Job satisfaction and turnover decision of employees in the Internet sector in the US

This paper proposes that high value on the work-life balance, compensation, career opportunity and fitness of culture and management style would improve job satisfaction. A turnover risk prediction model based on the random forest is constructed to understand the turnover risk feature and identify risk. Using a sample of 17,724 online reviews of employees from Glassdoor, the positive effect of antecedents, the job satisfaction variable as a mediator, and the unemployment rate variable as a moderator is verified. Finally, job satisfaction is identified as the most important feature for predicting turnover based on the random forest algorithm.

Automated summary

This paper proposes the importance of work-life balance, compensation, career opportunity, and culture and management style in improving job satisfaction. A turnover risk prediction model based on the random forest algorithm is constructed to understand the characteristics and identify risks of turnover. Empirical analysis using a sample of 17,724 online reviews from Glassdoor confirms the positive effects of antecedents, with job satisfaction as a mediator and the unemployment rate as a moderator. Finally, job satisfaction is identified as the most crucial feature for predicting turnover.