電腦化概念圖在科學學習上之應用研究(II)
No Thumbnail Available
Date
2000-09-30
Authors
張國恩
陳世旺
宋曜廷
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
近幾年來,概念圖(Concept maps)被廣泛的應用在教育的訓練、教學、及學生的評量上,且深受肯定。而過去概念圖的評量標準,有的僅考慮概念節點,有的則是以概念節點為主,聯結語為輔的方式來評量,但命題(Proposition)是構成知識的最小單位,這對由許多命題彼此相互連結所組成的概念圖而言,以往的評量方式,似乎稍嫌薄弱,因此本文提出一種以命題為主的屬性化概念圖(Proposition-based attributed concept maps)的觀念,也就是對概念圖中的每一命題給予一權值代表其重要程度。電腦化的概念構圖方式雖具有提供互動、立即回饋、修改容易和易於評量的優點,但為評量方便起見,使得學生必須使用由專家事先定義好的概念與聯結語來建構概念圖,而這樣的構圖方式往往會影響學生反省思考的機會,且無法真正反應出學生的學習狀態與學習成績,因此本文提出使用「開放式聯結語」(Relation-free)的概念與評分方法,讓學生能用概念圖表達出自己的想法,以增加學生反省思考的機會和真實反應出學生的學習狀態。到目前為止概念圖的評量結果都只是一個評估值,僅描述學生對某一知識主題的認識程度,學生學到了什麼、未學到什麼、及存在的錯誤概念為何,皆無法得知,因而無法提供適當回饋(Feedback)以幫助學生建構概念圖。因此本文提出一種以命題式屬性化概念圖為主的"質"的比對方法,它能記錄專家概念圖與學生概念圖之間的差異情形,並根據這些差異提供回饋以幫助學生建構概念圖及促進學生知識結構專家化。實驗結果顯示,本文所提出的評量與回饋方法確實能有效預測學生的學習成績並幫助學生建構概念圖及促進學生知識結構專家化,且使用「開放式聯結語」的構圖方式要比使用「固定聯結語」能反應出學生的學習狀態及學習成績。
In recent years, concept maps have widely been applied in education for training, instruction, as well as assessment. In the past, the assessment of concept maps only considered concept nodes or with assisted relation link between the two concepts. However, proposition is the essential element of knowledge, the above assessment are slightly frail. In this thesis, proposition-based attributed concept map is proposed. Each proposition in the attributed concept map is given a weight based on importance in a subject matter. In the conventional concept mapping system, students construct their concept maps by using the predefined concept nodes and relation links. The use of predefined concept nodes and relation links may reduce the chance of students to reflect by themselves. In this thesis, a computer-based concept mapping with relation-free is proposed. The system provides a mapping environment in which students can construct their concept maps by the use of their own relations and predefined concepts. Keeping students with freedom to use their own relations will increase the chance of students' self-reflection and get more learning. In the previous assessment, students' states, such as what is student learned, what is student unlearned, and what is student's misconception cannot reported and also cannot provide an appropriate feedback. A qualitative matching method is proposed to diagnose students' states. This matching method also provides an appropriate feedback based on the diagnosed states. In order to evaluate the proposed studies, an experiment was conducted. The experimental results revealed that the studies showed a good learning impacts on students, assisted students to efficiently construct their concept maps by using the assessment and feedback functions, and precisely diagnosed the students' learning states.
In recent years, concept maps have widely been applied in education for training, instruction, as well as assessment. In the past, the assessment of concept maps only considered concept nodes or with assisted relation link between the two concepts. However, proposition is the essential element of knowledge, the above assessment are slightly frail. In this thesis, proposition-based attributed concept map is proposed. Each proposition in the attributed concept map is given a weight based on importance in a subject matter. In the conventional concept mapping system, students construct their concept maps by using the predefined concept nodes and relation links. The use of predefined concept nodes and relation links may reduce the chance of students to reflect by themselves. In this thesis, a computer-based concept mapping with relation-free is proposed. The system provides a mapping environment in which students can construct their concept maps by the use of their own relations and predefined concepts. Keeping students with freedom to use their own relations will increase the chance of students' self-reflection and get more learning. In the previous assessment, students' states, such as what is student learned, what is student unlearned, and what is student's misconception cannot reported and also cannot provide an appropriate feedback. A qualitative matching method is proposed to diagnose students' states. This matching method also provides an appropriate feedback based on the diagnosed states. In order to evaluate the proposed studies, an experiment was conducted. The experimental results revealed that the studies showed a good learning impacts on students, assisted students to efficiently construct their concept maps by using the assessment and feedback functions, and precisely diagnosed the students' learning states.