Another Look at Reading First

Earlier this month the U.S. Department of Education released the final report from the implementation evaluation of Reading First. While the report didn’t get much media attention, some of its findings are noteworthy.

• First, it finds that Reading First schools are spending more time on reading, and are more likely to use reading strategies aligned with scientifically based reading research, than reading programs in non-Reading First schools. At the same time, it finds that non-Reading First schools are also increasing the amount of time they spend on reading, as well as their us of practices aligned with scientifically based reading research.
• Second, it finds modest but statistically significant evidence that third- and fourth-grade students in Reading First schools had improved their reading performances more rapidly than did students in non-Reading First schools.

Those findings are promising, but need to be interpreted with some caution. This implementation evaluation, commissioned by the Department’s Office of Planning, Evaluation, and Policy Development, seeks to describe how states and districts are responding to Reading First, as well as national trends in student achievement in Reading First and non-Reading First schools. It is different from the impact evaluation, of Reading First commissioned by the Institute of Education Sciences, which seeks to evaluate the effectiveness or results of the Reading First program. The interim report from the impact evaluation, released this past April, got considerable press when it found no statistically significant improvements in the reading comprehension of students in Reading First schools, as compared to their peers in schools that did not receive Reading First funds.

So how are policymakers and educators to interpret these results? For starters, these reports buttress the case for caution in using the results of the impact evaluation to justify cutting funding for the Reading First program. The implementation evaluation cannot show whether Reading First is effective, but it does show that Reading First seems to be “working” in the sense that it’s driving the changes in curriuclum and teaching strategies that policymakers intended. That’s good news, as is the fact that Reading First schools seem to be making gains in reading performance greater than those of non-Reading First schools. These results also square with other analyses, such as the Center on Education Policy’s study of schools’ and districts’ responses to Reading First.

Moreover, this report describes contamination or spill-over effects from Reading First that should shade how we consider the findings of impact evaluations. The evaluators found that not only Reading First schools, but also non-Reading First schools, increased their implementation of strategies aligned with teh principles of Reading First–such as extra assistance for struggling readers, or the use of curricula and instructional materials aligned with scientifically based reading research. This is due in part to the fact that, while states cannot award Reading First grants to all schools in a state, they can use Reading First funds to provide professional development to teachers working in schools that don’t receive Reading First funding. In the real world, such contamination and spill-over effects are often a good sign: They show that a program is working and that funding for the program is leveraging broader impacts beyond simply the schools or programs that receive grants. But when it comes to evaluating a program’s impact, such spill-over effects can undermine evidence of results, because some schools in the control group may be implementing the same approaches as those in the treatment group.

Policymakers and educators should keep both these factors in mind as we wait until the release of the final impact evaluation of Reading First, and should also consider them when evaluating the findings of that study.

Given the fact positive findings from the implementation and some other studies of Reading First, and the fact that the impact study is not complete, we believe it would be foolish for Congress to continue to defund the Reading First program at this time. As a recent report from the Foundation for Child Development documents, far too many American children aren’t learning to read proficiently by the end of third grade. That’s the problem Reading First was intended to solve, and even if that program has fallen short of some expectations, there’s still a real need for concentrated federal attention to and investment in high-quality research-based reading instruction to improve literacy outcomes in the early grades. Moreover, as the fiscal climate at the federal level becomes increasingly challenging, we can’t afford to lose any of the existing funding streams for young and early elementary school aged children. At least until there’s a concrete proposal to improve Reading First or repurpose funds to other programs demonstrated to have positive impacts on young children’s reading skills, Congress should continue funding Reading First at least at the current levels.