Digitally supported learning: Self-regulated learning in higher education

Video design: Carolin Hummel 

At the beginning of their studies, students are not well prepared to deepen their knowledge and to cope with large amounts of learning materials for examinations in self-regulated ways. We therefore offer a digital tutoring system (“CoTutor”) in a large first semester lecture (for students in a university teacher education program).

On the one hand, we want to improve our course offering and build up important study skills in the first semester. The student group is particularly heterogeneous. On the other hand, we are researching the use and learning effectiveness of this digital tutoring service in a real learning context. We are also working with the developers of the software and implementing functions based on learning psychology principles and findings. With the digital learning process data from the system, we can make learning visible.

As we also observe less sensible uses (e.g. massive learning shortly before the exam, obvious guessing and repetition), we provide detailed information on appropriate strategies for self-regulated learning.

Our focus is on supporting students in deepening their understanding and preparing for exams. Deepening understanding means that students link content, comprehend arguments, evaluate evidence, find examples and create application references when learning and working through the material. This type of development also goes beyond what is written in the textbook chapter. Students are expected to do this actively themselves, but are often unable to do so due to a lack of strategy and prior knowledge. For this reason, arguments, evidence, applications and links (including incorrect ones) appear as alternative answers in multiple-choice questions, which students are asked to think about. If the answer is incorrect, a detailed explanatory text appears as feedback and a learning opportunity. Exam preparation means that students are supported in reliably retrieving content from their long-term memory. Based on their own learning history, questions are therefore drawn again and answered repeatedly (“retrieval practice”).

Of particular interest to us is the actual, learning-effective use of the system by the students – using digital records of when, to what extent and with what success learning questions could be answered repeatedly in the system over the entire semester and the exam preparation period. Usage of the system can be monitored with digital learning process data if the data can be interpreted and are valid. In all years, learning success (exam performance) and learning prerequisites (grade point average of final school examinations, “Abitur” grade in particular) are linked to the digital learning process data. We investigate relations between individual prerequisites, use and impact of digital tutoring, making learning visible.

Further research relates to metacognitive assessments during learning (so-called judgments of learning and perceived mental effort) and their validity and effect on later exam success. In addition, we are introducing changes in the tutoring system (e.g. introduction of free texts to stimulate more generative learning activities; introduction of superficially modified answer alternatives to counter guessing and recognition) and investigating their effect. We also carry out experiments to teach elaborative learning techniques.

Corresponding researchers: Prof. Dr. Stefan Münzer, Samuel Wissel, Hatice Dedetas

Wissel, Janson, Ingendahl, Undorf & Münzer (under review). Item-by-Item Judgments of Learning Predict Learning Success in Higher Education.

Janson, Wissel, Schäfer & Undorf (under review). Judgments of learning in the wild – Establishing Ecological Validity with an Intelligent Tutoring System in a Field Study