WebAIM - Web Accessibility In Mind



后续调查是在2006年进行的October 2009,2010年12月,May 2012,January 2014,July 2015,和October 2017. Be sure to check out the results of these surveys for updated data and trends.


In December 2008 through January 2009, WebAIM conducted a survey of preferences of screen reader users. We received 1121 valid responses to the screen reader survey.

A few disclaimers and notices:

  • 由于四舍五入,总数可能不等于100%。
  • 该样本未控制,可能不代表所有屏幕阅读器用户。
  • 应注意解释这些结果。有些问题是技术性,我们了解许多参与者可能在技术上不太明媚。响应也可以基于用户与Web内容的体验,这些内容通常无法访问。万博体育官网网址如果筛选与Web内容的屏幕互补通常非常正,我们难道我们不禁怀疑响应可能不同。
  • We hope to conduct a survey of this nature again in the future. If you have recommendations or questions you would like asked, please让我们知道.



Which of the following best describes you?
响应 # of Respondents % of Respondents
由于残疾,我一直使用屏幕阅读器 1006 89.7%
I use a screen reader part of the time due to a disability 43 3.8%
I use a screen reader often, but do not have a disability that requires a screen reader 13 1.2%
I use a screen reader occasionally to perform accessibility evaluation 59 5.3%

Disabilities Reported

Disabilities Reported

响应 # of Respondents % of Respondents
失明 898 80.1%
低视力/视力受损 177 15.8%
认知 8 .7%
Deafness/Hard-of-Hearing 47 4.2%
Motor 24 2.1%
No disability 60 5.4%


Computer and Screen Reader Proficiency


响应 % of Respondents
Expert 22%
Advanced 44%
中间 27%
Beginner 8%

Pie Chart of Screen Reader Proficiency

响应 % of Respondents
Expert 17%
Advanced 41%
中间 32%
Beginner 9%

The responses for computer proficiency and screen reader proficiency were similar. Those who use screen readers for evaluation rated their screen reader proficiency much lower (80% chose Beginner or Intermediate) than those that always use screen readers (only 37% chose Beginner or Intermediate).

Screen Reader Usage





Upgrade Window % of Respondents
Immediately 41%
前6个月 25%
6-12个月 9%
1-2 years 9%
2-3 years 4%
3年以上 6%
没有回应 6%

Most respondents update their screen reader version soon after a new version is released (74.6% within the first year). However, a significant number of users may still be using screen readers that are 3 years old or older.

Screen Reader Customization

Screen Reader Customization

How customized are your screen reader settings? (e.g., changed verbosity, installed scripts, etc.)
响应 % of Respondents
A lot 29%
Somewhat 40%
Slightly 21%
Not at all 7%
没有回应 4%




78% of respondents reported using a screen reader on a desktop computer, 54% use a screen reader on a laptop, and 12% use a screen reader on a mobile phone. While many respondents report using a BrailleNote, PacMate, PDA or other devices, the totals for these has not yet been calculated and are not included in the numbers reported above. No respondents who use screen readers for evaluation reported using a screen reader on a mobile device.



浏览器 % of Respondents
IE6公司 33%
IE7公司 68%
IE8公司 2%
火狐 39%
Safari 6%

Respondents with no disability were nearly twice as likely to list Firefox as blind respondents - 66% to 37%.

注:The survey question did not ask for the predominant browser used, but for any browsers used with a screen reader. IE8 and Safari numbers were from free-form responses (i.e., they were not listed as available options to select). The percentage of Safari users is over double that of the overall population - this may be due in part to the fact that some in the Mac community actively solicited survey participation and encouraged respondents to indicate their Safari use, perhaps partially due to feeling snubbed because we didn't list them with IE and Firefox as direct choices.

Home Page Behavior


当第一次访问一个新的,陌生的主页,I'm most likely to...
响应 % of Respondents
浏览主页 46%
浏览或收听页面上的链接 35%
用搜索找到我要找的东西 13%
Look for a site map or site index 3%
没有回应 2%

Glaringly omitted from the list of possible responses was "Navigate using headings". Considering the航向导航的普及, this may have altered the results for this question.

Interestingly, the more proficient screen reader users are more likely to read through the home page and use links less often than less proficient screen reader users. This may be due to faster reading speeds for more experienced users. The home page strategies used were very similar regardless of proficiency, disability, or amount of screen reader use.

"Skip" Links


I use "skip to content" or "skip navigation" links...
响应 % of Respondents
Whenever they're available 22%
经常 16%
有时 28%
很少 19%
从未 10%
没有回应 4%

I prefer that the 'skip' link be called...

I prefer that the 'skip' link be called...
响应 % of Respondents
跳到内容 22%
跳到主要内容 28%
跳跃导航 6%
没有偏好 36%
没有回应 7%


Access keys


I use Access keys...
响应 % of Respondents
Whenever they're available 22%
经常 13%
有时 24%
很少 20%
从未 16%
没有回应 5%



Chart showing navigation by headings

响应 % of Respondents
Whenever they're available 52%
经常 24%
有时 14%
很少 5%
从未 2%
没有回应 3%

The responses to this question provided one of the greatest surprises to us. It is clear that providing a heading structure is important to screen reader users with 76% always or often navigating by headings when they are available. Use of heading navigation increased with screen reader proficiency with them being used always or often by 90.7% of expert users, 79.3% of advanced users, 69.9% of intermediate users, and 55.4% of beginners.



响应 % of Respondents
只要有空 26%
经常 25%
有时 31%
很少 11%
从未 4%
没有回应 5%

Over half of the respondent always or often use search when it's available. This emphasizes the importance of not only providing search functionality, but providing it in an easy-to-find and use manner.

Locating Search

Chart showing methods of finding search

响应 % of Respondents
Read through the page content until the search form is encountered 6%
Tab through page elements until the search form is encountered 8%
Find the word "Search" 18%
Jump to the first text/edit field on the page 25%
Jump to the first button on a page and go back one element 2%
跳转到页面中的第一个表单元素 36%
没有回应 4%

Respondents use a wide variety of techniques for finding the site search. Proficient screen reader users were more than twice as likely to jump directly to the form or text/edit field than less proficient users. Less proficient users were nearly three times more likely to use more manual methods (reading, tabbing, or finding) than more proficient users.


Chart showing use of site maps

If a site map is available, how often do you use it?
响应 % of Respondents
只要有空 8%
经常 10%
有时 26%
很少 27%
从未 24%
没有回应 5%

This response also provided a great diversity of responses, though a majority of respondents seldom or never use site maps. There was no marked difference in the use of site maps across screen reader proficiency or disability. In general, it appears that site maps may be beneficial, but are not commonly accessed by screen reader users.

Text-only versions


响应 % of Respondents
只要有空 24%
经常 13%
有时 24%
很少 23%
从未 13%
没有回应 4%


Screen Reader-only Content


If content is identified as being "for screen reader users", how often do you use it?
响应 % of Respondents
只要有空 38%
经常 15%
有时 25%
很少 13%
从未 4%
没有回应 5%

We found it interesting that users are much more likely to use content as identified as "for screen reader users" than they are to access text-only versions. Perhaps identifying content as being specific for this audience attracts higher levels of use.



How difficult are pop-up windows to you?
响应 % of Respondents
Very difficult 25%
有点难 28%
Not very difficult 26%
Not at all difficult 17%
No opinion 5%

The very evenly divided split among responses here is interesting. A closer analysis, however, reveals that pop-up windows are reported as very difficult twice as often by less proficient screen reader users than with higher proficiency. Alternatively, more proficient users were three times more likely to indicate that pop-up windows are not at all difficult. This shows that less proficient screen reader users (which represent 41% of respondents) have more difficult experiences with pop-up windows.

Web 2.0版

显示Web 2.0可访问性的图表

响应 % of Respondents
Very accessible 4%
Somewhat accessible 24%
Not very accessible 16%
Not at all accessible 2%
I don't know 54%

The majority of respondents don't know how accessible these technologies are, or quite possibly, they didn't know what we were asking. Firefox users were much more likely to give a favorable response, perhaps a reflection of Firefox support for ARIA, etc. Evaluators and those without disabilities were nearly twice as likely to indicate that these applications are not very or not at all accessible than those that always use screen readers or have disabilities. This may suggest that these applications are actually more accessible than evaluators believe them to be, or alternatively, that screen reader users with disabilities are less knowledgeable about the true inaccessibility of these technologies.


Images to enhance the mood or feel of a web page


响应 % of Respondents
由我的屏幕阅读器描述 59%
被我的屏幕阅读器忽略 31%
No opinion 10%

It's difficult to make a recommendation with such divided responses. As accessibility 'experts', we tend to recommend that developers not provide alternative text for such images, and this is partially reflected in the survey responses. We found a distinct difference between user groups. 66% of evaluators preferred that the image be ignored, compared to only 28% of those that always use a screen reader. Similarly, 65% of those with no disability preferred that the image be ignored, compared to 29% of those with disabilities. There were, however, only marginal differences in this preference when analyzed by screen reader proficiency - with more proficient users wanting descriptions slightly more often. This may be due to high reading rates and the ability of proficient users to sort out unwanted information.

Interpreting the results of this is difficult. It is clear that there is a disconnect between what evaluators/those without disabilities and full-time/disabled screen reader users want. In general, further analysis is needed before recommending to developers that such images always be given descriptive alternative text, particularly when we consider that the less proficient screen reader users tend to want less description.

Identification of photos


If a web page contains a photo of the White House, I prefer that the image be identified as...
响应 % of Respondents
白宫的照片 80%
白宫 12%
No opinion 8%

非常清楚的是,绝大多数屏幕阅读器用户更喜欢识别出这样的照片。有趣的是,那些没有残疾的人比那些有残疾的人更喜欢Briefer Alt文本的可能性。Briefer替代方案的趋势也略微增加,筛选读者熟练程度(与上面发现的装饰型图像上面发现的趋势似乎相反)。

注:这个不应该解释为用户prefer that all images be identified. In other words, this doesn't mean that alt="graphic of my house" is preferred over alt="my house". The results here refer to photographs only.



The Acme Corporation web site has a logo at the beginning of the page that links to their homepage. I prefer that the image be identified as...
响应 % of Respondents
Acme公司主页 24%
ACME CORPORATION LOGO.with link to homepage 32%
No opinion 7%

Once again, the overall numbers make it hard to recommend a strategy for logo alternative text. In general, those with disabilities, those that use a screen reader more, and those with higher screen reader proficiency all tended to prefer the more brief alternative texts more than those with no disabilities, less frequent use, and lower proficiency. 49% of evaluators preferred "Acme Corporation logo with link to homepage" compared to only 30% of those that always use a screen reader.


Chart showing ease of repeated links

响应 % of Respondents
Very easy 20%
有点容易 25%
有点难 31%
Very difficult 13%
I don't know 10%


Chart showing ease of Flash content

响应 % of Respondents
Very easy 3%
有点容易 11%
有点难 37%
Very difficult 34%
I don't know 15%

在所有的调查结果中,这个可能是最能说明问题的。71.5%的屏幕阅读器用户表示Flash很难,只有14.2%的用户表示Flash很容易。根据熟练程度、使用屏幕阅读器的时间和残疾程度,回答差别不大。从这些结果和许多的strong comments about Flash inaccessibility that screen reader users have significant issues accessing Flash content.


Chart showing ease of frames

Frames are...
响应 % of Respondents
Very easy 29%
有点容易 29%
有点难 19%
Very difficult 8%
I don't know 14%


注:调查问题没有区分frames that may require navigation and inline frames (iframes) which are read inline.

Acrobat / PDF.


响应 % of Respondents
Very easy 15%
有点容易 29%
有点难 31%
Very difficult 17%
I don't know 7%

Once again, the same number reported PDF as being easy as found it difficult. It's hard to infer anything from this except that user experiences with Acrobat files seem to vary greatly.


Two of the questions in WebAIM's recent screen reader survey asked users which sites they find more difficult or easier to use. Listed below are thetop tenanswers from each category.

Question: List a few of your favorite web sites used for search, reference, entertainment, shopping, or anything else.

Favorite web sites
Web site Number of respondents
Google 671
亚马逊 270
Wikipedia 95
Yahoo 79
eBay 68
BBC 52
Facebook. 43
YouTube 42
Audible 39
书呆子 38

Question: What are a few web sites or types of web sites that you would like to visit, but avoid because of accessibility issues?

Web site Number of respondents
基于闪存的网站 85
购物网站 56
亚马逊 46
Facebook. 45
News sites 39
聚友网 37
Yahoo 29
eBay 21
YouTube 21
Travel sites 21

Five sites are found in both top 10 lists of web sites (commonly used/commonly avoided):

Web site Easy Difficult
亚马逊 270 46
Facebook. 43 45
Yahoo 79 29
eBay 68 21
YouTube 42 21



Many users commented on difficult web sites in the form of general site types, such as flash-based or travel/airline sites. Even with different web sites grouped into categories, respondents were more likely to identify sites they visit than sites they don't visit. The top 10 used sites above were identified a total of 1397 times, whereas the top ten web sites that screen reader users avoided were identified only 400 times. Considering this difference, the number of times a site was listed as one that was avoided carries much significance. For example, while 43 respondents listed Facebook as a site they use and 45 respondents listed Facebook as a site they avoid. When you consider the prevalence of sites being listed, Facebook accounts for 11% of the web sites listed to avoid and only 3% of those that users find easy. In short, Facebook was much more likely to be listed as a site that was avoided than as one that was commonly used.

Comments from Screen Reader Users

"I don't use things such as MySpace because I hate the automatic flash player that's louder than your average train."


"I like visiting sites where I can shop, but the descriptions of items are most often not detailed enough for me to be sure of what I'll get."


"I just go and if it's a problem I don't go back. But I try to let their webmaster know how I feel."



"Image links which do not have alt tags are also very difficult, since the screen reader can not decipher what the link is."


Perhaps the most significant conclusion we can make from these survey results is that没有典型的屏幕阅读器用户. As developers, we sometimes view screen reader accessibility as JAWS or Window Eyes or VoiceOver (or whatever) compatibility. This survey emphasizes that screen reader accessibility is about real people - and people that have diverse abilities and preferences. As developers, we must do our best to accommodate the needs of this diverse group.

In general, these results suggest that following accessibility guidelines and standards, using technologies that support high levels of accessibility, and providing users with options is of the highest importance. The wide range of user responses makes it difficult to provide definitive recommendations for many things. It may also be interpreted that some things (such as relatively insignificant differences in alternative text or the wording of the "skip" link) really don't have much of an impact on screen reader users. On the other hand, the survey also indicates a very strong favorability toward headings and a very high level of difficulty with Flash content.