This is an update to this post from 2018 with the latest data from 2021's edition of WebAIM's Screen Reader survey.
With that out of the way, let us jump right into it!
Followed somewhat closely by NVDA with a 30.7% of participants and VoiceOver with 6.5%.
However, this is only the main screen reader surveyed users use. As we'll see, screen reader users will commonly use more than one screen reader in their daily activities.
As mentioned in the survey, these numbers make sense because 71.3% of respondents use more than one desktop/laptop screen reader.
So, usage of those overlap for different reasons like work/personal screen reader differences, affinity for different screen readers for different use cases like mobile vs desktop, etc.
When it comes to screen readers though, it's not only a matter of the tool you choose. It is also important under what operating system they are running and what web browser they are being used with. Let's see which are the most commonly found combinations.
The other top contestants in this category were:
So now that we know what tools are being used, let's see what are the preferred ways to navigate and discover the content of a website.
How screen reader users approach and navigate lengthy websites is a topic that I was especially interested in.
Given the nature of web accessibility, where different screen readers convey the same structures in slightly different ways, sometimes it is more important to craft an experience that provides extra context.
Sometimes, that can turn out to be as important as having all the items checked in the accessibility auditing tool of choice.
But don't forget those either :)
It is clearly the preferred option when it comes to moving through the content.
"Using the find function" follows with a discrete 13.9%. Following are users who prefer to read through the page with an 8.1%. After that come the users who prefer going through links take a 7.1% of the total and navigating through landmark/regions last with has a 3.2%.
That would tell us to test taking the approach of going through headers when manually testing the usability and accessibility of a site. Try to adapt the site to achieve an understandable experience through proper markup if explored in this way.
This might be more about me being old than anything, but I feel like I have heard more than once the assumption that screen readers can't work with Javascript.
99.4% of the surveyed users had Javascript enabled while using their screen readers so I guess there should be no more doubt about that.
Another thing I have found in the wild is that a lot of people think "accessibility" equals having an Apple device.
While Apple devices do provide a very good and accessible user experience, the survey shows that 91.1% of users use Windows on Desktop in combination with their preferred screen reader.
That is somewhat counterbalanced by the fact that on mobile, Apple is indeed the king with 71.5% of the votes going for VoiceOver (only available on iOS). That also matches the 71.9% of users choosing Apple devices for their usage of mobile screen readers.
Another topic that seems to be one of the first UX issues to be tackled when fixing a11y issues is having skip links. The question of the survey was about how often respondents use a "skip to main content" or "skip navigation" link when it is available on a page.
There's no choice that got a strong representation here. Going by most chosen to least we have the two top ones: "Sometimes use them" with a 28.4% and "Seldom use them" with 21.6%.
These two account half of the votes which is very interesting.
After those comes the "Always use them" option with a 16.8% and the two last ones "Often" and "Never" both got 14.4% of the votes each.
Based on those results, it seems to me that adding skip links usually comes very early in the improvements phase while not necessarily being the most impactful fix to apply.
That being said, having skip links is a very nice feature of a website that should not be overlooked, especially in crowded or complex pages. Since the fix is usually not that costly to implement, it's normal that it gets prioritized first even if it's not that relevant as it was or was perceived to be.
If you still want to know more about how to build your markup or app to be accessible and usable, I totally recommend you to check the original survey from WebAIM.
It contains many more facts and further detail. I honestly believe it to be a very valuable source of information when trying to grow your understanding of how to implement web accessibility in a way that actually helps people :).
Please ping me on Twitter if you have any comments or just liked this!