KMAN Publication Inc. (KMANPUB) Journal of Adolescent and Youth Psychological Studies (JAYPS) 2981-2526 7 Serial Number 47 2026 07 10 Elastic Net Regression for Predicting Psychological Well-Being among Adolescents from Resilience, Self-Compassion, Gratitude, and Perceived Social Support Elastic Net Regression for Predicting Psychological Well-Being among Adolescents from Resilience, Self-Compassion, Gratitude, and Perceived Social Support 1 14 10.61838/ EN Journal Article 2025 12 30 <table> <tbody> <tr> <td> <p><strong>Objective:</strong> This study aimed to examine the predictive roles of resilience, self-compassion, gratitude, and perceived social support in explaining psychological well-being among adolescents using Elastic Net Regression.</p> <p><strong>Methods and Materials: </strong>This cross-sectional predictive study was conducted among 1,284 adolescents aged 13 to 18 years from public secondary schools in Malaysia. Participants were selected through multistage cluster sampling from schools located in Selangor, Kuala Lumpur, Penang, and Johor. Data were collected using standardized self-report instruments, including the Ryff Psychological Well-Being Scale, Connor–Davidson Resilience Scale, Self-Compassion Scale–Short Form, Gratitude Questionnaire–Six Item Form, and Multidimensional Scale of Perceived Social Support. After data screening and handling of missing values through multiple imputation, descriptive statistics, correlation analysis, and Elastic Net Regression were performed. Predictor variables were standardized before modeling. The dataset was divided into training and testing subsets, and 10-fold cross-validation was applied to identify the optimal alpha and lambda values. Model performance was evaluated using R², Root Mean Square Error, and Mean Absolute Error. Bootstrap resampling with 1,000 iterations was conducted to assess coefficient stability.</p> <p><strong>Findings: </strong>Elastic Net Regression showed strong predictive accuracy, explaining 69.4% of the variance in psychological well-being in the training set and 68.1% in the testing set. The optimized model produced low prediction error indices, with RMSE values of 13.21 and 13.86 and MAE values of 10.18 and 10.73 for the training and testing datasets, respectively. Resilience emerged as the strongest predictor of psychological well-being, followed by self-compassion, perceived social support, and gratitude.</p> <p><strong>Conclusion: </strong>The findings indicate that psychological well-being among adolescents is strongly predicted by a combination of personal and relational protective factors. Resilience and self-compassion were the most influential predictors, while perceived social support and gratitude also made meaningful contributions.</p> </td> </tr> </tbody> </table> https://www.journals.kmanpub.com/index.php/jayps/article/download/5466/11076