Using Accessible Math Textbooks with Students Who Have Learning Disabilities
Preston Lewis, Steve Noble, Neil Soiffer · 2010 · Proceedings of the 12th International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS 2010) · doi:10.1145/1878803.1878829
Summary
This paper reports results from Project SMART (Supported Math Accessibility Reading Tool), a classroom study conducted at the University of Kentucky with 48 eighth grade students who had learning disabilities. The project provided digital versions of Pearson's Connected Mathematics 2 textbooks, converted from PDF to NIMAS-compliant XML with MathML encoding for mathematical expressions. Students used Read & Write Gold text-to-speech software paired with Design Science's MathPlayer plugin to have both text and math content read aloud. The control groups used the same textbooks in print format with traditional oral accommodations from staff readers. The study spanned three phases across fall 2008 and spring 2009 semesters at schools in Clark County and Shelby County, Kentucky, covering four math units: Say it with Symbols (algebra), Filling & Wrapping (area and volume), Samples and Populations (statistics), and Looking for Pythagoras. The research addressed three questions: whether the assistive technology met student access needs, whether accessible digital math improved test performance, and whether it affected student perceptions of their math abilities.
Key findings
Students using digital textbooks with MathML generally outperformed control groups on unit tests, with the most consistent improvements appearing in the algebra-heavy unit Say it with Symbols, which had the highest MathML density (10.95 expressions per page) — showing +14% to +16% improvement over controls across study sites. Survey results were strongly positive: 96% of students preferred the digital version over paper, 100% of teachers reported that words and symbols lighting up during reading helped students, 80% said computer-based math was easier to read, and 74% found it easier to complete math problems. Teachers noted that assistive technology eliminated decoding barriers so students could focus on content and reasoning. Customisable speech rules in MathPlayer proved important — default rules designed for blind users were distracting for sighted students with learning disabilities, and adjustments were needed so verbalisation matched how classroom teachers read expressions aloud. One unit (Looking for Pythagoras) showed negative results, but this was traced to a conversion error that stripped all MathML content from student copies.
Relevance
This study provides early but compelling evidence that making mathematical content digitally accessible through MathML benefits students with learning disabilities, not just those with visual impairments. The finding that MathML density correlated with improved outcomes suggests that the more mathematical notation a text contains, the greater the accessibility benefit of proper encoding. For practitioners, the paper highlights the critical importance of accurate content conversion — a single conversion error rendered one entire unit inaccessible. The study also demonstrates that speech rules for verbalising math expressions need to be tailored to the audience; what works for blind users may be counterproductive for sighted students with learning disabilities. This has direct implications for anyone implementing MathML in educational content today.
Tags: learning disabilities · mathematics accessibility · MathML · DAISY · print disabilities · assistive technology · K-12 education · text-to-speech · digital textbooks
Standards referenced: NIMAS · DAISY · MathML · Section 504 · ADA Title II