Are "Universal Design Resources" Designed for Designers?
Young Sang Choi, Ji Soo Yi, Chris M. Law, Julie A. Jacko · 2006 · Proceedings of the 8th International ACM SIGACCESS Conference on Computers and Accessibility (Assets '06) · doi:10.1145/1168987.1169003
Summary
This paper from Georgia Institute of Technology and UMBC investigates why universal design (UD) adoption in information and communication technology (ICT) remains slow, hypothesizing that the design resources meant to help designers create universally designed products are themselves poorly designed for their intended audience. The researchers conducted a two-part study: a heuristic evaluation of eight selected Universal Design Resources (UDRs) from different geographic regions and organizational types, and a web-based survey of the people who actually contributed to creating those resources. The eight UDRs evaluated ranged from legally binding standards like Section 508 and Section 255 to voluntary guidelines from the UK, EU, Ireland, and Australia, as well as academic educational resources. Four expert evaluators with backgrounds in UD, HCI, and human factors assessed each UDR against ten custom heuristics organized under three principles: addressing pertinent product design aspects, supporting the design process and designer psychology, and designing the document effectively. The heuristics were specifically developed to evaluate whether design resources meet the cognitive and practical needs of designers during the product development process.
Key findings
The heuristic evaluation revealed that most UDRs scored worst under Principle 2 (supporting design process and psychology), with heuristic H2.2 (supporting inevitable trade-off decision-making) generating the highest total severity scores — three catastrophic problems were found across the resources. This means the resources fail precisely where designers need the most help: making practical design trade-offs. Principle 3 (effective document design) scored lowest overall, with heuristic H3.1 (clear and appealing design) producing the lowest severity scores, suggesting the documents were at least visually acceptable. The survey comparing contributors to Section 508 (Resource A) and the NDA Accessibility Guidelines (Resource E) revealed a striking disconnect: Section 508 contributors rated their own professional expertise highly but rated the resulting resource poorly, while Resource E contributors had less confidence in their expertise but viewed their resource more positively. Notably, Resource A contributors rated 0 (no experience) for industrial design, interaction design, and graphic design — key skills for creating designer-friendly resources. Neither resource involved user evaluation with actual designers during development.
Relevance
This paper offers an important meta-level critique that remains highly relevant today: accessibility guidelines and standards are often created without considering the usability needs of the designers who must apply them. The finding that UDRs fail most at supporting design trade-offs is particularly significant — practitioners regularly face situations where accessibility requirements conflict with other constraints, and current resources offer little guidance for navigating these tensions. The research suggests that the gap between accessibility guidelines and actual accessible products is not just a matter of awareness or compliance, but partly a failure of information design. For organizations developing accessibility standards, training materials, or design systems, this paper argues that these resources should be treated as products themselves, subject to user-centered design processes including usability testing with their target audience of designers and developers. The work also highlights the importance of including people with industrial and interaction design expertise on standards development committees.
Tags: universal design · design resources · heuristic evaluation · usability · ICT accessibility · design process · guidelines
Standards referenced: Section 508 · Section 255 · WCAG 1.0 · COST 219