Students with disabilities are less likely than other students to complete their studies, go on to complete higher degrees, or secure graduate employment (Mamiseishvili & Koch, 2012). This disparity is evident when large datasets are used to examine success and completion rates, segmenting findings using demographic filters. In such studies, many of which predate the emergence of learning analytics as a field, disability is one variable among many (Tinto, 1997).

Learning analytics, with its goals of ‘understanding and optimising learning and the environments in which it occurs’ (Long & Siemens, 2011, p34) offers the possibility of identifying and removing barriers to accessibility in learning environments. There are two elements to this work. First, it is important that learning analytics tools do not introduce new accessibility issues. Second, analytics should be designed to increase access to learning opportunities.

Accessibility is often thought of in terms of the guidelines set out by the W3C web standards body, and that web accessibility means ‘people with disabilities can perceive, understand, navigate, and interact with the Web, and that they can contribute to the Web’ (W3C, 2018). By extension, learning analytics tools and dashboards can be considered accessible if people with disabilities can perceive, understand, navigate, interact with them and contribute to them. One aspect of this work is technical – designers will take accessibility into account when adding buttons and visual elements, deciding on colours, contrast, fonts and font size. Another aspect requires more thought about usability; does the tool cater for learners who need extra time to make responses; reduce cognitive load as far as possible; and remove triggers of anxiety? (Lister et al, 2020)

General principles and standards may be applied to technical and usability elements. When it comes to increasing access to learning opportunities for people with disabilities, solutions must be developed in the field of learning analytics. Three strands of work indicate potential ways forward. Work presented at LAK16 indicated some of the ways in which analytics might be used to contribute to disabled students’ learning, initially by using large datasets to identify courses on which students with declared disabilities had significantly lower success rates than other students (Cooper et al, 2016).

Development of conversational user interfaces (chatbots) has highlighted how language – use of jargon, overuse of abbreviations, and the introduction of confusing terms – can all present barriers to learners with cognitive disabilities, mental health issues, or some types of learning difficulty (Lister, Coughlan, Iniesto, Freear, & Devine, 2020). Elsewhere, work on serious games has pointed to ways in which learning analytics could be used to provide support for people with intellectual disabilities, personalising learning pathways and flagging when key areas of content have not been accessed (Nguyen, Gardner, & Sheridan, 2018).