Wei, X. (2024). Text-to-speech technology and math performance: A comparative study of students with disabilities, English language learners, and their general education peers . Educational Researcher , 53 (5), 285–295. https://doi.org/10.3102/0013189X241232995

Journal Article
Wei, X. (2024). Text-to-speech technology and math performance: A comparative study of students with disabilities, English language learners, and their general education peers. Educational Researcher, 53(5), 285–295. https://doi.org/10.3102/0013189X241232995

Tags

Disabilities Not Specified; K-12; Math; Middle school; No disability; Oral delivery; Text-to-speech device/software; U.S. context

URL

https://journals.sagepub.com/home/edr

Summary

Accommodation

Text-to-speech (TTS) technology was examined as a universal design for assessment (UDA) tool in large-scale math assessments. Extended time accommodation (ETA) was also examined in combination with TTS. Students had access to TTS as a universal tool, with some students also receiving ETA based on their Individualized Education Program (IEP) plans.

Participants

Nearly 3,000 students with disabilities (SWDs; n = 2,750), of whom 1,650 students were granted extended time accommodation and 1,100 were not, participated in the 2017 NAEP assessment. Additionally, there were 1,400 English language learners (ELLs; both with and without disabilities), as well as 23,920 general education (GE) students without extended time accommodation. All participants were grade 8 students who completed the 2017 National Assessment of Educational Progress (NAEP) math assessment.

Dependent Variable

Researchers utilized 2017 NAEP response data (students' answers to each math item), process data (interactions including TTS usage and timestamps), and survey data (items measuring perseverance and math interest/enjoyment) from the 2017 grade 8 NAEP math assessment. Accuracy and item response time were the primary outcome measures analyzed, while TTS usage patterns were analyzed in relation to item characteristics (word count, item difficulty, item type, item location) and student characteristics (demographics, math proficiency, survey responses).

Findings

TTS usage was associated with improved math item performance among ELLs (with and without disabilities) or SWDs who were granted extended time accommodation. TTS usage was positively associated with longer, more difficult math items, and higher usage was observed with multiple-choice or short-response formats compared to constructed-response items. Students using TTS generally took more time to complete items, suggesting that time constraints in timed digital assessments may limit the potential benefits of TTS for SWDs and ELLs in math problem-solving.