Analyzing the Work Performance of Night Shift Nurses Using Clustering, Statistical Tools and Data Analytics

This study examined the work performance of night-shift nursing personnel in a tertiary hospital setting, focusing on how fatigue, satisfaction, and collaboration influence behavioral and physiological performance indicators. A quantitative descriptive–correlational design was employed using a 15-item Work Performance Scale rated on a four-point Likert scale. Data were analyzed using descriptive and non-parametric statistics—including medians, interquartile ranges, and Spearman’s rank correlation—to assess performance patterns and relationships among variables. Complementary data-analytics and machine-learning methods such as K-Means clustering and decision-tree modeling were applied to uncover latent performance profiles and the relative contribution of key factors. Findings revealed that the overall composite mean of 2.69 corresponds to the descriptive interpretation “Often,” signifying that nurses consistently perform their professional duties despite the physiological demands of nocturnal work. The highest-rated domains were Teamwork and Collaboration and Satisfaction and Development, demonstrating strong interpersonal cohesion and positive work engagement. Conversely, Alertness and Concentration and Physical and Mental Endurance received lower mean values, suggesting that fatigue, disrupted circadian rhythm, and limited recovery periods continue to affect performance. Correlation analysis indicated a moderate positive association (ρ ≈ 0.54, p < 0.05) between concentration and physical endurance, while cluster analysis identified three behavioral profiles—Highly Engaged, Balanced, and Fatigued—reflecting varying adaptability to night-shift demands.