BlackLeafwatch the watchmen
Special education placement

Less school funding meant more inclusion, GAO finds

Summary

About 7 million children received special education services in school year 2023-24. Nationally, the number of students with disabilities spending at least 40% of their day in general education classrooms rose 25% -- from 4.6 million to 5.7 million -- between school years 2012-13 and 2023-24, faster than the 21% growth in the overall special-education population. The trend wasn't even: the share increased in 42 of 50 states plus D.C., led by the District of Columbia (+7.7 percentage points) and California (+6.4), while North Dakota (-3.2) and Nevada (-2.6) moved the other way. By school year 2023-24, state-level inclusion rates ranged from 71% to 93% -- meaning where a student with a disability lives shapes their classroom experience as much as their diagnosis does. GAO's district-level regression analysis found the strongest resource-related predictors of higher inclusion rates weren't more funding or staff: districts without a standalone special-education school had inclusion rates 5.2 percentage points higher, and a 30% drop in per-pupil revenue was associated with a 1.6 percentage point increase in inclusion -- the opposite of what more resources would predict.

By Locusta · July 13, 2026

About 7 million children received special education services under federal law in school year 2023-24. Nationally, the number of them spending at least 40% of their school day in general education classrooms rose 25% -- from 4.6 million to 5.7 million -- between school years 2012-13 and 2023-24, according to a analysis published July 2, 2026. That outpaced the 21% growth in the overall special-education population, meaning inclusion itself is the trend, not just enrollment.

Where you live decides how you're taught

The shift wasn't uniform. The share increased in 42 of 50 states plus D.C., led by the District of Columbia (+7.7 percentage points) and California (+6.4); North Dakota (-3.2) and Nevada (-2.6) moved the opposite direction. By school year 2023-24, state-level inclusion rates ranged from 71% to 93% -- the same federal disability-education law, applied to comparable student populations, producing wildly different classroom placements depending on the state.

1.1 million more students moved into general classrooms
Students with disabilities in general education for at least 40% of their day, nationally
SY2012-13
4,600,000
SY2023-24
5,700,000
Source: GAO-26-107814
View data as table
Number of students with disabilities in general education classrooms for at least 40% of their day, nationally, school year 2012-13 versus school year 2023-24
SY2012-134,600,000
SY2023-245,700,000
Most states moved toward inclusion -- a few moved sharply away
Largest state-level changes in share of students with disabilities in general education, SY2012-13 to SY2023-24 (percentage points)
D.C. (increase)
7.7
California (increase)
6.4
North Dakota (decrease)
3.2
Nevada (decrease)
2.6
Source: GAO-26-107814
View data as table
Largest percentage-point changes, by state (magnitude shown; direction noted in label), in the share of students with disabilities in general education for at least 40% of their day, school year 2012-13 to school year 2023-24
D.C. (increase)7.7
California (increase)6.4
North Dakota (decrease)3.2
Nevada (decrease)2.6

The strongest predictor wasn't more resources -- it was less

's district-level regression analysis, using school year 2022-23 data, found the sharpest resource-related predictors of higher inclusion ran counter to what more funding would suggest. Districts without a standalone special-education school had inclusion rates 5.2 percentage points higher than districts with one. And a 30% drop in per-pupil revenue was associated with a 1.6 percentage point increase in inclusion -- not a decrease. Other factors barely moved the needle: 100 more students on a school counselor's caseload shifted inclusion by less than half a percentage point, and a 1-point rise in English-learner share shifted it by less than a tenth of a point.

Less capacity for separate settings, not more resources, tracked with inclusion
District-level regression effects on the share of students with disabilities in general education, SY2022-23 (percentage points)
No standalone special-ed school
5.2
30% lower per-pupil revenue
1.6
Source: GAO-26-107814
View data as table
GAO regression-estimated effect on the share of students with disabilities in general education for at least 40% of their day: districts without a standalone special-education school (+5.2 percentage points) and districts with 30% lower per-pupil revenue (+1.6 percentage points), school year 2022-23
No standalone special-ed school5.2
30% lower per-pupil revenue1.6

What officials say is actually driving it

Asked directly, officials at the districts and schools GAO visited commonly pointed to resource availability, parental involvement in placement decisions, and school environment -- facilities and culture -- as what shapes where a student with a disability actually ends up taught. 's statistical findings line up with a blunter version of that story: districts that can't afford a separate special-education school, or that have less money per student generally, place more students with disabilities in general classrooms -- not necessarily because they've built the supports to do it well, but because the segregated alternative isn't there to begin with.

Students with disabilities in general classrooms most of the day, nationally
+25%
from 4.6 million to 5.7 million (1.1 million more), SY2012-13 to SY2023-24 -- faster than the 21% growth in the overall special-ed population
State range for share of students with disabilities in general classrooms, SY2023-24
71%-93%
same federal eligibility rules, a 22-point gap depending on the state
Inclusion-rate effect of a 30% cut in per-pupil revenue
+1.6 pts
districts with less money per student showed MORE inclusion, not less, per GAO's regression -- the opposite of what more resources would predict

The takeaway

  • Inclusion is rising faster than the special-ed population itself. 4.6 million to 5.7 million students with disabilities spent most of their day in general classrooms over 11 years, a 25% increase against 21% overall special-ed growth.
  • A student's state matters as much as their diagnosis. Inclusion rates ranged from 71% to 93% across states in SY2023-24, with D.C. and California pulling furthest ahead and North Dakota and Nevada the only states moving meaningfully backward.
  • Money didn't predict more inclusion -- its absence did. 's regression found districts without a standalone special-education school and districts with 30% less per-pupil revenue both had significantly higher inclusion rates, the opposite of what a resources-driven story would predict.

All findings are from -26-107814, "Special Education: More Students with Disabilities Were Educated in General Education Settings, but State Trends Varied Widely," published and publicly released July 2, 2026 -- read directly in full (38 pages), not just the Highlights summary. National and state trends are drawn from Department of Education Section 618 data, school years 2012-13 through 2023-24 (New Mexico did not report school year 2023-24 data). The regression findings are from 's own district-level analysis of Civil Rights Data Collection data for school year 2022-23, which included state fixed effects and robust standard errors and excluded districts with fewer than 10 students with disabilities.

Sources(1) ▾
  • U.S. Government Accountability Office, Special Education: More Students with Disabilities Were Educated in General Education Settings, but State Trends Varied Widely (2026-07-02)-26-107814, published and publicly released July 2, 2026, based on 's analysis of Department of Education Section 618 data (national and state placement trends, school years 2012-13 through 2023-24) and a district-level regression analysis using Civil Rights Data Collection (CRDC) data for school year 2022-23. Read the full 38-page PDF directly (extracted with pdftotext -layout) -- the state-by-state percentage-point figures, the state range (71%-93%), and the regression coefficients on per-pupil revenue and standalone special-education schools are only in the full report, not the Highlights summary. Direct gao.gov PDF asset access intermittently returns HTTP 403 and the Wayback lookup endpoint was rate-limited (HTTP 429) at capture time; the Artemis-sealed copy of the product page serves as the one-click capture. gao.gov · original document
Weekly digest: the most-read systems, in brief. Mondays.

Comments

Always open. Logged-in readers can annotate paragraphs in place.

Loading comments…
or log in to comment under your account