The way we teach most children to read sets them up to fail

English is a complex language with roots in many others, and the teaching of it should reflect this. AAP

A new batch of Australian five-year-olds has just started school, eager to learn to read and write. Unfortunately for them, English has one of the most difficult spelling systems of any language, thanks to the way it developed.

A patchwork of many languages

Words from Germanic Anglo-Saxon (woman, Wednesday) and Old Norse (thrust, give) were mixed with words from the church’s Latin (annual, bishop), and Norman French (beef, war). Pronunciation changed dramatically in England between 1350 and 1700 (The Great Vowel Shift), and scribes paid by the character added letters to words.

Science, technology and The Enlightenment added words, often based on Latin or Greek (anthropology, phone, school), wars and globalisation added even more, like “verandah” from Hindi, “tomato” from Nahuatl (Aztec) via Spanish, and “yakka” from Yagara (an Australian Indigenous language). Words are also continually being invented and added to contemporary dictionaries.

Words from other languages typically carry their spelling patterns into English. So, for example, the spelling “ch” represents different sounds in words drawn from Germanic (cheap, rich, such), Greek (chemist, anchor, echo) and French (chef, brochure, parachute).

English has 26 characters, but many more sounds. Shutterstock

Our originally Latin alphabet has only 26 letters for the 44 sounds in modern Australian English. To master our spelling system, children must grasp that words are made of sounds represented by letters, that sometimes we use two, three or four letters for a sound (f*eet, bridge, caught), that most sounds have several spellings (Her* f*irst nurse works **early), and that many spellings represent a few sounds (food, look, flood, broo*ch).

How should children be taught this complex code?

In his internationally acclaimed analysis of the effectiveness of teaching methods, Professor John Hattie assigns “effect sizes” ranging from 1.44 (highly effective) to -0.34 (harmful). Effect sizes above 0.4 indicate methods worth serious attention.

There are two main schools of thought about how to teach children to read and write, one focused on meaning (whole language) and one focused on word structure (phonics). Hattie’s meta-analysis gives whole language an effect size of 0.06, and phonics an effect size of 0.54.

But which type of phonics works best? The Clackmannanshire study provides convincing evidence for synthetic phonics. This starts from just a few sounds and letters in short words, and systematically adds and practises more sounds, spellings and syllable types, until children can read well enough to independently tackle the “real books” adults have been reading them.

Clackmannanshire is a disadvantaged area of Scotland, but by the end of primary school the children using this program were three years ahead of the national average on word reading, 21 months ahead on spelling and five months ahead on reading comprehension.

In 2005, Australia’s National Inquiry into Teaching Reading recommended that young children should be provided with systematic, explicit and direct phonics instruction, and that teachers be equipped to provide this. Similar inquiries in the US and UK agreed.

Are children being taught this way?

The short answer is no. The main reason is that few teachers are trained or equipped to teach synthetic phonics. They’re often taught at university by academics whose careers, publication records and reputations are based on whole-language teaching approaches, considered modern, progressive and child-centred. Phonics, conversely, is framed as old-fashioned, reactionary and teacher-centred, so is used less.

Children are typically encouraged to read “real books” containing long words and difficult spellings, and to guess unknown words from first letters and pictures. They try to write words that are too hard for them, and often the resulting spelling mistakes are put up on the wall for everyone to learn. They memorise lists of high-frequency words.

Phonics work in Australian classrooms typically focuses on initial letters and a few basic strategies, not sounds and their spellings in all word positions. There is little systematic instruction in word blending or segmenting (breaking words into parts, such as syllables), or in many of English’s 170 or so major spelling patterns. Australian curriculum requirements for English reinforce this mess-of-methods approach.

Many confused children learn to guess and memorise words rather than sounding them out. This seems to work at first, but by their third year of schooling lack of visual memory (disk full!) means they start to fail. The well-intended Reading Recovery program, about 80% whole language and 20% phonics, often fails to provide the boost these learners need.

Children who can’t read much by age nine are in serious trouble. By then, teachers expect them to have finished learning to read and to start seriously reading to learn. Yet the 2011 Progress in International Reading Literacy Study found that a quarter of Australian Year 4 students fell below international benchmarks in reading, with 7% scoring “very low”.

Using evidence in education

If large numbers of children were contracting a serious, preventable illness and you asked your doctor how to protect your child, you’d be rightly angry if the doctor didn’t understand the current medical research and thus recommended what s/he learnt at university, or had used before and preferred. You might contact the Medical Board to make a complaint or, if you had followed bad health advice, lodge a malpractice suit in the courts.

Evidence-based practice is deeply embedded in the culture of health professionals. Graduates are taught to read and understand the language of rigorous research and to turn to peer-reviewed academic journals and properly controlled experimental designs as the best sources of evidence. This doesn’t happen nearly enough in education.

Children’s opportunities are seriously compromised if they don’t learn to read and spell. They are much more likely to drop out of school early, be unemployed, suffer ill health and get on the wrong side of the law.

The vast majority of children will only learn to read and spell in the right developmental window when teachers are equipped with the best available methods, based on the best available evidence.


Alison Clarke co-authored this article. Alison is a speech pathologist at the Clifton Hill Child and Adolescent Therapy Group in Melbourne and is on Learning Difficulties Australia’s Council. Disclosure Statement: Alison Clarke is a speech pathologist in private practice who also runs the website www.spelfabet.com.au