This direct replication study compared the use of dichotomized likelihood ratios and interval likelihood ratios, derived using a prior sample of students, for predicting math risk in middle school. Data from the prior year state test and the Measures of Academic Progress were analyzed to evaluate differences in the efficiency and diagnostic…

The purpose of this paper is to showcase new research that seeks to provide guidance on the heterogeneity of treatment effects by utilizing the variance of demographic differences in state assessments. This study is focused on a simple randomized block design where students are nested within schools, and within each school students are randomized…

Given their significance to school violence, this study quantifies the association between bullying victimization and perceptions of safety separately for victimization where the type is not specified versus victimization that is physical in nature. Generalized liner mixed modeling was employed with 5,138 sixth- to eighth-grade students in 24…

This article considers the impact of a high-stakes testing and reporting requirement on students using publicly funded vouchers to attend private schools. We describe how such a policy was implemented during the course of a previously authorized multi-year evaluation of the Milwaukee Parental Choice Program, which provided us with data on voucher…

Schools with very few and relatively low-performing marginalized students may be most likely to trigger social identity threats (including stereotype threats) that contribute to racial disparities. We test this hypothesis by assessing variation in the benefits of a self-affirmation intervention designed to counteract social identity threat in a…

Background: Cluster-randomized experiments that assign intact groups such as schools or school districts to treatment conditions are increasingly common in educational research. Such experiments are inherently multilevel designs whose sensitivity (statistical power and precision of estimates) depends on the variance decomposition across levels.…