How the numbers in this tool were put together, and how much to trust them.
Every enrollment figure in this tool comes from one source: the Arkansas Department of Education's own student data system (the Statewide Information System, or SIS), as published in the state's public enrollment reports. I use the number the state itself publishes and treats as the official enrollment count for each district and each charter school system, for every school year from 2013–14 through 2025–26.
I should be upfront about one thing: I have not independently verified how Arkansas computes this number internally, and the state's own labeling is not a reliable guide. One ADE report that carries this figure is titled as an "October 1" enrollment count, but my testing found it is identical, to the student, to a figure that behaves like an average of enrollment over part of the year, not a single-day headcount. What I can say with confidence:
I chose this number because it is what Arkansas itself publishes and stands behind — not because I have audited its internal computation.
There is a second candidate source: the U.S. Department of Education's Common Core of Data (CCD), which publishes a fall headcount for every district in the country. I compared the two, district by district, for every overlapping year. The federal count runs consistently about 3.5–4% higher than the state's figure — every year, in a stable and predictable way. (One year, 2023–24, the two nearly converge; I found no explanation for that anomaly and flag it as unresolved.)
This gap is not an error in either source. They measure different things: the federal number is a snapshot of who is enrolled on a single day in the fall; the state number behaves like membership measured over time, which runs lower whenever students leave during the year. But it does mean the two cannot be mixed in one time series without a hidden few-percent jump wherever the sources switch. I tried two hybrid designs, found both baked that bias into long-term trend measurements, and abandoned them in favor of using the state source alone.
Two practical points also favored the state data. First, it matches what Arkansans actually see: the state's published district figures and news coverage of them agree with my numbers exactly (for example, Springdale's 21,097 and Little Rock's 18,964 for 2025–26). Second, the federal file simply lost track of two real, continuously operating Arkansas districts (Cross County, and the Imboden Charter district), while the state data covers both without interruption.
The tool divides 2013–2026 into three eras, shaded on every chart:
One timing detail worth being precise about: the EFA era starts with the 2022–23 school year, the year during which the LEARNS Act was passed. Fall 2022 enrollment was counted a few months before the law was signed, and the first EFA students didn't enroll until fall 2023 (the program then phased in: limited eligibility in 2023–24, expanded in 2024–25, and open to all students in 2025–26). So the era boundary marks "the school year Arkansas's education policy changed," not "the first year EFA money was flowing." Anyone attributing enrollment changes specifically to the EFA program should keep that offset in mind and should also read the "What this tool cannot tell you" section below.
Comparing a district to itself over thirteen years only works if the district is the same thing over those thirteen years. For 228 of Arkansas's 235 districts, it is. Seven districts absorbed a neighboring district somewhere along the way, and for those seven, this tool is careful not to pretend otherwise:
For these seven, the years before the merger show the district as it existed then: smaller, without the absorbed district's students. I never add the two predecessors' enrollments together to fake a longer history, because real students are lost in every merger transition and a stitched-together number would invent a history no district actually had. On these districts' charts, the merger year is marked with a visible seam, the pre-merger years are drawn in a distinct dashed style, and a note explains that the jump at the seam is the merger itself, not a sudden change in the underlying population. Their long-term growth classifications are computed only from years where the comparison is apples-to-apples. (Pine Bluff is the one necessary exception: its merger came so late, 2021, that it has no consistent-boundary years at all in the baseline era, so its baseline is measured on its own pre-merger history, clearly flagged wherever it appears.)
One more short-history case: the Jacksonville North Pulaski district didn't exist until 2016–17, so its pre-EFA history is genuinely shorter, and the tool flags it accordingly.
District boundaries on the map come from the U.S. Census Bureau's 2024 TIGER school district boundary files, simplified so the map loads quickly in a browser. Boundary shapes are current boundaries. For the seven merger districts above, the shape shown is today's (post-merger) boundary. One district, Imboden Charter, has no boundary in the Census file at all and appears as a point marker at its school's location instead, alongside Arkansas's charter schools.
The tool tracks all 22 public charter school systems reporting enrollment to the state in 2025–26. Charters are shown more simply than districts (no trajectory classifications, no baseline-versus-EFA comparisons) because most charters are too new for a pre-2020 baseline to mean anything. Where a charter system operates several campuses, the enrollment shown is for the whole system, not the single mapped location, and the table says so on the affected rows. Charters that opened mid-way through the EFA era show change figures computed from their own first year of data, and the table flags any figure that rests on a single year-over-year comparison.
Verified: For all thirteen years, the statewide total in this tool matches Arkansas's own officially published statewide enrollment count exactly, to the student. Ten years (2016–17 through 2025–26) were checked against ADE's State Report Card system and ADE's public enrollment-count reports; the three earliest years (2013–14 through 2015–16) were checked against ADE's "Enrollment Count by State" report; and the most recent year (2025–26) was re-confirmed against ADE's live public report during this audit (465,421, exact match). As an internal consistency check, the tool's district, charter, and other components sum exactly to that verified total in every year, with zero residual. Spot checks of individual districts against the state's published district-level report also match exactly.
Verified with limits: Individual district-level figures all come from the same state data file whose totals reconcile exactly. All 235 districts have now been individually cross-checked against ADE's current "Enrollment Count by District" report for the most recent year (2025-26), and every one matched exactly, to the student. That check covers one year for every district, not all 13 years for every district; earlier years still rely on the reconciliation and spot checks above, not a district-by-district second-source check for every year. The comparison against federal data, and the finding that the state's figure is not a single-day count, are my own analysis, reproducible from public data.
Not verified: The state's internal method for computing its enrollment figure (see the first section). The cause of the 2023–24 federal/state convergence anomaly. News-media attributions for a small number of figures were recorded second-hand and match my data exactly. Several of these (Springdale's 21,097, Little Rock's 18,964, the LEARNS Act's March 8, 2023 signing date, and the Pine Bluff–Dollarway merger) have since been independently re-confirmed against original external sources; the remainder have not.
Enrollment counts show what changed, not why. This tool does not track where departing students went (EFA participation, private school, homeschooling, or a move out of state) and does not include private school or EFA program data. A district losing students in the EFA era is a fact; the reason is not something these counts alone can establish. Claims about causes need evidence beyond enrollment numbers, and this tool deliberately does not make them.
Rush, Joel. Arkansas Public School Enrollment 2013-2026. Sidetown Labs, 2026. https://arkansas-enrollment.sidetownlabs.com/.
This data and analysis are licensed under Creative Commons Attribution-NonCommercial 4.0 (CC BY-NC 4.0). In practice: attribution is required, and commercial use is not permitted without permission.
Looking for more depth? The technical version covers the same ground with source-by-source verification detail: what was tested, how, and exactly which claims are confirmed, approximate, or open.