Metacognitive Problem-Based Learning Model as an Alternative to Improve Student Learning Outcomes in the Stoichiometry Solution Concept

A study has been conducted on applying the metacognitive-problem-based learning (M-PBL) model to stoichiometric solutions. The research aims to identify differences between models in learning grade-class outcomes. This quasi-experimental study involved students at SMAN 2 Banjarmasin. The research used a quasi-experimental design with a pretest-posttest nonequivalent control group. The sample from all populations (total sampling). Test instruments to measure students' learning outcomes. Questionnaires for measuring students' metacognition. The observer measures students' activityand qualitative explanations from the results of student and teacher interviews. The result shows there is a significant difference in learning outcomes by ANOVA test among M-PBL, M-DI, and DI classes; (2) the best student learning achievement is in the M-PBL group, which is shown by the Duncan test at α 0.05; and (3) the percentages of student learning process in the M-PBL group, M-DI group, and DI group were 77.82, 73.44, and 55.66%, respectively. The M-PBL model can improve student learning processes and outcomes, so it should be implemented in schools.