1. The Impact of School Improvement Grants on Practices and Student Outcomes: Findings from a National Evaluation Using a Regression Discontinuity Design
- Author
-
Society for Research on Educational Effectiveness (SREE), Dragoset, Lisa, Thomas, Jaime, Herrmann, Mariesa, Deke, John, James-Burdumy, Susanne, Graczewski, Cheryl, Boyle, Andrea, Tanenbaum, Courtney, Giffin, Jessica, and Upton, Rachel
- Abstract
The School Improvement Grants (SIG) program received over $3 billion through the American Recovery and Reinvestment Act of 2009. Through grants to states, SIG focused on turning around the nation's persistently lowest-achieving schools using one of four school intervention models, with the aim of substantially improving student achievement. The main research question for this study is: Did receipt of SIG funding to implement a school intervention model have an impact on the number of SIG-promoted practices used by low-performing schools and student outcomes? This question was answered using a novel research design--a regression discontinuity design (RDD)--and a large sample of schools from many states to provide rigorous evidence on the effectiveness of SIG. The study focused on the effect of SIG awards made in 2010, when over $3 billion in awards were made to all 50 states and DC. The authors selected a sample of 22 states and approximately 60 districts that were geographically diverse, and able to support estimating impacts using a regression discontinuity design (RDD). To efficiently support the RDD analysis, states and districts that had the largest number of schools eligible for SIG, and that had a high proportion of SIG-eligible schools actually receiving SIG funds to implement an intervention model were prioritized. To the author's knowledge, this is the first large-scale RDD that rigorously assesses the effectiveness of SIG using a large sample of schools from many states. The RDD tackles a host of methodological issues simultaneously (including multiple sites, multiple assignment variables, fuzziness, clustering of individuals within schools, and standard errors that take into account the bandwidth selection method). Although many of these issues are not new to the field, this study is the first to face them all at once. Because the analysis methods are based on simulation work and consultation with RDD experts, other education researchers will benefit from learning about how they were used to obtain the findings in this study. [SREE documents are structured abstracts of SREE conference symposium, panel, and paper or poster submissions.]
- Published
- 2017