Profiling Employees using Cluster Analysis in R

Imagine you are a HR manager of a big consulting company and that you are interested to profile the employees . The company collected data from 2000 employees. Below are the variables:

• SAT: satisfaction of the employee about the company
• LPE: last project evaluation
• NP: number of completed projects within the last 12 months 
• ANH: the average number of hours worked per month
• TIC: time spent in the company

Let's load dataset as "data" and understand the variables:

Now, lets normalize our variables first and then run hierarchical cluster analysis.

This will generate following dendrogram. Looking at the same, it can be concluded that there are three clusters in the data.

Lets generate three clusters solution

There are three clusters with 916, 538 and 546 members. Percentage wise cluster 1 has 45.8% employees, cluster 2 has 26.9% and cluster 3 has 27.3% employees. Lets profile the characteristics of employees of these clusters.

Following three clusters of employees can be described based on above analysis :

Low Performance segment: Employees of cluster 1 have low satisfaction level. They have not done many projects on average and are underutilized. It is also a segment where employees have been in the company for a shorter time than the average.

The High Performance segment: Cluster 2 are very satisfied employees. They have good evaluations and high utilization. They work on an average 4.53 projects in a year. In this cluster, employees are with the company for longer duration of time.

The Burned Out segment: Cluster 3 has a low level of satisfaction, the good evaluations, working on lot of projects, has a good utilization rate and has been in the company for a long time. They are those who worked the most but also they are the less satisfied. 

Hope you enjoy!!! Thank you!!!