Description of Project:

Reading Recovery is a highly effective, research-based, short-term literacy intervention of one-to-one tutoring for the lowest-achieving first-graders. Experimentally driven research has revealed that Reading Recovery improves student reading achievement, on average, by more than 30 percentile points. Such gains are critical if struggling readers are to have an opportunity to achieve at reading levels comparable to those of their more advantaged peers. The intervention is most effective when it is available to all students who need it and supplements good classroom teaching.

Through its established network of university training centers, Reading Recovery will be scaled up to target persistently low-performing schools, rural schools and schools serving sizeable proportions of English language learners. More than 700 teachers in about 300 schools in 20 states will serve more than a half-million children over the 5-year project, including 90,000 students who will receive direct one-to-one Reading Recovery lessons. Trained teacher leaders will implement professional development for two to four teachers per school so that all identified students can be served.

Five key project objectives are to (1) train 15 new teacher leaders in Years 1 or 2; (2) train 735 new Reading Recovery teachers each year for a total of 3,690 teachers; (3) ensure that trained Reading Recovery teachers work with more than 90,000 Reading Recovery students and more than 400,000 students in classrooms or in Title I small-group instruction; (4) conduct a rigorous program evaluation that includes both experimental and qualitative methodologies; and (5) provide high-quality oversight for the project, orchestrating activities across the partnering universities.

Description of Evaluation:

A multisite randomized controlled trial will be implemented to study the impact of Reading Recovery on student achievement. Students will be block randomly assigned to receive the program first or second, with the latter group serving as controls for the former group in randomly sampled schools. Longer-term program impacts will be assessed through a regression discontinuity design based on third- and fourth-grade state test scores of all students in selected schools. The evaluation will address both the first-year and longer-term effects of the intervention, and will include teacher interviews and logs to help evaluators understand teacher professional growth.