KMAN Publication Inc. (KMANPUB) Journal of Adolescent and Youth Psychological Studies (JAYPS) 2981-2526 7 Serial Number 50 2026 10 10 Predicting Future Anxiety Among Youth Based on Career Uncertainty, Social Support, and Psychological Flexibility Using CatBoost Predicting Future Anxiety Among Youth Based on Career Uncertainty, Social Support, and Psychological Flexibility Using CatBoost 1 19 10.61838/kman.jayps.6327 EN Journal Article 2026 07 17 <table> <tbody> <tr> <td> <p><strong>Objective:</strong> This study aimed to predict future anxiety among youth in Georgia based on career uncertainty, perceived social support, and psychological flexibility using the CatBoost machine-learning algorithm.</p> <p><strong>Methods and Materials:</strong> A cross-sectional predictive study was conducted among 472 young people aged 18–29 years from Georgia. Future anxiety was assessed using the Dark Future Scale, perceived social support using the Multidimensional Scale of Perceived Social Support, and psychological flexibility using the Psy-Flex; career uncertainty was assessed with a brief career uncertainty measure based on established career-decision indicators. CatBoost regression was used as the principal predictive method, with 80% of the data allocated to training and 20% to independent testing. Five-fold cross-validation was applied for hyperparameter tuning. Model performance was evaluated using the coefficient of determination (R²), root mean squared error (RMSE), and mean absolute error (MAE). Multiple linear regression was used as a benchmark model, and SHAP analysis was employed to determine predictor importance and direction of effects.</p> <p><strong>Findings:</strong> The final CatBoost model achieved an R² of .63, RMSE of 3.79, and MAE of 2.93 on the independent test set, outperforming multiple linear regression, which yielded R² = .49, RMSE = 4.46, and MAE = 3.49. Career uncertainty was the strongest predictor, accounting for 33.72% of model importance, followed by psychological flexibility at 27.84% and perceived social support at 19.63%. SHAP analysis showed that higher career uncertainty increased predicted future anxiety, whereas greater psychological flexibility and stronger perceived social support reduced predicted anxiety. The adverse effect of career uncertainty was particularly pronounced when psychological flexibility or social support was low.</p> <p><strong>Conclusion:</strong> Future anxiety among Georgian youth can be predicted with substantial accuracy using a machine-learning approach, with career uncertainty emerging as the principal risk factor and psychological flexibility and social support functioning as important protective predictors.</p> </td> </tr> </tbody> </table> https://journals.kmanpub.com/index.php/jayps/article/download/6327/12115