Investigating the Accessibility and Usability of Job Application Web Sites for Blind Users
Jonathan Lazar, Abiodun Olalere, and Brian Wentz
Journal of Usability Studies, Volume 7, Issue 2, February 2012, pp. 68 - 87
Article Contents
Recommendations
There were a number of usability problems on the employment application Web sites that were problematic for the blind participants in this usability study and kept the participants from independently submitting applications online. However, none of these usability problems were ones that were technically hard to solve or address. These were all commonly-known and understood problems, relating both to accessibility for blind users and general usability for all users. The solutions themselves are easy—such as creating textual equivalents for clickable image maps, accessible feedback for form errors, and clearly stating which fields are required and which data format should be used. For instance, if any of these employment application Web sites followed either Section 508 (Section508.gov, 1998) or the Web Content Accessibility Guidelines (W3C, 2011b), it is likely that most of the accessibility problems mentioned previously in the paper would have been addressed. Companies should ensure that their online employment processes are accessible and usable for users with disabilities.
If online employment application software is being purchased (such as solutions from Kenexa or Taleo) employers should request documentation that the software complies with Section 508, similar laws in other countries, or international standards. This can be done by asking for documentation of what methods were used to check for accessibility, or asking for a Voluntary Product Accessibility Template® (VPAT®) that documents the accessibility features (http://www.itic.org/index.php?src=gendocs&ref=vpat&category=resources). While it is possible that users with disabilities would face challenges in using the interface that are not covered under Section 508, the most basic accessibility problems documented in this study would have indeed been covered under Section 508 or similar laws.
If online employment application software is being developed or modified in-house, good user-centered design techniques should be used to ensure accessibility. These techniques include usability testing involving people with disabilities, expert inspections using assistive technology, and automated accessibility testing (software such as HiSoftware Compliance Sheriff, Odellus ComplyFirst, and Deque Worldspace). In addition, if the online employment process Web pages are going to be modified in any way, accessibility needs to be considered in the modifications.
Even though there is additional expense and time involved with user testing, we believe that it is important to have real users with disabilities test Web sites. We uncovered the following usability problems that would likely not be detected by automated software tools:
- Many of the usability challenges we consider to be serious for the blind participants in our study, such as no clear identification of when fields are required fields, free-text being preferred to look-up, and unclear data format preference, would definitely not be detected.
- While it’s likely that the inaccessible maps with no textual equivalent would be flagged, it’s unlikely that the inaccessible feedback when users entered incorrect information would be detected.
- User actions, such as entering incorrect information, were required before the inaccessible feedback was triggered. Without the actions, the feedback would not be evaluated.
- Data fields that do not make logical sense, such as requiring an “end date” to a job those participants marked as their current job, would not be detected.
Automated accessibility testing tools are necessary for evaluating and monitoring any large Web site, as there may be thousands of sub-sites and pages; however, those tools are not a replacement for user testing, especially when users with disabilities must perform tasks that involve a series of sub-tasks across multiple screens. User-based testing provides a much deeper understanding of accessibility and usability.
It is important to note that these participant attempts to submit applications were only the first step in the process of applying for a job. The entire process, once the individual submits the application, must also be accessible. If these applications were real applications (and not marked with “for training purposes only”), and if these applicants were chosen for interviews and further review, those future steps would also need to be accessible. For instance, there are reports of many employers requiring potential employees to take online aptitude tests. Are these online tests accessible? Are follow-up communications electronic? If so, are they accessible? And furthermore (and non-technically), when potential employees go for an interview, are those face-to-face meetings in accessible locations? Do the offices and buildings have Braille signage? This usability evaluation has only examined the initial attempts to submit an employment application online. Future work needs to evaluate the accessibility of the entire process.