HR-analytics – absences
Absences and particularly sick-leaves and other non-planned absences are one of the most important and consequential measure which relates to analytics in human resources. I will use simple 'absences' in the following to discuss all non vacation-like & unplanned absences usually due to sickness. Absences cost money and harm the organization and individuals alike. Growth of absences in a team may be due to problems in organizing the work or high overwork-situation. Maybe the good and committed employees have been working too hard for too long in a constant awaiting for the situation to get better? Or is it so that the absence-rate starts increasing after a new leader in a business unit takes charge? Is work organized so that there is some variance in the work duties? On the individual level the continuous increase in short period sick leaves may be a sign that the persons ability to work is deteriorating.
When analyzing absences one should start with looking at the current state of things, and then get the organizations past months and years as a reference point and to see the changes and trend. Finally large impact can be gained by forecasting absence-rates and levels.
A good starting point is to define the measures/indicators related to absences. Typically these include absence days per person, absence percentage (absence rate) and costs due to absences.
Absence days per employee
Measuring absences can be started with looking at how many absence days per employee there is (usually using FTE as the amount of employees). This reveals how absences affect your business and common sense can be used to look at the level: is the absence days per employee high or low? This figure also usually links well to costs.
Absence rate
Absence rate can be calculated by dividing absence-hours with the total amount of work hours the employee would have worked had she not been sick. And ofcourse the same figures can be calculated using days instead of hours. Calculating this KPI may not be so straight forward since in some HR-systems it can be difficult to say which exact workdays the employee was absent in comparison to which calendar days she was absent. In any case absence rate is one of the most important things HR should measure and follow.
Direct cost of absence
Direct salary cost of absences can be extracted from the salary payments system using the correct payment types. This will serve as the basis on top of which one must count also other side costs of paying the salary. Depending on the contract the employee may earn vacation time also on the absence-days, which should be taken into consideration also. Employee payment system may not provide the link to time & attendance system or other absence-information which creates challenges to the analysis (costs by absence reasons?). Obviously absences cause many different costs, for example having substitute employees, healthcare costs etc. which need to be taken into account.
Dimensions for absences
When analyzing sick leaves its not enough to just have to total for the whole company. One must be able to answer the questions: who is absent? Drill down in department hierarchy down to individual department unit to and compare the amounts and absence rate. Usually absences don't split evenly. Other typical dimensions are profession, age, length of employment, supervisor and type of employment (fixed term vs. permanent). Which group has the highest absence rate or has there been significant change in one of the groups? Ofcourse a rise in absenteeism may be due to natural cause like seasonal flu but it may be due to overwork or low job satisfaction which should also reflect in employee turnover. In this case HR-analytics can have an impact for example by hiring more employees to stop the overwork situation. It's also interesting to look at the average length of sick leaves: does most of the total amount of absences come from the long absence-periods or from many short few days absence periods?
Forecasting
Company should be able to predict absences. Some of the prediction may relate to personnel planning and employee reserves so that production will roll in all situations. Also interesting part of absence-analytics is make prediction for signs weather an employee is going to need a very long sick leave or even losing her ability to work? Prediction offers a chance to change the future which can yield profits in money and health. Predictions can be done by using machine learning algorithms.
In the end one should keep in mind that absences can be a personal issue and everyone will be sick sometimes. Employers must give enough time to recuperate.
