The way I did this was to have absences as SCD (Slowly Changing Dimension) type 2
(Google will explain it better than I can). The fact is a snap shot in time.
Here's an example of the absences dimension table, (simplified)
%keyAbsence Start Date End Date Person 1 08/01/2012 09/01/2012 Mark 2 15/07/2012 20/07/2012 Mark 3 14/09/2012 15/09/2012 Sarah
I used intervalmatch() to match start and end dates to a master calendar, allowing me to create the fact table.
You can then count users by their absences. So in your scenario create a chart with person as the dimension and an expression asIF(COUNT(DISTINCT %keyAbsence)>2,1)
he user can just select their own start date and end date..
Hope that is of some help if not It would be helpful if you could upload a photo of your data model or even upload the actual dashboard (but I'm guessing this is sensitive data, so probably not an option!)