Freiraum 2023_Feedback based on Analytics of Teaching and Studying meets Individual Coaching (FANTASTIC)

Project background

Learning Analytics uses data traces from students in the learning management system to analyze the use of learning materials and individual learning behavior. It therefore offers the potential to AI-generated feedback to reduce failure rates and increase student success. success. However, this can only succeed if both the feedback is accepted by the feedback is accepted by students and the associated recommendations for action are implemented.

Since self-regulated learning is often not sufficiently developed in first-year students in particular learning is often not sufficiently developed, human support is required to change their learning behavior requires human support. This is where peer coaching comes into play: Students are qualified as learning guides who (i) are trained in relationship building, (ii) receive a variety of coaching and learning methods and (iii) use this and (iii) use this together with the AI feedback to individually support the learning process. support the learning process individually.

The coaching provides additional information about subject-specific and methodological difficulties, which used in turn to offer or further develop differentiated learning material. This addresses subject heterogeneity in STEM subjects and enables individual learning paths.

Project objective

In a pilot project, initial experience is currently being experience with learning analytics. Suitable data tracks (indicators) that students leave behind in the learning management system when using learning just-in-time teaching in the learning management system are being correlated with examination model with the examination results. The model is trained with trained and tested with data from previous cohorts and tested in academic year 23/24 in a physics course as an AI-generated feedback and recommendation tool.

In the FANTASTIC project

1. Further develop and improve the AI tool and the learning analytics model, including the feedback derived from it, will be feedback derived from it for an extension to other courses,
2. The existing qualification for tutors (i) modularized, (ii) expanded to include new skills for individual coaching and (iii) sustainably implemented at the university beyond the end of the project,
the data analyses as well as the experiences of the peer coaches and teachers into the development of differentiated learning learning materials for students with heterogeneous prior knowledge in order to to support weaker students in particular and counteract drop-outs.

The aim is to combine these three levels of levels of support into a triad, (i) to motivate as many students as possible to students to accept the AI-generated feedback, and (ii) to provide them with professional and methodically in the individual improvement of their learning behavior. If this is successful, then the students not only experience self-efficacy, but also benefit in the long term from the acquired ability of self-regulated learning, which is becoming increasingly important in the volatile in the volatile modern working world.

Project procedure

The following procedure is planned for the three support levels of FANTASTIC are planned:

1. Based on the experience gained in the pilot project and the results of accompanying student accompanying student surveys, the feedback tool will be iteratively improved. In addition, compatibility with similar courses in terms of subject area and teaching methods will be similar courses in terms of subject area and teaching methods used and the tool will be adapted if necessary.

2. The existing qualification for specialist tutors needs to be expanded to provide the tutors with the necessary knowledge in the field of AI and data protection. to give. A modular concept is being developed for this. This will be tested in the tested and evaluated in the first half of the project. Depending on its success, it will then be second half of the project and the aim is to anchor it structurally.

3. A close exchange between the student coaches, the project staff and the teachers and teaching staff provides valuable information on where and which technical difficulties occur. This information is used to prepare differentiated learning materials to enable individual learning paths.

Innovation

Learning Analytics is just picking up speed in Germany. The innovative thing about the FANTASTIC project is that a great deal of attention is being paid to how the AI-generated feedback is given and what further support support services are provided to ensure that it is well received by students well accepted by students and increase its effectiveness. FANTASTIC is intended to contribute to gaining valuable experience as to whether and how AI can can contribute to improving teaching.