Human resource management systems research – how to gain impactful insights through formative measurement and hierarchical component models
本文主张在人力资源管理系统实证研究中采用形成性测量模型和层次组件模型，以更准确地反映系统概念，提供具体可行的建议，并通过PLS-SEM实现多层级概念化。
This paper emphasizes that paying greater attention to how human resource management (HRM) systems are conceptualized in empirical studies could provide more actionable insights and increase the impact of HRM systems research. We advocate the use of formative measurement models, arguing that this approach aligns better with the concept of HRM systems, and allows for a nuanced understanding of how each HRM practice and the system contribute to the outcomes of interest. In the same vein, we advocate the use of hierarchical component models, which allow a multi-level conceptualization representing HRM practices, the HRM system, and their intermediate levels of abstraction (e.g. ability, motivation, and opportunities as subcomponents of high-performance work systems). As a result, HRM systems research can move beyond general assertions and instead offer specific and actionable recommendations. We discuss and illustrate how these conceptual ideas can be implemented in partial least squares-structural equation modeling (PLS-SEM), and enriched by predictive model evaluation following state-of-the-art guidelines.