Providing Testing Supports to English Learners Taking the ACT
ACT provides supports on the ACT test to US students who are not proficient in English. These supports are designed to improve access and equity for those students whose proficiency in English might prevent them from fully demonstrating the skills and knowledge they have learned in school.
The following supports are available to eligible English learners (ELs):
- Extended time, not to exceed time and a half
- Approved word-to-word bilingual dictionary (no definitions)
- Test directions in the native language
- Testing in a familiar environment/small group
ACT is committed to evaluating the validity and efficacy of these testing supports. Two recent studies examined the performance, demographics, and high school experiences of ELs who took the ACT with supports between September 2017 and July 2019. Additional details about these analyses can be found in the following reports.
English Learners Research
This Data Byte focused on the programs and services that ELs received at school during the pandemic. Many ELs experienced disruptions, ranging from 36% to 44% for the three major EL programs available at school (i.e., EL instruction in an EL classroom, EL instruction in a regular classroom, and bilingual instruction in content classes). Additionally, most ELs experienced a transition to online learning during the pandemic, but EL supports were not prevalent during online learning.
Experiences and Perceptions of English Learners Testing with Language Supports (Schnieders, Moore, Herring, & Monroe, 2023)
This study surveyed 1,256 English learners to learn about their experiences and perceptions when taking tests with language supports. It focused on the types of supports ELs used in school and on the ACT, whether they thought the supports were useful, and why some ELs did not use supports when taking tests. The results indicated that ELs found language supports useful, and using language supports was associated with higher confidence about their ACT performance. However, some ELs took the ACT without supports due to lack of information about the availability of the supports or their eligibility for the supports. Recommendations were provided on how to better support ELs.
This Data Byte provides average ACT scores and percentages of students meeting each ACT College Readiness Benchmark for ACT’s 2021 graduating class by state and self-reported EL status.
English Learners who Take the ACT with Testing Supports: An Examination of Performance, Demographics, and Contextual Factors (Moore, 2021)
This study was ACT’s first look at the performance of ELs taking the ACT with ACT-approved testing supports. Demographic and contextual factors, along with limited English proficiency, played a substantial role in predicting the performance of ELs taking the ACT. ELs tended to score lower than non-ELs, took fewer core academic courses and fewer honors or AP courses, and had lower high school grades than non-ELs. ELs were more likely to be non-White, from a low income family, and/or a first-generation college student. This study highlights the importance of providing ELs with a rigorous education that includes instruction in both English language and core academic content to ensure that they have equitable opportunities and experiences as compared to their English proficient peers.
Score Gains and Validity Evidence for English Learners Testing with Supports on the ACT (Moore, Li, & Lu, 2021)
This study examined score gains for ELs who first tested without supports and retested with supports, as well as investigating technical aspects of test items and test scores of ELs testing with supports. Providing testing supports for ELs was found to have a positive effect on score gains, especially in reading, as well as stronger relationships between ACT scores and high school grades. Psychometric properties of scores for ELs who tested with supports generally supported score validity, with evidence of high classification accuracy, classification consistency, and comparable conditional standard errors of measurement. Some evidence of differential item functioning was found, but approximately half favored ELs and half favored non-ELs, suggesting minimal impact on total scores. Score reliability and standard errors of measurement were both lower for ELs who tested with supports, likely related to differences in score distributions. This study provides evidence that testing supports are providing a benefit to ELs and removing construct-irrelevant variance without conferring an unfair advantage, while pointing to the need for additional research to further examine the impact of the supports on predicting college performance.
This technical brief summarized score reliability for ELs and non-ELs across multiple assessments, subject areas, and grade levels, and found that reliability of ELs’ scores is consistently lower than reliability of non-ELs’ scores. Score reliability for ELs taking the ACT was comparable, and in some cases higher than reliability for ELs taking other assessments. Psychometric characteristics of ACT scores of ELs found high classification consistency and no evidence of differential item functioning.
Testing Supports for English Learners: A Literature Review and Preliminary ACT Research Findings (Moore, Huang, Huh, Li, & Camara, 2018)
This study examined ACT performance of ELs taking the ACT with state-allowed supports prior to the introduction of ACT-approved supports. ELs were found to have lower ACT scores and lower high school grades than non-ELs, as well as lower correlations between ACT scores and high school grades. Some evidence was found that providing testing supports had a positive effect on ELs’ ACT scores.
Gaining a Better Understanding of ELL Students: How Well Do Parents Speak English (Moore & Wilkinson, 2017)
This Data Byte surveyed how well ACT-tested students thought their parents spoke English. Sixty-seven percent of students whose first language was not English indicated that their parents spoke English well or very well.
Gaining a Better Understanding of ELL Students: The Number of Languages Spoken in a Home (Moore & Wilkinson, 2017)
This Data Byte surveyed percentages of respondents who reported speaking one language or two or more languages at home. The percentage of students who reported that they speak more than one language at home (95%) was substantially higher when students indicated that their first language was not English.
This Data Byte investigated how identification of ELs differ depending on how students are asked about English language usage. For example, the language students use most often at home may not be the same as the language that is easiest for them to speak.
Resources for Educators
EL Research Partnership Opportunity
ACT is currently recruiting colleges to submit first-year academic performance data to study relationships between ACT scores and college grades and provide evidence that the testing supports result in scores that can accurately predict ELs’ success in first-year college courses. Find additional information about how to participate in this study.
ACT Leadership Blog
Testing Supports for English Learners Taking the ACT: Improving Access and Performance
By Joann Moore, Senior Research Scientist, August 31, 2021
A recent blog by Joann Moore, senior research scientist at ACT summarized recent research examining performance of ELs taking the ACT with testing supports and highlighted the need for ELs to have access to a rigorous education to prepare them for college and career readiness.