置入運算思維於學習鷹架中對高中程式寫作課程中之自我效能與學習成效之影響

程式設計內容相互具有關聯性，然而程式概念抽象不易理解且複雜性高，因此多數學習者對於程式學習產生消極的信念，造成自我效能降低並影響學習表現。為了解決程式學習的困境，教學者多運用運算思維的概念來設計教學策略。本研究假設教學者提供學習鷹架來協助學習者與已知知識相互融會貫通，因後設認知鷹架能喚起學習者的已知知識，藉以輔助學習新知識，但其缺乏邏輯系統性的引導方式，所以本研究設計將運算思維流程置入於後設認知鷹架中來引導學習者學習程式設計。研究目的在於探討運算思維流程與傳統教學兩者鷹架引導方式對於學習者自我效能與學習成效之影響。研究方法為實驗法，實驗對象為普通高中一年級學生共39人，實驗組以運算思維流程之步驟來設計鷹架引導內容，對照組以傳統教學方式之程式敘述順序來設計鷹架引導內容，以學習成效測驗試題與自我效能量表作為量化研究工具，並在實驗結束後透過半結構式訪談收集質性資料。研究結果發現在進行實驗課程後，使用運算思維流程鷹架引導方式之學習成效與自我效能表現皆優於使用傳統教學鷹架引導方式，且自我效能與學習成效表現具有相關性。建議未來相關研究可深入探討影響自我效能與學習成效的其他因素。

關鍵字

程式設計學習；運算思維；學習鷹架；學習成效；自我效能

並列摘要

Programming content is interconnected. However, programming concepts are mostly abstract, difficult to comprehend, and highly complicated. As a result, most learners have negative impressions about learning programming, leading to lower self-efficacy as well as learning performance. To solve this dilemma, instructors employed the idea of computational thinking when devising teaching strategies. Metacognitive scaffolds can evoke learners’ known knowledge. Thus, this study hypothesized that by providing learning scaffolds, instructors, can help learners integrate new content with their existing knowledge so as to facilitate their learning of new knowledge. However, a logical and systematic scaffolding is lacking. Therefore, this study was designed to integrate computational thinking processes into metacognitive scaffolds to guide learners for studying programming. This study aimed to investigate the influences of the teaching scaffolds of computational thinking processes and traditional teaching approach on learners’ self-efficacy and learning outcome. The experimental method was selected as the study methodology and 39 first-year high school students were chosen as the study subjects. For the experimental group, we adopted the steps of computational thinking processes to design the scaffolding content. For the control group, the scaffolding content was designed following the programming narrative sequence of the traditional teaching approach. We selected learning outcome related test questions and self-efficacy scale as quantitative research tools, and garnered qualitative data through semi-structured interviews at the end of the experiment. The study findings indicated that after the experimental sessions ended, learning outcome and self-efficacy driven by scaffolding of computational thinking were better than those under the traditional teaching approach. Moreover, a correlation existed between self-efficacy and learning outcome performance. It is suggested that future research works should investigate other factors affecting learners’ self-efficacy and learning outcome.