Uncovering employee insights: integrative analysis using structural topic modeling and support vector machines

Online platforms that enable employees to voluntarily share their opinions and experiences about current and former employers present a valuable data source for investigating worker satisfaction. This user-generated feedback has the potential to provide insights that surpass the limitations of traditional survey methodologies. This study proposes a novel approach by integrating Structural Topic Modeling (STM) analysis with Support Vector Machine (SVM) techniques to scrutinize the robustness of STM findings, particularly concerning the relative significance of extracted topics. This research reveals several insightful observations based on analyzing employee reviews of a large Chinese tech company. Notably, the findings highlight the importance of intangible aspects within the work environment, such as cultural conflicts, leadership style, and perceived fairness, as significant contributors to satisfaction and dissatisfaction among the company employees. Furthermore, this study reveals inconsistencies with prior research on two significant aspects. First, while work-life balance is typically linked to job dissatisfaction, this study suggests that the negative consequences of work-life balance factors can be mitigated by favorable performance in some job satisfaction-related aspects. Second, while monetary rewards undoubtedly exert a considerable influence, they may fail to ensure employee satisfaction when other key aspects of the work experience are underperforming. This research not only contributes to the body of organizational research but also offers practical implications for enhancing employee satisfaction within global tech enterprises.

Text 118381.pdf - Published Version Available under License Creative Commons Attribution. Download (3MB) Official URL or Download Paper: https://journalofbigdata.springeropen.com/articles...

Actions (login required)

View Item