Project: IBM Attrition Analysis

IBM is an American multinational information technology company headquartered in New York, state. With operations in over 170 countries, the HR department is trying hard to keep their employees. In this project, we are using the IBM attrition and performance dataset to test out some potential factors for attrition

Data description:

####Source

• We obtained our original sample data set from Kaggle. -Kaggle dataset URL:
  https://www.kaggle.com/pavansubhasht/ibm-hr-analytics-attrition-dataset

Objective

• We intend to explore important questions regarding why IBM’s employees might be leaving the company and therefore need to be replaced with the talent pool available in the job market.

-Sample

1. Population of interest is IBM HR (management) group to help them identify the factors of why their employees in Research and Development, Sales and HR are leaving.
2. Our sample is a random sample out of the total IBM employees within 3 departments for the previous year this data was collected. So, this data is representative to determine IBM employee attrition

Summary and Conclusions:

• Contrary to our expectations from graphical and numerical descriptive analysis, we didn’t Education and stock options as statistically significant to predict monthly income.
• In order for IBM to retain their employees , since ‘monthly salary’ positively drives the attrition rate significantly , the company must consider the employees’ total work experience, their department, and their job role as important factors in deciding the salary ranges in which they pay their employees.
• Next, to control attrition, IBM must ensure that they do not subject their employees to increased ‘overtime’ and ‘business travel’ which may negatively impact the employees’ work life balance , especially if they are married and hence this may negatively impact their decision to stay with the company. Such employees will hence rate less on ‘relationship’ and ‘environment’ satisfaction and are more likely to leave the company soon.
• We also infer that factors such as these and others such as if the company is not investing on their employees in terms of vocational trainings offered over the year and their job involvement , the employees may feel dissatisfied and hence it may impact negatively on their job satisfaction levels
• Lastly, the company must understand that if the employee has worked for a high number of companies in their past career, the employee might prefer to change their firm every few years and this might have nothing to do with IBM. Hence, it is important to understand that high no. of companies worked before will impact negatively on attrition rate as this is more about such employees stability than the firm itself.

Implications:

• HR Management at IBM can highly benefit from our findings to keep a check on the key variables that we have found to be statistically significant in impacting the attrition rate at IBM so that they don't have to go back to the talent pool in market to recover their lost employees
• Since this dataset is publically available on Kaggle , our analysis if made public can be used and applied to bigger sample datasets by more tech companies like IBM in order to uncover key drivers for employees leaving their firms.