Analyzing Employee Exit Surveys

Last week I read this medium article curated on Towards Data Science, and it changed the way I thought about projects and the role of data scientists. The author of the piece explains their shift in mindset from “data scientists use data” to “data scientists make data useful”.

This statement changed my mindset as well, especially when I think of which projects would make a more meaningful impact in my portfolio. This was one of the first projects that came to mind.

In this project, I examined employee exit surveys from employees of the Department of Education Training and Employment (DETE) and the Technical and Further Education (TAFE) institute in Queensland, Australia. I was to take on the role of a data analyst, and answer two questions posed by my stakeholders.

1. Are employees who only worked for the institutes for a short period of time resigning due to some kind of dissatisfaction? What about employees who have been there longer?
2. Are younger employees resigning due to some kind of dissatisfaction? What about older employees?

The Data

The data of interest for the DETE surveys were:

• ID: An id used to identify the participant of the survey
• SeparationType: The reason why the person's employment ended
• Cease Date: The year or month the person's employment ended
• DETE Start Date: The year the person began employment with the DETE

The data of interest for the TAFE surveys were:

• Record ID: An id used to identify the participant of the survey
• Reason for ceasing employment: The reason why the person's employment ended
• LengthofServiceOverall. Overall Length of Service at Institute (in years): The length of the person's employment (in years)

Data cleaning summary

I needed the ID/Record Id to represent each record in the data set as a unique respondent, and I needed SeparationType/Reason for ceasing employment to identify which respondents cited dissatisfaction as their reason for resigning.

Before I could answer any questions, I needed to clean and combine both surveys and categorize employees by length of service and age. The TAFE surveys already had a column for length of service (LengthofServiceOverall), but for DETE I had to calculate it using the DETE Start Date and Cease Date.

To better understand the data, I converted the numerical value for length of service to a categorical value to answer the first question:

• New: Less than 3 years at a company
• Experienced: 3–6 years at a company
• Established: 7–10 years at a company
• Veteran: 11 or more years at a company

Are employees who only worked for the institutes for a short period of time resigning due to some kind of dissatisfaction? What about employees who have been there longer?

Percent of employees in each service category that resigned due to dissatisfaction.

I found that about 30% of new (< 3 years) employees were resigning due to dissatisfaction. However, about half of established (7–10 years) and veteran (11+ years) employees are resigning due to dissatisfaction.

Are younger employees resigning due to some kind of dissatisfaction? What about older employees?

Percent of employees in each age group that resigned due to dissatisfaction.

Overall, it appeared that only 20–30% of employees aged 20–25 resigned due to dissatisfaction. However, this proportion increases by 10% for employees aged 26–30. Employees in this age group are just as dissatisfied as employees aged 51 and older. Interestingly the percent of dissatisfied employees hovers around 35–43% for most of the data.

Conclusion

Personally, I would’ve expected newer and younger employees to resign in higher numbers compared to older and more established employees. Young people typically are less established and have more freedom to change careers. Newer employees have been in the company a short amount of time and would be less likely to feel obligated to stay on.

To answer the two questions posed by stakeholders:

• About 30% of newer employees that resign are doing so due to dissatisfaction. About half of more established employees are dissatisfied and resigning.
• Similarly, only 20–30% of younger employees are resigning due to dissatisfaction. This number goes up for employees 26–30 years old, declines for employees approaching 40, then increases again for employees older than 41.

It appears that dissatisfaction is most pronounced with years of service. As employees are in the job longer they become less satisfied.