Two standards way
Absence as a percentage
Absence expressed as number of days off
There are no rules about how to do it but the rising number of part-timers means that is becomes more complicated.

In a previous job absence levels went up by 0.1% when I arrived, because I did it 'correctly!

In reply to Peter Stanway:

Sorry Molly

Welcome to the Community

As Peter says, no way is "right" in an absolute sense. Where you have people whose hours vary or where you have populations working different durations, then working in %s will neutralise 99% of any calculation problems. %s of what?

Hours counted as for absenteeism / planned hours worked x100

Be clear as to what you will include for absenteeism purposes. Obviously holidays will be excluded, but what about territorial army and other absences of this nature. You may also want to do calculations both with and without maternity absence.
I've also included for each person the number of absence incidents in the year. In one company we excluded long-term absences (2 weeks or more) from one of the calculations because in many cases these absences could be planned for and had little disruptive effect on operational organisation.