A Study of the Relationship between Mechanics Knowledge Structure and Learning Achievements by the Application of Pathfinder Method for Tenth-Grade Students

• Main Authors: Pai-Hsiang Huang, 黃湃翔
• Other Authors: Shing-Ho Chiang
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/24763545584868916704

博士 === 國立高雄師範大學 === 科學教育研究所 === 92 === The purpose of this study is to apply Pathfinder Method to analyze mechanics knowledge structure for tenth-grade students. This study investigates both the predictive efficacy and discriminative efficacy of learning achievements by knowledge structure indices, with which it aims to evaluate the feasibility of applying Pathfinder Method in teaching assessment. In the meantime, this study also analyzes both the patterns difference of knowledge structure graph representation and the node activation in problem solving, with which it aims to provide the consultation for teaching diagnosis and teaching design. The subjects included 148 tenth-grade students from Kaohsiung district. The tools this study employed included the mechanics knowledge structure assessment, the foundation mechanics achievement test, the problem solving analysis with thinking aloud, and the questionnaire of understanding mechanics concept. The main findings and conclusions are made as follows: 1. The knowledge structure indices can effectively predict learning achievements and effectively discriminate subjects with different learning achievements. Among the three indices the most predictive of students’ learning achievements is the PFC index, which accounts for approximately 40.4% of the variance in students’ learning achievements (R=.635，R2＝.404). Furthermore, PFC index has the best predictive efficacy for high-achiever group (R=751); consequently, it shows that PFC index is the most representative of their mechanics learning achievement. 2. The variations in the patterns of knowledge structure graph representation for different learning achievement students indeed existed. The knowledge structures of high-achiever group are organized with superordinate concept in the center, more hierarchical, and displayed with tree structures that connect related concepts by concept definition. These variations influence students’ concept extracted and node activation path, and thus influence their performance in physics problems solving. 3. There are some mistaken or defective concept connections derived from the analysis of the knowledge structure graph representation and the connection-chain weight, and such connections are able to reflect the incorrect understanding of related concepts by students.。