XGBoost-Based Prediction of Family Resilience from Psychological Flexibility, Social Support, Family Cohesion, and Emotional Regulation
Applied Family Therapy Journal · 2026 · 7 (3) · 1-13
DOI: 10.61838/kman.aftj.5453 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Family Therapy Journal, 2026. Verfasst von Tendai Chikore und 2 weiteren Autor:innen.
Abstract (Originalsprache)
Objective
The present study aimed to predict family resilience based on psychological flexibility, social support, family cohesion, and emotional regulation using the XGBoost machine learning algorithm and to determine the relative importance of these predictors in explaining resilience outcomes among Canadian adults. Methods and Materials: This cross-sectional predictive study was conducted among 812 adults residing in Canada. Participants were recruited through community organizations, family support centers, and online platforms using a stratified convenience sampling method. Data were collected using the Family Resilience Assessment Scale (FRAS), Acceptance and Action Questionnaire-II (AAQ-II), Multidimensional Scale of Perceived Social Support (MSPSS), Family Adaptability and Cohesion Evaluation Scales IV (FACES-IV), and Emotion Regulation Questionnaire (ERQ). Following data preprocessing and standardization procedures, the dataset was divided into training (80%) and testing (20%) subsets. An Extreme Gradient Boosting (XGBoost) model was developed and optimized using five-fold cross-validation and grid-search hyperparameter tuning. Model performance was evaluated using the coefficient of determination (R²), root mean square error (RMSE), mean absolute error (MAE), and mean squared error (MSE). Feature importance and SHAP (Shapley Additive Explanations) analyses were performed to determine the relative contribution of each predictor.
Findings
The XGBoost model demonstrated excellent predictive performance, achieving an R² of 0.867 on the testing dataset, indicating that 86.7% of the variance in family resilience was explained by the predictors. Family resilience was positively correlated with psychological flexibility (r = .68, p < .001), social support (r = .72, p < .001), family cohesion (r = .79, p < .001), and emotional regulation (r = .64, p < .001). Feature importance analysis revealed that family cohesion was the strongest predictor (34.8%), followed by social support (28.7%), psychological flexibility (21.5%), and emotional regulation (15.0%). SHAP analyses confirmed these findings and demonstrated that higher levels of each predictor contributed positively to resilience predictions, with family cohesion exerting the largest overall influence.
Conclusion
The findings indicate that family resilience can be predicted with high accuracy using psychological flexibility, social support, family cohesion, and emotional regulation. Family cohesion emerged as the most influential determinant, highlighting the critical role of strong family relationships in fostering adaptive functioning. The study demonstrates the utility of machine learning approaches, particularly XGBoost, for identifying key resilience factors and provides evidence supporting multidimensional models of family resilience that integrate individual, relational, and social protective resources.
Zitieren
APA
Chikore, T., Macpherson, O., & Reinert, P. (2026). XGBoost-Based Prediction of Family Resilience from Psychological Flexibility, Social Support, Family Cohesion, and Emotional Regulation. Applied Family Therapy Journal, 7(3), 1-13. https://doi.org/10.61838/kman.aftj.5453
Vancouver
Chikore T, Macpherson O, Reinert P. XGBoost-Based Prediction of Family Resilience from Psychological Flexibility, Social Support, Family Cohesion, and Emotional Regulation. Applied Family Therapy Journal. 2026;7(3):1-13. doi:10.61838/kman.aftj.5453
BibTeX
@article{Chikore2026,
author = {Chikore, Tendai and Macpherson, Olivia and Reinert, Pascal},
title = {XGBoost-Based Prediction of Family Resilience from Psychological Flexibility, Social Support, Family Cohesion, and Emotional Regulation},
journal = {Applied Family Therapy Journal},
year = {2026},
volume = {7},
number = {3},
pages = {1--13},
doi = {10.61838/kman.aftj.5453}
}
EUVIMED