Paper Title
PROPOSING A MODEL FOR ASSESSING METACOGNITIVE REGULATION LEVELS DURING COLLABORATIVE PROBLEM-SOLVING

Abstract
Levels of metacognitive regulation in a collaborative learning were first proposed by De Backer et al. (2016), consisting of two levels i.e., deep and low level. However, preliminary research analyzing metacognitive regulation levels found that some verbal protocols of undergraduate students in a group cannot be classified as either deep or low-level based on the framework of De Backer et al. (2016). Thus, the current study aims to develop a valid hypothetical theory of metacognitive regulation level in collaborative problem-solving into three levels: high-level, middle-level, and low-level, based on logical deduction and theoretical studies. The framework was developed to achieve content and construct validity. Content validity reviewed the accuracy of the theories used as reference material. Construct validity reviewed the logicality of researchers' thinking in developing metacognitive regulation levels. Content and construct validity were obtained through revising the theory based on expert assessment. There are four experts examining the validity of the theory developed. We proposed a high-level metacognitive regulation that demonstrates increased task orientation activities, the planning of various alternative problem-solving strategies, and the ability to select effective problem-solving plans. It involves monitoring comprehension to build upon previously acquired understanding and engaging in reflective judgment as an evaluation of problem-solving endeavors. The explanation of middle-level and low-level are discussed. This framework is expected to provide a theoretical contribution to identifying the quality of undergraduate students' metacognitive regulation in collaborative problem-solving. Further research can be conducted to examine the empirical validity and the reliability of the theory. Keywords - Collaboration, Level, Metacognition, Problem-Solving