Measuring complexity of e-government services for people with low vision
Aritz Sala, Myriam Arrue, J. Eduardo Pérez, Sandra M. Espín-Tello · 2020 · Proceedings of the 17th International Web for All Conference (W4A) · doi:10.1145/3371300.3383350
Summary
This paper investigates how to measure the complexity of online government forms for people with low vision. The authors recognize that e-government services — accessed primarily through web-based forms — have the potential to greatly benefit people with low vision by allowing them to interact with public services from home using their preferred assistive technology. However, these forms are frequently inaccessible and overly complex for this population. The researchers evaluated five Spanish public e-services, all offering appointment-booking functionality but implemented with different form designs. They applied three complementary evaluation methods: the BOFS (Bespoke Online Form Selection) complexity metrics originally developed for older adults, the Barrier Walkthrough Method for people with low vision to assess WCAG conformance, and user testing with ten participants — five with low vision and five without visual impairments. The BOFS metrics measured structural complexity factors such as number of pages, questions, response options, and types of interaction elements (radio buttons, checkboxes, dropdowns, text inputs). The Barrier Walkthrough evaluation identified specific accessibility barriers including insufficient visual contrast, inflexible page layouts, and missing layout cues. The user testing captured real interaction data including task completion rates, time on task, scrolling behavior, erroneous clicks around form elements, and data entry errors.
Key findings
People with low vision completed only 80% of tasks compared to 100% for sighted participants, and took significantly longer — up to 6 minutes 31 seconds versus 2 minutes 23 seconds on the most complex form. The study revealed that existing BOFS complexity metrics, while useful, miss several factors critical for people with low vision. Insufficient visual contrast was a universal barrier, found in all five services, and correlated directly with higher erroneous click rates around input fields (19-21 errors in low-contrast services vs. 2 in adequate-contrast ones). Radio button implementation mattered enormously: tabular layouts generated many mis-clicks, while radio buttons with clickable adjacent text labels produced zero errors. People with low vision made 9 times more errors on required fields (9 vs. 1), suggesting that asterisk-based required field indicators are ineffective — textual notification that all fields are mandatory eliminated these errors entirely. CAPTCHAs caused significant problems, particularly when audio alternatives were presented as small, hard-to-locate links. Zooming by people with low vision introduced extensive horizontal and vertical scrolling, with one service requiring scrolling on 96% of visited pages.
Relevance
This research has direct implications for anyone designing or evaluating government web forms or any complex online form system. The findings demonstrate that accessibility compliance alone does not equal usability — even services claiming WCAG AAA conformance had significant barriers for people with low vision. For practitioners, the actionable takeaways include: make radio button labels clickable (not just the button itself), use adequate color contrast especially on form inputs, avoid CAPTCHAs or provide prominently displayed alternatives, indicate required fields with text rather than asterisks, and ensure forms are usable at high zoom levels without excessive scrolling. The study also highlights the need for disability-specific complexity metrics that go beyond general accessibility guidelines, particularly for the estimated 246 million people worldwide living with low vision.
Tags: low vision · e-government · form accessibility · complexity metrics · user testing · web accessibility · online forms
Standards referenced: WCAG 2.0 · WCAG 2.1 · WCAG 1.0