‘And + And’ rebrand won’t hide stale practice
Introduction and summary by Lifting Literacy Aotearoa
In this blog post, Professor Emeritus of Educational Psychology, James Chapman provides his response to details about the Reading Recovery (RR) “refresh” that were posted on the University of Auckland Tui Tuia Learning Circle website and published on their LinkedIn page on 27 April 2023. This was titled “Reading Recovery refresh: unpacking the AND + AND approach”.
Download a pdf of the blog here. Or read on below.
James Chapman is one of New Zealand’s most well respected and cited researchers on the science of reading and understanding the causes and responses to reading difficulties for children.
His main points are:
There is an abundance of well-designed, rigorous research that does not support RR as an effective intervention for struggling readers.
Claims by RR that the best approach is an “AND + AND” approach simply maintains the status quo with an even more unbalanced “balanced literacy” approach.
It would be truly refreshing if RR acknowledged the shortcomings and limitations of the three-cueing approach in particular.
None of the changes in the Reading Recovery “refresh” have come from Reading Recovery themselves (or the three universities that have the contract to train and support Reading Recovery teachers) - but appear to have been forced on them by the Ministry of Education.
The Ministry of Education is being naïve if it thinks that an organisation (and the universities that support it) that has shown such little enthusiasm for keeping up with the science of reading research and adjusting its approach accordingly, will suddenly turn itself around.
There are superior structured literacy alternatives to RR that could be put in place today and don’t require a leap in faith that the service being provided is truly aligned with the science of reading.
The Ministry of Education is acting in an anticompetitive way and continues to show favouritism to incumbent private proprietary business interests. The current contract with Reading Recovery has been in place for over 10 years. The Ministry has not explored contracting with alternative providers or indeed allowing schools to fund the training of their own literacy intervention teachers and provision of tier 2/3 literacy instruction in their schools.
He calls on:
The Ministry of Education to withdraw funding from Auckland, Otago, and Waikato universities for the training of RR teachers and to instead resource programmes (PLD and reading intervention teachers in schools) that demonstrate superior effectiveness with struggling readers.
Reading Recovery (and the universities that support it) to:
acknowledge that the abundance of well-designed, rigorous research has not supported RR as an effective intervention for struggling readers.
acknowledge that three longitudinal studies, in New Zealand, the UK and US, failed to find that the programme was more effective than not receiving RR at all.
acknowledge that it has frequently mischaracterised research findings to support its funding and continuation.
explicitly and openly reject the three-cueing approach.
acknowledge that those children struggling the most with learning to read benefit from instruction that strongly emphasises the development of alphabetic coding skills and word-level identification strategies.
acknowledge that it has been unacceptably slow to change its teaching model, thereby under-serving many struggling readers who enter the programme.
What schools tell us
At Lifting Literacy Aotearoa, schools who have made the shift to a structured literacy approach have repeatedly told us that they stopped using Reading Recovery in their schools, or plan to, because they realised it was inconsistent with and actually undermines a structured literacy approach. That is around 500, or 25% of all schools in New Zealand, and counting (see our Map of Structured Literacy schools).
[You may be interested to read another one of our blog posts, where Marianne Brown, a former whole language reading intervention teacher had her own aha! moment when her son was diagnosed with dyslexia and she found out about the science of reading research.]
But those schools are caught between a rock and a hard place. If they do not have Reading Recovery, they need to somehow fund the training and salary of their own reading intervention specialist/s.
Many of them choose to do so, through making tough budget decisions across their operating grant and furious fundraising efforts by their parents. But this is wholly inequitable, and many smaller, rural or low decile schools simply cannot make that happen and therefore have no or very limited intervention support to offer their most vulnerable and struggling students.
And there is also a need for literacy intervention teachers in intermediate and secondary schools that are not funded either. These quotes, from a recent (July 2023) survey of 300+ schools we ran are illustrative of what schools want to see:
“Really wish that Reading Recovery funding was available for all schools, not just those who have Reading Recovery, who would like to put the money towards allowing one of our MSL staff to be released to deliver tier 3 support. We don’t want Reading Recovery but we want that funding.”
“I would love to be able to tap into the funding offered for Reading Recovery and utilise it for Structured Literacy. We have a literacy specialist that takes small groups out and works on SL approaches to give them a little boost, tier 2 kids. To be able to pay for this through teachers salaries as Reading Recovery would be great; we put in 0.2 FTTE and MOE put in 0.2 FTTE, imagine the children we could support in literacy with a specialist literacy teacher for 0.4 FTTE in a small school like ours.”
“There is a lack of support to provide structured literacy as an intervention for senior school students. There needs to be easier access and a recognition amongst support services of the individual structured literacy programmes in schools and what happens when students still don’t make progress with explicit teaching.”
“Funding intervention teacher is a big challenge.”
“Challenges have been trying to give all of our tier 2/3 students intervention we just have so many and not enough TAs.”
Background on Reading Recovery contracts with the Ministry of Education
Reading Recovery is a proprietary educational service trademarked with the New Zealand Intellectual Property Office to the Marie Clay Literacy Trust. It was originally developed in New Zealand by Marie Clay at the University of Auckland in 1985. It has been, and remains, the only taxpayer funded reading intervention programme in New Zealand for decades.
The Ministry of Education decided in December last year to extend its contracts with the University of Auckland (Auckland UniServices), the University of Waikato and the University of Otago till the end of 2024 with two annual rights of renewal till 2026. These are worth a total of $4.1 million a year.
These contracts cover the costs of running RR nationally (by Auckland UniServices) as well as the costs to train Reading Recovery teachers as provided by the three universities. In addition, the Ministry also funds up to 271 FTTE for Reading Recovery teachers in schools (with schools also contributing from their own funds towards those costs). This funding comes to around $21 million a year. So that is a total taxpayer investment of at least $25 million a year for Reading Recovery.
For more information on the contracts signed with the universities, see documents released under the Official Information Act (OIA) and published on the Ministry of Education’s website (24 July 2023). For information on Reading Recovery funding overall see OIA release from 2018.
Lifting Literacy Aotearoa wishes to draw attention to the fact that one of the current (and longstanding) trustees of the Marie Clay Literacy Trust is Stuart McNaughton, Professor of Education, The University of Auckland and also Chief Education Scientific Advisor (since 2014) at the Ministry of Education. As Lifting Literacy Aotearoa has previously stated (see press release August 2020), we see Professor McNaughton’s involvement with Reading Recovery and the Marie Clay Literacy Trust as a serious conflict of interest with his Ministry of Education role. We do not believe this conflict has been managed appropriately by himself, the University of Auckland, the Ministry of Education, or the Prime Minister’s Chief Science Advisor.
Professor Emeritus James Chapman’s response to Reading Recovery “refresh”
Claims by RR that the best approach is an “AND + AND” approach actually maintains the balanced literacy status quo. It would be truly refreshing if RR acknowledged the shortcomings and limitations of the programme, and the three-cueing approach in particular.
In this blog I critique the so-called ‘refresh’ and explain why I am certain this new packaging covers the same old stale offering: a poorly considered, and proven ineffective intervention that diverts funds from provision of high quality programmes that actually improve literacy achievement for the long-term.
I begin with what the ‘refresh’ purports to include. Positive additions to RR’s teaching approach include reference to “alphabetics (phonics and phonemic awareness), fluency, comprehension, and writing activities”. Seems good!
Hard to fathom why it has taken RR three decades or so to accept what scientific research on reading has shown regarding the importance of phonemic awareness (and other) key skills involved in successful literacy learning.
Well, perhaps not really hard to fathom. RR has consistently shown a disregard for the abundance of research showing that these skills are crucial in learning to read, write and spell.
BUT…claims by RR that the best approach is an “AND + AND” approach simply maintains the status quo with an even more unbalanced “balanced literacy” approach.
It is misguided to mix the three-cueing “balanced literacy” teaching model and other features of whole language with explicit attention to foundational language skills. The two approaches are largely incompatible. Adoption of phonics and phonemic awareness could only be successful if RR rejected the emphasis on three-cueing.
Indeed, it would be truly refreshing if RR acknowledged the shortcomings and limitations of the three-cueing approach.
Three-cueing leads to weak readers
The three-cueing approach goes back to the 1960s and the very early days of the whole language movement. The argument, not based on research, is that three-cues are used to identify unfamiliar words in text: semantic uses (meaning), syntactic cues (grammar), and graphophonic cues (letters and sounds). It is based on the mistaken belief that beginning readers “sample” text and don’t need to focus on specific words, especially when they don’t know a word. Using context, or guessing, is enough, it is thought, to figure out an unknown word. Children are typically directed to use semantic and syntactic cues first, and graphophonic cues “as a last resort” (Marie Clay). Knowledge of phonics and the importance of word level identification strategies is downplayed.
This seriously disadvantages children who do not have well developed phonemic and phonological skills. And it teaches children to read the way weak readers read. This approach seriously disadvantages children who initially struggle with learning to read. Typically the sort of children who are placed in RR.
Instead, developing and struggling readers should be taught to focus on words first, and then see if their word decoding attempt makes sense in the text. In short, that’s what good readers do.
To unpack some of the other issues raised in the article and the reluctance of Reading Recovery to make changes, read on. I have structured this in a similar way in which their article does, by posing and answering a series of questions.
Is Reading Recovery based on research?
Reading Recovery claims the programme “is a proven intervention for literacy learning”. Proven?
What RR has proven is that it does not meet the needs of those children struggling the most with learning to read. RR has proven that it especially underserves the needs of Māori and Pacific children. RR has proven that its instructional approach is not based on contemporary scientific studies in reading. And it has proven that it is unable to reject the ideologically based adherence to the three-cueing doctrine.
Research supporting RR tends to be overwhelmingly “in-house”, mostly done by proponents of the programme. A large amount of rigorous research, published in established high-ranking, peer reviewed international journals, has not found RR to be beneficial for those children struggling the most with learning to read.
Robert Slavin and his colleagues in the US reported on a meta-analysis of intervention studies in 2011 and found that the RR programme was no better than programmes tutored by paraprofessionals (teacher aides) or volunteers. Similarly, in 2000 Batya Elbaum and her colleagues found in their meta-analysis of one-to-one intervention programmes that RR was no better than other programmes. They concluded their paper by stating that “it is particularly disturbing that sweeping endorsements of Reading Recovery still appear in the literature”.
Yes, it is!
More recently, two major studies of RR undertaken by proponents of the programme have not found evidence in support of RR. A very large longitudinal Federally funded study in the US found that children who were initially successful in the programme, performed worse than similar children who didn’t receive RR, 3 to 4 years following the programme. This study was initially touted by RR as proof the programme was working.
In the UK, a 5-year then 10-year follow-up longitudinal study found no advantage for RR children compared to similar struggling readers, also in RR schools, but who did not enter the programme. Yet, RR claimed this study as a success.
In Australia, an extensive examination of RR in New South Wales found that participation in the programme during Year 1 was associated with lower achievement in Year 3 for almost all RR children. The New South Wales state government announced in December 2017 that RR would no longer be funded.
Yes, that government withdrew funding from a programme that doesn’t work.
In New Zealand, my colleagues and I (Chapman, Tunmer & Prochnow, 2001) published results from a longitudinal study of RR in the journal Scientific Studies of Reading. Following exit from the programme, successfully discontinued RR children performed no better on reading assessments than struggling readers who didn’t get RR. These “successful” children were 12 months behind their peers one year later. Reading Recovery New Zealand largely ignored this study. So did the Ministry of Education.
Buckingham (2019) concluded in her review of major Reading Recovery studies in Australia, New Zealand, the UK and the US as follows:
The evidence for a sustained impact for Reading Recovery…ranges from negative to null. Where positive impacts have been found in the immediate post-intervention studies they are arguably not as large as might be expected given the extensive training given to Reading Recovery teachers and the intensity and duration of the program for students. A highly trained teacher, whose only role is to improve early reading and who works with a student on a one- to-one basis for 30 minutes every day for two terms, should be able to accelerate that student’s reading progress at a much higher rate than has been found in studies of Reading Recovery. (p. 8)
In short, Reading Recovery is not a programme that has been “proven” to be effective in meeting the needs of those children struggling the most with learning to read. At best, some children will benefit. These are usually children who have more advanced development in terms of their alphabetic knowledge and word-level decoding skills. But far too many children don’t benefit from this expensive programme because attention to foundational literacy skills is inadequate.
RR should acknowledge that the abundance of well-designed, rigorous research has not supported RR as an effective intervention for struggling readers.
RR should acknowledge that three longitudinal studies, in New Zealand, the UK and US, failed to find that the programme was more effective than not receiving RR at all.
RR should acknowledge that it has frequently mischaracterised research findings to support its funding and continuation.
Is word level information important?
Remember, it was the developer of RR, Marie Clay, who wrote that when coming across an unknown word in text, beginning readers need to use sentence-context cues as the main way of figuring out unknown words, and definitely “before they resort to left to right sounding out of chunks or letter clusters or, in the last resort, single letters” (emphasis added; Clay, 1998). She wrote further that word level information should be used mainly for confirming language predictions (guesses): “if a child has a bias towards letter detail the teacher’s prompts will be directed toward the message and the language structure” (1993, p. 42).
This is teaching children to read the way weak readers read. As Michael Pressley wrote in his 2006 text, Reading Instruction That Works, “the scientific evidence is simply overwhelming that letter-sound cues are more important in recognizing words…than either semantic or syntactic cues” (p. 21).
Indeed, this acknowledgement of the primacy of paying attention to word level information has even been included in the latest contract between the Ministry of Education and Reading Recovery providers. The Ministry of Education says that all Reading Recovery teachers “must teach learners to use letter-sound patterns as the primary technique for decoding unknown words”.
So how is Reading Recovery making sure this requirement is being followed by its tutors and teachers?
RR should acknowledge that the teaching model based on Clay’s views about reading was seriously misguided and has been long rejected by scientific studies of reading.
RR should explicitly and openly reject the three-cueing approach.
RR should acknowledge that it has been unacceptably slow to change its teaching model, thereby underserving many struggling readers who enter the programme.
Are lessons individually tailored?
RR persists in claiming that lessons are tailored to each individual’s needs. I say "phooey"! Each lesson is fairly standardised with little variation.
Data from Ministry of Education (MoE) National Monitoring Reports repeatedly show, year in and year out, remarkably similar percentages of children being unrecovered from RR—that is, “referred on” children. Roughly 15% of children placed in RR are “referred on”. This doesn’t include the children who are withdrawn from the programme because of slow progress. Those withdrawn children are not included in the data.
What is startling about the 15% of unrecovered children is that their scores on entry to the programme are so similar each year; scores on the Burt Word Test, the Writing Vocabulary Test, and Reading Book Level. It would be easy to use these scores to predict who would not benefit from RR, rather than placing struggling children into a programme that will not meet their needs.
More importantly, if RR truly tailored lessons to meet the needs of each child, they would have changed their instructional approach years (decades!) ago on the basis of understanding what these patterns of entry scores mean. Not only that, RR would have read and listened to research showing children struggling with learning to read who enter RR would benefit most from explicit attention to the development of alphabetic coding skills and word-level identification strategies. NOT guessing, looking at pictures, and getting mouths ready to say the unknown words!
The Ministry of Education data show clearly that RR does not tailor lessons anywhere near adequately.
In contrast, teachers using a structured literacy approach (particularly for tiers 2 and 3 - the level at which RR teachers operate) use a formative and data driven approach that responds to what the child before them knows and doesn't know, and they tailor their lessons accordingly.
And this diagnostic approach to teaching is supported by a very strong and well-evidenced explanatory model of how reading fluency and comprehension develops, e.g., The Simple View of Reading (Gough & Tunmer 1986).
RR should acknowledge that it hasn’t addressed the persistent fact that each year the 15% or so children who are unrecovered could have been identified on the basis of assessments done before being placed in the programme.
RR should acknowledge that each year it continues to fail a large number of “referred on” children by not tailoring lessons to meet their specific literacy learning needs.
RR should acknowledge that those children struggling the most with learning to read benefit from instruction that strongly emphasises the development of alphabetic coding skills and word-level identification strategies.
Does Reading Recovery support Māori and Pacific Children?
Disproportionately higher numbers of Māori and Pacific children enter RR. That’s not surprising given the inadequacy of Year 1 literacy instruction for most children in “balanced literacy”/whole language classrooms, especially disadvantaging Māori and Pacific children. Yet, further evidence that RR does not tailor each lesson to meet the needs of individual Māori and Pacific children is shown in the annual National Monitoring data—every single year.
Māori and Pacific children have higher rates of participation than Pākehā/European children in RR. They typically have more lessons and spend extra time in the programme than Pākehā/European children. And, these children are less likely to successfully complete the programme.
This persistent fact continues despite an NZCER Report on RR in 2005 that was supposed to focus on issues facing Māori and Pacific children in RR and make recommendations for changes. Changes? What changes? Nothing has changed. Sadly, this report did not recommend changes to the instructional model of RR for Māori and Pacific children, or other children for that matter.
In the most recent National Monitoring Report, for 2021, once again proportionately lower numbers of Māori and Pacific children successfully complete the RR programme, compared to their Pākehā/European peers. Same, same.
RR should acknowledge that it has not analysed the specific learning needs of Māori and Pacific children to tailor lessons to meet their needs.
RR should acknowledge that it has been seriously remiss in allowing the persistence of unacceptable discontinuation rates of Māori and Pacific children, especially given the additional lessons and time in RR they receive compared to Pākehā/European children.
RR should acknowledge that Māori and Pacific children would especially benefit from instruction that strongly emphasises the development of alphabetic coding skills and word-level identification strategies (as would all children).
Do Gains in Reading Recovery Last?
Not for many children who are discontinued. Two very similar studies conducted at Auckland and Canterbury Universities found that 40-45% of discontinued children were performing below the 25th percentile on standardised tests of reading, 2 to 4 years after exit from the programme.
Let me emphasise this again: 40-45% below the 25th percentile.
Even New Zealand PIRLS (Progress in International Reading and Literacy Survey) results for 2011 showed children who had been in RR were way behind their peers when they were in Year 5. (The PIRLS is an international 5-year cycle of reading assessments that focuses on the reading achievement and literacy experiences of 9-year-old children). In the 2011 survey, parents in New Zealand were asked whether their children had been in RR. Around 600 parents said “yes”. The data showed that by the time these children were in Year 5, they were far behind their peers, by a huge amount. The average score for children who didn’t get or need RR was 568. The average score for all New Zealand children in the survey was 531. The average score for children who had received RR was 493. So, 3 years after being in RR, these children were on average awfully far behind their peers. Sadly, this question hasn’t been repeated in the NZ section of the PIRLS survey since 2011....Mmmm!
Jennifer Buckingham concluded in her 2019 Australian policy paper on Reading Recovery: A Failed Investment: “numerous studies of Reading Recovery have no sound evidence that it has sustained positive effects on children’s reading achievement in the medium or long-term, despite its high cost and widespread use.” Buckingham reported an Australian study that showed a negative effect for programme participants two years following exit from RR.
A negative effect.
Clay claimed in her 1987 New Zealand Journal of Educational Research article that RR would “clear out of the remedial education system all children who do not learn to read” (p. 169). This claim is, and has been, totally without substance. And RR’s long-held boast that the programme would provide “an insurance against low literacy levels” is also demonstrably without substance.
RR should acknowledge that the programme has failed to attain even the most basic promise of providing sustained reading improvements for those children who successfully complete the programme.
But wait! Three New Zealand universities support Reading Recovery and train teachers!
What? How can that be? The University of Auckland, University of Otago, and Waikato University manage RR programme nationally and train RR teachers. These universities are proud of their research records and research aspirations. Yet, they train RR teachers to deliver a programme that is so lacking in supportive research. Worse, they train RR teachers to deliver a programme that overwhelmingly large numbers of studies show does not sustain the meagre gains that are made for some students on completion of the 20 weeks or so in RR.
And they continue to support a programme that clearly does not meet the needs of Māori and Pacific students or students in low decile/high deprivation schools.
Why would three of New Zealand’s important universities do this? Is it because of the money? Would these universities teach programmes in other applied disciplines that run counter to research? Would engineering schools be allowed to teach students in programmes that completely ignore research on earthquakes? Of course not. Would the Otago and Auckland medical schools allow teaching that runs completely counter to medical research findings? No, I’m sure they wouldn’t. So why continue to do such training of RR teachers in an intervention programme that is so lacking in supportive research? And so short on sustainable literacy learning outcomes for so many children who enter the programme. And so obviously out of date in terms of the teaching model.
Come on Auckland, Otago and Waikato! Your reputations are way above the need to deliver training of teachers in an over-hyped, seriously underperforming programme.
The Ministry of Education should withdraw funding from Auckland, Otago, and Waikato universities for the training of RR teachers.
What about the money?
Indeed! What about the money? Currently, over $25+ million of taxpayer funds is spent each year on the RR programme (approximately $4.1 million for running RR nationally and training teachers and $21 million to pay for RR teachers in schools). And for what gain? Let’s do the maths.
Each year, around 75% of children who are placed in RR successfully complete the programme and are discontinued.
But an average of 40-45% of those 75 children lose their gains in 2-4 years. Let’s say 40% of 75 children, which comes to 45 children likely having reasonable gains. We don’t know exactly how many children are withdrawn from RR because of inadequate progress, but a conservative estimate would be 5% of the total placed in RR each year. So 45 minus another 5 comes to 40. That is, on average only about 40 in 100 children going into RR each year have some lasting gains.
We don’t know how many struggling readers are not placed in RR because schools don’t think the programme will help. And they might make the stats look worse! Clay was against this, but it happens in New Zealand, Australia, the UK and the US.
Overall, at best, RR may provide longer term gains for around 40% of struggling readers. That’s a fail rate of around 60%.
And those who derive longer term gains are more often Pākehā/European children in higher decile/lower deprivation schools.
$25+ million each year for a return of around 40% at best, and a fail rate of around 60%.
It’s time for Reading Recovery to apologise
Apologise for not updating the programme when the overwhelming evidence showed that the RR teaching model was way out of date and persisting in teaching children to read the way poor readers read.
Apologise for not keeping to its promise to “clear out of the remedial system all children who do not learn to read”.
Apologise for failing to meet the needs of large numbers of Māori and Pacific children.
Apologise for not actually tailoring lessons to the specific needs of each child.
Apologise for mischaracterising research by claiming research support when results did not support RR.
Apologise for not making real and substantial changes in the recent “refresh”, including discarding the three-cueing approach.
Apologise to the many children, often now adults, and families who depended on RR to remedy reading difficulties, but who were seriously let down.
So, back to RR’s “AND + AND” approach. A bit of new stuff from scientific studies attached to the ongoing whole language with three-cueing approach. Is this nothing much more than a cynical way for RR to maintain funding from the Ministry of Education? And for the Ministry to avoid a fight with RR and its dwindling schools?
Reading Recovery and the Ministry of Education, it’s time to be honest with schools and parents
It’s time to be honest and acknowledge that RR will not assist children who are struggling the most with learning to read.
It’s time to be honest and acknowledge the overwhelming evidence that shows RR is “a failed investment”. RR did not act as an insurance against low literacy levels. New Zealand’s literacy levels have been tanking for decades now.
By the way, it’s not the fault of RR teachers. They are doing what they have been taught to do. They try their darndest.
Ministry of Education, it’s time to STOP funding a programme that continues to fail children during the programme, and many more children after completion of RR because the gains are fragile and just don’t stick.
Ministry of Education, in your concern for the achievement levels of Māori and Pacific students, heed the accurate conclusion of the Education Review Office, which stated in its April 2017 report: "New Zealand data indicates that Māori and Pacific students, and those from low decile schools were less likely to have been successfully discontinued from RR, and more likely to be referred on for further specialist help."
Don’t continue to support a programme that so badly discriminates against significant numbers of our children who struggle with reading.
This discriminatory practice is still the case today, yet you, Ministry of Education, continue to fund Reading Recovery. “AND+AND” tinkering won’t resolve the key issues. For goodness sake! Be bold. Accept the overwhelming scientific research data.
Resource programmes that already demonstrate effectiveness with struggling readers. Stop funding the programme that does not.
References:
Buckingham, J. (2019). Reading Recovery: A failed investment. Sydney, AU: The Centre for Independent Studies.
Chapman, J. W., Tunmer, W. E., & Prochnow, J. E. (2001). Does success in the Reading Recovery program depend on developing proficiency in phonological-processing skills? Scientific Studies of Reading, 5, 141–176.
Elbaum, B., Vaughn, S., Tejero Hughes, M. & Moody, S. (2000). How effective are one-to-one tutoring programs in reading for elementary students at risk for reading failure? A meta-analysis of the intervention research, Journal of educational psychology, 92(4), 605-619.
Jesson, R., & Limbrick, L. (2014). Can gains from early literacy interventions be sustained? The case of Reading Recovery. Journal of Research in Reading, 37, 102-117. DOI: 10.1111/1467-9817.12017.
May, H., Blakeney, A., Shrestha, P., Mazal, M., & Kennedy, N. (2023). Long-Term Impacts of Reading Recovery through 3rd and 4th Grade: A Regression Discontinuity Study, Journal of Research on Educational Effectiveness, DOI: 10.1080/19345747.2023.2209092
New South Wales Centre for Education Statistics and Evaluation. (2015). Reading Recovery: A sector-wide analysis. Sydney: Author. Retrieved from: https://www.cese.nsw.gov.au/publications-filter/reading-recovery-evaluation
Nicholas, K., & Parkhill, F. (2013). Is Reading Recovery sustainable two to four years after discontinuation? International Journal of Primary, Elementary and Early Years Education, 3-13. DOI: 10.1080/03004279.2012.759605.
Slavin, R., Lake, C., Davis, S., & Madden, N. (2011). Effective programs for struggling readers: A best-evidence synthesis, Educational Research Review, 6(1), 1-26
