結合線上診斷評量系統之適性補救教學研究
周雅釧(Ya-Chuan Chou), 黃志勝(Chih-Sheng Huang), 施淑娟(Shu-Chuan Shih), 郭伯臣(Bor-Chen Kuo)
Abstract
本研究主要目的在建立六年級數學領域「比與比值」單元的線上電腦化適性診斷測驗題庫及補救教學教材。透過實際的線上施測與補救教學的準實驗設計，同時探討此電腦化適性診斷測驗之省題率和精準度、適性選題下錯誤類型與子技能的預測精準度、與補救教材的教學成效。
研究結果顯示：接受此電腦化適性診斷測驗的受試者，比接受全測驗平均可節省39.22%的題目，平均預測精準度為93.22%。在此適性選題下，錯誤類型診斷的精準度平均達96.64%。且以補救教學結構做為補救教學基礎的實驗組，進步幅度大於50%。最後，依據研究結果提出相關建議，以做為後續研究的參考。This study aims to establish an adaptive remedial teaching material and a computerized adaptive diagnostic system on the mathematical unit of ”Ratio” on Grade 6. By on-line testing, we explore the economizing rate, the prediction accuracy of the computerized adaptive diagnostic mathematics test and the accuracy diagnose of the existence of bugs and sub-skills of students. We also evaluate the effect of the adaptive remedial teaching material via adaptive remedial instruction.
The major findings of this study are summarized as follows: The rate of saving items by the computerized adaptive diagnostic system is 39.22% and prediction accuracy is 93.22%. After analyzing the bugs, the prediction accuracy is 96.64%. By adaptive remedial instruction based on knowledge structure, the experimental group had improved above 50%. Finally, according to the experiment result we can render some related suggestions, and be the reference resources for follow-up studies.
研究結果顯示：接受此電腦化適性診斷測驗的受試者，比接受全測驗平均可節省39.22%的題目，平均預測精準度為93.22%。在此適性選題下，錯誤類型診斷的精準度平均達96.64%。且以補救教學結構做為補救教學基礎的實驗組，進步幅度大於50%。最後，依據研究結果提出相關建議，以做為後續研究的參考。This study aims to establish an adaptive remedial teaching material and a computerized adaptive diagnostic system on the mathematical unit of ”Ratio” on Grade 6. By on-line testing, we explore the economizing rate, the prediction accuracy of the computerized adaptive diagnostic mathematics test and the accuracy diagnose of the existence of bugs and sub-skills of students. We also evaluate the effect of the adaptive remedial teaching material via adaptive remedial instruction.
The major findings of this study are summarized as follows: The rate of saving items by the computerized adaptive diagnostic system is 39.22% and prediction accuracy is 93.22%. After analyzing the bugs, the prediction accuracy is 96.64%. By adaptive remedial instruction based on knowledge structure, the experimental group had improved above 50%. Finally, according to the experiment result we can render some related suggestions, and be the reference resources for follow-up studies.