6.1. Achievements of this Study

Whilst small in scale this study has echoed many aspects of earlier research, suggesting that:

• Users encounter problems with websites which may, or may not be related to compliance with accessibility guidelines.
• Websites which perform badly in accessibility tests may be poorly designed in many areas.

In addition, this research has suggested that:

• Most websites owned by SMEs do not comply with even elementary W3C recommendations.
• To many users, low accessibility makes very little difference to the completion of basic tasks on a website, but for a minority these tasks are extremely difficult or impossible.

Both user evaluation tasks were designed to be representative of what a typical user might want to achieve on any website and were not intended to be difficult, so these results should come as no surprise. However, the DRC warns against extrapolating beyond the data:

…not all the potential barriers will affect every user, as many relate to specific impairment groups, and a particular user may not explore the entire page…

DRC, 2004; p.24

Overall it would seem that visually–impaired users are more likely to use a range of assistive technologies and are more likely to struggle with tasks which should be easy for all users to complete. However, the results of this survey suggest that the difference is marginal and does not compare with the magnitude of the differences found by Nielsen and Pernice:

Our studies indicate that Web usability is about three times better for sighted users than for users with visual impairments.

Nielsen and Pernice, 2001; p.3

Previous work by Williams and Rattray considers the home page to be a good indicator of a site’s overall accessibility (Williams and Rattray, 2003; pp.713). This makes it reasonable to conclude that the findings of the first research phase are representative of the evaluated websites overall. But if anything, this study adds weight to the argument that automated tools are not to be relied upon. The ‘most accessible’ website as defined by automated evaluation also performed well in user evaluation, however a website with less overall violations was heavily criticised by users having only marginally missed out on AA–compliance. However, the contention that non–compliant sites would present problems to users was not borne out by the findings in the unequivocal fashion expected.

It would seem that for small businesses both website developers and commissioners have failed to embrace the recommendations made by the DRC cited in Section 2.4.1. The results of Phase One at the very least suggest a failure to consider the needs of disabled people in the context of best–practice. This is not surprising, as 50% of SMEs simply don’t see a business case for building accessibility into their websites (Higgs, 2006; Table 4.1.). In 2006 small businesses had never heard of PAS 78 (Higgs, 2006; Figure 4.5.) and from the results of this survey it could be suggested that three years later the situation remains much the same.