Screen Reader User Survey #7 Results

Introduction

In October 2017, WebAIM surveyed preferences of screen reader users. We received 1792 valid responses. This was a follow-up to 6 previous surveys that were conducted between January 2009 and July 2015 (see Related Resources).

A few disclaimers and notices:

• Totals may not equal 100% due to rounding.
• Total responses (n) for each question may not equal 1792 due to respondents not answering that particular question.
• The sample was not controlled and may not represent all screen reader users.
• We hope to conduct additional surveys of this nature again in the future. If you have recommendations or questions you would like us to ask, please contact us.

Demographics

Nearly 25% of respondents rely at least partially on the visual components when using a screen reader. 83.4% of those with disabilities rely exclusively on audio, compared to only 5.3% of those without disabilities (primarily testers). This is not entirely unexpected, but does indicate significant differences in usage between those with disabilities and those without disabilities.

Only 1.3% of those with disabilities rely exclusively on the visual output—many of these reported having cognitive or learning disabilities. Users of ZoomText, Narrator, and ChromeVox were much more likely to use the visual output than users of other screen readers.

Primary Screen Reader

Which of the following is your primary desktop/laptop screen reader?

Screen Reader	# of Respondents	% of Respondents
JAWS	811	46.6%
NVDA	555	31.9%
VoiceOver	204	11.7%
ZoomText	42	2.4%
System Access or SA To Go	30	1.7%
Window-Eyes	27	1.5%
ChromeVox	7	0.4%
Narrator	6	0.3%
Other	60	3.4%

The following chart shows historical trends for primary screen reader usage.

What happened in 2015? Essentially, the survey was distributed to a much broader audience, with many ZoomText and Window-Eyes users recruited to respond. Window-Eyes was also offered freely with Microsoft Office before the 2015 survey, but has since been discontinued. A much broader analysis from 2015 is available on the WebAIM blog.

This doesn't necessarily mean that the 2015 numbers were inaccurate. They certainly are accurate of respondents for that survey, which included more low-vision users than any other survey. 39% of respondents in July 2015 reported low vision, compared to only 20.4% on this survey. This difference in respondent demographics accounts for much or most of the decrease of respondents using ZoomText in 2017.

In short, there are three primary players—JAWS, NVDA, and VoiceOver. But we should not discount the continued impact of other screen readers, primarily ZoomText among the broader low vision community.

For survey simplicity, other specific screen readers were not offered as response options. The survey comments indicate that SuperNova was very common among "Other" screen readers.

Respondents with disabilities are more likely to use JAWS and less likely to use VoiceOver as their primary screen reader than respondents without disabilities. 10.4% of respondents with disabilities use VoiceOver compared to 22.6% of respondents without disabilities.

NVDA users reported higher levels of screen reader proficiency than users of other screen readers.

Screen Readers Commonly Used

Which of the following desktop/laptop screen readers do you commonly use?

Screen Reader	% of Respondents
JAWS	66.0%
NVDA	64.9%
VoiceOver	39.6%
ZoomText	6.0%
SA or SA To Go	4.0%
Window-Eyes	4.7%
Narrator	21.4%
ChromeVox	5.1%
Other	6.4%

See the commentary above regarding the July 2015 values.

Usage of JAWS, NVDA, and VoiceOver are all up since 2015, with Window-Eyes and ZoomText significantly lower. Of note is that Narrator, which has been significantly improved in Windows 10, was used as a primary screen reader by only 0.3% of respondents, but was commonly used by 21.4% of respondents.

68% of respondents use more than one desktop/laptop screen reader. This was up from 53% in July 2015. 36% use three or more, and 12% use four or more different screen readers. VoiceOver users most commonly use additional screen readers, which is notable since the other screen readers run almost exclusively on Windows.

Free/Low-cost Screen Readers

Do you see free or low-cost screen readers (such as NVDA or VoiceOver) as currently being viable alternatives to commercial screen readers?

Response	# of Respondents	% of Respondents
Yes	1,367	77.8%
No	184	10.5%
I Don't Know	206	11.7%

The positive perception of free or low-cost screen readers continues to increase. Positive responses to this question were 48% in October 2009, 60% in December 2010, 67% in May 2012, 74% in January 2014, and 78% now.

Only 66% of JAWS users answered "Yes" compared to an overwhelming 92% of VoiceOver users and 94% of NVDA users. Those that actually use free or low-cost screen readers have a much better perception of them than those who do not use them. Respondents with "Advanced" screen reader proficiency were also more favorable of free/low-cost screen readers.

Screen Reader Updates

Has your primary screen reader been updated in the last year?

Response	# of Respondents	% of Respondents
Yes	1,562	88.5%
No	203	11.5%

The vast majority (88.5%) of respondents indicated that their screen reader has been updated in the last year. This was 82.7% in 2014. 95.5% of NVDA users, 93.4% of VoiceOver users, and 85.9% of JAWS users updated in the last year.

Browsers

When using your primary screen reader, which browser do you use most often?

Browser	# of Respondents	% of Respondents
Firefox	719	41.0%
Internet Explorer 11	408	23.3%
Chrome	271	15.5%
Safari	184	10.5%
Internet Explorer 6, 7, or 8	71	4.1%
Internet Explorer 9 or 10	70	4.0%
Microsoft Edge	8	0.5%
Other	22	1.3%

For the first time, IE is no longer the most common browser among respondents. Internet Explorer (all versions) usage decreased to 31.4% from 53.5% in July 2015, 58.7% in January 2014, and 67.5% in May 2012. Firefox was used by 41% (up from 24.2% in 2014) of respondents. 31.4% represents a significantly higher IE usage than among the overall population (most statistics place it well below 10%). Usage of IE 6 through 10 was almost non-existent (1.6%) among those without disabilities, but remains at 8.8% among those with disabilities.

Usage of Internet Explorer was much higher in North America than elsewhere in the world. 35% of respondents from North America reported using IE as their primary browser, compared to only 23.4% in Asia and 24% in Europe.

Usage of Chrome more than doubled since July 2015, but was still well below usage by the overall population. Microsoft Edge usage was very low at .5%—notably below the overall population.

Screen Reader / Browser Combinations

There are many combinations in use, with JAWS with IE the most common, followed closely by NVDA with Firefox.

Operating System

Operating System

Operating System	# of Respondents	% of Respondents
Windows	1,304	72.8%
iOS	254	14.2%
Apple	141	7.9%
Android	55	3.1%
Linux	25	1.4%
Other	13	.7%

Operating system data above was detected from the system used to complete the survey. Respondents using iOS and Android nearly tripled since 2015. Respondents without disabilities were almost 4 times more likely to use Apple than respondents with disabilities, whereas users with disabilities were more likely to respond using iOS devices.

Braille Output

Do you use braille output with your screen reader?

Response	# of Respondents	% of Respondents
Yes	516	33.3%
No	1,034	66.7%

Because it would not generally be expected that users without disabilities would use Braille, they have been omitted from these data. Braille usage at 33.3% was up slightly from 27.7% in May 2012. 48.7% of VoiceOver users used Braille compared to a much lower 35.1% of JAWS users and 29.9% of NVDA users.

Mobile Screen Readers

Web Accessibility Progress

Which of the following best describes your feelings regarding the accessibility of web content over the previous year?

Response	# of Respondents	% of Respondents
Web content has become more accessible	711	40.8%
Web content accessibility has not changed	703	40.4%
Web content has become less accessible	327	18.8%

Respondents have a slightly more positive perception of the state of web accessibility now than they did in 2015. Respondents without disabilities tend to be more positive about recent progress (51.7% thought it has become more accessible) than those with disabilities (39.6% thought it has become more accessible).

Impacts on Accessibility

Which of the following do you think would have a bigger impact on improvements to web accessibility?

Response	# of Respondents	% of Respondents
Better (more accessible) web sites	1490	85.3%
Better assistive technology	257	14.7%

Over time, more respondents have answered "better web sites" to this question—68.6% of respondents in October 2009, 75.8% in December 2010, 81.3% in January 2014, and now 85.3% on this survey. This change perhaps reflects improvements to assistive technology. It certainly indicates that users expect site authors to address accessibility issues.

Social Media Accessibility

In general, how accessible are social media web sites to you?

Response	# of Respondents	% of Respondents
Very Accessible	253	14.9%
Somewhat Accessible	921	54.3%
Somewhat Inaccessible	246	14.5%
Very Inaccessible	83	4.9%
I Don't Know	192	11.3%

Compared to responses from previous surveys, respondents are increasingly positive about the accessibility of social media sites - 69.2% find them very or somewhat accessible compared to 55.2% in 2012 and 60.3% in 2015. 73.1% of respondents with advanced screen reader proficiency rate social media sites as very or somewhat accessible, compared to only 62.8% of respondents with beginner proficiency.

Landmarks/Regions

How often do you navigate by landmarks/regions in your screen reader?

Response	# of Respondents	% of Respondents
Whenever they're available	307	18.0%
Often	213	12.5%
Sometimes	491	28.8%
Seldom	364	21.3%
Never	332	19.4%

The frequent use of landmarks and regions has continually decreased from 43.8% in January 2014, to 38.6% in July 2015, to 30.5% on this survey. It's difficult to know the reasons for this. It could be due to infrequent or improper usage of landmarks/regions in pages. Or perhaps because other mechanisms are continually better. 45.4% of JAWS users reported always or often using landmarks in July 2015 compared to only 28.5% now just 2.5 years later.

Finding Information

When trying to find information on a lengthy web page, which of the following are you most likely to do first?

Response	# of Respondents	% of Respondents
Navigate through the headings on the page	1180	67.5%
Use the "Find" feature	252	14.4%
Navigate through the links of the page	118	6.8%
Navigate through the landmarks/regions of the page	69	3.9%
Read through the page	128	7.3%

While reliance on headings as the predominant mechanism for finding page information had notably increased between 2008 and 2014, responses to this question are largely unchanged since 2014. While 30.5% of respondents indicate that they always or often use landmarks when they are present, only 3.9% use this method for finding information on a lengthy web page. Those with advanced screen reader proficiency are much more likely to use headings (73% use headings) than those with beginner proficiency (42% use headings) who are more likely to read through the page.

Heading Structures

Which of the following page heading structures is easiest for you?

Response	# of Respondents	% of Respondents
One first level heading that contains the site name	95	6.6%
One first level heading that contains the document title	858	60.0%
Two first level headings, one for the site name and one for the document title	476	33.3%

Preference for a single <h1> that presents the document title has significantly increased from 37.1% in 2010 to 60% in 2017. A single <h1> for the site name was by far the least desired.

"Skip" Links

When a "skip to main content" or "skip navigation" link is available on a page, how often do you use it?

Response	# of Respondents	% of Respondents
Whenever they're available	273	15.8%
Often	283	16.4%
Sometimes	480	27.8%
Seldom	374	21.6%
Never	319	18.4%

When compared to July 2015, the frequent usage of "skip" links has decreased from 37.8% to 30.2%. 54.9% of respondents without disabilities always or often use "skip" links compared to only 29.6% of respondents with disabilities. This represents a very significant disparity in usage.

It is important to note that while usage has decreased among screen reader users, "skip" links still provide notable benefit for other keyboard users.

Problematic Items

The survey asked respondents to select their most, second most, and third most problematic items from a list. In giving each selected item a weighting, the following chart shows the overall rating of difficulty and frustration for each item.

In order, the most problematic items are:

1. CAPTCHA - images presenting text used to verify that you are a human user
2. Screens or parts of screens that change unexpectedly
3. Links or buttons that do not make sense
4. The presence of inaccessible Flash content
5. Lack of keyboard accessibility
6. Complex or difficult forms
7. Images with missing or improper descriptions (alt text)
8. Missing or improper headings
9. Too many links or navigation items
10. Complex data tables
11. Inaccessible or missing search functionality
12. Lack of "skip to main content" or "skip navigation" links

CAPTCHA remains the most (by a notable margin) problematic item indicated by respondents. The order and indicated difficulty for the items in this list are largely unchanged over the last 8 years, with one notable exception—"Screens or parts of screens that change unexpectedly". This item has moved from 7th most problematic in 2009 to 5th most problematic in 2012 to 2nd most problematic in 2017. This is likely a result of more complex and dynamic web applications.

Respondents with disabilities were nearly twice as likely to rank CAPTCHA and unexpected screen changes as problematic items than respondents without disabilities, who generally indicated that keyboard and forms accessibility were much more problematic than their peers with disabilities. 10.9% of respondents with disabilities rated keyboard accessibility as their single most problematic item, compared to 39.6% of respondents without disabilities. This suggests some notable disparities in perception of difficulties between these two groups.