KMAN Publication Inc. (KMANPUB) Journal of Assessment and Research in Applied Counseling (JARAC) In Press 2026 10 01 CatBoost Classification of High-Risk Perfectionism Using Conditional Self-Worth, Fear of Evaluation, and Cognitive Fusion CatBoost Classification of High-Risk Perfectionism Using Conditional Self-Worth, Fear of Evaluation, and Cognitive Fusion 1 15 EN Journal Article 2025 12 14 <table> <tbody> <tr> <td> <p><strong>Objective:</strong> This study aimed to classify high-risk perfectionism among Indonesian university students using a CatBoost machine-learning model based on conditional self-worth, fear of negative evaluation, and cognitive fusion.</p> <p><strong>Methods and Materials: </strong>This cross-sectional predictive study was conducted among 436 university students in Indonesia during the 2025–2026 academic year. Participants were selected through convenience sampling from public and private universities. High-risk perfectionism was defined according to the upper quartile of perfectionistic concerns scores, while the remaining participants were classified as non-high-risk. Data were collected using standardized self-report instruments measuring perfectionistic concerns, conditional self-worth, fear of negative evaluation, and cognitive fusion. After screening the dataset for incomplete responses, invalid cases, and response irregularities, the data were divided into stratified training and testing subsets. A CatBoost classifier was trained using five-fold cross-validation, and model performance was evaluated using accuracy, balanced accuracy, sensitivity, specificity, precision, F1-score, ROC-AUC, PR-AUC, confusion matrix indices, and calibration statistics. Feature importance and SHAP values were used to interpret the contribution of predictors to model output.</p> <p><strong>Findings: </strong>The CatBoost model demonstrated strong classification performance in identifying high-risk perfectionism. In the independent test set, the model achieved an accuracy of 0.885, balanced accuracy of 0.863, sensitivity of 0.818, specificity of 0.908, precision of 0.750, F1-score of 0.783, ROC-AUC of 0.926, PR-AUC of 0.812, Matthews correlation coefficient of 0.711, and Brier score of 0.104. The confusion matrix showed that 27 of 33 high-risk students were correctly classified, while 89 of 98 non-high-risk students were correctly identified. SHAP analysis indicated that cognitive fusion was the strongest predictor, followed by conditional self-worth and fear of negative evaluation.</p> <p><strong>Conclusion: </strong>The findings suggest that high-risk perfectionism among Indonesian university students can be accurately classified using an interpretable CatBoost model and that cognitive fusion, conditional self-worth, and fear of negative evaluation are important psychological indicators for screening and preventive counseling.</p> </td> </tr> </tbody> </table> https://www.journals.kmanpub.com/index.php/jarac/article/download/5659/11253