Not how teachers teach it. How do teachers themselves work out the code?

There is a substantial body of research into how children learn to read and spell, the role of phonemic awareness and phonics in that process, and how these skills can be taught. The DfE has validated 44 systematic synthetic phonics (SSP) programmes, aligned with its core criteria for what should be taught. But far less attention has been given to whether teachers have the underlying knowledge needed to use them successfully and to map words for reading and spelling throughout the school day.
 

Source: Department for Education, Choosing a phonics teaching programme

https://www.gov.uk/government/publications/choosing-a-phonics-teaching-programme/list-of-phonics-teaching-programmes

What happens, for example, when a child writes:

sed

wont

enee

woz

bickoz

What does the teacher think the child knows?

Which parts of the child's mapping are plausible?

Which phonemes has the child identified?

Which graphemes have they selected to represent them?

What does the response reveal about the child's developing knowledge of the alphabetic code?

And, crucially, what does the teacher do next?

Far less attention has been given to a different question that surely underpins effective phonics instruction and responsive, in-the-moment support throughout the day, particularly for children at risk of reading and spelling difficulties:

Can teachers themselves accurately analyse the code in the words they are expected to help children read and spell?

Because if teachers are expected to interpret children's attempts, identify what they already know, recognise what is missing and decide what support is needed next, their own ability to map the word accurately has to come first.

But there is another part of the question.

When a teacher maps a word, how do they decide on the mapping?

How do they determine:

Which phonemes? How many?

Which graphemes? How many?

Which phoneme value does each grapheme represent?

What happens when their own pronunciation differs from a standardised pronunciation?

What informs their decision when a correspondence hasn't been explicitly taught in the phonics programme they use?

And critically:

Are they identifying the sounds they actually hear, or are the letters they can see and the GPCs they already know influencing what they think they hear?

This isn't primarily a question about how teachers teach phonics.

It is a question about the teacher's own phonological–orthographic analysis and the decisions underlying it.

England: teachers have been tested on the sound structure of words

Stainthorp: no consistent system for segmenting words

Rhona Stainthorp's 2004 study is particularly relevant.

Thirty-eight graduate trainee primary teachers at the University of London completed a phonological-awareness assessment. Their responses were then used as the basis for teaching about the sound structure of words, and the assessment was repeated six months later.
 

Stainthorp reported that the trainees:

 

“did not use a consistent system for segmenting words into component sounds.”

Although performance improved substantially at the second assessment, many trainees still hadn't developed sufficient understanding of word sound structure to teach children about phonemes with certainty.

This is important.

It demonstrates that literate graduate trainee teachers could read and write English successfully themselves while not consistently analysing words into their component sounds.

But Stainthorp's primary question wasn't:

Why did this teacher decide that these were the component sounds?

The study identified instability and inconsistency in teachers' phonological analysis. It didn't set out to investigate the full decision-making process underlying individual responses.

For example, did the teacher's knowledge of the spelling influence the sounds they identified? Were they drawing on GPCs they had previously been taught? Were they analysing their own pronunciation or an assumed pronunciation of the word?

Those questions take us beyond whether the answer was correct.

Source: Stainthorp, R. (2004). W(h)ither Phonological Awareness? Literate trainee teachers' lack of stable knowledge about the sound structure of words. Educational Psychology, 24(6), 753–765.

https://doi.org/10.1080/0144341042000271728

University of Reading research record:

https://centaur.reading.ac.uk/id/eprint/31738
 

England: adults have also been tested on phoneme–grapheme mapping

Connelly: graphophonemic awareness

Vincent Connelly's 2002 UK study moves even closer to the relationship between phonology and orthography.

The research examined graphophonemic awareness, requiring participants to match phonemes with the letters or letter groups representing them in written words.

Participants were 93 UK undergraduate students.

Initial performance was poor, with particular difficulty involving graphemes containing more than one letter. Brief instruction in phonic generalisations improved performance.

This goes beyond testing phonemic awareness alone because it directly concerns the relationship:

PHONEMES ⇄ GRAPHEMES

However, the participants weren't practising classroom teachers making code-mapping decisions while supporting children's reading and spelling.

Nor was the principal research question why a participant selected a particular mapping.

Source: Connelly, V. (2002). Graphophonemic awareness in adults after instruction in phonic generalisations. Learning and Instruction, 12(6), 627–649.

https://doi.org/10.1016/S0959-4752(01)00034-2

ERIC record:

https://eric.ed.gov/?id=EJ662126

Seeing the spelling can influence what we think we hear

This becomes particularly important when investigating teacher code-mapping decisions.

Highly literate adults already have established orthographic representations of familiar words.

Research by Ehri and Wilce examined the influence of orthography on readers' conceptualisation of the phonemic structure of words.

In other words, learning how a word is written can affect how a literate person consciously thinks about its sound structure.

That presents an important problem when assessing teacher phonemic awareness.

If a teacher can see a written word and is asked:

“Which sounds can you hear?”

their answer may not arise solely from analysis of their speech.

Their existing knowledge of the spelling may influence their judgement.

That means a teacher could potentially identify the letters or familiar graphemes first and then attribute the phonemes they expect those graphemes to represent.

This is very different from:

hearing or producing the word → identifying its constituent phonemes → mapping those phonemes to graphemes.

Source: Ehri, L. C., & Wilce, L. S. (1980). The influence of orthography on readers' conceptualization of the phonemic structure of words. Applied Psycholinguistics, 1, 371–385.

https://doi.org/10.1017/S0142716400009802

This isn't only an English finding

Research internationally has also identified gaps between teachers' assumed, reported and demonstrated knowledge of spoken and written language.

Moats: explicit knowledge of spoken and written language

Louisa Moats' work has been particularly influential in showing that being a proficient adult reader doesn't necessarily mean possessing the explicit linguistic knowledge required to analyse language for teaching.

Her research examined teachers' knowledge of areas including phonology, orthography and morphology.

Source: Moats, L. C. (1994). The missing foundation in teacher education: Knowledge of the structure of spoken and written language. Annals of Dyslexia, 44, 81–102.

https://doi.org/10.1007/BF02648156

Moats and Foorman subsequently examined the measurement of teachers' content knowledge of language and reading.

Source: Moats, L. C., & Foorman, B. R. (2003). Measuring teachers' content knowledge of language and reading. Annals of Dyslexia, 53, 23–45.

https://doi.org/10.1007/s11881-003-0003-7

Australia: knowing what a phoneme is isn't the same as finding the phonemes

Fielding-Barnsley investigated the phonemic-awareness and phonics knowledge of 162 Australian pre-service teachers.

The findings revealed an important distinction between knowing the definition of a phoneme and being able to apply that knowledge to actual words.

A teacher can know the answer to:

What is a phoneme?

without necessarily being able to answer:

Which phonemes are in this word?

And if the word is being mapped from speech, another question immediately follows:

Which phonemes are in this word as this person actually pronounces it?

Source: Fielding-Barnsley, R. (2010). Australian pre-service teachers' knowledge of phonemic awareness and phonics in the process of learning to read. Australian Journal of Learning Difficulties, 15(1), 99–110.

https://doi.org/10.1080/19404150903524606

Australia: perceived knowledge versus demonstrated knowledge

Stark, Snow, Eadie and Goldfeld investigated the language knowledge and self-rated ability of practising Australian teachers.

Teachers' knowledge of fundamental linguistic constructs was variable.

Importantly, teachers' perceptions of their own ability didn't reliably predict their demonstrated knowledge.

That has an obvious implication for teacher code mapping.

Competence cannot necessarily be assumed from a teacher's confidence, experience or familiarity with phonics instruction.

Source: Stark, H. L., Snow, P. C., Eadie, P. A., & Goldfeld, S. R. (2016). Language and reading instruction in early years' classrooms: the knowledge and self-rated ability of Australian teachers. Annals of Dyslexia, 66, 28–54.

https://doi.org/10.1007/s11881-015-0112-0

Australia: assessing what you may know less about

Carson and Bayetto examined Australian teachers' phonological-awareness assessment practices alongside their self-reported and actual knowledge.

Their title captures the problem:

The Challenge of Assessing What You May Know Less About.

This raises an important question.

Teachers are expected to listen to children's responses, assess their phonological and phonics knowledge and decide what they need next.

But:

What if the teacher's own analysis of the word is wrong?

Return to the child's spelling:

bickoz

Before deciding what this tells us about the child's spelling knowledge, the teacher needs to analyse the child's spoken word and the written representation the child has produced.

The same applies to:

sed

wont

enee

woz

The child's attempt isn't merely something to mark right or wrong.

It is evidence of how that child is currently connecting speech and print.

Interpreting that evidence depends, at least partly, on the teacher's own ability to analyse those relationships.

Source: Carson, K., & Bayetto, A. (2018). Teachers' Phonological Awareness Assessment Practices, Self-Reported Knowledge and Actual Knowledge: The Challenge of Assessing What You May Know Less About. Australian Journal of Teacher Education, 43(6).

https://doi.org/10.14221/ajte.2018v43n6.5

US research: analysing English word structure

Spear-Swerling and Brucker investigated teacher-education students' knowledge of English word structure.

Their assessment included graphophonemic segmentation and other tasks requiring participants to apply their knowledge to actual words.

Again, this moves beyond asking whether teachers know terminology.

It asks whether they can perform the analysis.

Source: Spear-Swerling, L., & Brucker, P. O. (2003). Teachers' acquisition of knowledge about English word structure. Annals of Dyslexia, 53, 72–103.

https://doi.org/10.1007/s11881-003-0005-5

Cheesman, McGuire, Shankweiler and Coyne also assessed first-year teachers' knowledge of phonemic awareness, including their ability to analyse words into phonemes.

Source: Cheesman, E. A., McGuire, J. M., Shankweiler, D., & Coyne, M. (2009). First-Year Teacher Knowledge of Phonemic Awareness and Its Instruction. Teacher Education and Special Education, 32(3), 270–289.

https://doi.org/10.1177/0888406409339685

What has previous research established?

It would therefore be wrong to claim that nobody has investigated teachers' ability to analyse words.

They have.

Previous research has examined whether teachers, trainee teachers and other literate adults can identify and count phonemes, segment words into component sounds, relate phonemes to graphemes, identify letters or letter groups corresponding to phonemes, demonstrate graphophonemic awareness and analyse English word structure.

Some of the findings are striking.

In England, Stainthorp found that graduate trainee teachers didn't use a consistent system for segmenting words into component sounds.

Connelly found difficulties when UK adults were required to relate phonemes to graphemes.

International research has identified gaps between teachers' perceived knowledge and their demonstrated knowledge.

And research into orthographic influence provides good reason to distinguish between identifying the phonemes actually present in speech and reporting phonemes influenced by knowledge of the written word.

So there is already evidence that teacher knowledge cannot simply be assumed.

But what informed the answer?

This is where a different research question emerges.

A teacher produces a particular mapping.

It is judged correct or incorrect.

But:

Why did they map the word that way?

An incorrect answer tells us that something differed from the expected analysis.

It doesn't necessarily tell us what caused that difference.

Did the teacher:

listen to their speech and identify the phonemes they actually produced?

start with the written letters?

recognise a familiar grapheme and retrieve its taught phoneme value?

use the GPCs emphasised by their phonics programme?

assume particular letters had to form one grapheme?

allow the spelling to influence the sounds they thought they heard?

analyse their own pronunciation or an assumed standardised pronunciation?

recognise that accent could change the phonological analysis?

know what to do when the word contained a correspondence they hadn't previously been taught?

These possibilities represent different underlying knowledge and different decision-making processes.

Even a correct answer doesn't tell us how they got there

Two teachers could produce exactly the same final mapping through different processes.

One might work:

SPEECH → PHONEMES → GRAPHEMES

They say or hear the word, identify its constituent phonemes and then determine which graphemes represent those phonemes.

Another might effectively work:

SPELLING → FAMILIAR GRAPHEMES → EXPECTED PHONEMES

They recognise familiar written patterns and use their existing GPC knowledge to infer the phonemes they expect to be present.

They may arrive at the same answer.

But they haven't necessarily demonstrated the same thing.

This becomes especially important when a teacher needs to listen to a child's pronunciation, rather than analyse a familiar written word.

What happens outside the phonics lesson?

A phonics programme can provide a planned sequence of GPCs, words and lessons.

But children don't encounter words only during phonics lessons.

They encounter words they want to read and spell throughout the school day.

They write words that haven't been taught.

They attempt spellings using the code knowledge they already have.

They encounter grapheme–phoneme correspondences outside the programme sequence.

They pronounce words differently.

They ask:

“How do I spell this?”

They get stuck reading a word in a book, a maths problem, a science activity or something they have written themselves.

At that point, the teacher can't rely solely on following the next scripted phonics lesson.

They need to analyse the word in front of them and the speech of the child in front of them.

That requires teacher knowledge.

The DfE already expects the underlying analysis

The Department for Education's Reading Framework describes the alphabetic code as the relationship between the sounds of speech and the letters or groups of letters representing them.

It states:

 

“Understanding that the letters on the page represent the sounds in spoken words underpins successful word reading.”

The framework also makes clear why teachers themselves need to understand the alphabetic code.

Source: Department for Education, The Reading Framework: teaching the foundations of literacy

https://www.gov.uk/government/publications/the-reading-framework-teaching-the-foundations-of-literacy

The DfE's Writing Framework describes spelling in the opposite direction. Children identify the constituent phonemes in spoken words and then write the graphemes representing them.

Source: Department for Education, The Writing Framework

https://www.gov.uk/government/publications/the-writing-framework

So the underlying process is bidirectional:

SPEECH ⇄ PRINT

Teachers are expected to support it.

The question is whether we know that they themselves can do it accurately and consistently.

What hasn't been established?

Existing research provides important evidence about teacher knowledge and word-analysis performance.

Stainthorp demonstrated inconsistency in trainee teachers' segmentation of words into component sounds.

Connelly directly investigated adults' ability to relate phonemes and graphemes.

Other research has demonstrated gaps in teacher knowledge and shown that orthography can influence conscious phonological judgements.

But a more specific question remains:

What informs teachers' code-mapping decisions?

The issue isn't only whether the final mapping agrees with a predetermined answer.

It is how the teacher arrived at it.

That means investigating the potential influence of:

their own pronunciation

the spelling they can see

previously taught GPCs

phonics programme knowledge

assumptions about grapheme boundaries

standardised pronunciation

and their understanding of the English alphabetic code itself.

It also means asking teachers to explain their decisions.

Why did they identify those phonemes?

Why did they group those letters as a grapheme?

Why did they assign that phoneme value?

Would they map the word differently if somebody pronounced it differently?

What informed the answer?

From “What answer did they give?” to “How did they decide?”

Previous research gives us good reason to assess teacher code-mapping knowledge.

But identifying the final answer is only part of the picture.

When teachers identify the phonemes and graphemes in a word:

What are they attending to?

What knowledge are they drawing on?

Does seeing the spelling influence what they believe they hear?

What happens when their pronunciation differs from a standardised pronunciation?

What happens when the word falls outside the correspondences explicitly taught by their programme?

And when two teachers map the same word differently:

Why?

Understanding that process could help distinguish between difficulties involving phonemic awareness, orthographic knowledge, grapheme identification, knowledge of the alphabetic code, pronunciation, orthographic interference and the framework being used to analyse the word.

That takes us beyond asking:

Can teachers map words?

It asks:

How do teachers decide how to map them?

Earlier research has already shown that teachers don't necessarily analyse the sound and spelling structure of English words consistently.

The next question is why.

References

Stainthorp, R. (2004). W(h)ither Phonological Awareness? Literate trainee teachers' lack of stable knowledge about the sound structure of words. Educational Psychology, 24(6), 753–765.

https://doi.org/10.1080/0144341042000271728

Connelly, V. (2002). Graphophonemic awareness in adults after instruction in phonic generalisations. Learning and Instruction, 12(6), 627–649.

https://doi.org/10.1016/S0959-4752(01)00034-2

Ehri, L. C., & Wilce, L. S. (1980). The influence of orthography on readers' conceptualization of the phonemic structure of words. Applied Psycholinguistics, 1, 371–385.

https://doi.org/10.1017/S0142716400009802

Moats, L. C. (1994). The missing foundation in teacher education: Knowledge of the structure of spoken and written language. Annals of Dyslexia, 44, 81–102.

https://doi.org/10.1007/BF02648156

Moats, L. C., & Foorman, B. R. (2003). Measuring teachers' content knowledge of language and reading. Annals of Dyslexia, 53, 23–45.

https://doi.org/10.1007/s11881-003-0003-7

Fielding-Barnsley, R. (2010). Australian pre-service teachers' knowledge of phonemic awareness and phonics in the process of learning to read. Australian Journal of Learning Difficulties, 15(1), 99–110.

https://doi.org/10.1080/19404150903524606

Stark, H. L., Snow, P. C., Eadie, P. A., & Goldfeld, S. R. (2016). Language and reading instruction in early years' classrooms: the knowledge and self-rated ability of Australian teachers. Annals of Dyslexia, 66, 28–54.

https://doi.org/10.1007/s11881-015-0112-0

Carson, K., & Bayetto, A. (2018). Teachers' Phonological Awareness Assessment Practices, Self-Reported Knowledge and Actual Knowledge: The Challenge of Assessing What You May Know Less About. Australian Journal of Teacher Education, 43(6).

https://doi.org/10.14221/ajte.2018v43n6.5

Spear-Swerling, L., & Brucker, P. O. (2003). Teachers' acquisition of knowledge about English word structure. Annals of Dyslexia, 53, 72–103.

https://doi.org/10.1007/s11881-003-0005-5

Cheesman, E. A., McGuire, J. M., Shankweiler, D., & Coyne, M. (2009). First-Year Teacher Knowledge of Phonemic Awareness and Its Instruction. Teacher Education and Special Education, 32(3), 270–289.

https://doi.org/10.1177/0888406409339685