Putting Web Site Quality and Accessibility into Context - Accessibility

Before looking at the survey results for accessibility, I want to relate an important experience I had with accessibility in 2004. I was the project manager for a Web site that provided an online version of an Australian national survey of all students graduating from higher education. The purpose of the survey was to gather statistics to help direct the future of higher education in Australia. The online version was launched with a reasonable degree of testing. In terms of accessibility, given the survey was run on behalf of a government organization, it was a legal requirement for the survey to adhere to Level 1 of the Web Content Accessibility Guidelines (WCAG) standard.

We tested the site prior to launch and used online validation tools to confirm the Web site met the standards. In terms of the automated testing, we ensured it passed all tests. For the warnings generated, we evaluated each of them and then decided if we needed to make changes or if the warnings did not affect the accessibility of the site.sdfasdf The Web site was launched in 2003. In early 2004, we received an e-mail stating the site did not adhere to accessibility standards and that if it was not rectified legal action would be taken. The person who made this claim was the same person that took the Australian government to court over the Sydney 2000 Olympics Web site stating it discriminated against him (he had a visual impairment), and he won the case. We knew we had to take this claim seriously.

On further investigation, we discovered that Level 1 was not sufficient; the Web site had to adhere to Level 2 and in some cases Level 3. It took us some time and effort to adjust the site accordingly. We also engaged assistance from the Victorian Royal Society of the Blind (VRSB) to ensure that the changes made worked. What we learned was that simply meeting the guidelines and passing the automated testing didn’t mean the Web site was actually accessible in a practical sense. For example, on a particular page, we had a number of form elements. We adjusted the code to meet the accessibility standards, but when we tested that page with a speech reader (JAWS) we found it didn’t make sense; the order in which the elements were read out wasn’t logical. Questions weren’t immediately followed with the possible answers even though they were laid out that way in the code. This finding was backed up by feedback from VRSB who explained to us the difference between pure compliance and practical compliance, that is, a Web site that passes automated tests and a Web site that is usable by a person with disabilities, in this case blindness. In time, we struck a balance and resolved the issue. But what is clear is that making a Web site accessible is not a straight forward task.

On the other hand, I recently launched a Web site that had a requirement to adhere to Level 1 WCAG standards. In this case, it was done, and automated tests were passed. In terms of the practical nature of those changes and if the Web site was practical for a disabled person to use, this was not tested. The client only cared if it passed the automated testing and the superior could be informed that the Web site was compliant. It was just a matter of being able to tick off a box.

Levels of Accessibility

The term “accessibility” is used often, but what exactly it refers to is unclear. In fact, when asked what level of accessibility respondents’ Web sites adhere to, the responses show a clear lack of understanding of the differences in accessibility. In fact, more than 50% of respondents say they don’t know which standard their Web site adheres to, and 18% say their Web site doesn’t adhere to any accessibility guidelines (see Graph 14). However, before getting indignant about this apparent lack of standards, there are some important ethical questions to ask. Does it matter if a Web site is accessible? If it is accessible, is it OK only to deal with certain types of accessibility, for example, mobility issues and not visual impairment? Is there a need for legislation that all sites provide a level of accessibility? To which standards should we be adhering given there are different standards such as the Americans with Disabilities Act, Section 508, and the W3C standards.

Graph 14 - What level of accessibility?

Let’s start with the first question: does it matter? Ethically, as a Web professional I would say yes; from a commercial perspective, if I’m the client and I’m paying for the Web site, I may not see the value in paying extra to making my Web site adhere to accessibility guidelines. Perhaps I sell motorcycles, and my customers don’t have visual or physical impairments. To pay for a feature that is almost certain not to impact the success of my Web site is not a wise commercial decision. Should adherence be a legal requirement, like having wheelchair access to a building? That’s a different case, but so far accessibility is not a legal requirement for all Web sites. Looking at the survey results, there seems to be a wide range of requirements when it comes to what percentage of the target markets requires the Web site to be accessible. The results range from 18% saying their entire target market requires adherence to accessibility to 13% saying less than 10% have this need (see Graph 15).

The results are fairly evenly distributed, but what is interesting is that no respondents say that none of their target market has a need for accessibility. What this indicates is that even if we want to take a purely commercial view and ignore accessibility, we could find that we are ignoring part of our target market. In asking this question, we find that only 17% of respondents even know what proportion of their target market requires accessibility (see Graph 5).

This points to a strong case of it’s better to be safe than sorry. If there is a question of whether you should be adhering to accessibility guidelines, the important thing is to be informed and to make the effort to find out how much it matters and what percentage of your target market it affects; then at least you can make an informed decision. In terms of types of impairment, if I know that my Web site is going to be accessed by people with visual impairments, and I make sure it is accessible in that manner but I don’t care about people with mobility issues, is that OK? Once again, ethically questionable but commercially you could justify it. Finally, should we simply make it law to make all commercial Web sites adhere to accessibility guidelines? I think the answer is clear that this is a good thing, and we should support any government that chooses to make this law.

Accessibility Testing

The complexity of accessibility is clear. We can’t simply run an automated test and hope that it means a Web site is accessible. There are numerous factors to consider, including the types of impairment, how best to accommodate for them, and how to test to ensure the Web site is accessible from a practical perspective. The obvious solution is to get your Web site tested by people with disabilities. They will tell you if your site is really accessible. Does this happen in the real world? Unfortunately, the answer is almost never. According to our survey, 35% don’t use disabled people to test their site, 25% don’t know if they do so, and only 3% can actually say that they do so “a lot” (see Graph 7).

Graph 7 - does your organisation used disabled people to test accessibility of your site?

This doesn’t bode well for organizations that actually want their site to be accessible, but it is in l ine with the level of testing done for other quality factors of Web sites. In short, we don’t put much effort into making sites accessible, and if we do, we rarely actually test our sites using people with the disabilities for which we are trying to cater.