← All Miami-Dade stories

Miami-DadeMethodology and data qualityeasy

The federal chronic absenteeism number for Miami-Dade elementary schools is lower than the state's because it leaves out pre-kindergarten

Thesis

Reporters comparing Florida's chronic absenteeism figures with the U.S. Department of Education's will find the federal counts lower at most elementary schools. The two are the same submission: at the schools with no pre-kindergarten enrollment the counts agree exactly at about four in five and within one student, the federal perturbation, at more than nine in ten. At schools with pre-K, the annual-count gap has a regression slope of about 0.66 against October pre-K membership. This supports a population-definition difference but does not directly estimate pre-K chronic-absence prevalence: annual enrollment and a one-day October count have different denominators, and the fitted intercept is not zero. EDFacts counts kindergarten through grade 12; the state file counts everyone enrolled.

Supporting findings

  • 2022-23 comparison of schools matched by NCES id; the gap regresses on pre-K membership with r above 0.85. The exact counts, shares and slope are on the chronic absenteeism metric page and re-measured on every harness run.
  • Schools whose pre-K count is itself suppressed (one to nine pre-K students) are in neither group, because their pre-K membership is unknown.
  • The state also publishes a second definition, absent 21 or more days, and the federal civil-rights survey a third (15 or more days); none of the three agree on levels.
  • The state file is served as the metric; the EDFacts copy is read only as the cross-publisher check, with the population difference documented on the metric page.

Reporting directions

  • Use the state's file for any elementary-school comparison; use EDFacts only for subgroup breakdowns the state does not publish.
  • Ask the district for its own pre-K attendance policy: roughly two in three pre-K students are chronically absent by the state's definition, which is a story in itself.

Where the numbers are

Methodology & replication recipe — the queries and sources behind the findings above.

More on methodology and data quality