本研究嘗試以國小高年級數學領域之「小數乘除」為主題,探討結合知識結構之貝氏網路推論工具的評量診斷模式,並嘗試建製一套有效且快速的診斷與補救教學系統。診斷測驗採以適性化出題,並搭配以電腦多媒體動畫為主的補救教學,希望藉由本系統讓學生可以在最短的時間內,接受個別化的診斷測驗,並施予適性化、即時化的電腦補救教學。以期能同時達到評量、診斷、補救教學的功能。 研究結果發現: 1. 結合知識結構之貝氏網路推論工具的評量診斷模式,可有效應用於診斷學生錯誤類型與子技能。 2. 以學生試題結構為依據之電腦化適性出題模式,能達到節省題目,縮短施測時間的目的。 3. 透過電腦適性化補救教學後,學生在「小數乘除」概念的測驗表現上具有顯著差異,平均分數進步且學習成效明顯。
The main purpose of the research is to explore the diagnostic assessment model on the base of Knowledge Structure Theory to build a convenient and effective diagnosis system. We use Bayesian networks for modeling assessment data and identifying bugs and sub-skills about “Decimal Multiplication and Division” of Mathematics in Grade 6. The multimedia computer is devised for adaptive diagnostic testing and computerizes adaptive remedial instruction with the system. Students can receive not only individual diagnostic tests, but adequate and in-time computerized adaptive remedial instruction in limited time. Evaluation Diagnosis and remedy can be achieved simultaneously. The results: 1. The Bayesian networks evaluation mode with Knowledge Structure Theory apply effectively to the diagnosis of students’ mistakes and sub-skills. 2. Using the computerizes adaptive testing model on the base of Knowledge Structure Theory , we can shorten the testing time with lesser questions. 3. The progress of students is significant after taking the computerize adaptive remedial instruction.