Random Forest Prediction of Family Functioning Based on Emotional Availability, Family Communication, Psychological Well-Being, and Coping Self-Efficacy
Applied Family Therapy Journal · 2026 · 7 (3) · 1-13
DOI: 10.61838/kman.aftj.5462 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Family Therapy Journal, 2026. Verfasst von Justas Rimkus und 2 weiteren Autor:innen.
Abstract (Originalsprache)
Objective
The present study aimed to develop and evaluate a Random Forest machine learning model for predicting family functioning based on emotional availability, family communication, psychological well-being, and coping self-efficacy among Canadian adults. Methods and Materials: This cross-sectional predictive study was conducted among 768 adults residing in Canada who were recruited through community organizations, social media platforms, and online advertisements. Data were collected using the Family Assessment Device (FAD), Brief Emotional Availability Scale (BEAS), Family Communication Scale (FCS), Ryff’s Psychological Well-Being Scale (PWB), and the Coping Self-Efficacy Scale (CSE). Following data screening and preprocessing procedures, descriptive statistics and Pearson correlations were calculated. The dataset was randomly divided into training (80%) and testing (20%) subsets. A Random Forest regression model was developed using five-fold cross-validation and hyperparameter optimization procedures. 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 employed to determine the relative contribution of each predictor variable.
Findings
Significant associations were observed between family functioning and all predictor variables. Family communication demonstrated the strongest negative correlation with family dysfunction (r = -.76, p < .001), followed by emotional availability (r = -.71, p < .001), psychological well-being (r = -.68, p < .001), and coping self-efficacy (r = -.64, p < .001). The Random Forest model achieved strong predictive performance, yielding an R² of .847 on the testing dataset and a cross-validation R² of .836. Feature importance analysis identified family communication (37.2%) as the most influential predictor, followed by emotional availability (28.6%), psychological well-being (20.4%), and coping self-efficacy (13.8%). Permutation importance and SHAP analyses confirmed the stability of this predictor hierarchy and revealed that communication and emotional availability exerted the greatest impact on model predictions.
Conclusion
The findings indicate that family functioning can be predicted with high accuracy using emotional availability, family communication, psychological well-being, and coping self-efficacy. Relational factors, particularly communication quality and emotional availability, emerged as the most influential determinants of family functioning. The results support the application of machine learning approaches in family psychology and highlight key targets for interventions designed to strengthen family relationships and promote adaptive family functioning.
Zitieren
APA
Rimkus, J., Langlois, S., & Khoury, L. (2026). Random Forest Prediction of Family Functioning Based on Emotional Availability, Family Communication, Psychological Well-Being, and Coping Self-Efficacy. Applied Family Therapy Journal, 7(3), 1-13. https://doi.org/10.61838/kman.aftj.5462
Vancouver
Rimkus J, Langlois S, Khoury L. Random Forest Prediction of Family Functioning Based on Emotional Availability, Family Communication, Psychological Well-Being, and Coping Self-Efficacy. Applied Family Therapy Journal. 2026;7(3):1-13. doi:10.61838/kman.aftj.5462
BibTeX
@article{Rimkus2026,
author = {Rimkus, Justas and Langlois, Sophie and Khoury, Lina},
title = {Random Forest Prediction of Family Functioning Based on Emotional Availability, Family Communication, Psychological Well-Being, and Coping Self-Efficacy},
journal = {Applied Family Therapy Journal},
year = {2026},
volume = {7},
number = {3},
pages = {1--13},
doi = {10.61838/kman.aftj.5462}
}
EUVIMED