If you want to assess the state of absenteeism within the organization, you often start by looking back. How much absenteeism was there in the past year? To calculate the absenteeism rate, you compare the number of days someone was absent in a year with the number of days they could have worked that year. This is usually calculated in calendar days, but it can also be in work or roster days.
However, this doesn’t give the complete picture. You also want to know how many employees do and do not step over the threshold to report sick. For this, you look at how often employees report sick on average: the reporting frequency. This is determined by dividing the number of sick reports by the number of employees. If the reporting frequency is 1, employees report sick on average once a year.
But be aware, some employees did not report sick at all that year. This is called the zero absenteeism rate. The importance of zero absenteeism is often underestimated, sometimes even not processed in systems. Yet it is a very important factor for the conclusion.
Also, look at how long the absenteeism cases last on average: the average absenteeism duration. For this, you look at the closed absenteeism cases. For the other cases, you do not yet know how they will develop. The distinction between short and long cases is arbitrary but good to make. Every absenteeism specialist knows that a relatively large share of short or long absenteeism cases often has a completely different underlying issue.
By combining all this data, you can see if there is a problem. Once you have determined what absenteeism looks like within the department or organization, you can compare it internally with your own goals, with other departments within the organization, or externally with benchmark figures. It is also good to look at the trend in the organization and predict what will happen if the organization does nothing; will the figures get better or worse? You can outline this very simply in Excel, but it can also be done with a complex predictive model where the prediction is based on a combination of variables over time. These variables can include: age development, development of the share of frequent absentees, development of the male-female ratio, development of team reception, development of the share of employees working with a roster, development of the share of employees in direct contact with customers, and so on. The latter is obviously more accurate, but it remains a prediction.
Based on the previous analyses, the organization decides whether it has a problem it wants to address. The need to do so is greater as data, of course, shows that absenteeism affects other results. For example, customer satisfaction, because there are not enough people to provide services. Or financial results, because absenteeism costs money. Or the reputation as an employer, making it harder for the organization to attract applicants for vacancies.
If all four KPI’s are wrong, the organization has an overall problem, a so-called absenteeism culture, which it will have to tackle from all sides: both aimed at reducing the number of employees who are absent and shortening the duration of absenteeism. However, it may turn out that the reporting percentage is low and zero absenteeism is high, but the average absenteeism duration is long. Then the threshold to return after reporting sick is apparently very high, so you focus mainly on the employees who are sick and try to shorten the absenteeism process.
It may also be that the reporting frequency is high, but the absenteeism duration and zero absenteeism are very low. Then employees are not sick for long, but they report sick very often. Then it is better to investigate the cause of getting sick rather than staying sick. Also, check whether the problem occurs throughout the organization or only in a part or specific group.
Example calculation
Suppose the organization has a reporting frequency of 1. That is the average reporting frequency in the Netherlands, so it seems to be fine. But it makes a big difference whether there is a zero absenteeism rate of 20 percent or 80 percent. If the zero absenteeism rate is 20 percent, 80 percent of the employees reported sick that year. The average reporting frequency of 1 is therefore caused by 80 percent of the employees. They reported sick an average of 1.25 times this year. If the zero absenteeism rate is 80 percent, the reporting frequency of 1 is caused by 20 percent of the employees. A small group reported sick an average of 5 times that year. That can be a problem.
Absenteeism brings many costs for an employer, but these are often difficult to calculate exactly. Still, it is good to make an estimate of these costs. This allows you to determine how much a good and targeted absenteeism policy can yield. And how the organization can best insure itself or what provisions it should make. You can also use absenteeism costs to make a business case. For example, you state: if we implement this solution, we expect absenteeism to decrease by X percent, which will yield X euros. You then compare this to the costs.
