“Every” Student Succeeds? Academic Trends at the Intersection of (Long-Term) English Learner and IEP Status

Abstract

In this study, we describe long-term trends in students’ academic performance, documenting persistent and systematic disparities in academic outcomes across several student subgroups. We leverage administrative data from longitudinal student records to examine academic outcomes for English learners (ELs), students with Individualized Education Plans (IEPs), and dual-identified students (ELs with IEPs) enrolled in one U.S. state within the last decade, spanning the years 2009 to 2019. Adopting a lens of intersectionality and using descriptive statistics in a cohort-sequential design, we examine reclassification rates that shape the long-term EL (LTEL) subgroup. Grouping students based on whether they were (ever) identified as an EL or as having a disability, we also report on academic content (Reading/ELA and Mathematics) proficiency rates across time for these subgroups. Summarizing the records of over half a million students starting school across six years (2009–2014) and four elementary grades (K-3) nested in 24 independent cohorts, our findings show that, on average, dual-identified students’ rates of LTEL identification are almost twice that of ELs without disabilities, and that only a very small proportion of dual-identified students achieve proficiency in academic content. Our results raise questions and concerns regarding the adequacy and alignment of federal, state, district, and school-level policies and resources that govern the education of English learners, students with disabilities, and especially dual-identified students.

Acknowledgements

We are grateful for helpful comments from Daniella Molle, H. Gary Cook, Laurene Christensen, Katherine Edwards, Alicia Kim, Michael Nicholson, and three anonymous reviewers. We also thank the WIDA Research Subcommittee for their guidance and support, and the staff of the State Education Agency that supported the data sharing, making this study possible. The views expressed in this article are the authors’ and do not represent the University of Wisconsin, WCER, or WIDA.

Disclosure statement

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Notes

1 Non-Regulatory Guidance: English Learners and Title III of the Elementary and Secondary Education Act (ESEA), as amended by the Every Student Succeeds Act (ESSA, 2015). Retrieved from:https://www2.ed.gov/policy/elsec/leg/essa/essatitleiiiguidenglishlearners92016.pdf

2 About 12% of students served under IDEA Part B also qualify for English learner (EL) services (U.S. Department of Education, 2021). In 2021, approximately 1.6% of students enrolled in public elementary and secondary schools were dual-identified as a student with a disability and an English learner. (OSEP Fast Facts: Students with Disabilities who are ELs Served under IDEA Part B). Retrieved from: https://sites.ed.gov/idea/osep-fast-facts-students-with-disabilities-english-learners

3 The proportion of ELs in special education can vary widely relative to the non-EL population in different states, up to 9 percentage points higher (e.g., New Mexico) or lower (e.g., New Jersey). (Office of English Language Acquisition, 2021). Retrieved from: https://ncela.ed.gov/files/fast_facts/20201216-Del4.4-ELsDisabilities-508-OELA.pdf

4 Statewide special education rates also vary widely. New York state has the largest proportion of students with disabilities (19.9%), followed by Pennsylvania (19.8%), Maine (19.7%), and Delaware (19%). States serving the lowest proportions of students with disabilities include Texas (10.8%), Hawaii (11.2%), and Idaho (11.7%) (National Center for Education Statistics, 2020). Retrieved from: https://nces.ed.gov/programs/digest/d20/tables/dt20_204.70.asp

5 Furthermore, until recently, many states did not have consistent and systematic procedures in place for determining the eligibility of reclassification for ELs who were unable to take all four domains of the ACCESS for ELLs assessment due to a disability, precluding them from receiving an overall composite proficiency level – a required component driving reclassification decisions for many English learners (Porter, Cook, & Sahakyan, 2019).

6 Meanwhile, by requiring that numbers of LTELs be reported, federal law incentivizes states, districts, and schools to monitor and prevent accumulation of ELs who are in language support programs for extended periods of time.

7 Supplementary analysis (available upon request) confirms that most students are identified with an IEP during their first three years in school. The rates of first-time IEP identification were not substantially different across cohorts.

8 Due to confidentiality, we are unable to provide a more detailed description of the state’s demographic profile, state-specific rules and regulations around the education of EL and IEP students (such as identification and state-specific reclassification processes and criteria), or any other potentially identifiable information. Data sharing implemented according to DUA191485.

9 ACCESS for ELLs is the collective name for WIDA's suite of summative English language proficiency assessments. ACCESS is a high-stakes assessment taken by English learners in kindergarten through grade 12, in WIDA Consortium member states. Is given annually to monitor students' progress in acquiring English language proficiency in academic contexts. ACCESS assesses the four language domains of Listening, Speaking, Reading, and Writing, based on which an overall composite proficiency level is calculated for the student. This score is typically the main criteria determining the student’s eligibility for exit from EL status (reclassification)

10 The sizes of the first through third grade cohorts are larger in 2009 than in the following years because it is the first year for which we have data; 2009 cohorts therefore include students who started school in earlier grades. As we show in the findings, despite the differences in cohort sizes, the rates computed for the outcomes included in the study were remarkably consistent across cohorts.

11 The kindergarten and 2009 starting cohorts have a larger sample size, for both subgroups, due to the design of the study and the nature of school enrollment, for never- and ever-EL students. While the heavier weight of the K and the 2009 starting cohorts does not affect our main findings, the weighted average of the sample, not presented due to space limitations, would skew towards the average of that of the kindergarten and 2009 starting cohorts.