Know more about turnover

HR Analytics: Predicting Employee Churn in R

Anurag Gupta

People Analytics Practitioner

Understanding the data

glimpse(org)

Rows: 2,291
Columns: 12
$ emp_id            <chr> "E11061", "E1031", "E6213", "E5900", "E3044"...
$ status            <chr> "Inactive", "Inactive", "Inactive", "Inactiv...
$ turnover          <int> 1, 1, 1, 1, 1, 0, 0, 1, 0, 1, 0, 1, 1, 0, 1,...
$ location          <chr> "New York", "New York", "New York", "New Yor...
$ level             <chr> "Analyst", "Analyst", "Analyst", "Analyst", ...
$ date_of_joining   <chr> "22-03-2012", "09-03-2012", "06-01-2012", "2...
$ last_working_date <chr> "11-09-2014", "05-06-2014", "30-04-2014", "0...
$ gender            <chr> "Male", "Female", "Female", "Female", "Femal...
$ department        <chr> "Customer Operations", "Customer Operations"...
$ mgr_id            <chr> "E1712", "E10524", "E4443", "E3638", "E3312"...
$ cutoff_date       <chr> "31-12-2014", "31-12-2014", "31-12-2014", "3...
$ emp_age           <dbl> 22.49, 22.42, 22.24, 22.32, 22.14, 22.67, 22...

Calculating turnover rate

$$ \text{Turnover rate} = \frac{\text{Number of employees who left }}{\text{Total number of employees}} $$

$$ \text{Turnover rate} = \frac{\text{Count of all 1's }}{\text{Count of all 1's + Count of all 0's }} = {\text{mean(turnover)}} $$

where

$$ \text{1 means Inactive ; 0 means Active} $$

Count Active and Inactive employees

# Count Active and Inactive employees

org %>% 
  count(status)

# A tibble: 2 x 2
  status      n
  <fct>     <int>
1 Active    1881
2 Inactive   410

Calculate turnover rate

# Calculate average turnover rate 

org %>% 
    summarize(turnover_rate = mean(turnover))

      turnover_rate
1         0.1789612

Calculate turnover rate at each level

df_level <- org %>%  
    group_by(level) %>% 
    summarize(turnover_level = mean(turnover))

df_level

# A tibble: 7 x 2
  level             turnover_level
  <fct>                      <dbl>
1 Analyst                   0.215 
2 Assistant Manager         0.0365
3 Director                  0     
4 Manager                   0.0435
5 Senior Manager            0     
6 Specialist                0.149 
7 Vice President            0

Visualize the turnover trends using ggplot

# Visualize the results
library(ggplot2)
ggplot(df_level, aes(x = level, y = turnover_level)) + 
    geom_col()

Let's practice!

HR Analytics: Predicting Employee Churn in R