Although self-regulated learning (SRL) is seen as highly relevant for successful college learning, college students oftentimes show a lack in SRL abilities. Therefore, it seems necessary to foster SRL in this group of leaners. In order to evaluate such training and to foster SRL in an optimal way, a valid assessment of this competence and its development is necessary. As different methods for the assessment of SRL show benefits and points of criticism, the present study used a multimethod approach to investigate convergence between and across different measures as well as their predictive validity for achievement. SRL was conceptualized of cognitive, metacognitive, and motivational components. Seventy college students were assessed with two broad SRL-measures (questionnaire, strategy knowledge test) and two task-specific SRL measures (microanalyses, trace data) within a standardized laboratory setting. Moreover, GPA of college entrance diploma was gathered as an indicator of general achievement level. Results indicate moderate to high relations between the different components of SRL (cognition, metacognition, and motivation) within one assessment level and no relations between the different assessment methods within one component. With regard to achievement, we found that every component is predictive for achievement but only if measured with different assessment methods. The results are discussed with regard to their implications for future research and the use of different assessment methods for SRL.

No abstract