Lost in Translation: Challenges and Barriers to Sign Language-Accessible User Research
Amelie Unger, Dieter P. Wallach, Nicole Jochems · 2021 · Proceedings of the 23rd International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS '21) · doi:10.1145/3441852.3476473
Summary
This experience report describes the challenges hearing researchers face when conducting user research with sign language (SL) users, based on the authors' experience running remote focus groups with German Sign Language (GSL) users as part of the AVASAG project — which is developing an automated system to translate German text into GSL via an avatar for travel information and services. The team conducted three focus groups with 3-4 GSL users each (10 participants total), lasting 90-120 minutes via video conference. The research team included both hearing HCI-design researchers and deaf GSL experts, with a native GSL user serving as moderator. The authors frame their work through Ability-Based Design, focusing on participants' SL communication abilities rather than their hearing disabilities. They describe how integrating deaf team members from the outset was essential — without this cross-language team composition, hearing researchers would have designed procedures that lacked appropriate communication, information, and interpretation for SL users. The focus groups yielded nearly 2,000 interpretative evaluation units about barriers within public transportation, demonstrating both the effectiveness of the approach and the pressing need of this population to share their experiences.
Key findings
The paper identifies several critical lessons for conducting SL-accessible research. First, sign language knowledge must be embedded in the research team from the beginning, not just during data collection — all preparation materials including consent forms, questionnaires, email communications, and video conference instructions need to be available in sign language, not just written text. Participants explicitly stated they preferred communicating in GSL rather than German text because it allowed them to express themselves more freely and accurately. Second, attention capabilities in remote SL focus groups are more limited than in spoken language sessions because participants must maintain constant visual focus on the screen. Unlike spoken discussions where participants can look away and still hear, averting visual attention in SL discussions causes direct information loss. The authors found that 90-120 minute sessions pushed the maximum attention range and should be split into shorter segments. The paper articulates a fundamental dilemma: there is growing demand for including SL users in research, but HCI lacks the accessible tools and validated procedures to do so properly. This manifests in two directions — communication towards participants (translating research instruments into SL, whose scientific validity cannot yet be determined) and communication from participants (SL content is visual and cannot be easily documented, quoted, or stored in text-based academic formats without subjective interpretation and information loss).
Relevance
This paper exposes a systemic barrier in accessibility research itself: the methods and tools researchers use to study DHH users are often not accessible to those users. For practitioners conducting any form of user research with SL communities, the lessons are immediately applicable — all participant-facing materials must be available in sign language, deaf team members should be involved from project inception, and standard research instruments like validated questionnaires may need complete redesign rather than simple translation for SL populations. The deeper challenge the paper raises — that academic publishing and research documentation are inherently text-based, creating structural exclusion of SL users from the research enterprise — calls for a fundamental rethinking of how the scientific community documents, stores, and disseminates knowledge produced in visual languages. Until sign language is treated as equally valid as written text in research contexts, the field will continue to underrepresent the perspectives and needs of SL users.
Tags: deaf and hard of hearing · sign language · user research · focus groups · language accessibility · ability-based design · research methods · German Sign Language · inclusion