Introduction

In July 2010 the online learning management system (LMS) Blackboard from Blackboard Inc was awarded the Nonvisual Accessibility Gold Certification by the National Federation of the Blind in the United States (Disabled World 2010). While it is laudable that this LMS, one of the most widely used in universities across the world, was acknowledged for its inclusive design, it also raises a number of uncomfortable questions. Blackboard was launched in 1997. Thirteen years is a long time to wait for an accessible version of the software. Blackboard continues to be the only LMS to have been accredited this level of certification. At a time when the possibilities of eLearning and online education are in the public spotlight through interest in the development of the massive open online course (MOOC), and at a time when enrolments in online courses are rising at a much higher rate than those in traditional face to face learning and teaching in higher education, it is disturbing to find this limited and belated approach to access for people with disabilities. eLearning holds many possibilities for inclusion for people with disabilities, however the online platforms utilized must provide access for all students.

In many cases, for students who study fully online, university staff may not meet them until their graduation ceremony. Roberts, Crittenden and Crittenden (2011) found that the majority of these students with disabilities chose not to disclose they had a disability. While this is one of the benefits afforded by studying online, they also found that these students did not request accommodations to help with access to course material that was presented in an inaccessible format. Students with disabilities can become invisible online. This means that more care and thought needs to be put into employing universal design practice in developing online learning material. As Jaeger (2012) notes "For persons with disabilities, unless technological design and implementation meaningfully focus on inclusion, the internet may become a new means of increased marginalization in society". Twenty-seven per cent of Americans live with a disability that interferes with activities of daily living (Fox 2011). This group of people is currently underrepresented in tertiary education (Sachs & Schreuer 2011; Wentz, Jaeger & Lazar 2011) although it is growing in number particularly with the increased use of eLearning serving to promote inclusion for this group of people (Fichten, Ferraro, Asuncion, Chwojka, Barile, Nguyen, Klomp & Wolforth 2009). As Jaeger (2012) points out 'disabled' is the only minority group that people may join over the course of their life. Only fifteen per cent of people with disabilities are born with them. All people should be seen as only temporarily able-bodied. The global proportion of people with disabilities in the population is rising due to both age and environmental factors (Vincente & López 2010). Inclusive design that facilitates access for people with disabilities helps everyone.

Given this, making eLearning accessible should be a priority for universities. This paper briefly explores the current rising rates of online learning in higher education. It examines how disability is activated differently online and the impact of this on learning and teaching through the internet and the accessibility of two of the most popular learning management systems, Blackboard and Moodle, and the different approaches, benefits and problems associated with each system. It then explores the eLearning environment beyond the structure of a LMS to a broader digital campus that includes social networks, video hosting sites and micro blogging, where students and staff are increasingly expanding the learning and social environment in higher education. It also questions the legal and moral responsibilities of universities to make all their online activities accessible to all students, regardless of disability.

eLearning

eLearning can involve students studying fully online and also a blend of online and face-to-face education. Teaching can involve the use of formal online learning management systems such as Blackboard, WebCT and Moodle, web-based lecture technologies, such as Lectopia or the massive open online course (MOOC) sites such as Coursera or edX. It can also take advantage of other less dedicated online platforms such as video hosting sites like YouTube and social networks like Twitter and Facebook. In 2012 more than one in three students in the United States (33.5%) were taking at least one online course and online enrolments were growing at a rate of 6.1 percent in an environment where overall enrolments were growing at a rate of only 1.2 percent (Allen & Seaman 2014).

There have been reports of resistance to learning and teaching online for both staff and students (Williams 2006; Woo, Gosper, McNeill, Preston, Green & Phillips 2008). Some, such as Stanford University president John Hennessy have predicted the decline or replacement of more traditional face-to-face learning and teaching in favor of courses moving online (Perry 2012). Others, such as Cary Nelson, the former president of the American Association of University Professors have, in response to the recent growth and interest in MOOCs, warned of the collapse of the university as an institution in the face of the transformations that this type of online learning herald (Walsh 2013).

While eLearning and particularly the adoption of MOOCs may threaten to disrupt the traditional way that universities have operated the adoption of eLearning presents a number of advantages. The practice allows greater flexibility for both students and staff (Heijstra & Rafnsdottir 2010). It also provides a platform to more effectively and cheaply distribute learning materials, especially to a geographically dispersed cohort of students. As Craig, Wozniak, Hyde and Burn (2009) observe, it can be used as a means to bypass overcrowded campuses. It can also potentially provide better learning outcomes. Gosper, Green, McNeil, Phillips, Preston and Woo (2008) observe that students find online technologies such as web-based lecture technology, help them to learn and achieve better results. Chen and Chiou (2012) report that students in blended learning environments outperform students who are only studying in a face-to-face context, and they feel a stronger sense of community. Birch and Williams (2011) similarly find students using online material perform better. Chamberlin and Lehmann (2011) point to the advantages that networks such as Twitter can offer in a higher education context, and Allen (2012) observes the potential advantages of Facebook in an educational context.

Fichten et al (2009) note that eLearning can promote inclusion for students who are unable to attend class and also for students with print impairments who can more easily access course notes and handouts made available digitally. This inclusion of people with disabilities is significant. Many of the technologies that are used as parts of online learning platforms have their origins in systems to promote the inclusion of students with disabilities. However not all of the platforms or their individual elements are inclusive.

Disability and the Internet

The social model of disability argues that disability is located in social practice rather than an individual body. A person may have a particular impairment, but it is the impact of decisions made by society that causes it to be a disability (see Oliver 1996; Finkelstein 1980). A person who uses a wheelchair may have a specific mobility impairment, but it is the lack of wheelchair ramps on a university campus that causes disability. However disability is activated differently online. Impairments that might encounter significant disabling environments in the analogue world, such as for a wheelchair user, may have less impact when using the internet. Other impairments such as print impairments related to vision, cognition, and manual dexterity and, increasingly, with the use of video and audio through the internet, people with hearing impairments may find different online environments can be significantly disabling (see Ellis & Kent 2011; Goggin & Newell 2003).

As Annable, Goggin and Stienstra (2007) question:

If there is much more acceptance of disability as a social, rather than purely medical, phenomenon, and greater public support for the removal of barriers and for an end to discrimination and exclusion of people with disabilities, why are information technologies — often the newest, most heralded ones — still disabling?

While accessibility and the internet is an ongoing struggle it can also provide significant opportunities for many people with disabilities. For people with many impairments, access to the internet is not a disabling experience. The network provides many opportunities for social interaction and options regarding disclosure that might not be available in an analogue environment interacting with people face to face. For people who have problems navigating the analogue world the internet provides many opportunities for work, leisure and commerce. As Dobransky and Hargittai (2006) note, access to digital communications technology can increases a sense of independence and self-determination for people with disabilities, and allow people to take advantage of online support without leaving their homes. Similarly Guo, Bricout and Huang (2005) observed that "the internet can remove barriers inherent in the physical environment and reduce discrimination towards disabled people." However, Dobransky and Hargittai (2006) caution:

The increasing spread of the Internet holds much potential for enhancing opportunities for people with disabilities. However, scarce evidence exists to suggest that people with disabilities are, in fact, participating in these new developments. Will the spread of information technologies (IT) increase equality by offering opportunities for people with disabilities? Or will a growing reliance on IT lead to more inequality by leaving behind certain portions of the population including people with disabilities?

For other people with disabilities the internet can prove a difficult environment to access. As the internet becomes more a part of everyday life, the impact of digital disability on people's lives increases. Increasingly being unable to access the internet can be seen as a form of disability in itself (Tănăsescu, Stegăriou & Păunescu 2010). Dobransky and Hargittai (2006) found that people with a disability are less likely to have access to information technology. Similarly Fox (2011) found in the United States people with disability are significantly less likely to use the internet.

Much of the disabling nature of the internet can be attributed to ignorance rather than deliberate discrimination. This is still wrong. Foley and Ferri (2012) state "We contend that technology should be conceived of as a global, accessible and inclusive concept, not one that requires a qualifier based on who it is for." However as Wentz, Jaeger and Lazar (2011) observe:

Digital information is not inherently accessible or inaccessible, but the choices made by those developing and implementing technology determine whether a technology ultimately will be accessible or inaccessible. This is particularly true in the online environment, given the rapid pace of technological change and introduction of new Web-enabled technologies, as online technologies are often obsolete before they are made accessible

Ellis and Kent (2011) outlined three stages of accessibility to online environments for people with disabilities. In the first instance an online platform or technology will be accessible, but not distributed widely. In the second stage it becomes more widely distributed, but as part of the redesign that leads to this popularity it often becomes no longer accessible. A third stage is reached where there is enough political pressure placed on the platform to make it accessible; in this last case designers/programmers retrofit access measures into it. Ellis and Kent argue that to avoid this cycle repeating with each new platform, a fourth stage must be reached where accessibility and universal design are built in at the earliest stages of development. While these stages will not fit to each and every situation they do map well onto eLearning over the past twenty years.

eLearning and Disability

In higher education in the United States and United Kingdom, the percentage of students with a disability is between eight and fourteen percent. This contrasts with eighteen percent in the working age population (Sachs & Schreuer 2011). In Australia the representation of students with a disability has been reported as low as four percent (Ellis 2011). eLearning has great potential to help both existing students with disabilities in their studies and also facilitate a more equitable representation of this group of people in higher education. However as Seale (2013) has observed "the relationship that disabled university students have with both their technologies and institutions is poorly understood." In order for this potential to be realized the eLearning platforms need to be as accessible as possible for students with a range of different impairments. Seale and Cooper (2010) describe accessibility in this context:

Broadly speaking, accessibility in relation to e-learning (e.g. virtual learning environments, digital repositories, multimedia, web portals and discussion boards) is understood as ensuring that learners are not prevented from accessing technologies or content and experience offered by technologies on the grounds of their disability.

As well as technical accessibility Sachs and Schreuer (2011) found that the attitude of faculty towards people with disabilities influences the success or failure of those students. It will also affect the likelihood of students, particularly those with invisible disabilities, from disclosing and requesting any necessary accommodation. As Seale (2013) observed non-disabled students are viewed in the context of what they can do with technology, whereas students with a disability are viewed in terms of what they cannot do. Sachs and Schreuer's study focused on on-campus students, once students are online the nature and number of what can be considered invisible disabilities grows. Roberts, Crittenden and Crittenden (2011) found that this unwillingness explicitly to disclose a disability and request accommodation also is a feature of students studying fully online. Guglielman (2010) also cautions that eLearning courses need to address both technical and pedagogical aspects of accessibility and inclusion.

The reduced access to information technology experienced by people with disabilities, noted above by Dobransky and Hargittai (2006), creates an initial barrier to this type of learning. Those people with access to technology then encounter a number of problems that have been documented by Fichten et al (2009). These include the accessibility of websites and learning management systems, the accessibility of digital audio and video content and alternatives, inflexible time limits built into online exams, the accessibility of PowerPoint presentations, and also course material in inaccessible PDF formats and the lack of access to needed adaptive technologies. Van de Bunt-Kokhuis and Bolger (2009) also highlighted problems with the inaccessibility of online chat rooms, and particularly the incompatibility of screen readers with these forums for students with vision impairments. Kelly (2009) found that almost one-third of students who used assistive technology to access online educational material found that this material was unreliable or inconsistent if it could be accessed at all.

Many online courses are not designed with accessibility in mind (Roberts, Crittenden & Crittenden 2011). This means that students who do not disclose that they have a disability maybe disadvantaged. It also means that when students do request accommodation to access the learning environment it requires a process of design-redesign to accommodate the students, adding addtional costs. Roberts, Crittenden and Crittenden (2011) suggest that courses should be designed to be accessible from the beginning. Implementing universal design principles at the outset avoids costs caused by the need to engage in a digital retrofit, and serves to include those students who would otherwise be excluded by an unwillingness to request accommodations. Following Ellis and Kent (2011) disability access needs to be built into the design process at the beginning, not retrofitted. These technical moves to provide access will need to be done in conjunction with a pedagogical approach to course design that is inclusive for people with disabilities and the overcoming of ableist bias or discrimination on the part of teaching staff.

Accessible eLearning provides a number of affordances to students and also teaching staff with disabilities. For many there potentially are more choices in terms of disclosure for a variety of impairments. The reluctance of students to disclose that they have a disability, even when necessary to seek access to inaccessible content found by Sachs and Schreuer (2011) and Roberts, Crittenden and Crittenden (2011) points to the value this choice has for many students with disability. Learning material that is made available online can have far more options for accessibility than analogue content, electronic text can be read aloud and translated to braille, audio files can be electronically transcribed as text. Finally the eLearning environment can provide students with a much greater degree of flexibility, lectures can be listened to when circumstances allow rather than at a set time and place. Tutorial discussions take place on asynchronous discussion boards again free of the time and place restrictions of their analogue counterparts. These three features of disclosure, accessibility, and flexibility provide great potential for people with disabilities to engage in a higher education learning environment. However these benefits are not just limited to students with a disability, accommodations made for accessible course design help all students. Technology such as web-based lecture systems are valuable for both students with disabilities and the broader student population (Williams & Fardon 2005). Text made available as an audio file can be listened to in different settings. Subtitles can be used to read the content of a video presentation when sound is not appropriate. Information that is less fixed to a specific format can be accessed in multiple ways and is more easily searchable.

Disability and Learning Management Systems

While these potential benefits, and impediments, for students with disabilities can be applied in a variety of different eLearning contexts most universities make use of some form of formal learning management system to facilitate both blended and fully online learning and teaching. In 2009 Cooper and Heath critiqued learning management systems for adopting a one-size-fits-all approach to accessible eLearning through adopting a compliance approach to the initial Web Content Accessibility Guidelines of 1999 (WCAG 1.0) from the World Wide Web consortium (W3C). While these have been updated in 2009 to WCAG 2.0 there is still a tendency to see accessibility as an afterthought or a potential legal liability to overcome. There are two different approaches to accessibility that are linked to understanding of online ownership and design that can be contrasted by looking at two of the most widely used learning management systems, Moodle and Blackboard.

Moodle, originally an abbreviation of Modular Object-Orientated Dynamic Learning Environment was developed as an open source learning platform by Martin Dougiamas at Curtin University. Moodle was first released in August 2002, and is an open source-project that is freely available. Individual users are invited to modify and add modules to the system, but on the proviso that these additions and changes are similarly made freely available to others. Moodle currently has nearly seventy million users involved in studying seven and a half million separate courses (Moodle 2015). Moodle should be accessible to the majority of assistive technology, at least at the level of its HTML code and Elias (2010) has observed that there are many accessibility modules available to be used with individual course design. However in 2012 Calvo, Iglesias and Moreno found that Moodle, following the W3C guidelines, was not accessible to people with visual impairments. Elias (2010) also found that while there are many accessibility modules available, they are rarely used.

Part of the problem related to accessibility through Moodle is that as each person designing a course sets out their own Moodle web interface. Often operating without an existing accessible template these are then constructed in a way that prevents assistive technology such as screen readers being able practically to navigate the page. According to Asakawa, (2005) this technical accessibility aspect was a common feature of the Web following an update to Section 508 of the United States Rehabilitation Act in 1998 and the subsequent dot com bubble when, Cooper and Heath (2009) note, designers would pay attention to the letter of the law as set by WCAG 1.0, rather than its intent. Thus alternative text is provided even for inconsequential images, such as a shadow behind an actual picture that adds to the visual aesthetic of a web page, but as alternative text merely clutters the page unnecessarily for a person who is accessing it through a screen reader. This can be replicated in a Moodle site where there are a large number of unnecessary elements that, while technically accessible, also clutter navigations and descriptions of the page used by screen readers and other assistive technology. In the case of contemporary Moodle sites this poor design is a consequence of a lack of awareness of the course designers of good accessible design principles.

Blackboard is one of the first web based learning management systems, having been launched in 1997. The system currently has about half of the market in these systems in higher education (Edutechnica 2014). In 2010, the system was awarded its Gold-level certification from the National Federation of the Blind. As Jaeger (2012) describes the system's level of accessibility it "was primarily inaccessible when it was launched and only became disability-friendly ten years later." This belated conversion to an accessible platform was achieved as a function of Blackboard's top-down development. The company acts as a central source for its system and as such is in a good position to be able to retrofit its accessibility. While this ability to retrofit has now made this platform more accessible, it also precludes other parties developing accessibility modules in a similar way to Moodle.

Both these learning management systems now have the potential to be able to provide a more accessible platform. However in both cases, even with an accessible basis for the LMS, the course content that is hosted through these sites may have its own accessibility issues as outlined by Fichten et al (2009) above. As Guglielman (2010) cautions, "disabled students can access to the e-learning platform but not to contents, resources, activities, collaboration and interaction tools." This problem, in part, as Keller (2010) acknowledges, stems from challenges to making university material accessible due to growth in complexity of digital media in recent years.

Disability and the Digital Campus

Stienstra, Watzke and Birch observed in 2007 that "The concept of the LMS has not evolved sufficiently to keep pace with the changing landscape of academic technology, especially with modes of interaction and collaboration fostered by popular online social networks like Facebook and Twitter." Increasingly, university learning, teaching, and student contact occurs though social networks including Facebook, Twitter, Skype, YouTube and other online networks and web 2.0 applications (see Baran 2010; Kent 2014; Tay & Allen 2011). Dabbagh and Kitsantas (2012) have cautioned that "higher education institutions are still primarily relying on traditional platforms such as course and learning management systems (CMS/LMS) that do not capitalize on the pedagogical affordances of social media." Similarly Wodzicki, Schwämmlein and Moskaliuk (2012) observe that, "Social media open up multiple options to add a new dimension to learning and knowledge processes. Particularly, social networking sites allow students to connect formal and informal learning settings."

These online platforms and tools are not just used in the context of learning and teaching. Students, both on-campus and fully online also make use of social networks and web 2.0 tools as a venue to socialize, a place for informal learning and community-building. The online university campus mirrors the role of its analogue counterpart as a place for both learning and teaching and for students to socialize and access support (Ellis & Kent 2014; Leaver 2014). However these online networks also have accessibility problems for people with disabilities. Boudreau (2011) explored the accessibility levels of Twitter, Facebook, Linked in, YouTube and Google+. These networks were assessed for a number of accessibility features including section headings, color contrasts, labels and form fields, keyboard navigation, text equivalents for images, multimedia, language, and validation. None of the networks met even thirty percent of the accessibility criteria set, as Boudreau questioned "could they do worse if they tried."

While there are clearly many accessibility issues associated with this broader digital campus beyond the online classrooms there are also as Hollier (2012) observes many work-arounds being used by the disability community to access many of these sites. Knowledge of, and literacy with, these sites can allow student with disabilities, and other students to be directed to more accessible options, particularly when these are being deployed in formal learning and teaching. However what is the role of educators and institutions in relation to these online spaces that occur outside the university login? Who is responsible for accessibility in these informal settings, both legally and morally?

Legal Responsibilities

Goggin and Newell (2003) observed how important the role played by legal requirements is in providing accessibility for people with disabilities to digital technology. However as Jaeger (2012) has noted "the conceptions of disability under the law, exemptions from compliance, limited enforcement, and the ability of the law to keep pace with technological development all hinder the impact that the laws have had thus far." Wentz, Jaeger and Lazar (2011) also note the nature of disability laws, particularly in the United States, differ from other civil rights legislation. To activate the legislation a person has to prove both that they are disabled and are being discriminated against. It is up to the individual being discriminated against to prosecute the law. This means in practice that these laws have been harder to prosecute than those relating to other civil rights. To force the application of a relevant law a person who is blind would, in many cases, have to launch a complaint against an inaccessible website that they are, by definition, unable to experience (see Worthington 2000). However, while this tends to delay the implementation of anti-discrimination laws against people with disabilities, it does not stop them. The relatively recent accessibility of Blackboard is a reflection of the growing awareness of universities to their responsibilities to provide an accessible environment both on an analogue campus and online for people with disabilities.

In 2010, a coalition of disability rights organizations successfully sued a group of universities who were trying to introduce Amazon's Kindle eBook reader for textbooks. The Kindle actually had accessibility features such as speech to text, but they were disabled, and when they were later activated again, the controls to activate them were not accessible (Foley & Ferri 2012). The United States Departments of Justice and Education issued a joint statement on the case:

As officials of the agencies charged with enforcement and interpretation of the ADA and Section 504, we ask that you take steps to ensure that your college or university refrains from requiring the use of any electronic book reader, or other similar technology, in a teaching or classroom environment as long as the device remains inaccessible to individuals who are blind or have low vision. It is unacceptable for universities to use emerging technology without insisting that this technology be accessible to all students. - Perez and Assistant (2010)

This raises important questions about what level of responsibility the university has. Clearly in this case issues related to educational resources are important. However, prominent institutions such as the California State University system, that has in the past successfully pressed large organizations, such as Apple to make its iTunes U academic network more accessible, admits that it is not compliant with its legal accessibility responsibilities (Keller 2010). An institution's legal responsibilities in relation to the extended digital campus are yet to be explored. As Ingeno (2013) observes "the line between what is and isn't discriminatory is often blurred in an online setting." As learning and teaching increasingly migrates beyond the closed login of the learning management system these issues will become further interconnected.

Conclusions

eLearning is a growing area of the higher education landscape. This teaching practice holds great potential to be an avenue of inclusion for people with disabilities in that context. However, this potential is endangered by the relative inaccessibility of the online environment that is currently used both in terms of formal learning management systems and also the other social and web 2.0 tools that are used in conjunction with these systems. Pioneers of eLearning in pre- and early-web internet history would present information in a necessarily simple interface that was accessible to many of the assistive technologies available for people with disabilities to use online. While these online offerings were accessible they were not widely used. When Blackboard was launched in 1997 eLearning began to become far more widely distributed, but it used an inaccessible platform. In 2010, Blackboard receiving its Nonvisual Accessibility Gold Certification seemed to herald learning management systems moving from Ellis and Kent's (2011) widely distributed but inaccessible phase to one of being retrofitted for accessibility.

How these questions of accessibility will play out, when looking at the broader environment of the digital campus, is still being determined. Like the different approaches to accessibility in LMSs the way that different platforms and networks are structured will influence how they can be adapted for greater accessibility. In 2009, having previously been highly inaccessible, Facebook was, like Blackboard, retrofitted to make the social network easier to access (Hollier 2012). The same was not done at the then rival MySpace, but given the way individual users were encouraged to customise their own pages, implementing any change would have been more of a challenge. Twitter, while having significant accessibility issues through its main interface, could be accessed through the Easy Chirp site, which has been developed by a third party as an accessibility work-around (Ellis & Kent 2010), analogous with the process that leads to the accessibility modules developed for Moodle.

Access to higher education and equality of access for people with disabilities is an important moral obligation for universities. Beyond this, in many countries, including the United States, it is also a legal requirement. Although as Seale and Cooper (2010) note while many teachers in higher education understand that eLearning should be accessible, not all are aware how to make it accessible. This problem is exacerbated by the inherent difficulties in prosecuting civil rights legislation when it comes to people with disabilities. In this context, it is important to observe, as Stienstra, Watzke and Birch (2007) do, the important role played of disability advocates, government and industry in driving moves towards greater accessibility. The work of the American Foundation for the Blind and others in preventing the deployment of an inaccessible system in the Kindle and the work of the California State University system in lobbying for Apple's iTunes U service to be more accessibility-friendly (Ellis & Kent 2014) are both potent examples of this process in this context.

The need for those identified by Stienstra, Watzke and Birch (2007) to push for greater inclusion comes from the need to retrofit inaccessible online design. Both Roberts, Crittenden and Crittenden (2011) and Ellis and Kent (2011) have argued for the advantages that can be had, and pitfalls that can be avoided, if courses are designed to be accessible from the beginning. This is important for both the formal online spaces of learning management systems, but also for the broader internet environment used in higher education for learning and teaching. Sadly there are limited signs at this time that Ellis and Kent's fourth stage of digital accessibility, especially in terms of eLearning and educational design is about to be reached. But there are signs of hope, Denise Wallace, vice-president of legal affairs at Dillard University states, of the university's ambition "The goal is not to wait until someone comes and self-identifies; the goal for equal access is to make it accessible from the beginning" (Ingeno 2013). While this is still just a goal and driven, in this case, by a response to the Americans with Disabilities Act, as much as a sense of social justice, this still points favorably to a future where eLearning is more an opportunity and less a barrier for people with disabilities.