Tailoring Human Resource Policy in a Government Bank: Machine Learning Insights into Job Rotation, Resilience, and Adaptability in Nueva Ecija

Abstract

This quantitative, descriptive study leverages machine learning, specifically agglomerative hierarchical clustering, to generate actionable insights into job rotation, resilience, and adaptability among 70 employees in a government bank in Nueva Ecija, Philippines. Survey data from employees with job rotation experience across eleven branches were analyzed using standard descriptive statistics, t-tests, and Spearman correlations, with clustering techniques employed to segment the workforce. The clustering analysis tested various distance and linkage methods. While the Euclidean-average linkage combination achieved the highest cophenetic correlation coefficient, confirming its statistical validity, visual inspection favored the Euclidean-complete linkage method. This approach yielded five interpretable and diverse employee groups, supporting calls to balance quantitative cluster validity with qualitative interpretability. Key findings indicate that job rotation is generally associated with above-moderate resilience and adaptability. However, challenges related to geographic relocation were evident, with male employees reporting higher difficulty adjusting to new environments than females. Familial and contextual factors, specifically marital status, number of children, and residence-work location differences, positively correlated with both resilience and adaptability, highlighting their significant role in employee coping mechanisms. Furthermore, a strong positive relationship between resilience and adaptability was confirmed. Based from the cluster-specific insights, policy recommendations include developing structured job rotation programs, providing targeted commuter support, and implementing cluster-specific interventions to address group-specific needs, thereby optimizing organizational effectiveness and enhancing workforce well-being.