Abstract
This study employs a data-driven approach to explore the landscape of workplace environments and its implications for employee well-being. By analyzing a large-scale dataset comprising over 1.3 million observations spanning from 2017 to 2021, we identify nine key workplace factors of job demands and job resources, using factor analysis. These factors encompass dimensions such as workload, emotional burden, organizational integrity, job autonomy, and surrounding support. Subsequently, employing Gaussian mixture models (GMM), we classify employees into ten distinct workplace clusters, ranging from Grade D (the most challenging) to Grade A1 (the most favorable). Our findings reveal significant variations in employee well-being across these clusters, with higher grades associated with better mental health, work engagement, job satisfaction, and workplace cohesiveness. Additionally, we examine the impact of workplace cluster changes on employee well-being, highlighting the importance of understanding how shifts in the workplace environment affect employee outcomes. Our study contributes to the literature by providing a comprehensive understanding of workplace dynamics and offering valuable insights for organizational management and policy formulation.