Predicting Employee Attrition: the Impact of Hybrid Work

Abstract

The continuity of contributions of its workers in driving the company ahead is critical to the growth of any corporation. There has been a significant increase in employee attrition in recent decades as more people leave their jobs to seek better ones elsewhere. Organizations are becoming more concerned about rising employee attrition. Employee attrition may have a large impact on an organization. It might result in decreased performance, higher expenses, and challenges in recruiting and training new employees. A thorough understanding of the causes of employee attrition will assist management to make adjustments to enhance the organization's work culture for future workers by minimizing attrition. This research applied five different machine learning algorithms to predict employee attrition. Among these algorithms, Support Vector Machine, and Gaussian Naive Baye achieved the highest accuracy. Additionally, we applied multivariable regression analysis to learn the significance of each factor and to check whether the hybrid work plays any role in attrition likelihood.  Hybrid work is when employees split their time between working from home and coming into the office. It gives workers more freedom over when, where, and how they get their jobs done, and can increase employee satisfaction, engagement, and lower level of attrition. We empirically investigated whether the hybrid work can mitigate the adverse impacts of various attrition factors. The results show that personal, workplace, management, and compensation factors are important reasons for attrition. We also find that employees may sacrifice some benefits if they can choose a hybrid workplace. Moreover, according to our findings, hybrid work can reduce the negative impacts of personal and family factors of attrition.  Some causes of attrition may be managed, such as the company's work environment, benefits, or salary structure; others, such as location, competition, or industry standards, are more difficult to manage but still important to increase attrition. This research hopes to provide recommendations for the former by discussing its causes and ways to minimize attrition.