Identification of Gifted English Learners: An Empirical Examination of Two States

• In Event: 42.076 - Talent Ignored: Investigating the Underidentification of English Learners for Gifted Programs

Abstract

Objectives/Purpose

The goal of the current study was to examine the extent to which EL students are under-identified as gifted using state data, and explore the role of district level practices in EL identification.

Perspectives/Theoretical Framework

Research that specifically and empirically examines the extent of EL identification does not currently exist. Research on student identification typically focuses on identification rates between racial groups. Further, this research utilizes school level data as a proxy for student level identification, which can have negative implications for the interpretation of findings (Freedman, 1999). Regarding best practices, there is some evidence that certain practices can help enhance EL identification rates such as professional development (Esquierdo & Arreguín-Anderson, 2012) and assessment of students in their native language (Harris et al, 2007). However, research that empirically examines the relation between practices and student-level EL identification could not be located.

Methods

To examine the degree to which EL students were identified as gifted, we conducted a series of three-level (student-school-district) logistic regression models, where students’ identification status at grade 5 was the outcome variable. Student race, eligibility for free and reduced lunch, gifted status and achievement scores were utilized for student level data. These variables were aggregated to the school and district levels and used to represent school and district estimates for each state.

Data Sources

Three waves of data from state databases from 2011 to 2014 were utilized to model student identification for gifted programming. State 1 contained 69,938 students nested within 1,034 schools across 181 districts and state 2 contained 93,671 students nested within 1,381 schools nested across 193 school districts. District data were also utilized to explore the relation between district practices and student identification. Four district practices were of particular interest in the current study: 1) assessing ELs in their native language; 2) modifying the identification process for underserved populations; 3) offering professional development on the use of referrals, nominations or rating scales; and 4) offering professional development on how to recognize giftedness in traditionally underserved populations.

Results

While EL status was negatively related to gifted identification in State 2, EL status was positively related to identification in State 1 after controlling for students’ 3rd grade math and reading achievement, student demographics, school and district SES, school and district achievement, and the percentage of gifted and underrepresented students in the school and district. Additionally, modifying the identification process for underserved populations and offering professional development around the use of referrals, nominations and ratings scales was positively related to EL gifted identification.

Discussion and Significance

The results reveal a surprising finding: the under-identification of EL students may be a function of context and, in the current study, varies by state. Further, there is evidence that district practices are both positively and negatively related to the identification of EL students. This study is the first of its kind to use student level data to examine the phenomenon of EL identification for gifted. Implications for the research and practice will be discussed.

Authors

• Rashea Hamilton, University of Connecticut - Storrs

• D. Betsy Mccoach, University of Connecticut

• Michael Shane Tutwiler, University of Rhode Island

• William Estepar-Garcia, University of Connecticut