Long-Term English Learners' Performance: In-Depth Examination of an English Language Proficiency Assessment

• In Event: Validity and Utility of English Language Proficiency Assessments: New Evidence in the Common Core Era

Abstract

Objectives

This study examines linguistic and cognitive constructs reflected in English language proficiency (ELP) standards of ELP assessments that are related to English learners’ (EL) becoming long-term English learners (LTELs). Using students’ performance data on an ELP assessment, the ELPA21 summative assessments, this study investigates English language knowledge and skills in elementary school that appear to be particularly challenging for LTELs.

Theoretical Framework

LTELs typically refer to students who remain as ELs for more than five years (Every Student Succeeds Act [ESSA], 2015; Linquanti & Cook, 2015; U.S. Department of Education, 2016). LTELs generally show lower academic performance, language proficiency, and significant gaps in reading and writing skills compared to their peers (Callahan, 2006; Menken et al., 2012; Olsen, 2010; Sahakyan & Ryan, 2018). Many ELs in secondary school belong to this group, and their EL status may limit their opportunities to learn and access to advanced classes and college (Callahan & Shifrer, 2016). While research indicates some external factors contributing to ELs’ becoming LTELs such as inappropriate programs and shows how to better serve LTELs (Menken et al., 2012; Thompson, 2015; Thompson & Rodriguez-Mojica, 2018), little is known about LTELs’ English language development in their pre-LTEL period and how to prevent ELs at risk from becoming LTELs (Shin, 2020; Shin & Hansen, 2021).

Data Sources & Methods

We used student- and item-level ELPA21 summative assessment data. We focused on students initially classified as EL in kindergarten (2015-16) and followed them until the end of Grade 5 (2020-21). To examine what an item measures, we used item metadata, including the ELP standards and item targets and the task type to which an item belongs. Additionally, we qualitatively coded item features for descriptions of functions (cognitive, linguistic) involved in the development of English language and identifiable across the language domains (listening, reading, speaking, writing) and for the range of tasks included on the ELPA21 summative assessment. Using the qualitative codes, we conducted quantitative analyses to identify which types of items best predict ELs’ becoming LTELs.

Findings

Findings include analyses using the Grade 2 ELP assessment. Reading is the most significant predictor of reclassification by Grade 5, indicating it is the most challenging domain for LTELs in Grade 2. Additionally, performance on some task types in Grades 2 and 3 are more strongly associated with reclassification by Grade 5. Since each task type is aligned with one or more ELP standards, findings suggest that task type-level performance could help identify students who are at risk of becoming LTELs and identify the areas they need the most help.

Significance

The findings are useful for identifying those ELs at risk of becoming LTELs so that interventions can be applied in earlier grades to reduce the chances of those students becoming LTELs. The findings can guide practitioners and policy makers to develop EL curriculum and intervention programs that emphasize and strengthen areas or skills of English language, especially for ELs at risk for becoming LTELs.

Authors

• Nami Shin, University of Kansas

• Edynn Sato, Sato Education Consulting, LLC

• Jenny C. Kao, University of California - Los Angeles

• Eunhee Keum, University of California - Los Angeles