<![CDATA[Since overconfidence bias negatively impacts the student learning process, it is crucial to understand how to measure it effectively. While the global metacognitive judgment is commonly used to assess overconfidence, it lacks specificity, making it difficult to identify the particular concepts in which students exhibit overconfidence. This study proposes the use of local metacognitive judgment to measure overconfidence and examines the validity inferences of this instrument through Rasch analysis. Additionally, the study aims to compare overconfidence across gender, country, and grade level. Two studies in science learning were conducted: the first involved 532 Indonesian and Korean high school students (both male and female), who answered 24 items on Genetics and Evolution with a confidence scale; the second study included 376 undergraduate students from various semesters (grade levels) who completed 25 Genetics items with a confidence scale. Rasch analysis was used to assess the validity and reliability of the local metacognitive judgment. The results indicated that the instrument is psychometrically valid and highly reliable. Furthermore, the findings revealed that Indonesian students exhibited significantly higher overconfidence than Korean students, males were more overconfident than females in both countries, and undergraduate students showed peak overconfidence when first learning genetics concepts at university. These findings are consistent with previous research using global metacognitive judgment. Therefore, the local metacognitive judgment is a valid tool that offers practical advantages over global approaches, as it provides item-specific overconfidence data, enabling educators to identify and address overconfidence in specific concepts rather than relying on overall scores.]]>
