What’s a grapheme, and why does it matter?

A grapheme is the written symbol (one or more letters) that represents a speech sound (a phoneme) in a word. In English, graphemes can be one letter (m, a), two letters (sh, oa), or more (tch, igh). For more information, see our interactive guide to English Graphemes. Learning to read isn’t about “guessing” words from pictures or context. It's about mapping these graphemes to the sounds in spoken language so print reliably connects to speech and meaning. This letter-sound mapping is the engine that gets reading started and keeps it improving. That isn’t philosophy; it’s what decades of research show (see Castles, A., Rastle, K., & Nation, K, 2018).

The big idea: reading depends on mapping spellings to sounds

Professor of psychology and distinguished researcher Linnea Ehri has shown that children become skilled readers by forming bonded connections between a word’s spelling (graphemes), its pronunciation (phonemes), and its meaning - a process called orthographic mapping. When children know common grapheme–phoneme correspondences and can pull apart and blend the sounds in words, they can map words to memory quickly and read them automatically later (see Ehri, 2014; Ehri, 2017).

The “self-teaching” payoff

Once children can translate unfamiliar letter strings into sounds (decode), every successful attempt becomes a mini-lesson that teaches the specific spelling of that new word. David Share called this the self-teaching hypothesis: decoding is the primary route by which readers build their internal store of word spellings. In other words, systematic knowledge of graphemes doesn’t just help with today’s lesson—it multiplies learning across every text a child encounters (see Share, 1999; Share, 1995).

What the strongest scientific evidence tells us about how to teach reading effectively

Multiple large reviews and national inquiries converge on the same message: explicit, systematic teaching of grapheme–phoneme correspondences improves reading and spelling outcomes, especially in the early years and for students at risk.

• U.S. National Reading Panel (2000): Systematic phonics instruction produces significant benefits for K–1 students and for older struggling readers, supporting decoding, spelling, and comprehension (see nichd.nih.gov; Shanahan, 2005)

• UK Rose Review (2006): Recommended systematic synthetic phonics as the prime approach to teaching decoding/encoding, replacing multi-cueing models (see Five from Five).

• Australia’s National Inquiry into the Teaching of Literacy (Rowe Report, 2005): Advised explicit, systematic phonics within a comprehensive literacy program; Australian guidance continues to reflect this evidence base. (see The Reading League)

• “Ending the Reading Wars” (Castles, Rastle & Nation, 2018): A landmark review of the science of reading instruction, explaining why phonics is central in alphabetic systems like English and integrates evidence across reading development from novice to expert (see Reading Rockets).

Why grapheme teaching is essential in English (not just “nice to have”)

English has a deep, morpho-phonemic orthography. That means the language is based on a complex relationship between letters and sounds.

In addition to single letters representing individual speech sounds, such as:

s like in snake:

…many important graphemes contain multiple letters. For example:

ch like in cheese:

and igh like in light:

…one sound can also be spelled several ways. For example:

ai like in rain:

and ay like in play:

…one grapheme can represent different sounds. For example:

ea like in sea:

and ea like in head:

…adding to the complexity, English words are made up of units of meaning (morphemes) which can affect the way words are spelled and pronounced! Compare knife and knives, as an example.

Because of this complexity, children learning to read and spell in English benefit from being taught graphemes, and that they can consist of one or many letters. This is exactly the knowledge base Ehri identified as enabling orthographic mapping, and the reason systematic, cumulative coverage of grapheme patterns outperforms incidental exposure (Joyner Library, sites.dundee.ac.uk).

Graphemes, phonemic awareness, and comprehension: how they fit

Reading comprehension rests on two pillars: decoding and language comprehension (the “Simple View of Reading”). Strong oral language and knowledge are crucial, but without efficient decoding driven by grapheme knowledge, the system stalls. Decoding frees up cognitive resources so children can focus on meaning, and it kick-starts the positive cycle where more reading leads to more vocabulary and knowledge, avoiding Matthew Effects in which early difficulties snowball (Five from Five, University of Warwick, Reading Rockets).

What high-quality grapheme instruction looks like

Research-aligned practice typically includes:

1. Systematic scope and sequence covering the full set of common graphemes (single letters and digraphs/trigraphs), with explicit teaching and frequent review (nichd.nih.gov, dera.ioe.ac.uk).

2. Cumulative decoding and encoding: children practice reading and spelling with words that use only taught graphemes (decodable texts), then widen as knowledge grows (nichd.nih.gov).

3. Integrated phonemic awareness focused on blending, segmenting, and (as knowledge grows) manipulating phonemes in tandem with graphemes to drive orthographic mapping (Joyner Library).

4. Plenty of successful practice to harness the self-teaching effect—lots of short decoding “wins” across varied words and texts (ScienceDirect).

5. Explicit teaching of advanced code (alternative spellings, conditional rules, morphology like -ed, -s, -tion) so children can eventually tackle the real English writing system, not a simplified subset (sites.dundee.ac.uk).
   

What about “balanced literacy” or cueing?

Approaches that nudge children to rely on pictures, context, or first-letter guesses do not build the stable grapheme knowledge needed for lasting progress and can delay accurate word recognition. Policy and practice reviews in Australia, the US and the UK have shifted away from multi-cueing towards explicit code instruction for this reason (dera.ioe.ac.uk, Five from Five, Learning Difficulties Australia).

For children with reading difficulties

Students with dyslexia or weak phonological skills need clear, systematic teaching of graphemes and phonemes. When instruction targets these mappings and provides guided practice with cumulative decodable reading and spelling, outcomes improve and the widening-gap “Matthew Effects” can be mitigated. (nichd.nih.gov, University of Warwick)

Practical takeaways for classrooms and homes

• Name it and map it: When a new pattern appears (tch, igh), call it a grapheme, link it to its sound, and practice with reading and spelling (Joyner Library).

• Choose texts that match what’s been taught: Early on, use decodables so children can apply their grapheme knowledge successfully and often (nichd.nih.gov).

• Keep revisiting the code: Spiral back to previously taught graphemes in quick, daily review to strengthen memory and automaticity (Joyner Library).

• Build language alongside decoding: Talk, read aloud, and teach vocabulary and knowledge every day (Five from Five).

Key references (open access where possible)

• Castles, A., Rastle, K., & Nation, K. (2018). Ending the Reading Wars: Reading Acquisition From Novice to Expert. Psychological Science in the Public Interest. Explains why phonics (grapheme–phoneme knowledge) is central and how it integrates with broader reading development. (sites.dundee.ac.uk, IDA Southern California Tri-Counties, Reading Rockets)

• Ehri, L. C. (various). Chapters and reviews on orthographic mapping—how grapheme–phoneme connections bind words to memory and support automatic word reading. (Joyner Library, wsra.org, ufli.education.ufl.edu)

• National Reading Panel. (2000). Teaching Children to Read—meta-analytic evidence that systematic phonics benefits decoding, spelling, and comprehension. (nichd.nih.gov, files.eric.ed.gov)

• Rose, J. (2006). Independent Review of the Teaching of Early Reading—UK review recommending systematic synthetic phonics as the prime approach to decoding/encoding. (dera.ioe.ac.uk)

• Rowe, K., for the Australian National Inquiry into the Teaching of Literacy (2005). Recommendations for explicit, systematic phonics within a comprehensive literacy program. (research.acer.edu.au, The Reading League)

• Share, D. L. (1995; 1999/2008). The self-teaching hypothesis showing how decoding unfamiliar words drives orthographic learning. (ScienceDirect)

• Stanovich, K. E. (1986). Matthew Effects in reading—why early decoding success/failure widens achievement gaps over time. (University of Warwick)