Short answer
What does the term science of reading refer to?
The science of reading is the body of research from psychology, education and linguistics on how people learn to read and why some find it hard. Its central findings are that reading comprehension depends on two broad abilities, decoding written words and understanding language, and that most children learn to decode best when they are taught letter-sound relationships explicitly and in a planned order, known as systematic phonics. It is not a single program or product. It also stresses vocabulary, background knowledge and fluency, so it is broader than phonics alone.
- The simple view of reading describes comprehension as the product of decoding and language comprehension.
- Scarborough's reading rope shows word recognition and language comprehension as strands woven together.
- In a meta-analysis of trials with struggling readers, systematic phonics was the only approach with a significant effect.
- Writing systems differ: English spelling is harder to learn to decode than more consistent spellings.
- Researchers still debate how large the advantage of phonics is and what else should be taught alongside it.
If your child is learning to read in the United States, the UK or Australia, you have probably heard the phrase science of reading at a school meeting, in the news or in a podcast. It is often presented as a dramatic correction, after years in which many classrooms taught children to guess words from pictures and context. That story is partly true. It is also simplified, and the phrase is now attached to products that have little to do with the research.
This guide focuses on the research itself: the main models, what the evidence on phonics shows, where researchers disagree, and practical points for parents, especially parents of children who are struggling. For the learning difficulty itself, see what dyslexia is.
What is the science of reading?
The science of reading is not a curriculum, a brand or a teaching method. It is a shorthand for decades of research on reading: how the brain links print to speech, how children move from sounding out words to reading fluently, what predicts difficulty, and which kinds of teaching help. The work spans cognitive psychology, linguistics, education and neuroscience, and it has been carried out in many languages.
A widely cited review by Castles, Rastle and Nation summarised the field under the title Ending the Reading Wars. Its core message is that learning to read is not natural in the way learning to talk is. Spoken language develops in almost every child without formal teaching; reading is a cultural invention that has to be taught. For alphabetic writing systems like English, that teaching should include explicit instruction in how letters represent sounds, and then move on to building fluency, vocabulary and comprehension.
Science of reading model 1: the simple view of reading
One of the most useful models in the science of reading is also one of the oldest. In 1986, Philip Gough and William Tunmer proposed the simple view of reading: reading comprehension is the product of decoding and language comprehension. Written as a formula, R = D × C. Decoding means turning print into words; language comprehension means understanding those words when they are spoken.
The multiplication is the important part. If either factor is zero, reading comprehension is zero. A child who decodes perfectly but does not understand the language will not understand the text; a child with a wide vocabulary who cannot decode the words will not either. The model is a simplification, but it gives parents and teachers a clear first question: where is the difficulty?
| Profile | Decoding | Language comprehension | What it often looks like |
|---|---|---|---|
| Good reader | Strong | Strong | Reads accurately and understands well |
| Decoding difficulty (typical of dyslexia) | Weak | Strong | Understands stories read aloud, but struggles to read the words |
| Comprehension difficulty | Strong | Weak | Reads words fluently, but struggles to explain what the text meant |
| Mixed difficulty | Weak | Weak | Struggles with both; often needs support with language and decoding |
Science of reading model 2: Scarborough's reading rope
Hollis Scarborough's reading rope, published in 2001, adds detail to the simple view. It shows skilled reading as a rope made of many strands. The lower strands, word recognition, become increasingly automatic with practice. The upper strands, language comprehension, become increasingly strategic. In a skilled reader, the strands are tightly woven together.
| Strand group | Strand | In plain words |
|---|---|---|
| Word recognition | Phonological awareness | Hearing the sounds inside words |
| Word recognition | Decoding | Linking letters to sounds and blending them |
| Word recognition | Sight recognition | Recognising familiar words instantly |
| Language comprehension | Background knowledge | What the reader already knows about the world |
| Language comprehension | Vocabulary | How many words the reader understands |
| Language comprehension | Language structures | Grammar and sentence structure |
| Language comprehension | Verbal reasoning | Inference, understanding metaphor |
| Language comprehension | Literacy knowledge | How books and different kinds of text work |
The rope is a reminder that the science of reading is not only phonics. Phonics is how children build the lower strands; the upper strands come from talk, stories, reading aloud and learning about the world. Both are needed. For the first strand, our guide to phonological awareness explains the skill and how to practice it.
Phonics and the science of reading: what the evidence shows
The best-known conclusion of the science of reading is that explicit, systematic phonics helps children learn to decode. Systematic means the letter-sound relationships are taught in a planned order, from simple to complex, rather than when they happen to come up. Explicit means the teacher shows the pattern directly instead of hoping children work it out.
Several lines of evidence support this. The National Reading Panel's meta-analysis found that teaching phonemic awareness improved reading and spelling, and worked best when linked to letters. For children who already struggle, a meta-analysis of randomised controlled trials found that systematic phonics was the only approach with a statistically significant effect on reading; approaches such as auditory training, coloured overlays and medication did not show reliable effects. And theory explains why: each time a child sounds out an unfamiliar word, they lay down the spelling of that word in memory, a self-teaching process that gradually builds instant word recognition.
Language matters. English has one of the least consistent spelling systems among alphabetic languages. A study comparing European languages found that after a year of school, children learning English could decode far fewer words than children learning languages such as Finnish, Italian or German. That makes explicit phonics especially important in English, and it is one reason why English-language debates about reading have been so heated.
Science of reading myths
| Myth | What the research actually says |
|---|---|
| It is just phonics | Phonics is the decoding half. Vocabulary, background knowledge, fluency and comprehension are part of the same research and equally necessary. |
| It is a program you can buy | It is a body of research. Programs can be more or less aligned with it, but none of them is the science of reading. |
| Children learn to read naturally if surrounded by books | Books matter for language and motivation, but reading is not acquired like speech. Most children need explicit teaching of the code. |
| Good readers guess words from pictures and context | Skilled readers recognise words from their letters. Guessing is a sign that decoding is not yet secure. |
| Phonics is only for struggling readers | Systematic phonics helps beginning readers in general; struggling readers need more of it, for longer. |
| Any multisensory branded method is proven | A meta-analysis of Orton-Gillingham-branded interventions did not find a significant effect. The principles of explicit, sequenced teaching are better supported than any brand. |
| Decodable books are the whole answer | They help children practice phonics on real text, but research on the books themselves is limited. |
The myth about guessing is worth a closer look. Many classrooms taught children to work out unknown words by asking whether a guess made sense, sounded right and looked right, often called three-cueing. Cognitive research points the other way: context helps with meaning once a word is read, but word recognition itself depends on processing the letters. A child who relies on pictures and first letters may look fluent in early books and then hit a wall when the pictures disappear.
What the science of reading still debates
It would be easy to present the science of reading as settled. It is not, and an honest picture includes the disagreements.
- How big the phonics advantage is. Some researchers, notably Jeffrey Bowers, have argued that the evidence for systematic phonics over other approaches is weaker than often claimed: effects in meta-analyses are moderate, shrink over time, and many comparisons are against approaches that already included some phonics. Others reply that the overall pattern still clearly favours systematic phonics. Both sides agree children need to learn letter-sound links.
- How much phonics, and for how long. Research says less about the ideal dose, how quickly to move children on, or how to balance phonics time against vocabulary and knowledge building.
- How to teach comprehension. There is more agreement that background knowledge and vocabulary matter than about how best to teach them.
- Branded methods. Many programs that describe themselves as science of reading aligned have little direct evidence of their own. Our page on Orton-Gillingham covers this in detail.
- Other languages. Much of the research is in English. In more consistent writing systems children decode sooner, and fluency becomes the main challenge earlier.
None of these debates overturn the central findings. They are about emphasis, dose and how to apply the research well. Be cautious of anyone who presents a single program, app or technique as the science of reading, or who uses the label to sell methods the research does not support.
What the science of reading means for parents
You do not need to become a reading researcher to use these ideas. A few practical points cover most of it.
- Ask how phonics is taught. A good answer describes a planned sequence of letter-sound patterns, daily practice and books that match what has been taught, such as decodable books.
- Notice guessing. If your child looks at the picture or the first letter and guesses, gently prompt them to look at every letter and sound it out.
- Keep reading aloud. Reading to your child builds vocabulary and background knowledge, the upper strands of the rope, long after they can read on their own.
- Practice little and often. A few minutes of sound games or phonics on most days helps more than a long weekly session.
- Watch for persistent difficulty. If your child finds sounds, letters and decoding hard despite good teaching, ask the school about extra support and consider an assessment.
If your child is already struggling, the science of reading points to more intensive, explicit and systematic teaching of the same skills, delivered in small steps with lots of review. Our guide on how to help a child with dyslexia turns that into a home plan, and dyslexia accommodations covers what to ask for at school.
The science of reading and dyslexia
Much of the science of reading grew out of research on dyslexia, and the two are closely connected. Dyslexia is a difficulty with accurate and fluent word reading and spelling, most often linked to a weakness in the sound structure of words. In the simple view of reading, it is the classic decoding difficulty profile: language comprehension is often fine, but print is hard to turn into words.
The practical implication is reassuring. The same kind of teaching that helps most beginning readers, explicit phonics built on phonological awareness, is also what helps children with dyslexia, with more time, more repetition and smaller steps. Tools that the research does not support, such as special fonts and coloured overlays, are not a substitute. Our dyslexia friendly font page explains what the font research found.
