Analysing Disability Descriptions and Student Suggestions as a Foundation to Overcome Barriers to Learning

Abstract

Artificial intelligence can support increasingly complex conversational interactions and also has the potential to interpret meanings from free text input and make recommendations based on patterns in data. There are important opportunities to apply this to real-world problems faced in access to education. In this paper, we summarise existing research and trends linking disability support and technology, then report on a survey conducted with students with disclosed disabilities (n = 138) to explore what systems might need to do to effectively understand disabled students in their own words, and provide suggestions of technologies, strategies and resources that could be relevant to overcoming barriers to learning. Through thematic analysis, five approaches that students used to talk about their disabilities are identified (medical, functional, support, experiential and administrative), and three major types of suggestions they make around what supported them and may be useful to others are also identified (external tools, university support and practices, concerns and solutions). The survey approach and the findings of the analysis provide a potential foundation for the effective design of systems that could crowdsource a knowledge base around disabilities, hold conversations to understand disabilities and barriers and make relevant recommendations to individuals as to how they could overcome these barriers.