電腦適性測驗與資訊輔助補救教學融入教學之研究—以國小柱體與錐體單元為例

The Study of integrating computerized adaptive testing and the remedial instruction with IT into teaching─ Take cube and awl for example

許逸偉(Hsu-I-wei)

指導教授 ： 施淑娟 郭伯臣

亞洲大學/資訊學院/資訊工程學系碩士在職專班/碩士(2010年)

摘要

摘 要 本研究以知識結構為基礎，結合貝氏網路，研發一套運用數位數學教材與電腦適性測驗。首先依據教材內容，建立專家知識結構模式，再以能力指標6-s-05、子技能、錯誤類型等相關內容編製電腦適性診斷測驗試題，以進行紙筆測驗預試。接著依照預試施測資料結果來分析學生的知識結構順序，再依此順序結構建立電腦適性診斷測驗系統的選題規則及題庫。同時，再編製以知識結構為基礎之數位數學教材供實驗組進行教學與補救教學活動使用，並於正式施測後，根據受試者作答所得資料加以分析，利用貝氏機率統計推論方法，觀察在適性與全測之作答情況下，對子技能與錯誤類型之辦識率。 研究結果簡要摘述如下： 一、實驗組使用本套指導教材進行教學後，實驗組前測分數為87.28分，控制組為77.94分，經檢定達顯著差異，顯示本套數學指導教材具有學習成效。 二、實驗組經過補救教學後，前測平均分數是87.28分，後測平均分數是94.76分，經檢定達顯著差異，表示研究者自編之補助教材，的確能幫助學生之學習。 三、電腦適性測驗能有效節省49％以上之題目，且能達到94％以上的預測精準度。 四、以貝氏網路推論電腦適性測驗，在適性和完整作答情況下，對錯誤類型與子技能的一致性，前測達93.1%，後測達95.4％。 五、比較不同學習風格之學生在學習成就之表現，而知學習風格與學生學習成就無明顯相關。

關鍵字

知識結構、貝氏網路、電腦適性診斷測驗、柱體與錐體、學習風格

並列摘要

Abstract

This research aims to establish a knowledge structure and Bayesian networks based mathematical teaching materials and computerized adaptive testing. First, we analyzed the content of the textbook, and established the expert knowledge structures of the content. According to the expert knowledge structures, indicator, sub-skills, and mistaken types which can be calculated in the Bayesian networks, items were designed. After the pre-test, ordering theory is used to decide the students' knowledge structure and those parameters were used in the item bank. At the same time, establish a knowledge structure based mathematical teaching materials which can be used by experimental teaching. After experiment, analyze the data by Bayesian probability statistics method, inspect the prediction accuracy of the sub-skills and mistaken types in the situations of adaptive test and completely tested.
Some findings are briefly outlined as follows:
1. After taking the experimental teaching, experimental group students' average grades were better than control group’s significantly (87.28>77.94). It shows that this teaching material is valuable for learning.
2. After experimental group students taking the remedial instruction, they have significant progress on their average grades (post-test > pre-test,94.76> 87.28). It shows that this teaching material is valuable for remedy.
3. The rate of saving items by Bayesian networks computerized adaptive testing (BNAT) is above 49% and prediction accuracy can reach over 94%.
4. With regard to the effect of BNAT, the prediction accuracy of the mistaken types and sub-skills is 93.1% in pre-test and 95.4% in post-test.
5.The compare the performance of the achievement about students among the different learning styles,and there was no sibnificant correlation between the learning styles and the achievement about students.

並列關鍵字

knowledge structure, bayesian networks, computerized adaptive test,cube and awl，learning styles

參考文獻

林立敏、白曉珊、郭伯臣、劉育隆(2007)。數位個別指導教材研發與適性補救教學模式之研究--以國小五年級數學「因數與倍數」單元為例。網際網路技術學刊，8卷2期，頁191-197。

李俊儀（2004）。資訊科技融入數學教學模組之開發與研究－以國中平面幾何基礎課程教學為例。花蓮師院學報，19，119-142。

教育部（2003）。國民中小學九年一貫課程綱要數學學習領域。台北：教育部。

黃燕輝（2004）。資訊科技融入數線教學成效之探討--以國小三年級學童為例。國立台北師範學院數理教育研究所碩士論文。

張英傑（2001）。兒童幾何形體概念之初步探究。國立台北師範學院學報第十四期，491-528。