Emotional Intelligence and Academic Achievement of High School Students: A Data-Driven Decision Support and Predictive Analysis Framework

Emotional intelligence (EI) has grown in importance for educational decision making in the recent past because it involves emotional and self-regulatory skills that could be useful in addition to traditional academic measures. This study creates an interpretable predictive framework to analyze the relationship between EI and academic achievement of high-school student profiles. Descriptive statistics, correlation analysis, multiple linear regression, ridge regression, k-nearest-neighbours regression, held-out validation, permutation importance and support-priority grouping were combined in the analysis. EI showed a moderate positive association with academic achievement (r = 0.578). Ridge regression had the highest generalization with a five-fold cross validation RMSE of 6.49 and held-out RMSE of 5.98 with 49.4% of the variance in achievement explained. Based on the results of the permutation analysis, EI was the most important predictor, followed by study hours, parental support, stress and attendance. The decision-support layer also divided profiles into three groups of priority: high-support, monitoring, and low-support, each having its own achievement and EI pattern. Results show how this interpretable predictive workflow can transform emotional and educational markers to academic-support priorities without superseding professional judgment.