Please join us for this semester’s first EPE Research Brown Bag talk by Joe Waddington–Assistant Professor in the Department of Educational Policy Studies & Evaluation. The talk will be held on Wednesday February 24th from noon – 1pm in TEB 122. Waddington’s talk summarizes work on school choice in Indianapolis that he is conducting with Mark Berends at Notre Dame. An abstract of the talk is below:
School or “School Type” Effects? Examining the Heterogeneity in Student Achievement and Engagement Outcomes between Schools of Choice in Indianapolis
Abstract
Researchers studying the impacts of school choice policies on K-12 student outcomes have largely focused on the effectiveness of a network (e.g., KIPP Schools), type (charter, magnet, or private schools), or policy (e.g., schools participating in a voucher program) vis-à-vis traditional public schools. Until recently, researchers have paid less attention to reasons why average effects are mixed despite evidence that choice schools of all types are heterogeneous (e.g., Gleason et al., 2010; Clark et al., in press). Building on our previous research, we unpack the school-level differences in longitudinal impacts on achievement and engagement outcomes for students transferring to charter, magnet, Catholic and other private schools by addressing these questions:
- When students switch to a choice school, how do the impacts on student achievement and engagement differ based on the demographic (e.g., racial/ethnic or socioeconomic) composition of schools?
- How do these impacts differ based on the academic (e.g., mean achievement; proportion of English Language Learners or special education students) composition of schools?
We analyze seven years (2009-2015) of longitudinal, student-level demographic, achievement, attendance, and suspension records from the Indiana Department of Education (IDOE) for all students attending traditional public and choice schools in grades 3-8 in Indianapolis. We supplement these data with school-level information from federal and state sources and aggregated student-level data. We use student fixed effects models (achievement outcomes) and conditional logistic regression models (engagement outcomes) to identify our main school type effects for students switching from traditional public to choice schools. We then include a series of interaction terms between the school type indicators and school-level demographic and academic composition measures to parse out differences in effects between and within school types.
During the seminar, we will present both a descriptive analysis and the preliminary results of our work after estimating these models. We will assess the differences across schools within and between school types, contextualize any heterogeneous effects, and discuss limitations of using school-level administrative data. Our study offers new evidence for the growing research base that examines how school choice effectiveness differs across urban schools.