KMAN Publication Inc. (KMANPUB) Applied Family Therapy Journal (AFTJ) 3041-8798 7 Serial Number 32 2026 03 01 Machine-Learning Identification of Family Adaptation Patterns During Economic Stress Based on Coping Strategies and Couple Communication Machine-Learning Identification of Family Adaptation Patterns During Economic Stress Based on Coping Strategies and Couple Communication 1 10 10.61838/kman.aftj.5213 EN Journal Article 2026 01 11 <p><strong>Objective:</strong> This study aimed to identify distinct family adaptation patterns during economic stress by applying unsupervised machine-learning to the interplay of coping strategies and couple communication behaviors.</p> <p><strong>Methods and Materials:</strong> A quantitative, cross-sectional design was employed with married couples from Malaysia experiencing varying degrees of economic stress. Data were collected via self-report instruments measuring perceived economic pressure, coping mechanisms (problem-focused, emotion-focused, and avoidant), and couple communication patterns (constructive and demand-withdraw). An unsupervised K-Means clustering algorithm was applied to the multidimensional data to segment the couples into naturally occurring adaptation profiles, followed by post-hoc statistical analyses to evaluate the characteristics of each profile and their associations with objective socio-economic factors.</p> <p><strong>Findings:</strong> The K-Means clustering analysis ( ) successfully delineated three distinct family adaptation patterns. Cluster , the “Resilient-Constructive Adapters” ( , ), exhibited high problem-focused coping and constructive communication, reporting the lowest perceived economic stress ( , ). Cluster , the “Distressed-Avoidant Adapters” ( , ), was characterized by heavy reliance on avoidant coping and demand-withdraw communication, experiencing the highest levels of economic stress ( , ) and showing a disproportionate representation from the lowest income bracket ( ). Cluster , the “Ambivalent-Emotion-Focused Adapters” ( , ), displayed moderate communication quality with a high reliance on emotion-focused coping and social support seeking, reporting moderate economic stress ( , ). A statistically significant association was found between cluster membership and objective household income ( , ).</p> <p><strong>Conclusion:</strong> Unsupervised machine-learning successfully revealed that families navigate financial hardship through interconnected behavioral typologies, highlighting the necessity for tailored interventions that address both objective financial realities and dyadic relationship dynamics to foster resilience.</p> https://www.journals.kmanpub.com/index.php/aftj/article/download/5213/9466