Answers are grounded in this paper and cite their sources.
In this three-state study, the authors estimate the magnitudes of achievement gaps between English learner (EL) students and their non-EL peers, while avoiding typical caveats in cross-sectional studies. The authors further compare the observed achievement gaps across three distinct dimensions (content areas, grades, and states) and report patterns of EL–non-EL achievement gaps within and across states. The study findings suggest that linguistic barriers and long-term EL designation may contribute to the observed achievement gaps. The findings further suggest that the differences in the stringency of state reclassification criteria may influence the reported size of the EL and non-EL achievement gaps between states. Notes 1Results from all analyses for each state, content area and grade level combination are presented in a report available in public (CitationKim & Herman, 2008, available at http://www.cse.ucla.edu/products/reports/R738.pdf). For sets of HM analyses estimating achievement gaps, see chapter 3, Appendix CH3 of the cited report (Tables A1–A7). For sets of HM analyses estimating the relationships between ELP and annual state assessments, see chapter 3, Appendix CH3 (Tables B1–B7). 2In reading in the upper grade (eighth grade), unlike all other combinations, the achievement gap between current ELs and non-ELs of State A is rather similar to that of State B (instead of the gap of State A being greater than that of State B). Further examination of all states helped us find that this is due to an anomalous designated level of meeting proficiency in upper-grade reading in State B. This article compares how well ELs just meeting reclassification criteria would perform to the state-designated level of meeting proficiency in the annual state assessment for all students. While in the other subject, grade, and state combinations, non-ELs tended to perform similarly to the level of meeting proficiency. However, in upper grade reading in State B, non-ELs tended to perform better by far than the designated level of meeting proficiency, which results in the greater gap in outcome than expected. Given that the level of meeting proficiency in state annual assessment is beyond the scope of this article, we just consider this as an exceptional result. 3The exceptions would be a few students whose scores are differentially missing across content areas (e.g., students who took ELA test but did not take math test, or vice versa). 4Because data were available with regard to the relationships between ELP and content-area assessments, we conducted additional analysis on one state concerning when EL students are identified. The base cohort, a group of students who were designated as EL students in the 2002–03 school years, comprised the majority of the EL population. The research results show that later cohorts who were designated as EL students in the 2003–06 academic school years performed better in the content-area assessments given the same level of ELP assessment than the base cohort who were designated as EL students in the 2002–2003 academic year, controlling for FRL status. We used only the results from the later cohorts to be consistent with the year of the annual state assessment scores, which are the outcomes of the analysis.
No takes yet. Share an insight, caveat, or question.
Kim et al. (2009) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: