Many universities lack the resources to provide all students with equal access to practical laboratory work. Online laboratories offer one way to address this challenge. Significant advances in AI technologies in recent years, particularly in natural language processing (NLP) tools like ChatGPT, have created opportunities to meaningfully integrate these technologies and enhance the educational experience of online labs. This conference submission presents work conducted within the KICK 4.0 project, which focuses on online laboratories in engineering education. The project also addresses new competency requirements for students that are becoming increasingly relevant with the growing use of NLP technologies.
This work-in-progress submission explores our approach to making NLP technologies usable in online labs in higher engineering education. Our goal is to identify the potential and limitations of NLP technologies for laboratory-based instruction from both the instructor’s and the learner’s perspectives. We recognize that NLP technologies are still relatively new to both lecturers and students, and their successful use depends on how well users are trained to leverage the technology’s capabilities. Our aim is to help both groups significantly improve their skills in using NLP technologies, particularly in the context of online lab instruction.
To achieve this, we first conducted a detailed requirements analysis in collaboration with engineering instructors and students at our university. The findings of this analysis are discussed in detail in this submission and connected to existing research. Preliminary results show that the acceptance of NLP technologies depends largely on the quality of the feedback and on the perception of the results’ quality. It is therefore important to generate high-quality feedback, and the user plays a significant role in shaping the results by posing the right questions or priming the system accordingly. For students, the primary focus is on the content and language of the answers, while for teachers, the focus is more on the transparency of how the feedback is generated and the reliability of those answers.
Based on our insights, we developed an evaluation framework and established criteria to assess the practicality and effectiveness of the tested NLP systems in our instructional context. Our instructional intervention was applied in an undergraduate thermodynamics lab course, and the introduction of NLP was evaluated accordingly. The results of our work will highlight instructional design factors that positively influence both teachers’ and students’ ability to effectively use NLP technologies in lab courses. Additionally, we will discuss factors that may limit the successful integration of NLP and identify areas for potential improvement.
The full paper will be available to logged in and registered conference attendees once the conference starts on June 22, 2025, and to all visitors after the conference ends on June 25, 2025