RE: Discussion: WCAG 2.1 Conformance of Small Multiple (Multi-Panel) Scientific Figures

Hi Godwin,

This does seem like it should be treated as a complex image whose inherent meaning is baked into the image as a whole. So providing an alt text summary + extended description/data table would likely be enough for conformance. There may be specific and rare cases where providing individual images would make sense (either with or without the overall image) but that would be entirely context dependent.

Having said that, I'd be interested in gauging user feedback and seeing what improvements might be needed and technically achievable that go beyond minimum conformance - I see COVE<https://www.cdc.gov/cove/data-visualization-types/small-multiples.html> has some interesting features that you linked to. For example, and just off the top of my head, providing an option for viewing individual zoomed in versions of each individual component without having to zoom the whole image (which potentially triggers awkward 2D scrolling).

I work with educational publishers and will periodically look at complex content in subjects like STEM. At this level, there's very little published guidance, especially with good examples, not least because the level of knowledge needed to understand the content itself can be very high. Relying on basic principles is important - what's the image context, what info do users need to know, what other information is provided on the page, etc. But there are a lot of judgement calls, reliance on subject-matter experts and some compromises. There's definitely a need for more user research and probably new techniques too.

On a side note, you may be interested in a couple of initiatives that can help with guidance around this kind of specialised and complex content:


  *   Publishing Accessibility Action Group<https://www.paag.uk/training-learning/> (PAAG) - although this is UK-based it includes members and contributors from all kinds of publishers, suppliers, and accessibility specialists from across the globe. There's a mailing list and bimonthly meeting.
  *   Scholarly Image Taxonomy<https://stm-stec-ai-alt-text-website-ezzd.onrender.com/> - developed by the International Association of Scientific, Technical & Medical Publishers (STM). It's still in its early stages but worth following to see how it develops. I don't think they've captured this specific type of image, but they are keen to get feedback from the community.


Kind regards,

Nick

---
Nick Bromley
Director & Accessibility Consultant
Red Kite Digital Accessibility Ltd

From: Godwin Choy <godwinchoy@gmail.com>
Sent: 26 February 2026 20:53
To: w3c-wai-ig@w3.org
Subject: Discussion: WCAG 2.1 Conformance of Small Multiple (Multi-Panel) Scientific Figures

Hello WAI Interest Group,

I'd like to raise a topic that I believe affects many scientific and data-driven organizations but lacks clear WCAG guidance: the accessibility conformance of "small multiple" figures.

For those unfamiliar, a small multiple (a term popularized by Edward Tufte) is a series of small, simplified graphs arranged in a grid layout so that viewers can compare trends across all panels simultaneously. Here are some public references:
- Juice Analytics overview: https://www.juiceanalytics.com/writing/better-know-visualization-small-multiples
- Wikipedia overview: https://en.wikipedia.org/wiki/Small_multiple
- U.S. federal data visualization standards: https://xdgov.github.io/data-design-standards/components/small-multiples
- CDC COVE small multiples guidance: https://www.cdc.gov/cove/data-visualization-types/small-multiples.html

This is a foundational visualization technique in scientific publishing, public health, economics, climate science, and many other fields.

The Problem I Encountered:
In a scientific publishing context, a figure consisting of 12 panels in a 3×4 grid was created for comparative trend analysis. The figure is a raster image (JPG), spans the full width of the page content area (1140px), and scales responsively on desktop and mobile devices. During an accessibility evaluation, the evaluator recommended separating the figure into 12 individual enlarged images, each with its own long description.

I believe this defeats the figure's purpose. The analytical value of a small multiple depends on the spatial arrangement - viewers scan across panels to perceive patterns. Each panel is intentionally simplified and smaller than a standalone chart because you are not meant to examine one panel in isolation. Breaking it apart is like cutting a map into pieces and stacking them.

The figure already has: alt text, a detailed figure caption, a long description summarizing trends across all panels, an accompanying data table, sufficient contrast and resolution, and responsive design.

The Guidance Gap:
I've searched the WCAG 2.1 Understanding documents, Techniques, and WAI Tutorials extensively. The closest example is the SC 1.1.1 "data chart" example - but it describes only a single bar chart. There is no documented example or guidance for multi-panel composite figures like small multiples.

My interpretation is that:
- SC 1.1.1 Situation B applies: treat it as a complex image, provide short + long text alternatives and a data table
- SC 1.4.10 Reflow exception applies: the matrix layout is essential to meaning, similar to maps and diagrams
- SC 1.4.4 is met: text labels are legible at 200% zoom, and system-level assistive technology zoom provides additional access

But because there is no documented precedent for this specific image type, evaluators may reasonably interpret the absence of an example as non-conformance.

Questions for the Community:
- Has anyone else encountered this issue with small multiples or multi-panel scientific figures during accessibility evaluations?
- How have you handled it? Did you keep the figure intact or split it apart?
- Does anyone have experience with WCAG conformance assessments of complex scientific visualizations more broadly?
- Would it be valuable to propose that the AGWG add guidance or an example addressing this visualization type?

I have also submitted a formal clarification request to public-agwg-comments@w3.org<mailto:public-agwg-comments@w3.org> on this topic.

I'd welcome any perspectives, experiences, or guidance.

Thank you,
Godwin