探索空間能力與領域特定知識對科學學科表現之影響

Abstract

本研究旨在選定地球科學課程中，與空間思考有關的「太陽視軌跡」為主題，分別以「空間能力測驗」、「標準化測驗」、「天文繪圖測驗」三項測驗代表「領域廣泛空間能力」、「領域特定知識」及「科學學科成效」指標，探討空間能力、領域特定知識與科學學科表現三者間的關係。另外，本研究使用眼球追蹤法記錄受試者在解題時的注意力分配及解題歷程，以釐清領域廣泛的空間能力與領域特定知識在科學學科表現中扮演的角色。進一步透過晤談，以內容分析法比較教師與在領域特定知識及科學學科表現不同的三組學生（高分組、中分組、低分組）在解答空間形式的問題時的認知模式、解題行為、知識結構、策略使用的差異，以進一步釐清領域特定知識在科學學科表現上的作用。本研究受試對象為具有基礎先備知識的大一及大二學生，測驗及晤談資料的有效樣本共40人，含理組19人（男性9人；女性10人）及文組21人（男性10人，女性11人），扣掉眼動資料缺漏的兩位學生後，眼動的有效資料為38人。另有12位在高中任教的地球科學教師參與研究，這些教師提供本研究的專家資料來源。 研究結果發現： 1..領域特定知識較佳者在解答單一選擇題形式的標準化試題的歷程上，傾向較能辨認解題關鍵區域且花費較多的時間比例在這些（尤其是圖片解題關鍵）區域上、在題幹區域花費較長時間，以形成問題表徵、似乎不完全採用前向思考；也會搭配後向思考的解題策略。2.領域廣泛的空間能力與解題時的注意力分配有關，空間能力較差者相較於中等程度以上者分配較少的時間在圖片區域上。3.無論教師或高分組、低分組學生，在解答「天文繪圖測驗」時皆傾向找出一致的通則解答題目，但低分組傾向直覺的心智運作，教師能以科學專有名詞明確說明這些解題技能及策略的用意、能靈活採用科學模型搭配思考、追求精確數值而非模糊地陳述性質。4.以領域特定知識為中介變項的分析顯示，領域廣泛的空間能力並非直接影響學生的科學學科表現，而是透過領域特定知識，影響最終的科學學科表現。另外，在科學學科成效表現不同的高、中、低分組的主要差異是在陳述性知識及程序性知識的豐富程度上。 最後，本研究對實務上的教學及教材設計提供以下建議：1.對低空間能力者的學習教材應給予視覺空間上的援助、2.大部分學習者對教材的需求是促進概念之間的連結、3.專業知識的養成主要仰賴長期記憶的知識品質。The study aimed to examine the interactions among spatial ability, domain-specific knowledge, and student’s science achievement in the context of ‘The apparent path of the sun’ which is believed to be one of the topics requiring spatial thinking in the secondary earth science curriculum in Taiwan. Three assessments were used to assess domain-general spatial ability (The Purdue Visualization of Rotations Test), domain-specific knowledge (a test adopted from the entrance examination for high schools and colleges in Taiwan) and science achievement (a self-constructed test on the topic of the apparent path of the sun). Eye-tracking method was employed to record the visual attention distribution and the process of problem solving to clarify the roles of domain-general spatial ability and domain-specific knowledge in student’s science achievement. By interview and content analysis on participants’ knowledge structures and problem solving strategies, the effects of domain-specific knowledge on science achievement were explored. Forty university students who were in non-earth-science majors but had taken introductory earth science lessons in high schools voluntarily participated in the study. In addition, twelve teachers who were teaching earth science in high schools were invited as the expert group who provided the criteria for further distinction between experts and novices. The study results are shown below: First, students who showed better domain-specific knowledge could tell better the key information related to problem-solving, attend more to these (especially figure) areas, and spend more time on the problem description to form the problematic representation. They seemed to adapt both thinking forward and backward strategies. Second, students with less domain-general spatial ability would spend less time on figures than those with medium and high spatial ability. Third, most participants tended to find a coherent strategy to draw the apparent paths of the sun with four different latitudes, regardless of whether they are teachers, high or low science achievement students. However, teachers could use scientific terms to explicitly elaborate the meanings of their strategies and skills, and could coordinate better the scientific model (i.e., Celestial sphere model) in solving problems. Moreover, their answers were more accurate (with quantitative value. Fourth, further statistical analysis indicated that domain-specific knowledge fully mediated the effect of domain-general spatial ability on student’s science achievement. Moreover, students' knowledge structures, rather than their spatial abilities, determined their performance of the domain-specific problem solving involving spatial thinking. Based on the study findings, three suggestions for the design of teaching and learning materials were provided. First, materials offering visual-spatial assistance are needed for lower spatial ability students. 2. For most students, making connections between relevant concepts are most critical in solving domain-specific spatial problems. 3. Developing the expertise in science relies primarily on the quality of domain-specific knowledge stored in long-term memory.