PDF Accessibility of Research Papers: What Tools are Needed for Assessment and Remediation?
Open accessAbstract
Trillions of documents online are in PDF format, but only a small amount of these PDF documents include the necessary markup to make them accessible for people with disabilities. This paper presents the results of three related data collection efforts: a survey (with 61 participants), interviews (with 6 participants), and usability testing (with 6 participants), to learn more about what tools are needed for content contributors, to assist them in the assessment and remediation of accessibility in PDF documents. The paper provides suggested features and usability needed for software tools to support PDF document accessibility, as well as implications for content creators, scientific publishers, as well as the creator of the PDF format, Adobe.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The complexity of science figures makes it challenging to write effective alt text, and it remains a significant problem.
“The complexity of science figures also makes it challenging to write effective alt text; therefore, it remains a significant problem [54, 65, 79, 79].”
From How University Disability Services Professionals Write Image Descriptions for HCI Figures Using Generative AI (2026) · cites this paper as [65]
There is a general lack of awareness concerning accessibility and disability-related issues.
“There is also a general lack of awareness concerning accessibility and disability-related issues (García-González et al., 2021;Jembu Rajkumar et al., 2020;Schultz & Azadbakht, 2021).”
From Exploring the role of generative AI in higher education: Semi-structured interviews with students with disabilities (2025) · cites this paper as Jembu Rajkumar et al., 2020
Even correctly generated borne-accessible PDFs often still required additional remediation.
“Second, Rajkumar et al. [30] noted that even correctly generated borneaccessible PDFs often still required additional remediation.”
From iTagPDF: Towards Finally Automating PDF Accessibility (2026) · cites this paper as [30]
No citation lists for this paper
No citation lists for this paper yet — checked again daily.
References 11 · Cited by 28
- Authors
- Jembu Rajkumar, A., Lazar, J., Jordan, J. B., Darvishy, A., & Hutter, H.-P.
- Affiliations
- ZHAW Zurich University of Applied Sciences (CH)
- Venue
- Proceedings of the Annual Hawaii International Conference on System Sciences
- Published
- 2020
- ISSN
- 1530-1605 (ISSN-L)
- 2572-6862
- OpenAlex
- W3004643255
- Cited by
- 28
- References
- 11
- Citations by year
- 2026: 3 · 2025: 6 · 2024: 3 · 2023: 9 · 2022: 3 · 2021: 3 · 2020: 1
- FWCI
- 5.41
- Type
- conference paper
- Language
- en
- Open access
- green
- hdl.handle.net
- Licence
- cc-by-nc-nd
- Locations
- Hawaii International Conference on System Sciences (HICSS) (publishedVersion)
- Journal of the Association for Information Systems (submittedVersion)
- Zürcher Hochschule für Angewandte Wissenschaften digital collection (Zurich University of Applied Sciences) (submittedVersion, cc-by-nc-nd)
- ScholarSpace (University of Hawaii at Manoa) (submittedVersion, cc-by-nc-nd)
- ZHAW Digital Collection (cc-by-nc-nd)
- Topics
- Social Sciences › Human Factors and Ergonomics › Digital Accessibility for Disabilities
- Physical Sciences › Computer Science › Human-Computer Interaction › Usability and User Interface Design
- Physical Sciences › Computer Science › Computational Theory and Mathematics › Mathematics, Computing, and Information Processing
- Keywords
- PDF accessibility, research papers, people with disabilities, usability, content creators, scholarly publishing
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
