Understanding Employee Attrition Factors in the Information Technology Sector: A Text Analytics Perspective

Recently online reviews on websites have provided a precious source of data for businesses. Some of these websites collect customers’ opinions about products and services provided by these businesses, some others such as Glassdoor and Indeed are the websites on which the employees write their opinions about their company. Although this data source is usually used by job seekers to find proper job opportunities, it has been used recently by researchers and business owners for discovering the employees’ satisfaction and dissatisfaction factors. In this study we collected 825129 comments people left on Glassdoor and Indeed about their current or previous companies in IT section. First, we have applied the Latent Dirichlet Allocation (LDA) topic modelling technique to find out the most cited attrition factors in the comments. We recognized that “Personal Development”, “Financial and Professional Development” and “Cultural Development” are the most frequent factors mentioned by employees in IT companies. Then, by applying a novel topic-based sentiment analysis technique we have tried to figure out the polarity of comments about each of three found factors. The trend of these factors in terms of their frequency and polarity were tracked through the time and also among different IT sections. The assessment of current and former employees’ comments showed that these two groups have some behavioral differences which can help IT companies to unravel the reasons of their recent high attrition rate. Some behavioural differences were also detected among different IT sections. Further analysis showed that there is a statistically significant correlation between the polarity of these factors and quantitative rating review of companies and also their market capitalization.

Name thesis:degree_name

M.Sc. Management

Level thesis:degree_level

Masters

Discipline thesis:degree_discipline

Faculty of Business

Department dc:contributor.department

Faculty of Business Programs

Grantor

Brock University

Year dc:date.issued

2023

Author dc:creator

• Monshi, Mina

• Text Mining
• Topic Modeling
• Topic-Based Sentiment Analysis
• Information Technology
• Employee Attrition

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/10464/17835

OAI identifier oai:identifier

oai:brocku.scholaris.ca:10464/17835