How common is dyslexia?
The best global estimate comes from a 2022 systematic review and meta-analysis that pooled prevalence studies of primary school children from the 1950s to June 2021. It put the prevalence of developmental dyslexia at 7.10%, with a 95% confidence interval of 6.27% to 7.97%. In a typical class of 30, that is about two children.
The rate barely changed with the writing system: 7.26% in alphabetic scripts and 6.97% in logographic scripts such as Chinese, and 7.13% in regular spelling systems against 7.55% in irregular ones; none of these differences was statistically significant. The authors note that most studies were small and used different definitions, so treat the figure as an estimate with a range, not a precise count.
National guidelines quote ranges for the same reason. The German clinical guideline summary gives 3–11% of children and adolescents for reading and/or spelling disorder, and the French national research institute Inserm estimated at least 3–5%. Where a study draws the line (the bottom 5%, 10% or 15% of readers) largely decides the number it reports.1,6,7
| Study population | Estimate | Source and data year |
|---|---|---|
| Germany (guideline summary) | ≈8% | combined reading and spelling disorder, 20166 |
| Brazil, grades 2–6 | 7.5% | reading impairment, N = 1,618, 20168 |
| Global, primary school | 7.1% | meta-analysis, studies to 2021, published 20221 |
| France, age 10 | 6.6% | longitudinal study cited by Inserm, 20077 |
| Netherlands | 3.6% | estimate cited by the Dutch government, 20199 |
| Spain, Canary Islands | 3.2% | psychometric criteria, N = 1,050, 200910 |
How many children in the US have dyslexia?
There is no official US count of dyslexia. Schools report students under the Individuals with Disabilities Education Act (IDEA), and dyslexia falls under its "specific learning disability" (SLD) category. In 2022–23, 2,408,000 students aged 3 to 21 were served for a specific learning disability. That was 32.0% of the 7,526,000 students served under IDEA and 4.9% of all public school students, making SLD the most common disability category.
Two things to keep in mind. SLD also covers difficulties in writing and maths, so the figure is not dyslexia alone. And it counts only students who qualify for special education, so a child who has dyslexia but gets support only in the general classroom, or none at all, is not in it.3,11
| School year | Share of enrollment | Source |
|---|---|---|
| 1980–81 | 3.6% | NCES Digest table 204.303 |
| 1990–91 | 5.2% | NCES Digest table 204.303 |
| 2000–01 | 6.1% | NCES Digest table 204.303 |
| 2010–11 | 4.8% | NCES Digest table 204.303 |
| 2022–23 | 4.9% | NCES Digest table 204.303 |
Dyslexia statistics in England (UK)
England's Department for Education records the primary type of need of every pupil with special educational needs. In 2025/26, 180,217 pupils had "specific learning difficulty" (SpLD) as their primary need: 158,833 on SEN support and 21,384 with an education, health and care (EHC) plan. That is 2.2% of the 8,359,455 pupils in the data set (our calculation), up from 179,554 the year before.
SpLD includes dyslexia, dyscalculia and dyspraxia, and only a pupil's primary need is recorded, so a child whose main recorded need is something else does not appear. Independent schools are not in this data set. Of the pupils with SpLD as primary need, 95,557 were boys and 84,660 girls.4
Dyslexia statistics by country
The table puts the best published prevalence estimate next to the closest official count for each country. They measure different things: a prevalence study tests a sample of children against a research definition, an official count records children a school system has identified and placed in a category. Where we found no figure, the table says so.3,4,6,7,12,10,13,9,8,1
| Country | Prevalence estimate | Closest official count | What the count covers |
|---|---|---|---|
| United States3 | No national prevalence study found | 2,408,000 students, 4.9% of public school students (2022–23) | Specific learning disability under IDEA: reading, writing and maths |
| England (UK)4 | No national prevalence study found | 180,217 pupils, 2.2% (2025/26) | Specific learning difficulty as primary need; includes dyscalculia and dyspraxia |
| Germany6 | 3–11% of children and adolescents; about 8% combined reading and spelling disorder (2016) | No official figure found | – |
| Austria | No national figure found | No official figure found | – |
| Switzerland | No national figure found | No official figure found | – |
| France7,12 | At least 3–5%; French studies 6–8% (Inserm, 2007) | No official count found; Assurance Maladie states 3–5% of children (2024) | All specific learning disorders of reading, writing, spelling and maths |
| Belgium | No national figure found | No official figure found | – |
| Spain10,13 | 3.2% met psychometric criteria (Canary Islands, N = 1,050, 2009) | About 250,936 students supported for learning disorders (2024–25, our calculation) | 26.8% of 936,328 students with "other specific needs"; all learning disorders |
| Mexico | No national figure found | No official figure found | – |
| Netherlands9 | 3.6% dyslexia; 8.8% severe reading or spelling problems (cited 2019) | 15,932 final-exam candidates with dyslexia adjustments (2019) | Exam adjustments only; the government reported no national count of dyslexia care |
| Brazil8 | 7.5% reading impairment, grades 2–6, four cities (2016) | No official figure found | – |
| Worldwide1 | 7.1% of primary school children (meta-analysis, 2022) | No global count exists | – |
Why official counts are lower than prevalence estimates
If about 7% of children have dyslexia, why does England record specific learning difficulty for 2.2% of pupils? The first reason is under-identification. In a US study of 491,103 beginning second-graders, only 1 in 4 boys and 1 in 7 girls who met the study's criteria for reading impairment had been identified by their school as having a learning disability.
The second reason is definitions. Official categories are built for school funding and support, not for counting dyslexia: the US category includes writing and maths, England records only the primary need, Spain counts all learning disorders together, and the Netherlands told its parliament in 2019 that it had no clear view of how many children receive dyslexia care.
The third reason works in the other direction: who decides. In the Spanish Canary Islands study, teachers flagged 27.9% of children as having learning difficulties, but only 8.6% met psychometric criteria and 3.2% met them for dyslexia. Teacher concern is a reason to look closer, not a count.14,4,9,10
At what age is dyslexia identified?
England's data show how late recording can come. Among 5-year-olds, 0.53% of pupils had specific learning difficulty recorded as their primary need; by age 12 it was 3.29%, more than six times as many (our calculation from the DfE age data, 2025/26). The difficulty is usually present much earlier: children at family risk show weaker language and speech-sound skills before school.
Recording rises through primary school because a reading difficulty becomes visible once reading is taught and compared with classmates, and because assessment takes time. Screening and support do not need to wait for that: our dyslexia symptoms by age list shows the early signs, and the free dyslexia screening test checks the skills behind reading.4,2
Is dyslexia more common in boys or girls?
More boys than girls have dyslexia, but the gap is smaller than school referral figures suggest. In the 2022 meta-analysis, prevalence was 9.22% in boys and 4.66% in girls. Four epidemiological cohorts in New Zealand and the UK found reading disability in more boys than girls in every sample, with odds ratios from 1.39 to 3.19.
The ratio grows with severity. In 491,103 US second-graders it peaked at 2.4 to 1 on a broad reading fluency measure and 1.6 to 1 on a narrow decoding measure. A twin study found that the boys' over-representation at the low end is explained by differences in the average and the spread of reading scores between the sexes, partly linked to processing speed and inhibitory control.
Girls are easier to miss. In a large twin sample, reading disability went with inattention in both sexes, but with hyperactive and impulsive behaviour only in boys; that behaviour is more disruptive and more likely to prompt a referral.1,15,14,16,17
| Cohort | Boys | Girls | Odds ratio |
|---|---|---|---|
| Dunedin (NZ) | 21.6% | 7.9% | 3.19 |
| Christchurch (NZ) | 20.6% | 9.8% | 2.38 |
| ONS (UK) | 17.6% | 13.0% | 1.43 |
| E-Risk (UK) | 18.0% | 13.0% | 1.39 |
Dyslexia across languages
Dyslexia exists in every language studied, and the 2022 meta-analysis found similar rates in regular and irregular spelling systems. What changes is how it shows. In a study of word reading at the end of the first school year, children in most European languages read familiar words with more than 95% accuracy, against about 75% in Portuguese, French and Danish and 34% in English (rising to 76% in the second year).
Because reading is accurate early in regular languages such as German, Spanish or Dutch, dyslexia there shows mainly as slow, effortful reading and poor spelling rather than many reading errors. A comparison of German and English children with dyslexia found the same core: slow reading, trouble with made-up words and a slow, letter-by-letter decoding process.
A study of 1,114 children with dyslexia and 1,138 controls in 6 languages (Finnish, Hungarian, German, Dutch, French, English) found that phoneme awareness and rapid naming were the strongest predictors everywhere, and that their impact was larger in the more complex spelling systems.1,18,19,20,21,22
Family history and heritability
Family history is the strongest known risk factor. A meta-analysis of 21 independent samples of children with a parent or sibling with dyslexia found that 45% (95% CI 39%–51.2%) went on to have dyslexia, against 11.6% of children without that family history. Individual studies ranged from 29% to 66%.
Twin studies point the same way. A meta-analysis of 49 twin studies with more than 38,000 people found that genetic influences explained more of the differences in reading and reading-related skills than either shared or individual environment. There is no single dyslexia gene; many genes each add a small effect, and family history raises the risk rather than deciding the outcome.2,23,24
Dyslexia with ADHD, dyscalculia and language difficulties
Dyslexia often comes with other difficulties. A review of studies from 2001 to 2011 found a mean co-occurrence rate of 45.1% between learning disabilities and ADHD; that figure covers learning disabilities in general, not dyslexia alone.
For maths, a meta-analysis found that students with a maths disability were 2.12 times more likely to also have a reading disability (odds ratio, 95% CI 1.76–2.55). In a representative sample of 1,454 third-graders, combined literacy and arithmetic deficits were 4 times more frequent than chance would predict. Our sister site dyscalculia.io covers maths difficulties, and our page on dyslexia and dyscalculia explains the difference.
Language difficulties overlap too. In one study, 55% of children with a specific reading disability had impaired oral language, and 51% of children with a specific language impairment had a reading disability.25,26,27,28
Effect of reading interventions
Dyslexia is lifelong, so these figures do not describe dyslexia going away. They describe how much reading skills improve with teaching. Studies report this as an effect size: an effect of 0.3 means the average child who received the intervention ended up about a third of a standard deviation ahead of the average child in the comparison group, a modest but real gain.
Across 22 randomised controlled trials, phonics instruction had an average effect of g = 0.32 on reading (95% CI 0.18–0.47) and 0.34 on spelling, and it was the only approach whose effect was statistically confirmed. A 2023 meta-analysis of 53 studies of children aged roughly 5 to 11 with or at risk for dyslexia found g = 0.33 on standardised reading tests; more hours of intervention went with larger effects, while grade level made no significant difference.
Earlier is not automatically better in these data. Small-group reading interventions in kindergarten to grade 3 averaged 0.54 on standardised measures of foundational skills, with no difference by grade, and interventions in grades 4 to 12 averaged 0.21 on standardised measures. Measures and study designs differ, so compare these numbers with care; what they share is that structured teaching helps at every age.
Some popular approaches have no good evidence. Coloured overlays and lenses cannot be endorsed according to a systematic review, special dyslexia fonts did not improve reading rate or accuracy in two studies, and a joint report of US paediatric and ophthalmology bodies found no scientific evidence that vision training or coloured lenses work. Our page on how to help a child with dyslexia turns the evidence into a daily plan.5,29,30,31,32,33,34,35
| Meta-analysis | Effect size | Scope |
|---|---|---|
| Kindergarten–grade 3, small groups | 0.54 | 72 studies, standardised foundational skills, 201630 |
| Grades K–5, with or at risk for dyslexia | 0.33 | 53 studies, standardised reading, 202329 |
| Phonics, randomised trials | 0.32 | 22 RCTs, reading, 20145 |
| Grades 4–12, struggling readers | 0.21 | standardised measures, 201531 |
Dyslexia in adults: what the numbers show
We found no authoritative national estimate of how many adults have dyslexia in the US, UK or any of the other countries above. Because dyslexia is lifelong, the childhood estimates are the best available guide, but they were measured in children.
What official data do show is adults and older students using adjustments. In the Netherlands, 15,932 final-exam candidates sat their exams with dyslexia adjustments in 2019, down from 20,933 in 2016; in 2018 about 8% of candidates used them. In the US, 21% of undergraduates reported a disability in 2019–20, a figure that covers all disabilities. How dyslexia shows in adult life is covered in dyslexia in adults, and the free dyslexia test for adults takes about 6 minutes.
If you need the cost and route of a formal assessment by country, the table on how dyslexia is diagnosed lists them; we do not repeat it here.36,9,37
How these numbers were compiled
We use only peer-reviewed studies (with a PubMed or DOI record) and official statistics from national agencies, ministries and parliaments. Fact pages from charities are not used as a source for numbers. Every figure shows its own data year; the year in the title is the year this page was updated. Where we calculated a share ourselves from official counts, the text says so.
When we could not find a figure for a country, the table says "no official figure found" rather than borrowing one. Last updated 2026-10-05. Next review planned for 2027-04-05, or earlier when the US Digest of Education Statistics or England's special educational needs release publishes a new year.3,4,13,9,7,12
- US: NCES Digest of Education Statistics table 204.30 and Condition of Education, school year 2022–23.
- England: Department for Education, Special educational needs in England, 2025/26, type-of-need and age-and-gender data sets.
- Spain: Ministry of Education statistics on students with specific support needs, 2024–25.
- Netherlands: answers to parliamentary questions on dyslexia, 2019. France: Inserm collective expertise (2007) and Assurance Maladie (2024).
- Prevalence, sex ratio, family risk, co-occurrence and intervention effects: the meta-analyses and cohort studies listed under Sources.