Initially, instructors and learning researchers jointly analyse individual students‘ performance along three dimensions: questionnaires addressing students metacognitive and motivational status are evaluated together with cognitive results from weekly homeworks. This analysis then allows to assign students to different groups of different competency levels using relatively simple rules. Thereby, statistical analysis is used to show the most reliable differences between groups of students, e.g. between different cohorts or just during the course of a semester.
Optimizing tests and analysis will then allow to find more and more sophisticated differences between different groups and even individuals along the three dimensions. These differences, which can be related to deficits in problem solving, can then be addressed by offering tailormade tasks. At the same time statistical analysis of student performance is also used to determine the effects of learning support implemented during an academic semester.
Aimed at identifying individual support needs, initial assessments and questionnaires are the starting point of personalized support, as conceived by the project researchers. Consequently, first different groups of students would receive variants of homework tasks focusing on specific knowledge gaps and skill development needs. The project scientists construct variants by using, for instance, contextually modified task definitions, additional hints to solution approaches, background information, or preparatory tasks of a narrower scope that help practice particular skills.
With time, support will be faded out to the extent that competencies develop, strengthening a student’s competence for self-regulated learning and competence development. Electronic implementation focuses on Moodle with specific add-ons or chatbot apps allowing intuitive DeALS-access for students and instructors.
The plan for the future is to generalize the DeALS-Phys approach to other disciplines. Whilst researchers apply the DeALS-Phys approach to physics and a diverse student body in this project, results and ideas will also be available for transfer to other disciplines or universities. To facilitate dissemination, researchers will mainly use a standard cloud-based platform (Moodle). The learning support addresses cognitive, metacognitive, and motivational processes that underlie academic competence development in any scientific discipline. Similarly, the DeALS approach contributes to evaluating and optimizing instructional design in any domain.
Prof. Christian Stamov-Roßnagel
Julie Direnga
Milos Kupresak