Web Accessibility Baseline Testing
Access to Web-based services has become an essential part of higher education. Providing equal access to all users is more than just an ideal goal at universities, it’s the law. The Americans with Disabilities Act of 1990 (ADA) requires equal access for all persons with disabilities. The Vocational Rehabilitation Act of 1973, as amended in 1998 in Section 508, extends that requirement to information technology resources at all institutions that receive federal funds, directly or indirectly.
Universities across the country now find themselves needing some form of baseline testing to demonstrate compliance with state and federal laws. As a researcher in Web accessibility at a mid-level university, baseline testing is the tool I used to measure and document compliance. Baseline testing, however, goes beyond basic record keeping – properly used, it is a proactive approach to identify websites in need of improvement, and then provide the means to track changes over time.
The Website Inventory: What’s in your Web presence?
The starting point for baseline testing is an inventory of an institution’s websites. In my research, the inventory includes the Uniform Resource Locator (URL), the website’s relationship to the university (i.e. department, division name) and host server name and location. Inactive websites are a problem for automated testing, so I eliminate URL redirects and abandoned websites in the inventory process. Every URL is then assigned a reference number and added to the database, which I found surprisingly diverse. Websites in my university’s Web presence include campus units, institutes and centers, auxiliaries and foundations, athletics, faculty and staff websites, student clubs and organizations, and third-party websites.
Accessibility Software Checkers: What’s your Accessibility Health Score?
The next step in baseline testing is to check compliance of the inventory using a software tool. An automated accessibility checker is a program or enterprise-wide service that can be configured to examine live web pages in a website and test accessibility against a predetermined set of guidelines. I recommend selecting an enterprise-wide service for this step, the advantage being the ability to automatically test a single page or all pages of a website. Accessibility software checkers behave similarly to browser search engines, examining all written content including hyperlinks, source code, attachments and digital content, and then presenting a score when finished. This score is referred to by some workers in the field as the Accessibility Health Score (AHS). Good software checkers allow the selection of individual or multiple rule sets, are flexible in programming, and run quickly and efficiently. All software checkers, however, work from checkpoints, and human judgment is required to validate results and intervene where necessary.
Prioritizing Scores: Factoring in Website Importance
Accessibility Health Scores are an indication of compliance, but do not identify websites in need of remediation. A low accessibility health score alone does not say how much a website is accessed or reveal the importance of the information contained within the website. Remediation should start with websites providing critical services and serving the highest number of students. For example, a website for Irish Folklore could have a high AHS, and yet only serve a small percentage of the student population (with the exception of Saint Patrick’s Day), whereas a website for learning services in support of English classes could have an identical AHS, but serve a larger percentage of the student population and, therefore, have higher priority for remediation.
To prioritize websites by use and demand and compare against accessibility health scores, analytics are required to identify an institution’s highly trafficked websites. The data in my research came from a study termed a student services lifecycle (SSL). The SSL is a campus study that identified and ranked campus services used by students from their time as prospective students seeking admission through their graduation. SSL studies consider variations in the mission of the institution, learner types, educational offerings and service delivery. Services students access every semester, such as registration, financial aid, the library and bookstore, are highest in demand and of greatest importance to the student population. How a university communicates with its students is also a factor within a SSL study. Communications through websites vary in many ways: email, electronic forms, registration, service portals for administrative tasks, virtual learning environments, and more traditional “brochure ware” websites that advertise and highlight a university’s programs. The SSL assists in identifying websites most critical to students, and provides the means to augment accessibility health scores to rank websites by need of remediation.
In incorporating the SSL study into a baseline report, each service in the SSL is given a score based on usage. Services are then correlated to the websites providing the service and the websites are assigned a ranking based on importance and demand. The rankings from the SSL are then combined with the accessibility health scores to produce an overall score, which at my institution I termed the Serviceable Accessibility Index (SAI).
The Final Report Card: Serviceable Accessibility Index
Websites with low SAI values reflect low accessibility health scores and high SSL ratings, meaning, they represent the poorest compliance in providing access to the disabled and have the greatest importance in the institution’s Web presence. SAI values clearly show the proportion of websites that are out of compliance, and form the key metric in Web accessibility baseline reports.
Federal regulations require universities to assure equal access for all who use their electronic services. Baseline testing is an essential part of this process, which is still evolving. By modifying standard AHS values with website ratings and analytics to produce SAI scores, baseline testing becomes a proactive approach to improving user access. Using this approach, institutions are able to plan and track their remediation efforts more efficiently, while at the same time documenting their compliance with federal regulations.
- California State University. (2009). Policy for the provision of accommodations and support services to students with disabilities.
- Brajnik, G. (2008). A comparative test of web accessibility evaluation methods. Paper presented at the Proceedings of the 10th international ACM SIGACCESS conference on computers and accessibility, Halifax, Nova Scotia, Canada. (.pdf)
- Harper, K., & Dewaters, J. (2008). A quest for website accessibility in higher education institutions. The Internet and Higher Education, 11(3-4), 160-164.
- Thompson, T., Burgstahler, S., & Comden, D. (2006). Research on web accessibility in higher education. Information Technology and Disabilities.
- Vigo, M., Arrue, M., Brajnik, G., Lomuscio, R., & Abascal, J. (2007). Quantitative metrics for measuring web accessibility. Paper presented at the Proceedings of the 2007 International Cross-Disciplinary Conference on Web Accessibility (W4A), Banff, Canada.
- U. S. Access Board. (2012b). Electronic and information technology accessibility standards (Section 508).
- U.S. Access Board. (2012c). The Rehabilitation Act Amendments of 1973, as amended.
Accessibility Image by Courtney Carmody; illustration by Jeff Stevens.