Retention in the Transition to Higher Education: Chat Bots and Student Mental Health

Mitchell, Alexander ORCID: https://orcid.org/0000-0001-5630-2620 and Scott, Michael ORCID: https://orcid.org/0000-0002-6803-1490 (2023) Retention in the Transition to Higher Education: Chat Bots and Student Mental Health. In: Proceedings of the SICSA Workshop on the Pedagogical Impacts of ChatGPT and Knowledge-Driven Large Language Models, August 24, 2023, Aberdeen, Scotland.

Preview

Image (detailed poster explaining the subject of the title) WoPICGPT 2023 Poster (Portrait).pdf - Accepted Version Available under License Creative Commons Attribution. Download (368kB) | Preview

Abstract / Summary

Mental health presents an obstacle to an increasing number of students. It is widely known that when people first join an institution of higher learning they encounter anxieties, uncertainties, frictions, and obstacles. Many such challenges go beyond the cognitive domain, into the affective domain, with far-reaching implications for mental health and well-being. However, the nature of the specific barriers different people face and the emotional journeys that they undertake are poorly understood; at least, in terms of realising valid and reliable statistical models with predictive utility. How can universities better identify and support students with their mental health and well-being in the early stages of their studies? This research leverages chatbots (integrating large language models) to not only signpost students to relevant support services but also to diagnose and alert stakeholders to high-risk cases. The data from such tools offers a new lens into this important question. This poster illustrates research-in-progress that uses clustering and logistic regression on queries submitted to the chatbot to determine the likelihood of retention, as students undertake their transition into higher education. Thus, highlighting risks to consider and laying the foundations for interventions that enact positive change and improve the success of new learners.

Item Type:	Conference or Workshop Item (Poster)
Subjects:	Education Technology
Courses by Department:	The Games Academy > Computing for Games
Depositing User:	Michael Scott
Date Deposited:	12 Sep 2023 14:18
Last Modified:	23 Nov 2023 19:40
URI:	https://falmouth-test.eprints-hosting.org/id/eprint/5074

Actions

View Item (login required)