HR Analytics For Predicting Employee Attrition with Logistic Regression

Today most of the organisations are failing to retain the valuable employees. As high Employee attrition rate can never be positive sign for an organization, it is necessary to identify, organize and manage the factors affecting Employee attrition.  Present paper focused on developing an algorithm to predict employee attrition rate by considering the factors affecting employee attrition. Data considered for the study has been collected from secondary source Kaggle Data sets. Simple random sampling Technique is used. Data containing 35 different attributes or factors of 1,500 employees belonging to IT Companies were selected. Python is used as a tool for conducting Data analytics. An exploratory data analysis constructed to see the effect of every factor against employee attrition. Feature Engineering was performed to select the most influencing factors affecting Employee attrition rate. Feature selection gave the five most influencing factors driving towards employee attrition rate. Further a machine learning algorithm Logistic regression was applied as its accuracy rate is high as compared to other. Based on the predicted Employee attrition rate, we have calculated Employee turnover rate. Major factors affecting the Employee attrition rate are identified as Environment satisfaction, Job involvement, Job satisfaction, Relationship satisfaction, Work life balance. Employee turnover rate is calculated by predicting Employee attrition rate. Total turnover rate is 16.122%. There could be lot of factors or the variants that could affect employee attrition. Present paper helps to study how various significant factors influence the attrition of employees and what kind of working environment drive towards attrition.