Size Matters (Spacing Not): 18 Points for a Dyslexic-Friendly Wikipedia
Luz Rello, Martin Pielot, Mari-Carmen Marcos, Roberto Carlini · 2013 · Proceedings of the 10th International Cross-Disciplinary Conference on Web Accessibility (W4A) · doi:10.1145/2461121.2461125
Summary
This paper investigates the effect of font size and line spacing on readability and comprehension for people with dyslexia, tested in the real-world context of Wikipedia articles rather than isolated text samples. The authors conducted an eye-tracking study with 28 participants with confirmed dyslexia diagnoses (ages 14-38, mean 21.36) who read six Spanish Wikipedia articles about animals. The study tested six font sizes (10, 12, 14, 18, 22, and 26 points) and four line spacing values (0.8, 1.0, 1.4, and 1.8). Previous research had studied these parameters in isolation using decontextualized text, making it unclear how findings would generalize to real websites with navigation bars, images, sidebars, and other contextual elements. The experiment used a hybrid-measures design: font size was a repeated measure (each participant read all six sizes) while line spacing was between-group. Readability was measured through eye-tracking fixation duration, comprehension through multiple-choice questions (both literal and inferential), and subjective perception through Likert-scale ratings. The study used actual Wikipedia HTML modified via a Chrome plug-in to change CSS properties, preserving the authentic reading context.
Key findings
Font size had a significant effect on both objective readability and subjective perception, while line spacing did not. For fixation duration (the objective readability measure), a significant effect of font size was found (F(5,130)=2.44, p<0.05). Pairwise comparisons showed that fixation durations for 18-point text were significantly shorter than for 10-point (p<0.01) and 26-point (p<0.05) text, indicating 18 points as the optimal size. Crucially, readability improved as font size increased up to 18 points, but larger sizes (22, 26) showed no further improvement and even slightly increased fixation durations. For comprehension, font size also had a significant effect (F(5,130)=3.72, p<0.01), with 18-point text yielding the highest comprehension scores. Subjective perception ratings mirrored these results — participants rated 18-point text as easiest to read and understand. Line spacing showed no significant effects on any measure: not fixation duration (F(3,24)=2.06, p=0.13), not comprehension (F(3,24)=0.54, p=0.66), and not subjective ratings. These results differ substantially from previous recommendations of 12 or 14 points for dyslexic readers, likely because prior studies used isolated text rather than real website contexts.
Relevance
This study provides an evidence-based, specific recommendation — 18-point font size — for designing web content accessible to readers with dyslexia. The finding that line spacing has no significant effect is equally valuable, as it contradicts commonly cited guidelines recommending increased line spacing (1.4 to 2.0) for dyslexic readers. For web developers and designers, the practical implication is clear: prioritize adequate font size over line spacing adjustments. The recommendation of 18 points is notably larger than WCAG's general minimum guidance, suggesting that dyslexia-specific considerations require going beyond baseline accessibility requirements. The study's strength lies in its ecological validity — testing on actual Wikipedia pages rather than isolated text ensures the findings are relevant to real-world web reading. However, the sample size of 28 participants and restriction to Spanish-language Wikipedia articles are limitations. The finding that readability plateaus at 18 points (with no benefit from going larger) is useful for designers balancing readability against screen real estate and layout concerns.
Tags: dyslexia · typography · readability · eye tracking · font size · line spacing · text presentation · cognitive accessibility · web accessibility
Standards referenced: WCAG