Alt Text - Writing Alt Text for Complex Images
Adding Alt Text to Complex Images
Some images are really complex and require special considerations when creating alt text.
What We Do With Complex Images
Complex images are difficult to write appropriate alt text for. These are images that have text, charts, graphs, or other images that convey a lot of information to readers. There is no one right way to write alt text for complex images, because every complex image is different, and they all have different contexts.
For our job in making USU’s Canvas and other websites accessible, we should always include alt text, but it is okay if our efforts aren’t perfect. If there is a specific image (or other part of a document, for that matter) that is taking more time to fix than it’s worth, causing issues in a document, or doesn’t seem quite right, reach out to get some help! Another set of eyes can be beneficial, and we’re more than happy to help figure things out so that we as an Accessibility Team can be effective.
Images With Lots of Text
A general recommendation for images with text is to use the image text as the Alt Text, but some complex images have too much text for this to be effective. In cases where there is too much text in an image (for example, if the image includes several paragraphs of text), then it should not be an image, and should simply be a part of the text.
In some cases, the text from an image should be summarized or described. If there is an image with too much text that for some reason must be an image, the alt text should give a summary of the most important parts of the text.
Example 1
Ineffective Alt Text:
“Hey! Can I try that?”
“A cartoon person pointing to a plug connected to the earth.”
Effective Alt Text:
“Hey! Can I Try That?: A student handbook for choosing and using assistive technology.”
“The cover of a student handbook for choosing and using assistive technology titled ‘Hey! Can I Try That?’”
Example 2
This image contains too much text to effectively convey it in Alt Text, so we have two options to make this document accessible. One option is to insert the image, mark it as decorative, and type out the text in the document (following formatting as much as possible). If the image can be easily transferred into text, the second option is to simply omit the image and replicate its text and formatting (the best that you can). Both of these are good options—it is up to you as the editor to decide what is most appropriate for a particular image.
Charts and Tables
Some images containing text are tables. In that case, it would not make sense to use the image text as the alt text. For many complex images, it is best to incorporate them into the text. If there is a table or list that can be replicated within the main document, it would be more effective to simply add it there in place of the image.
Another idea is to have the image and its text next to each other in the document so that the text is conveyed but the visual is still available. This is especially helpful for large flowcharts or other text-filled images that also convey information visually.
In most cases for charts, the text should be summarized in a way that describes the information the chart gives you. This is difficult to describe, but look at the examples below to give a better idea of what summaries should look like for charts.
Example 1
Ineffective Alt Text:
“Domain of Medicine and Domain of Public Health flowchart.”
(Close, because it has some of the text that it should, but does not convey any information in the image)
“World, world region, nation, state/province, community, household, individual. Domain of public health, domain of medicine.”
(While this one does include all of the text in the image, reading it sounds almost nonsensical. It doesn’t tell the listener anything about why it says what it says.)
Effective Alt Text:
“The domain of medicine includes the individual and the household, while the domain of public health includes the household, community, state or province, nation, world region, and world.”
“Domain of Medicine: individual and household. Domain of Public Health: Household, community, state or province, nation, world region, and world.”
Example 2
Ineffective Alt Text:
“Benefits control criteria hierarchy”
“A chart with arrows connecting benefits control criteria to control criteria to economic, political, and social subcriteria.”
Effective Alt Text:
“The goal is Benefitting Control Criteria Hierarchy, which is connected to Economic (Income, Improve services, efficient Usage of facilities), Political (Conflict Resolution, Muslim Support, Religious Authority) and Social (Exposure to other culture, Tolerating other cultures, jobs) factors."
“Economic, political, and social factors are all essential to benefitting control criteria hierarchy. In turn, income, improving services, and efficient use of facilities are all essential to economic factors; conflict resolution, muslim support, and religious authority are essential to political factors; exposure to other culture, tolerating other cultures, and jobs are essential to social factors.”
Graphs
Some complex images cannot be effectively transferred to text or tables, like graphs (and some charts). The best course of action in this case is to describe the information in the graph or other image in the alt text the best you can. Remember to describe the content, not the visual aspects.
Example 1
Ineffective Alt Text:
“A graph of global average temperatures.”
“A red, black, and blue squiggly line”
"Graph of Global Average Temperature from 1880 to now"
Effective Alt Text:
“A graph showing that global average temperatures have risen substantially since 1880.”
“From 1880 to 1900, global average temperatures declined slightly, but have risen steadily since about 1910.”
"Graph of temperature vs. baseline from 1880 to 2020. The baseline temperature is zero, and from 1880 to 1910, on average the temperature was .2 to .4 degrees lower than baseline, and since 1910, the temperature increased substantially, to around 1 degree greater than the baseline.”
Example 2
Ineffective Alt Text:
“A pie chart of favorite colors.”
“Blue, red, green, and yellow circle”
Effective Alt Text:
“A pie chart showing that most people’s favorite color is blue, followed by red, green, and then yellow.”
“A pie chart of people's favorite colors. 45% said blue, 35% said red, 15% said green, and 5% said yellow.”
In a case like this, a table may also be appropriate, depending on the context and how important the information is to the document:
Maps
Maps can be another confusing part of complex alt text. Maps aren’t always difficult, because sometimes all that needs to be written is “map of…”
However, maps are often supposed to convey more than just the fact that it is a map. In these cases (like maps with different colors indicating pieces of information about certain regions), it is important to convey that in the alt text.
Example 1
For a map like this one, our initial response is probably similar to “The United States divided into five regions”. However, to be effective, we need to be more specific. If the text around the image tells about each region, the states that are included, etc., then it would be appropriate to write alt text that simply says “Map of the United States showing the west, southwest, midwest, southeast, and northeast regions.”
However, if the map is the only way of telling the reader what each region entails, then the alt text should give more information than before- for example, “Five regions of the United States. The Western region spans from the west coast to Colorado, as well as Alaska and Hawaii. The Southwestern region includes states from Arizona to Texas and Oklahoma. The Midwestern region spans from the Dakotas to Ohio, and the Southeastern region spans from Louisiana to the Southern half of the east coast. The Northeastern region spans from Pennsylvania to Maine.”
While this is a longer description, if it is the description the image and surrounding context warrants, then it’s okay for it to be a little longer than usual.
In other contextual cases, it may be necessary to be extremely specific in terms of what’s in the image. For example, if the document included things that required students to reference the above map and know which states were in which region, it may be helpful or necessary to include a table next to the map classifying each state as belonging to a particular region.
It is important to note that this will not be necessary very often. Most documents include info relevant to the image in the surrounding contexts, and usually that level of detail isn’t necessary. However, it is a good idea to be familiar with many different options, because different images require different adaptations. Familiarity with these options will help you more effectively judge what is necessary and most beneficial.
