Enrollment change needs two comparable observations
Compare enrollment for the same entity and a defined pair of school years. Absolute change is ending enrollment minus starting enrollment. Percentage change divides that difference by starting enrollment, then multiplies by 100. A zero or unavailable starting value cannot support the usual percentage-change formula.
A worked example
A district moving from 10,000 to 9,500 students has a change of −500 students, or −5%. A district moving from 1,000 to 950 has the same percentage change but loses 50 students. These hypothetical examples show why both the count and the percentage belong in a comparison.
Check the district and school population
Renamed schools, new campuses, closures, district reorganizations, and changing grade spans can change a total without reflecting the same continuing population. Join using documented identifiers, review unmatched records, and keep the inclusion rule constant. Do not mix a sum of selected schools with an official district total without checking their coverage.
What this release summarizes
The enrollment trends summary presents derived school-level classifications for the 2020-to-2024 comparison in release v2026.07. Counts of shrinking, stable, or growing schools are not district enrollment growth rates. Consult NCES Common Core of Data files for year-specific files and the CCD definition for the directory universe.
What a trend does not explain
Enrollment change alone cannot establish a cause. Attendance boundaries, grade configurations, population changes, and program choices require additional evidence. Avoid describing a declining total as a measure of instructional quality or forecasting future enrollment from two observations.
Need the trend field for every school?
Start with the free Austin ISD sample, then choose district, metro, state, or national coverage.
Email me the sample →