Using Embedded Assessment and Database to Prob Problem-Solving Strategies under Computer Simulation Environment: The Case of Kinematic Graph

• Main Authors: Chen Po-Yu, 陳柏宇
• Other Authors: Chien Cheng-Chih
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/37066852131485753071

碩士 === 國立臺北教育大學 === 教育傳播與科技研究所 === 95 === This research is dedicated to explore students' problem-solving strategies in computer simulated environment. Students' solving process could be collected by embedded assessment and database. Embedded assessment and database are a combined tool. Students are allowed to solve problems in computer simulated form in light of "Assume-Observe-Explain" course. These problems are embedded in webpages. While users are operating, therir solving process would be recorded in database automatically. The outcome shows that embedded assessment and database could reflect senior hight school students' simulator-using models and solving strategies. According to developmental order, solving strategies could be analyzed into 3 stages, "Trial at will", "Solving in scienticifc way" and "Strategic solving". The last 2 stages include recognizable solving stragegies. In addition, embbeed assessment and database could reflect most students' misconceptions. Consequesnce of interview reveals that database records, dialogue of interview, paper-and-pencil score, these 3 sources have common points. They all unveil interviewers' consistent misconceptions. In order to simplify database records, this study designs "validity rate" to represent solving validity in one computer simulation problem. Accompany with "regression" tool, it divides solving process into 3 strategic stages. In the part of statistics, using ANOVA method to explore relationship between "Solving strategy and paper-and-pencil score" and "Solving strategy and validity rate". Solving strategies are recognized in artificial way, not automatic way mentioned above. It's checked one by one. The outcome reveals these 2 sets are in significant relationship. The more compact the solving strategy is, the higher paer-and-pencil score and validity rate are. That is, computer simulation problem's operational model and kinematic graphs' concept are correlated.