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…

Most educational climate research is conducted among (day school) students who spend the bulk of their young lives outside of school, potentially limiting the amount of climate variance that can be captured. Boarding school students, on the other hand, spend much of their lives at school and thus offer a potentially unique perspective on…

This study used administrative data from Wisconsin Virtual School to identify patterns of student engagement in online courses (defined as the amount of time students were logged in to their course each week and how this varied over time). The study also examined whether the patterns were associated with course outcomes (defined as the percentage…

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.…

Post-recession Federal policy initiatives, such as secondary/postsecondary career pathways and gainful employment higher education accountability standards, prioritize the alignment of education practices with market-driven outcomes. Using longitudinal student record data merged from college and state K-12 data systems with the Unemployment…

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.…