Reading Instruction in Australian Schools – March 2004

Dr Brendan Nelson
Minister for Education, Science and Training

Dear Dr Nelson
As researchers, psychologists, linguists and educators who have studied the processes underlying the development of reading, and who are familiar with the scientific research literature relating to the acquisition of reading, we are writing to you to express our concerns with the way in which reading is typically being taught in Australian schools. We would like particularly to draw to your attention the continuing discrepancy between the model of reading development that forms the basis for most of our current school curricula and teaching methods, and the model of reading development that is emerging as a result of the research into reading that has been undertaken over the past twenty to thirty years.

Reading instruction in Australia is based largely on the whole language approach, which makes the assumption that learning to read is like learning to speak, and requires only exposure to a rich language environment without any specific teaching of the alphabetic system and letter sound relationships. However, the research on reading development has shown clearly that this is not the case, and that the ability to read is a complex learned skill which requires specific teaching.

In the United States there has been widespread public debate about different approaches to the teaching of reading, which has led to a series of government-funded reports designed to examine the scientific evidence relating to how children learn to read, and what strategies are most effective in teaching reading. All of these reports have come up with essentially the same conclusion; that mastery of the alphabetic code is essential to proficient reading, and that methods of instruction that teach this code directly are more effective than those that do not.

In Australia there has been little public debate about different approaches to the teaching of reading, and little change in teaching practices that over the past twenty years have been based predominantly on the beliefs and assumptions associated with whole language. The claim has been made that the dichotomy between different approaches to the teaching of reading is false, and that elements of both major approaches (whole language and phonics) are used to teach children how to read. The debate between the two approaches has therefore been dismissed as divisive and unproductive.

While there have been some initiatives in some states and in some individual schools to modify teaching methods to incorporate a greater emphasis on phonological awareness and phonics instruction, our view is that there is as yet little evidence of a major shift in the fundamental assumptions underlying the teaching of reading in Australia. The view that children learn to read by being exposed to literacy activities from an early age persists, and systematic teaching of the alphabetic principle is therefore believed to be unnecessary, since most children will pick it up through exposure to reading. In cases where children do not learn to read, their failure is blamed on their parents or their background rather than on ineffective teaching methods, and calls are made for greater efforts to introduce reading to children at an earlier age, either through access to preschool programs or by providing their parents with free books to read to their children at home. Neither of these solutions addresses the fundamental problem that poor reading skills are in most cases associated with ineffective teaching practices, and that the children who are most disadvantaged by ineffective teaching are those from less advantaged backgrounds whose parents are unable to make up for the deficiencies of the school by teaching their children how to read.

One inevitable consequence of ineffective teaching is the need for remedial programs to assist children who are failing. In Australia and New Zealand the widespread adoption of whole language approaches to the teaching of reading was followed by the introduction of Reading Recovery, an expensive one-to-one remedial program designed to assist children who are falling behind in reading. In some Australian states, up to 20 per cent of children participate in Reading Recovery, despite the lack of any clear evidence as to the long-term benefits of this program. Effective initial teaching of reading would substantially reduce the need for costly remedial programs for failing students.

We believe that the time has come for a review of the approaches to reading instruction adopted in our schools, and a critical examination of the assumptions underlying these approaches.

Given the emphasis that is now being placed on evidence-based policy, we ask that consideration be given to setting up an independent review to examine the research evidence relating to the teaching of reading, and the extent to which current practices are based on this evidence. In view of the entrenched positions of many people within the education establishment, we believe that such a review should seek advice from a wide range of people including those with knowledge and expertise in the fields of language development, cognitive science and reading research.

Yours sincerely,

Professor Vicki Anderson
Dept. Psychology
Royal Children’s Hospital/
University of Melbourne

Associate Professor Judy Bowey
School of Psychology
University of Queensland

Dr Lesley Bretherton
Deputy Director and Clinical Co-ordinator
Department of Psychology
Royal Children’s Hospital
Parkville, Victoria, Australia, 3052

Ruth Brunsdon
Clinical Neuropsychologist
Rehabilitation Development and Developmental Cognitive Neuropsychology Research Unit
The Children’s Hospital at Westmead

Professor Brian Byrne,
Professor of Psychology,
University of New England.

Dr Anne Castles
Senior Lecturer
Department of Psychology
University of Melbourne

Professor Max Coltheart DSc FASSA FAA FBA
ARC Federation Fellow and Scientific Director
Macquarie Centre for Cognitive Science
Macquarie University, Sydney

Associate Professor Veronika Coltheart
Psychology Department and Macquarie Centre for Cognitive Science
Macquarie University, Sydney

Dr Linda Cupples
Director, Speech Hearing and Language Research Centre
Department of Linguistics
Macquarie University

Dr Marion M. de Lemos
Honorary Fellow
Australian Council for Educational Research

Dr. Ruth Fielding-Barnsley
Lecturer / Learning Support
School of Learning and Professional Studies
Queensland University of Technology

Dr Janet Fletcher
Director, Child Study Centre
School of Psychology
The University of Western Australia

Dr Steve Heath
Child Study Centre
School of Psychology
The University of Western Australia

Dr Kerry Hempenstall
Senior Lecturer
Psychology and Disability Studies
RMIT University

Dr John Hogben
Senior Lecturer
Child Study Centre
School of Psychology
The University of Western Australia

Dr Teresa Iacono
Senior Research Fellow
Centre for Developmental Disability Health Victoria
Monash University

Dr Pamela Joy
Senior Clinical Neuropsychologist, Child Development Unit
Head, Developmental Cognitive Neuropsychology Research Unit
The Children’s Hospital at Westmead

Dr.Suze Leitão
Speech pathologist and lecturer (Human Communication Science)
School of Psychology,
Curtin University of Technology, WA

Dr Genevieve McArthur
NHMRC Howard Florey Centenary Fellow
Macquarie Centre for Cognitive Science
Macquarie University

Professor Philip Newall
Audiology Section
Speech, Hearing and Language Research Centre
Macquarie University

Dr Lyndsey Nickels,
QEII Research Fellow,
Macquarie Centre for Cognitive Science (MACCS),
Macquarie University, Sydney

Dr Kristen Pammer
School of Psychology
The Australian National University, Canberra

Professor Margot Prior, AO FASSA, FAPS
Department of Psychology,
University of Melbourne and Royal Children’s Hospital.

Dr Karen Smith-Lock
Research Fellow
Macquarie Centre for Cognitive Science (MACCS)
Macquarie University

Dr Geoffrey W. Stuart
Senior Fellow
Department of Psychology
University of Melbourne

Professor Kevin Wheldall,
Director, Macquarie University Special Education Centre
Macquarie University


Explanatory Notes to Accompany Letter to Dr Brendan Nelson, March 2004

On research into reading development
The major breakthrough in terms of understanding the processes involved in learning to read is generally attributed to the pioneering work of Isabelle Liberman and her colleagues in establishing the link between the ability to identify individual sounds (or phonemes) in words and the ability to read (see, for example, Liberman, 1973).

A comprehensive review of the research literature on the mental processing that underlies skilled reading and on how reading should be taught, undertaken by a group of leading experts in the field under the aegis of the American Psychological Society, was published in 2001 in the November issue of the journal Psychological Science in the Public Interest (see Rayner, Foorman, Perfetti, Pesetsky and Seidenberg, 2001). A more general article on the same topic by the same authors, under the title How should reading be taught, was published in the following year in the March issue of the Scientific American.

On reading instruction in Australia
The predominance of the whole language approach to the teaching of reading in Australia has been documented in several sources, including the 1992 Report of the House of Representatives Standing Committee on Employment Education and Training (The Literacy Challenge: Strategies for early intervention for literacy and learning for Australian children), and the 1994 review of literacy instruction in Australian primary schools, by van Kraayenoord and Paris (1994).

On the public debate on reading in the United States
Following the publication in 1955 of the book by Rudolf Flesch entitled Why Johnny Can’t Read, there has been continuing debate in the United States with regard to the relative merits of different approaches to the teaching of reading, and particularly the role of phonics in reading instruction. This led to a number of studies and government-funded reports designed to investigate the scientific evidence relating to the effectiveness of different approaches to reading instruction. The most notable of these reports were Jean Chall’s book Learning to Read: the Great Debate (1967), Marilyn Adams book Learning to Read: Thinking and Learning about Print (1990), the report Preventing Reading Difficulties in Young Children, edited by Catherine Snow, Susan Burns and Peg Griffin (1998), and the Report of the National Reading Panel, Teaching Children to Read: An Evidence-Based Assessment of the Scientific Research Literature on Reading and its Implications for Reading Instruction, published in 2000. This last report provided the impetus for the passing of the No Child Left Behind Act in the United States in 2002, which allocated one billion dollars to Reading First, a program designed to improve reading achievement through the adoption of effective teaching practices based on scientific research, as documented in the report of the National Reading Panel. A summary of these reports as well as the main findings of the National Reading Panel is provided in the ACER report Closing the gap between research and practice: Foundations for the acquisition of literacy (de Lemos. 2002).

On the public debate on reading in Australia
There has been relatively little public debate in Australia regarding the effectiveness of different approaches to the teaching of reading (the whole language versus phonics debate, or the ‘reading wars’). Some passing references to the debate have surfaced in the media, but apart from the article by Jane Cadzow published in the Age Good Weekend in October last year, there appears to have been little public concern on this issue. As Jane Cadzow remarked in her article: ‘What’s odd in Australia is the apparent lack of community concern. Overseas, the row over teaching methods has made headlines and prompted petitions, but here hostilities have been largely confined to literacy conferences and the columns of academic journals’. In part, the reason for this lack of interest can probably be attributed to the strategy adopted by supporters of the whole language approach, who tend to side-step the issue by claiming that teaching instruction in Australia incorporates elements of both approaches, and that the debate about different teaching methods is therefore irrelevant. For example, in the Final Report of the NSW Inquiry into Early Intervention for Children with Learning Difficulties, the Committee makes the comment that: ‘It is difficult for this Committee to get to the bottom of the debate between exponents of either the ‘whole word’ or ‘phonics’ approach to literacy pedagogy. Many literacy experts and education departments argue that the dichotomy is false and that elements of both methods are used to teach children to read: it is therefore a divisive and unproductive debate’ (see paragraph 6.28 of the NSW report).

On the causes of poor reading skills
A good overview of the issues raised by Rayner, Foorman, Perfetti, Pesetsky and Seidenberg (2001) in their article on How psychological science informs the teaching of reading is provided in the summary article which appeared in the online Newsletter on the website of the Reading Reform Foundation in the UK. Because of its direct relevance to the issues raised in our letter, this summary article is quoted in full below:

The greatest continuing problem of the public schools is their failure to teach many children how to read.
Most of the academic and behavioural problems had by children in the course of their school careers stem from poor reading. Children who start as poor readers tend to fall further behind their peers every year, not grow out of it.
Poor reading skills stem in large part from faulty teaching practices. In particular, teachers fail to systematically teach new readers how to “sound out” words, i.e., they fail to teach phonics. Without decoding skills, many children stumble, guess, acquire bad reading habits, and get discouraged.
Following World War II, the “whole-word” teaching method was popular. Also called the “look-say” approach, it taught reading by using repetitious materials that emphasized 50-100 words, e.g. “Run, Spot, run” from the famous Dick and Jane series. Phonics was an add-on, not an essential.

In more recent years, a teaching method that minimizes both decoding and repetition became popular. Called “whole-language” (or “literature-based instruction” or “guided reading”), it stressed student interest and enjoyment. It used so-called “embedded phonics” and worked even less well than the “whole-word” approach.
How could schools not notice that their methods weren’t working?
Fortunately, many children come to school with literacy skills acquired at home. With them, any teaching method seems to work. Children who lack such advantages do less well but their failure is easily blamed on their parents and backgrounds. So instead of recognizing the problem, schools argued that their methods worked for many students and for those who failed, better pre-school enrichment was needed.
A larger impediment was at work too: defective teacher training. Virtually every teacher and administrator trained in a school of education has been taught to idealize naturalistic forms of teaching and to frown on their opposite regardless of learning outcomes. Reading instruction that teaches discrete skills in an orderly sequence – i.e. uses phonics – was, therefore, considered substandard despite its superior results.
Whole language, therefore, was very attractive to educators despite its ineffectiveness with children who need the most help in learning to read. It was naturalistic and unstructured, and reading experts in the schools of education assured that it was a “best practice.” That it was ineffective with disadvantaged students was said to be the result of insufficient time and attention to reading, not ineffective teaching.
Whole-word and whole-language reading methods have dominated the schools of education because education professors have historically considered it more important for students to be exposed to preferred forms of teaching than it is for them to gain specific knowledge and skills. In their view, reading instruction using explicit, systematic phonics may be effective but it is “unnatural” and therefore entails the risk of detrimental side effects. That the use of ineffective reading instruction exposes the child to the risk of a far greater handicap than any side effects imagined by phonics opponents is largely ignored.
The Call for Proven Methods
In the mid-eighties, California’s Department of Education mandated whole-language reading instruction statewide. By the mid-nineties, reading scores had fallen to the point that they became a public scandal and a major political issue. In 1995, the California State Assembly relied on outside experts to develop and pass a bill mandating the use of phonics-based reading instruction.
In 1993, Massachusetts enacted legislation that resulted in state curriculum becoming infused with whole-language. Eventually the new curricula came to the attention of linguistics researchers at leading academic institutions in the state, whereupon a protest signed by 40 leading scholars was sent to state educational authorities and the guidelines rewritten.
In 1997, a National Reading Panel (NRP) was authorized by Congress and convened by the U. S. Department of Education to examine the research on reading instruction and make recommendations. The NRP’s report was published in 2000 followed by the American Psychological Society report on which this Briefing is based.
Both reports are authoritative and both conclude that phonics-based reading instruction is indispensable. The interesting and engaging reading activities called for by whole-language reading methods are a useful adjunct but not a substitute for reading instruction that systematically and explicitly teaches decoding skills.
The deficient reading outcomes of public schooling are essentially a product of ineffective teaching stemming from defective teacher training. Reform will require significant retraining at all levels of the schooling establishment beginning with the schools of education.
From: Newsletter of the Reading Reform Foundation, No. 50, Spring Term 2003, at

On Reading Recovery
There have been many claims and counter claims relating to the effectiveness of Reading Recovery. A major problem in interpreting the evidence is the poor design of most of the evaluation studies, the lack of comparable comparison groups (in terms of randomised control and experimental groups), and the variability in the quality of the instruments used to assess reading achievement, which in many cases are based on teacher assessments rather than standardised tests. In general, it can probably be said that those studies which claim to show positive effects are usually based on flawed designs and inadequate measuring instruments, while studies based on more rigorous designs and using standardised testing instruments tend to show little or no long term effect.

Concerns regarding the failure of Reading Recovery to deliver on its claims to improve student achievement prompted a group of leading US reading researchers to produce a three-page critique of Reading Recovery, under the title Evidence-Based Research on Reading Recovery, which was sent to members of the US Congress and circulated by email. A major concern of these researchers is that Reading Recovery fails to meet the needs of the lowest performing students because its developers have failed to integrate the findings of independent scientifically-based reading research into their program. More specifically, the program does not include explicit instruction in phonological awareness and the use of spelling-to-sound patterns in recognising unfamiliar words in text. A further concern is the excessive cost of the program because of its reliance on one-to-one intervention by highly trained teachers, when the research evidence has indicated no advantage of one-to-one over small group instruction, and that the same effects of one-to-one intervention can be achieved through the use of trained volunteers rather than specialist Reading Recovery teachers.

Evidence from Australia, while limited, supports these general conclusions. For example, the ACER evaluation of the Literacy Advance strategy implemented by the Catholic Education Commission of Victoria found that ‘participation in Reading Recovery does not appear to be a significant influence on literacy achievement in Year 3….The benefits of Reading Recovery observed when these students were in Year 1 in 1998 did not endure over time’ (see Ainley, Fleming and McGregor, 2002, page 86). Similarly, an earlier evaluation of Reading Recovery undertaken in NSW found that children with poor metalinguistic skills are less likely to benefit from Reading Recovery, and urged caution in the more widespread adoption of this program in NSW without further evaluation of its effectiveness (see Center, Wheldall, Freeman, Outhred, and McNaught, 1995; Center, Wheldall and Freeman, 1992).

On the need for a review of reading instruction in Australian schools
A number of reports and enquiries into literacy development and achievement have been undertaken in Australia in recent years. However, these reports and enquiries have not addressed the fundamental question of the effectiveness of the strategies used to teach reading, or the validity of the assumptions on which our teaching methods are based. It is for this reason that we believe a new approach to the investigation of reading instruction in our schools is required.

What needs to be addressed is the critical question of whether the teaching of reading in Australia is based on scientific knowledge relating to how children learn to read, and whether the methods used to teach reading in our schools are based on empirical evidence as to the strategies that are most effective in teaching reading. Such a review would not need to duplicate the extensive reviews of the literature that have been undertaken elsewhere, but would use this information as the basis for examining the extent to which current practices in Australia reflect the advances that have occurred in reading research in recent years. This would require an examination of the curricula relating to the teaching of reading in each state and territory, a survey of practices in schools to determine the extent to which practices incorporate the essential elements of effective reading instruction, including systematic instruction in the alphabetic principle and letter sound relationships, and the appropriateness and adequacy of teacher training in terms of providing teachers with up-to-date knowledge of the research literature relating to reading, as well as adequate training in the application of this knowledge to the implementation of effective teaching programs in the classroom.

Marion de Lemos, PhD
Honorary Fellow
Australian Council for Educational Research
March, 2004

Adams, M. J. (1990). Beginning to read: Thinking and learning about print. Cambridge, Mass.: MIT Press.
Ainley, J., Fleming, M., and McGregor,M (2002). Three years on: Literacy advance in the early and middle primary years. Melbourne: Catholic Education Commission of Victoria.
Cadzow, Jane (2003). The war of the words. Good Weekend, October 4.
Center, Y., Wheldall, K., and Freeman, L. (1992). Evaluating the effectiveness of Reading Recovery: A critique. Educational Psychology, 12, 263-274.
Center, Y., Wheldall, K., Freeman, L., Outhred, L. and McNaught, M. (1995). An evaluation of Reading Recovery. Reading Research Quarterly, 30, 240-263.
Chall, J. (1967). Learning to read: The great debate. New York: McGraw-Hill. (Second edition, 1983; Third edition, 1996).
de Lemos, M. (2002). Closing the gap between research and practice: Foundations for the acquisition of literacy. Melbourne: ACER.
Baker, S et al (2002). Evidence-Based Research on Reading Recovery. Letter to members of US Congress signed by 32 reading researchers. (at
Flesch, R. (1955). Why Johnny can’t read. New York: Harper and Row.
House of Representatives Standing Committee on Employment Education and Training. (1992). The literacy challenge: Strategies for early intervention for literacy and learning for Australian children. Canberra: Australian Government Publishing Service.
Liberman, I. Y. (1973). Segmentation of the spoken word and reading acquisition. Bulletin of the Orton Society, 23, 65-77.
National Reading Panel (2000). Teaching children to read: An evidence-based assessment of the scientific research literature on reading and its implications for reading instruction. (Posted 13 April 2000, accessed 15 April 2000, last modified 16 May 2000).
New South Wales Parliament, Legislative Council Standing Committee on Social Issues. (2003). Realising potential: Final report of the Inquiry into Early Intervention for Children with Learning Difficulties. Sydney: Legislative Council Standing Committee on Social Issues.
Rayner, K., Foorman, B.R., Perfetti, C.A., Pesetsky, D. and Seidenberg, M.S. (2001) How psychological science informs the teaching of reading. Psychological Science in the Public Interest, 2 (2), 31-74.
Rayner, K., Foorman, B.R., Perfetti, C.A., Pesetsky, D. and Seidenberg, M.S. (2002). How should reading be taught. Scientific American, March, 286 (3), 84-91.
Reading Reform Foundation (2003). Article on the paper How should reading be taught by Rayner, Foorman, Perfetti, Pesetsky and Seidenberg (2002). Reading Reform Foundation Newsletter No. 50, Spring Term 2003 (at
Snow, C.E., Burns, S., and Griffin, P. (1998). Preventing reading difficulties in young children. Washington, DC: National Academy Press.
van Kraayenoord, C.E. and Paris, S.G. (1994). Literacy instruction in Australian primary schools. The Reading Teacher, 48, 3, 218-228.

Read more