The study aimed to investigate the relationships among assessment content, learning outcomes, and self-confidence performed by elementary students in Taiwan. Over three hundred fifth graders from seven different schools participated in this study. A measurement tool was developed by the researchers based on the revised Bloom’s taxonomy of educational objectives related to astrology concepts. At first, the measurement of science concepts including the test item, reliability, item difficulty, and item discrimination was examined from the pilot study and validated by ten experts in science education. Secondly, several statistical methods were applied to analyze the factors stated above, such as descriptive statistics, one-way analysis of variance test, and chi-square test for independence. The findings indicated that the students’ cognitive levels of the science test had a more significant impact than the knowledge types between self-confidence level and the student learning outcome. The lower the cognitive level of the test was, the higher level of student performance and self-confidence was. Meanwhile, there was less probability of the condition of lack of confidence or overconfidence under the lower cognitive level. The relationships between the students’ confidence levels under the knowledge of high-order-thinking level and the application level were not significant. In conclusion, it is highly recommended that teachers raise the students’ cognitive levels of science learning gradually by implementing different types of assessment design. It is suggested to analyze the relationship among the meta-cognitive knowledge, learning outcome, and self-confidence exhibited by the higher graders than elementary students.
Ying-Feng Wang, National Taichung University of Education, Taiwan
Powered by WP LinkPress