“How to Describe Complex Images for Accessibility”
A DIAGRAM webinar presented by Bryan Gould, NCAM
Q & A Summary – October 30, 2012
QUESTION: “Are there any specific guidelines for writing descriptions for images in assessment items? (That is, guidelines on
how to describe an image without giving the answer)”
QUESTION: “Do photos get described also or does the caption serve as the only required description?”
ANSWER: It depends on context. Many photos are sufficiently described/labeled by their caption. For example, “Maya Angelou” or “Brooklyn Bridge”. Yet, sometimes photos need more description.
QUESTION: “Do we need to note which states weren’t in the union at the time (Hawaii and Alaska?)”
ANSWER: In general, you should only describe information that is present in the image. In other words, you shouldn’t describe what you can’t see (with minor exceptions).
QUESTION: “Do you have the alt-text that describes all the Illustrations/Cartoons?”
ANSWER: There was no alt text for the images in my slides. I used them just to show the approach. For more examples of image description, please visit these resources:
QUESTION: “Still not sure how I would describe the water life cycle in that image.”
QUESTION: “What is being done to make social media (Facebook, Twitter, Skype) more accessible? And what are the implications with the new graphically-heavy Windows 8?”
ANSWER: This question is beyond the scope of what the DIAGRAM Center addresses, but it’s a good question! See this October 4, 2012 blog posting for a brief discussion of accessibility in social media.
QUESTION: “What about describing specialized symbols and discrete graphics. Many of my clients want to know ‘what it looks like’ in order to communicate with their sighted colleagues.”
ANSWER: If you know that the intended audience would benefit from a certain style of description or having something specific described, then yes, do it. That’s great information to have and good guidance.
QUESTION: Several questions asked about the CDC hepatitis poster.
ANSWER: As I said during the presentation, while I would have trimmed some of the descriptions even more, the description was still quite good. Thanks for paying such close attention!
QUESTION: “In slide 20 (bar chart) should Bill & Linda list year (2009)?”
ANSWER: Yes. Slide 20 contained a typo – now fixed.
QUESTION: “Where can I get the font used in your slides?”
ANSWER: I use “Times,” white text on blue background.
QUESTION: Several questions asked about context, like this one: “What is your best advice when you are not a subject matter expert and the image you have to describe includes no text? For example, images in a foreign language textbook?”
ANSWER: If you really have no idea what the image is, then you might be misleading the reader by guessing. There is a certain level of trust required on the part of the reader that the image descriptions are meaningful and as accurate as possible. In other words, if you don’t know what the image is or why it’s there, how helpful can your image description be? Better to pass the work onto someone who does know. For this and other reasons, there is a movement to have image descriptions written in the creation/production process (i.e., by text authors and publishers.)
QUESTION: “If an image is essentially a graphic version of information presented in words in the text, is it necessary to summarize? It may simply repeat what’s already described/explained.”
ANSWER: It’s best to put the image in some sort of context, for example: “a poster” or “a billboard.” After that, if the words are most
crucial, then that’s what the description should focus on.
QUESTION: “Would you translate or use exact quotes? Things like 5X (five times) and other typographical shortcuts, I mean.”
ANSWER: It depends on context, of course. I lean towards whichever is more direct/less ambiguous.
QUESTION: “Do you have specific suggestions for maps that students need to find information on to answer a question?”
ANSWER: In general, the description should focus on the information the student is required to discover. If the image is in an assessment, then the image description must maintain the test construct and not “give away the answer.”
QUESTION: “Is there a way to find out what screen readers do well – such as how good parentheses work?”
ANSWER: Test them! Screen readers and screen reader users are all slightly different and they vary by version and the type of document being read. For an extended answer I recommend this WebAIM article on screen readers.
QUESTION:“Are there any specific terms or processes that visually impaired persons use that we should try to employ within our descriptions? I’m wondering if there are certain techniques or methods that the majority of visually impaired students learn that we could use to explain images in frameworks they already use.”
ANSWER: In general, the answer is no, there are not “universal” techniques or methods for describing complex images. This is probably due to the low incidence of blindness and the fact that most people who are b/vi experienced at least some of their schooling and/or professional training with sight, meaning that image description was not necessarily a requirement throughout their education. These trainings and other efforts by NCAM and the DIAGRAM Center are intended to provide best practice guidance, especially as more texts and tests are delivered digitally.