Centre for Scholarship and Innovation
We propose the development of OELAssist, a comprehensive system designed to revolutionise the student experience within the Open Engineering Lab (OEL) by addressing challenges faced by students during experiments. The primary issue we intend to investigate, and resolve is the occurrence of difficulties that hinder successful experiment completion, leading to attainment gaps and potential attrition.
The primary objectives of this project are to reduce attainment gaps, increase student retention, and promote progression for students engaging with OEL modules.
Our approach consists of two integral stages. In the first stage, we will collect and analyse data on student interactions with an experiment to identify when and where problems arise. Leveraging machine learning techniques, we will create sequential rules to establish a baseline for successful experiment completion. This initial phase is crucial for understanding student behaviour and challenges.
In the second stage, we will develop a system to detect issues in real-time and provide timely feedback to students. This feedback will not only indicate where students went off course but also offer remedial advice, supplemented with relevant theory and instructional materials. To achieve this, we will employ generative AI models, tailored to the course content and domain knowledge.
Anticipated outcomes include a reduction in attainment gaps, increased student retention, and improved progression rates for those engaging with OEL modules. By offering personalised support, we expect to enhance the learning experience, making it more accessible and rewarding for all students.
The impact of OELAssist on staff and students is significant. For students, it means having access to a robust support system that can identify and address their difficulties promptly. This can boost their confidence, motivation, and ultimately, their success in engineering experiments. For staff, the project offers insights into student challenges and an automated means to provide tailored assistance, thereby improving teaching effectiveness.
The OELAssist project aims to revolutionise the OEL student experience by leveraging data-driven insights and generative AI to address difficulties faced during experiments. The anticipated impact includes increased student success, reduced attainment gaps, and enhanced teaching effectiveness, aligning with eSTEeM's goals of improving STEM education.