Web Watch: An Accessibility Analysis of UK University Entry Points

The Special Educational Needs and Disability Act (SENDA) came into effect on 1 ^st September 2002. The Act removes the previous exemption of education from the Disability Discrimination Act (1995), ensuring that discrimination against disabled students will be unlawful. Institutions will incur additional responsibilities in 2003, with the final sections of legislation coming into effect in 2005 [1].

The implications of the Act will be of much interest to institutional Web managers who will be concerned that inaccessible Web pages will render their institution liable to claims from disabled students who are unable to access resources due to accessibility barriers.

Although the SENDA legislation covers much more than Web resources, Web developers are fortunate that the World Wide Web Consortium (W3C) has, for several years, been active in addressing the accessibility of Web resources through the work of the W3C Web Accessibility Initiative (WAI) [2]. Work carried out by the WAI includes the development of a set of guidelines for authors of Web pages. A number of organisations have developed tools which with help with the analysis of the accessibility of Web pages, based on the WAI guidelines.

In this WebWatch survey the Bobby accessibility checker [3] is used to report on the accessibility of the entry points to UK University Web sites.

About The Survey

The survey was carried out using the Web-based Bobby tool to analyse the accessibility of entry points of UK University Web sites. The results of the survey were recorded, including information on compliance with Bobby's A and AA guidelines and the numbers of Priority 1 (P1) and Priority 2 (P2) errors.

It should be noted that Bobby is only one of a range of accessibility testing tools. Bobby is used in this survey as it is probably the most well-known. A list of other accessibility testing tools is available on the W3C WAI Web site [4].

The Findings

Only four entry points appeared to provide AA compliance by having no Priority 1 or Priority 2 errors. These were the University of Bristol, the University of East London, the University of Sheffield and Writtle College.

Although the number of entry points which provide AA compliance is disappointing it is pleasing to note that seventy entry points appeared to provide A compliance by having no Priority 1 errors.

Discussion

Methodology

This survey addressed only accessibility aspects which could be detected by use of software. A more thorough accessibility audit would require use of manual testing. However this would be time-consuming to carry out.

However it can be said that although the survey could class pages as accessible which manual testing could prove to be inaccessible, the main conclusion of the survey is that the vast majority of pages have been found to be inaccessible to some extent.

It should be noted that a number of entry points contained frames or redirects to other pages. In a number of cases this meant that further analysis would be required in order to measure the accessibility of the page. Where redirects were implemented, the accessibility of the destination of the redirect was analysed. However in this survey elements of framed pages were not analysed.

Main Findings

A number of errors were caused by missing ALT attributes in IMG tags. The use of ALT attributes is important for access by users with speaking browsers or with images switched off (which may be the case for Web sites stored on devices with limited memory capacity, such as PDAs. As can be seen from Figure 1, the use of meaningful ALT attributes can allow a Web site to be used effectively (note click on the image to see a larger version).

-
Figure 1: University of Bristol Entry Point With And Without Images

For some of the errors reported by Bobby it may be more difficult to understand the accessibility problems. In order to highlight a number of accessibility problems which Bobby can find, we will review the errors found on entry points with only a few errors, as listed in the following table.

Although it appeared that Writtle College had no P1 or P2 errors subsequent analysis revealed that the HTML tages were generated by JavaScript. Since the Bobby tool does not process JavaScript, its findings were based on the page containing the JavaScript source code and not the output of the JavaScript.

The findings for the University of Wales Institute, Cardiff (UWIC) reported a single occurrence of one P2 error:

Use a public text identifier in a DOCTYPE statement

This error indicates that the document does not contains a DOCTYPE at the start of the document of the form:

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">

The DOCTYPE is needed in order for the document to comply with HTML standards. If a document does not comply with HTML standards, although it may look fine in most browsers it may be difficult for conversion tools to convert the document to alternative formats, including accessible formats for use by specialist devices.

The findings for the University of Wales Institute, Cardiff (UWIC) also illustrate a failing of automated tools. Although this would appear to have a high degree of accessibility, manual analysis reveals that the page contains a single image. This page is, to a certain extent, inaccessible. Although the page contains textual information requesting users to click for the next screen the quotation from the Sunday Times and the Award For Excellence are not available to users with text browsers.

The findings for the University of Central Lancashire reported two P2 errors: the missing DOCTYPE, as discussed above, and the need to:

Use relative sizing and positioning (% values) rather than absolute (pixels).

Absolute sizes make it difficult for users to resize elements on the screen, such as the size of text.

It should be noted that the entry point for the University of Central Lancashire Web site contained an image-based interface with minimal use of ALT attribute. However the page did provide access to a text-based Web site which was generated dynamically using the Betsie software [5].

The findings for the London School of Economics also showed one P2 errors: four instances of the need to use relative sizing. No fewer than 72 instances of this error were found on the Newman College Web site.

The Northern School of Contemporary Dance appeared to be missing only the DOCTYPE statement. However the entry point in fact contains a Shockwave page and users with a Shockwave plugin cannot access the Web site.

The entry points for the remaining Web sites listed in the Table contained a small number of instances of errors which have been mentioned previously, in particular the need to use relative sizing and positioning.

From these examples it can be seen that, in a number of cases, pages which appear to have a high degree of accessibility are, in fact, inaccessible. On the other hand pages which contain certain errors, such as a missing DOCTYPE, can be made to comply with accessibility guidelines by making relatively simple changes.

Other errors, such as replacing absolute positioning by relative positioning, may be subject to more debate with Web teams. This change, for example, is likely to alter the appearance of the page. However this change will ensure greater flexibility for the page. If an institution wishes to provide maximum flexibility it will be necessary to make such a change.

Conclusions

Many institutions will be aware of the SENDA legislation and of the work of the W3C WAI. Institutions will, I am sure, wish to minimise accessibility barriers to their Web sites. However the findings of this survey of entry points for institutions, the most important page on a Web site, show that there is clearly still much work to be done.

Appendix 1: Results

A summary of the findings is given in the following table.

The information in the table was initially collected on 7^th August 2002. On this date the AA compliance data was obtained. The survey was repeated on 2^nd September 2002, when A compliance data was obtained.

Note that a number of the entry points were found to contain a redirect. In cases in which the address of the redirected page was displayed in the browser the new location was analysed by Bobby.

References

1. The Special Educational Needs and Disability Act, Sophie Corlett, Skill
   http://www.techdis.ac.uk/resources/skill01.html
2. Web Accessibility Initiative (WAI), W3C
   http://www.w3c.org/WAI/
3. Bobby, Watchfire
   http://bobby.watchfire.com/
4. Evaluation, Repair, and Transformation Tools for Web Content Accessibility, W3C
   http://www.w3.org/WAI/ER/existingtools.html
5. Betsie, BBC
   http://betsie.sourceforge.net/

Author Details

Brian Kelly
UK Web Focus
UKOLN
University of Bath
Bath
BA2 7AY

Email: b.kelly@ukoln.ac.uk

Brian Kelly is UK Web Focus. He works for UKOLN, which is based at the University of Bath