# How Many US Kids Are Homeschooled? (2026) Author: Sonakshi Arora Author URL: https://www.monstermath.app/blog/author/sonakshi-arora Published: 2026-10-28 Category: Homeschooling Statistics Category URL: https://www.monstermath.app/blog/category/homeschooling-statistics Meta Title: How Many US Kids Are Homeschooled? (2026) Meta Description: About 3.4% of US students are homeschooled, and one in five of those families cites special needs the school couldn't meet. The federal figures, explained. Tags: statistics, homeschool Tag URLs: statistics (https://www.monstermath.app/blog/tag/statistics), homeschool (https://www.monstermath.app/blog/tag/homeschool) URL: https://www.monstermath.app/blog/how-many-us-kids-are-homeschooled-2026 **_TL;DR:_** _The latest published NCES national survey estimated that 3.4% of US K–12 students - about 1.77 million children - were homeschooled in 2022–23. More recent state data show that homeschooling has continued to grow in many places, although a directly comparable national estimate for 2025–26 is not available. In 2022–23, 21% of homeschooled children's parents cited unmet special educational needs as one reason for homeschooling._ If you're weighing up homeschooling for a neurodivergent child, it helps to know how many families have made the same decision and why. The answer involves more caveats than most roundups admit, so this article gives you the figures along with what they do and don't measure. Most of the numbers below come from two federal sources: the National Center for Education Statistics, which runs the official homeschooling count, and the Census Bureau, which captured the pandemic years. For the most recent picture we've also used state registration data compiled by Johns Hopkins. ## The headline numbers Survey year Homeschooled students Share of students ages 5-17 1999 850,000 1.7% 2003 1,096,000 2.2% 2007 1,520,000 3.0% 2012 1,773,000 3.4% 2016 1,690,000 3.3% 2019 1,457,000 2.8% 2022-23 1,765,000 3.4% Figures for 1999 through 2019 come from the NCES Digest of Education Statistics \[2\]. The 2022-23 figures come from the Parent and Family Involvement Survey \[1\]. Both count homeschooled children aged 5 to 17, as a share of all children that age who are in school or homeschooled. ## How many kids are homeschooled right now? The latest official figure is 3.4% of students in 2022-23, or about 1,765,000 children \[1\]. That's up from 2.8% in 2018-19. A wider figure is worth knowing too. NCES counts 5.2% of students as learning at home, up from 3.7% in 2018-19 \[1\]. That's the 3.4% who are homeschooled, plus another 1.8% in full-time virtual school whose parents don't call it homeschooling. Which figure gets quoted makes a big difference, so it's worth checking. 3.4% is homeschooling. 5.2% is homeschooling and virtual school counted together. ## Is homeschooling growing? Less obviously than the headlines suggest. The rate climbed through the 2000s, fell back by 2019, and by 2022-23 was level with where it stood in 2012 \[2\]. That federal count stops at 2022-23. For something more recent, the Johns Hopkins Homeschool Research Lab tracks the registration counts that 30 states publish each year. Of the 22 states that had reported for 2025-26 by September 2026, 17 showed increases and five showed declines \[5\]. The trend across three years is the clearer signal. The share of reporting states showing growth has fallen from 90% in 2023-24, to 80% in 2024-25, to 77% this year. And across those 22 states, the growth rate itself dropped from 6.3% last year to 1.5% \[5\]. As Johns Hopkins researcher Angela Watson puts it, "Homeschooling is still growing, but not the way it was." These state counts and the federal survey measure different things, so the two sets of figures aren't directly comparable. ## The pandemic spike, and who it reached During the first year of the pandemic the Census Bureau tracked homeschooling month by month. Among households with school-age children, the share of homeschooling went from 5.4% in April 2020 to 11.1% that October \[4\]. That survey counts households rather than individual students, so it isn't comparable to the 3.4% figure above. The variation underneath is the interesting part. Among households where the respondent identified as Black, homeschooling rose five-fold, from 3.3% to 16.1% \[4\]. Massachusetts went from 1.5% to 12.1%. Plenty of other states barely moved at all. ## Why parents homeschool Parents can give more than one reason, so these don't add to 100%. From the 2022-23 survey \[1\]: Reason given Share of homeschooling parents Concern about school environment (safety, drugs, peer pressure) 83% Desire to provide moral instruction 75% Dissatisfaction with academic instruction elsewhere 72% Emphasis on family life together 72% Desire to provide religious instruction 53% Desire for a nontraditional approach 50% Child has other special needs the school can't or won't meet 21% Child has a physical or mental health problem lasting 6+ months 15% Child has a temporary illness 2% ## How many kids are homeschooled for special needs? This is the figure most relevant to families with neurodivergent children, and it's the one most roundups bury. In 2022-23, parents of 21% of homeschooled kids cited special needs the school couldn't or wouldn't meet as one of their reasons \[1\]. Another 15% cited a physical or mental health problem lasting six months or more. These categories overlap, and neither tells us how many homeschooled children are neurodivergent. Neither share is new. In 2019 those same two reasons came in at 23.1% and 15.6% \[3\]. Parents have been citing special needs at about this rate for years. One nuance. Asked to name the single most important reason, only 7% of parents chose special needs and 5% chose a health problem \[1\], against 28% who chose concern about the school environment. So for many of these families, special needs could be just one reason among several. ## Does "special needs" mean neurodivergent? Not necessarily. The survey's special-needs category covers many kinds of educational need. It doesn't say which children are autistic, have ADHD, or have dyslexia. It runs the other way too. A neurodivergent child might be homeschooled for reasons that have nothing to do with their diagnosis: flexibility, family preference, or the school environment. So the 21% is useful for understanding why parents homeschool. It isn't a count of neurodivergent children. ## What these numbers can't tell you Four limits worth knowing, especially if you're citing these figures elsewhere. **The published figures don't break down by diagnosis.** NHES does collect information about children's health and disability. But no published national table splits homeschoolers by specific diagnosis, and the likely reason is sample size: homeschoolers are a few hundred cases in the survey, too few to divide by individual conditions and still meet NCES reporting standards. So if you come across a precise figure for autistic homeschoolers in the US, it's worth asking where it came from. **Registration is patchy.** Roughly a quarter of states don't require homeschooling families to register at all, so state records can't fill the gap. **The definition has an attendance threshold.** NCES counts children homeschooled for all or some classes but excludes anyone attending a public or private school more than 25 hours a week. **It's all parent-reported.** One parent answers about one selected child. Nobody verifies what's happening. Unschooling is even thinner. There is no reliable count at all, and researchers in the field say so directly, which is why our piece on [unschooling for neurodivergent kids](https://www.monstermath.app/blog/unschooling-for-neurodivergent-kids-can-it-actually-work) talks about the evidence rather than the numbers. ## What to take from the numbers Homeschooling covers about 3.4% of US students, roughly 1.77 million children, which is where it stood in 2012 as well. Around one in five homeschooling families gives special needs as one of their reasons, and that share has been steady for years. So if school hasn't worked for your child, this is a well-travelled route rather than an unusual one. What these numbers won't tell you is whether it works. They count who is doing what, not how children fare. ## FAQs ### How many kids are homeschooled in the US? About 3.4% of K-12 students, or roughly 1.77 million children, in the 2022-23 school year according to the National Center for Education Statistics. That's up from 2.8% in 2018-19 and matches the previous peak of 3.4% recorded in 2012. ### Did homeschooling really double during the pandemic? Household rates did, briefly. The Census Bureau's Household Pulse Survey found 5.4% of households with school-age children homeschooling in April 2020 and 11.1% by October 2020. But that measures households rather than students, and the question wording changed between the two waves, so it isn't directly comparable to the 3.4% student figure. ### How many homeschooled kids have special needs? In 2022-23, 21% of homeschooling parents said one of the reason for homeschooling was that their child had special needs the school could not or would not meet, and 15% cited a physical or mental health problem lasting six months or more. ### How many homeschooled children are autistic or have ADHD? There's no reliable national figure. The federal homeschooling statistics don't provide a count of homeschooled children with autism or ADHD, and homeschoolers make up too few cases in the survey to split reliably by individual diagnosis. Any precise number you see is an estimate. ### Is homeschooling still growing? Slowly, and less than it was. The federal rate doubled between 1999 and 2012, then flattened, and the 2022-23 figure of 3.4% matches 2012's. More recent state registration counts compiled by Johns Hopkins show 17 of 22 reporting states growing in 2025-26, but the growth rate across those states fell from 6.3% to 1.5% in a single year. ### How many kids are unschooled in the US? There's no reliable figure. Unschooling is a philosophy rather than a reporting category, so no federal survey tracks it, and researchers who study it state plainly that firm prevalence data doesn't exist. ## References 1. Sempeles, E., & Cui, J. (2024). Parent and Family Involvement in Education: 2023 (NCES 2024-113). _National Center for Education Statistics_. [https://nces.ed.gov/pubs2024/2024113.pdf](https://nces.ed.gov/pubs2024/2024113.pdf) 2. National Center for Education Statistics. (2022). Number and percentage of homeschooled students ages 5 through 17, by selected characteristics: Selected years, 1999 through 2019. _Digest of Education Statistics_, Table 206.10. [https://nces.ed.gov/programs/digest/d22/tables/dt22\_206.10.asp](https://nces.ed.gov/programs/digest/d22/tables/dt22_206.10.asp) 3. National Center for Education Statistics. (2021). Percentage and percentage distribution of homeschooled students ages 5 through 17, by reasons their parents gave for homeschooling: 2019. _Digest of Education Statistics_, Table 206.15. [https://nces.ed.gov/programs/digest/d21/tables/dt21\_206.15.asp](https://nces.ed.gov/programs/digest/d21/tables/dt21_206.15.asp) 4. Eggleston, C., & Fields, J. (2021). Census Bureau's Household Pulse Survey shows significant increase in homeschooling rates in fall 2020. _U.S. Census Bureau_. [https://www.census.gov/library/stories/2021/03/homeschooling-on-the-rise-during-covid-19-pandemic.html](https://www.census.gov/library/stories/2021/03/homeschooling-on-the-rise-during-covid-19-pandemic.html) 5. Watson, A. R. (2026). Homeschool Growth: 2025-2026. _Johns Hopkins Institute for Education Policy_. [https://education.jhu.edu/edpolicy/policy-research-initiatives/homeschool-hub/homeschool-growth-2025-2026/](https://education.jhu.edu/edpolicy/policy-research-initiatives/homeschool-hub/homeschool-growth-2025-2026/) --- This blog is powered by Superblog. Visit https://superblog.ai to know more. ---