One of the world’s most widely used reading intervention programs for young children has taken another hit to its credibility.
Reading Recovery — a one-on-one tutoring program for first graders — has long been controversial because it’s based on a theory about how people read that was disproven decades ago by cognitive scientists. A 2019 story by APM Reports helped bring public attention to the fact that reading programs based on this theory teach the strategies struggling readers use to get by. In other words: Children are taught to read the way that poor readers read.
Now, a new, federally funded study has found that, by third and fourth grade, children who received Reading Recovery had lower scores on state reading tests than a comparison group of children who did not receive Reading Recovery.
“It’s not what we expected, and it’s concerning,” said lead author Henry May, director of the Center for Research in Education and Social Policy at the University of Delaware. May delivered the findings at an April gathering of education researchers in San Diego.
At least 2.4 million students in the United States have participated in Reading Recovery or its Spanish-language counterpart since 1984, when the program first came to America from New Zealand. The program is also used in Australia, Canada and England, among other countries.
The new research could prompt schools to reexamine their investment in Reading Recovery, and consider other ways to help struggling first-graders.
May was the principal investigator of an earlier federally funded study of Reading Recovery, one of the largest ever randomized experiments of an instructional intervention in elementary schools. That study, which began in 2011, found evidence of large positive gains in first grade, as has other research. The program’s advocates have pointed to that research as evidence that the instructional approach is effective and based on sound science.
But whether the initial gains last and translate into better performance on state reading tests remained a question. This new study on the long-term impact of Reading Recovery is the largest, most rigorous effort to tackle that question, according to May.
The fact that students who participated in Reading Recovery did worse in later grades than similar students who did not get the program surprised May.
“Was Reading Recovery harmful? I wouldn’t go as far as to say that,” he said. “But what we do know is that the kids that got it for some reason ended up losing their gains and then falling behind.”
In a written response to the study, the Reading Recovery Council of North America, the organization that advocates for the program in the United States, disputed some of the research methodology and maintained that their program is effective. It also said: “Reading Recovery has and will continue to change in response to evidence gathered from a wide range of studies of both students having difficulties with early reading and writing and their teachers.”
At one point, Reading Recovery was in every state. But school districts have been dropping the program – today, it’s in nearly 2,000 schools in 41 states.
In fact, the first district to implement the program in the U.S. recently decided to stop using it.
Leslie Kelly, executive director of teaching and learning at Columbus City Schools in Ohio, said the decision to drop Reading Recovery is part of a larger effort to bring “the science of reading” to the district. She said she and her colleagues realized that their approach to reading instruction, including Reading Recovery, didn’t align well with that science.
Her advice to other districts that are still using Reading Recovery is to take a close look at the program’s effectiveness: “Do your research. Read a lot, and really look at do you have evidence of impact? That’s really the key. Do you have evidence of impact, and how do you know? And if you don’t have evidence of impact, you have to ask yourself why and then what are you going to do about it?”
Critics of Reading Recovery have long contended that children in the program do not receive enough explicit and systematic instruction in how to decode words. In addition, they say, children are taught to use context, pictures and other clues to identify words, a strategy that may work in first-grade books but becomes less effective as text becomes more difficult. They say kids can seem like good readers in first grade but fail to develop the skills they need to be good readers in the long run.
May said this could explain his latest research findings. “If you don’t build up those decoding skills, you’re going to fall behind, even though it looked like you had caught up in first grade.”
He said the results could also be explained by the fact that about 40% of the students who received Reading Recovery got no further intervention after first grade. “Because the kids didn’t get the intervention that they needed in second and third grade, they lost those gains,” May said. “I think that’s a plausible hypothesis.”
But the study also found that the students who were in Reading Recovery were more likely than the comparison group to receive extra help for reading after first grade. Advocates for Reading Recovery have justified the program’s high cost — estimated to be up to $10,271 per student — by saying that the program reduces the need for further reading intervention.
This new research comes as schools and states are looking for ways to help students recover from the disruptions of the pandemic, including disruptions to their reading development. May’s findings are something for policymakers and school leaders to consider as they make decisions about what programs to invest in.
Emily Hanford is a senior correspondent and Christopher Peak is a reporter for APM Reports, the documentary and investigative reporting group at American Public Media. This story was adapted from their earlier reporting. A collection of stories from APM Reports on how kids learn to read can be found here.