Identifying factors for employee retention using computational techniques: an approach to assist the decision-making process

Abstract

In the today’s competitive environment, employee retention is a challenge faced by many industries. This work aims to identify the factors that influence employee retention. This is done using employees’ feedback and various computational techniques. A survey is conducted within multiple sectors to collect data. The questionnaire is divided into two parts: the first part includes demographic information, whereas the second part contains questions pertaining to employees’ job description and their satisfaction. The questions on the second portion are based on theories like Herzberg’s duality theory, expectancy theory, social cognitive theory, and sociocultural theory. These theories are further linked with factors like motivation, recognition and reward, bullying and work harassment. Later, the frequent items mining technique from the domain of data mining is utilized to identify the frequent factors from an employee perspective toward better retention rates. A test is also conducted to ensure the reliability of the data. The obtained results indicate it to be 87% reliable. A comparison between two frequent items mining methods indicates four times quicker performance of the k Direct Count and Intersect (kDCI) method in identifying key retention aspects from the data. A tool is utilized for analysis of variance (ANOVA) and exploratory factor analysis (EFA) tests to find factors crucial for retaining employees. The result identifies that work environment, reward and recognition, work performance, supervisory support, and income have high impact on employee retention.