Skip to content
0

Tackle increasing Vacancy: Can you reduce employee turnover?

YES! We can help reduce Employee Turnover. Let's get to it!!

📖 Background

You work for the human capital department of a large corporation. The Board is worried about the relatively high turnover, and your team must look into ways to reduce the number of employees leaving the company.

The team needs to understand better the situation, which employees are more likely to leave, and why. Once it is clear what variables impact employee churn, you can present your findings along with your ideas on how to attack the problem.

Table of Contents

  1. Executive Summary
  2. The Dataset
  3. Introduction
  4. Problem Statement and Goal of Analysis
  5. Analysis Plan
  6. Exploratory Data Analysis
  7. Predictive modelling
  8. Discussions
  9. Recommendations
  10. References

(Invalid URL)

Executive Summary

Human Resource (HR) Analytics, also known as People Analytics is a data-driven approach to manage people at work. Tackling the problem of employee churn falls under retention in HR analytics. This project covers mainly three of the types of data analysis including; Descriptive, Predictive and Prescriptive Analysis.

  • In order to understand the factors responsible for employee turnover and find ways to reduce the number of employees leaving the company, detailed analysis was carried out on the employees dataset. Some of the variables present are:
    • "satisfaction" - a measure of employee satisfaction from surveys.
    • "avg_hrs_month" - the average hours the employee worked in a month.
    • "department" - the department the employee belongs to.
  • Questions for the Purpose of Analysis include:
    1. Which department has the highest employee turnover? Which one has the lowest?
    2. Investigate which variables seem to be better predictors of employee departure.
    3. What recommendations would you make regarding ways to reduce employee turnover?
  • To bring to the fore from the analysis, the results show prospectively that:
  1. The Department with the highest number of employee turnover as seen above is the Sales department, while the departments with the least is the Finance and IT department.
  2. The Percentage Employee Turnover is between 0.31 and 0.26, with the highest percentage from the IT department, this gives us insight into the rate of employee turnover/churn: highest in the IT department and lowest in the finance department
  3. Top three variables of employee churn prediction are Satisfaction, Review, Average Number of Hours per month

Finally, my recommendations are that:

  1. The Company's human resource team and board needs to focus more on employees with tenure less than 6 years as the trend of employee churn tends to be higher on and below 6 years tenure.
  2. A more robust method for conducting interviews for it staff, as this would help uncover more underlying issues/factors causing turnover.
  3. The Company could give incentives like bonusesnand other benefits.
  4. Using a Structured onboarding process.
  5. Promotion of Employees to new levels could positively impact employees chances of staying at the company.
  6. A Thorough review of the average number of working hours and possible asking emloyees for their preferences.
  7. Look into the Sales Department and try to encourage them, as they have the highest number of employee churn.

(Invalid URL)

💾 The Dataset

The department has assembled data on almost 10,000 employees. The team used information from exit interviews, performance reviews, and employee records.

  • "department" - the department the employee belongs to.
  • "promoted" - 1 if the employee was promoted in the previous 24 months, 0 otherwise.
  • "review" - the composite score the employee received in their last evaluation.
  • "projects" - how many projects the employee is involved in.
  • "salary" - for confidentiality reasons, salary comes in three tiers: low, medium, high.
  • "tenure" - how many years the employee has been at the company.
  • "satisfaction" - a measure of employee satisfaction from surveys.
  • "avg_hrs_month" - the average hours the employee worked in a month.
  • "left" - "yes" if the employee ended up leaving, "no" otherwise.

(Invalid URL)

Introduction

Human Resource (HR) Analytics, also known as People Analytics is a data-driven approach to manage people at work. It enables HR leaders to develop data-driven insights to inform talent decisions, improve workforce processes and promote positive employee experience.

Some of the problems addressed by HR Analytics include:

  • Hiring/Assesment
  • Retention (our use case)
  • Performance evaluation
  • Learning and Development
  • Collaboration/team composition
  • Others

Problem Statement and Goal of Analysis

Employees are leaving the company, can we leverage data science to figure out why and derive actionable insights?

This analysis aims to:

  • To Discover the reasons and factors that are responsible for employee turnover.
  • Data-driven prediction of employee liable to churn.
  • Recommend ways to tackle employee churn based on insights derived from analysis.

Analysis Plan

The project deals with analyzing and predicting employee churn from a given dataset making it a supervised machine learning problem, specific to classification(binary) as we are dealing with a yes or no target variable. Our Dataset is under 10000 and is thus a medium-sized dataset. In order to understand the factors responsible for employee turnover and also make predictions, a systematic approach will be employed including:

  1. Exploratory Data Analysis: to visualize the relationships between the different independent variables with the dependent variable. This will also lend credence to our prediction process and expectations.
  2. Predictive modelling: using the historical data to help predict future outcomes based on employee turnover variables. i.e understanding variables that could help inform us of employee churn
  3. Discussions: bring the whole ideas and insights gained from the analysis together.
  4. Recommendations: based on insights gained prescribe possible actions and key variables to look out for and try to improve to prevent employee churn
# Import the necessary libraries
from IPython.display import display
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

# Read in the data
df = pd.read_csv('./data/employee_churn_data.csv')
df.head()

data = df.copy()

Exploratory Data Analysis (EDA)

# Examining the data
df.info()
  • the dataset consists of 9540 observations/employee_records
  • No missing values
  • some incorrectly imputed data types
# Set FOnt Scale and Style
sns.set(font_scale=1.25)
sns.set_style('darkgrid')
# Department with sorted number of employee turnovers: from highest to lowest
no_turnover = df[df.left == 'yes'].groupby('department').left.count().sort_values()

no_turnover.to_frame()
# Department with sorted number of employee turnovers: from highest to lowest

fig, ax = plt.subplots(figsize=(8, 6))
no_turnover.plot.barh(ax=ax, title="Plot of Employee Turnover for each Department", colormap='Dark2')
plt.xlabel("Number of Employees that left")
plt.ylabel("Departments")
plt.show()
‌
‌
‌