When Empirical Isn’t

The recurring theme I have encountered today is commenters relying on empirical measures that aren’t really as empirical as one might think. One example of that is William Galston’s Wall Street Journal op-ed. In it he’s complaining about what actually may be a problem—enduring truancy after the COVID lockdowns:

About 15% of students nationwide in 2019 were chronically absent, meaning they missed 10% or more of the school year, or about 18 days, according to the American Enterprise Institute. Full data aren’t yet available from the 2023-24 school year, but fragmentary statistics from local jurisdictions aren’t encouraging.

Students from poorer families are more likely to be chronically absent from school, but even in the nation’s richest districts, chronic absenteeism was nearly twice as high in 2023 as in 2019. Surprisingly, the length of time a school was closed isn’t a reliable predictor of absenteeism. Chronic absenteeism in 2023 stood at 28% of students for districts that remained closed the longest, not far ahead of 25% of students for districts that reopened the fastest.

A district’s racial makeup is a better predictor of what percentage of students miss school, but even in majority-white districts, chronic absenteeism rose significantly, from 13% in 2019 to 22% in 2023, compared with 17% to 30% in majority nonwhite districts.

That sounds pretty alarming but how serious is this issue really?

The problem is that “chronic truancy” is not consistently defined or applied. In Illinois “chronic truancy” is defined as nine days per academic year of unexcused absences. In Washington, DC it’s ten days. But there is no distinction between nine days per year, nine days per month, or never attending school at all.

I think we should be able to agree that outright avoidance of school is a problem. How serious is just not going to school for one day per month? That’s all that DC’s definition of chronic truancy requires.

A real chronic truancy problem has all sorts of implications among them that it calls into question the entire strategy of individuals improving their prospects in life through education. If you won’t go to school, that’s not going to happen. I should add that for the last 50 years teachers have been complaining about being saddled with addressing every social problem rather than just “readin’, ritin’, and ‘rithmetic”.

3 comments… add one
  • steve Link

    Per what I think is an official DC site its 10 days per year. Also, right after they define that at the end of section 1 they go on to define chronic absenteeism. Truancy is an unexcused absence. Absenteeism include excused and unexcused days. 10 excused days still seems like a lot to me but 10 days without an excuse seems like enough to indicate there could be a real problem.

    https://attendance.dc.gov/page/attendance-faqs#:~:text=A%20student%20who%20is%20truant,truant%20student%20under%20the%20law.

    Steve

  • Per what I think is an official DC site its 10 days per year.

    That’s exactly what I said.

  • Drew Link

    I think we need to separate the issues. Absenteeism is absolutely a problem. Parsing the incidence is useful: inner city kids are more prone. Further, in a class of twenty, it’s really, say, six. This is real live empirical (original) research from someone who actually taught in the inner city. Not some Pavlovian citation of an (agenda driven) , ahem, “study.”

    Switching gears, Talking about precision, eg 9 vs 10 days, is just jacking off. Or other precision issues. Let’s not let false precision replace experience, wisdom and intuitive capabilities.

    Lack of precision, and the importance of the absentee issue you cite are not mutually exclusive issues in analysis, but they are different.

    I have no doubt Covid created an opportunity (read: excuse) for some. But this issue is structural. And though not exclusive to the poor, it is dominated by the poor.

    I really find all of this to be yet another manifestation for the hysterical, and void of cost/benefit response to Covid.

    Almost makes you think Covid was political, and not scientific……

Leave a Comment