Descriptive statistics

HR Analytics: Predicting Employee Churn in Python

Hrant Davtyan

Assistant Professor of Data Science American University of Armenia

Turnover rate

# Get the total number of observations and save it
n_employees = len(data)    
# Print the number of employees who left/stayed
print(data.churn.value_counts())  
# Print the percentage of employees who left/stayed
print(data.churn.value_counts()/n_employees*100)

0    76.191746
1    23.808254
Name: churn, dtype: float64

Summary

Stayed	Left
76.19%	23.81%

Correlations

import matplotlib.pyplot as plt
import seaborn as sns
corr_matrix = data.corr()
sns.heatmap(corr_matrix)
plt.show()

heatmap

Let's practice!

HR Analytics: Predicting Employee Churn in Python