The most insightful line in CAP's new report Teacher Absence as a Leading Indicator of Student Achievement is surely:
Yet very little is known about the properties of this new school-level measure.
If, for example, teacher absence is a leading indicator of student achievement, you might expect the state with the lowest rate (Utah, 21% absent more than 10 days) to have significantly higher 8th grade NAEP scores than the state with the highest (RI, 50%). Alas, they don't.
This is the kind of think tank report where nobody really disputes the basic points -- a lot of teacher absences aren't good and they're influenced by direct policy choices (rules about taking absences) but also an indicator of problems with the culture of a school -- but you know in the end it is just going to be a hammer used to make teaching a worse job.
Also, there's pretty obviously a lot of potential slop in the data (e.g., If a teacher quits after two weeks, how many days of absence is it? If their long-term sub misses 10 days, is that two absent teachers for one job? etc...) and not a lot of explanation of its rigor.