>This study employed a cross-sectional correlational design with a sample of Iranian EFL learners enrolled in language institutes and university language centers in Tehran. Participants completed standardized self-report measures assessing grit (perseverance of effort and consistency of interests), academic hope (agency and pathways), and academic persistence. Data were analyzed using supervised machine learning models suitable for psychological tabular data, with academic persistence specified as the outcome variable and grit and academic hope components as predictors. Model performance was evaluated using explained variance and error-based indices. To ensure interpretability, explainable AI methods, specifically SHapley Additive exPlanations (SHAP), were applied to identify global and local feature contributions and to clarify the direction and magnitude of predictor effects.</p> <p><strong>Findings: </strong>The machine learning model explained a substantial proportion of variance in academic persistence, indicating strong predictive performance. Explainable AI analyses revealed that academic hope–agency was the most influential predictor of persistence, followed by perseverance of effort, academic hope–pathways, and consistency of interests. Higher levels of agency beliefs and perseverance of effort were associated with higher predicted persistence scores, while consistency of interests showed a comparatively weaker contribution. The results demonstrated non-linear and differential effects of motivational components on persistence, highlighting individual variability in predictor influence.</p> <p><strong>Conclusion: </strong>The findings indicate that academic persistence among Iranian EFL students is primarily driven by agency-based hope and sustained effort, and that explainable AI provides a powerful framework for uncovering nuanced motivational mechanisms beyond traditional linear models.</p> </td> </tr> </tbody> </table> https://www.journals.kmanpub.com/index.php/jayps/article/download/5074/9212