本研究之目的為：一、探討行動取向師資培育模式的理論基礎。二、建構行動取向師資培育模式的變項。三、實地試驗行動取向師資培育模式的效能。四、探討知識轉變為行動的歷程。 本研首先透過文獻分析，探討行動取向師資培育模式的理論基礎，建構行動取向師資培育模式的知識、計畫、表現、省思和精熟等五大變項，再用「個案實驗研究法」，選擇花蓮師範學院大學部三年級兩位修習「教育實習」之師資生為樣本，分為實驗組和控制組，進行「知識輸入」實驗處理後，實施六回合計畫、表現、省思試教實驗。資料蒐集的方法有知識引出訪談、教學設計、放聲思考think aloud與轉譯、試教錄影與轉譯、省思錄影與轉譯及研究者實地觀察記錄。資料分析的方法有資料的持續比較、路徑發現分析、精熟情形分析、資料的次數分析、百分比分析等。本研究的重要發現有： 1.教師的知識應分為「領域通用或領域間通用的知識」（一般教育學知識）和領域專用的知識（學科知識）兩大類；領域專用的知識應再分為領域內通用的內容知識、單元╱課專用的內容知識、領域內專用的教學法知識、領域內專用的教學步驟知識；兩類知識交融產生「領域專用的教育學知識或學科教育學知識」。 2.獲取與記憶之學院設計安排的「識字與寫字」教學相關知識、實驗處理的知識，確實成為發展教學行動表徵與採取行動的基礎。 3.一節的「識字與寫字」課，需交融領域間通用的的知識、領域內通用的內容知識、單元專用的知識或常識、領域內專用的教學法知識交融轉換生產教學行動表徵。 4.「識字與寫字」的教學設計需要提取領域內通用的內容知識和單元專用的內容知識，這些知識影響教學行動表徵的生產量。 5.本研究中「設計的教學行動表徵」有84%以上產生具體行動表現。 6.知識越充足計畫量就越多、教學表現量也越多，省思量也隨著增加。 7..經六回合的計畫、表現、省思之後，各個項目都有進步和精熟化的表現，以本研究自編的「陳述性知識與程序性知識11等級量表」評估知識的進步與精熟等級分別從實驗前的3.4和3.5，提昇為7.9和8.1。 最後本研究針對師資培育方案規劃、課程發展、教學、督導、與評鑑，以及未來繼續研究之建議。 / The purpose of this study was to investigate an action-learning model for teacher education. In addition, the study intended to provide implications for the linking knowledge and action of teacher education. The current researcher reviewed literature to analyze the rationale of an action-learning model for teacher education and to construct the five variables of knowledge, design, performance, reflection and mastering for the model. The experiment method was used. The subject was two student teachers. One student teacher studied how to teach elementary school students’ Chinese vocabulary and handwriting from a videodisk. Both the two student teachers teach elementary school students’ Chinese vocabulary and handwriting for six sessions. Knowledge elicitation interview, instructional design, thinking aloud and video recoding were used for datd collection. The showing your work analysis, protocol analysis, constant comparison analysis, the pathfinder analysis, frequency analysis and percentage analysis were used for data analysis. The findings from this study are the following: 1.The pedagogical domain-specific knowledge or pedagogical content knowledge was blended by the knowledge of domain-general and the knowledge of specific-specific. In order to analyze the pedagogical domain-specific knowledge from instructional design, the knowledge of general-general is named of intra-general-general. The knowledge of domain-specific should be divided into inner-domain-general content, inner-lesson-specific content, inner-domain-general pedagogy, and inner-domain-general process. 2.The programmed knowledge learned from teacher education program and from experiment treatment were able to develop the representation of teaching action and teaching action. The two kinds of knowledge were useable. 3.Teaching elementary school students’ Chinese vocabulary and handwriting needs to blend the knowledge of intra--domain general and the knowledge of domain-specific（The knowledge of inner-domain-general content, inner-lesson-specific content, inner-domain-general pedagogy, inner-domain-general teaching process）. 4.To design the teaching of elementary school students’ Chinese vocabulary and handwriting need to blend the knowledge of general-general and the knowledge of specific-specific. The more the knowledge of these two areas induces the more production of representation of teaching action. 5. The subjects had transferred 84% representation of teaching action into their teaching in the current study. 6.The more knowledge of intra-domain general and the knowledge of domain-specific could produce the more production of the representation of teaching action. The more production of the representation for teaching action could transfer the more performance in teaching. The more performance of teaching could induce the more reflection of teaching. The current study not only provides implications for teacher education but also suggests directions for further research.

本研究探討設計導向的知識翻新學習對師培生的學習有何影響，包括探討其學習結果與學習歷程，同時本研究也分析各學習分組間的差異。研究對象為修習教學媒體與操作課程之44位師培生，以小組為單位進行課程，每組4至5人，共分為10組，採個案研究法。 資料來源包括：(1)師培生三次教案設計；(2)師培生三次教案設計之組內討論；(3)師培生三次教案設計後同儕回饋。資料分析方式說明如下：其一、將三次教案設計根據改編的教案評量表進行教案品質評分，透過相依樣本無母數檢定，比較其成果是否有所改變。其二、根據Mishra和Koehler（2006）所提出之七項教學知識架構，將三次基於教案設計的小組討論及同儕回饋進行編碼，以了解各組想法及回饋中所包含的教學知識有何改變；並以想法類別編碼分析各組師培生討論之想法有何差異。其三、利用集群分析各學習小組，比較不同集群之間學習成果與學習歷程之異同，並藉由質性分析探討各小組如何討論及設計教案，並剖析差異原因。 主要研究結果如下：(1)師培生整體教案設計品質提升，其中又在教學創新及創意表現向度（Z= 2.87，p < .01）與教學架構向度（Z=2.55，p < .05）上有顯著成長。(2)在各組教案設計討論過程中，發現包含一項教學知識的想法數量有顯著減少趨勢（Z=-2.81，P<.01）；在各組收到的回饋中也發現，回饋包含兩項教學知識者於期末則有明顯成長趨勢（Z= 2.19，p < .05）。(3)利用集群分析將各組師培生教案設計所討論的想法進行分類，結果得到兩大集群，其一在認知及後設認知類型想法中產出較多，命名為想法高產出組；另一群則相反，命名為想法低產出組。此外研究還發現，想法高產出組在教案設計之創新及創意面向（t=2.33，p<.05）、教學媒體及科技面向（t=2.80，p<.05）顯著高於低想法產出組，也較低產出組能產出更多認知及後設認知類型討論想法。 綜上述而言，本研究發現設計導向的知識翻新學習歷程有助於師培生教學媒體教案設計之翻新，並提升師培生在討論及回饋中融合運用教學知識之能力。此外亦發現，能提出較多高階類型想法組別，其教案設計之創新程度、科技使用品質及給予他組回饋的表現上，相較之下也較佳。本研究建議教師在教學上應同時兼顧師培生討論數量及品質，鼓勵學生不斷提升討論的想法層次，並適時介入低產出組學生討論之中，協助學生不斷改進其教案設計的想法，以提升其教案設計所需的相關教學知識。 / The purpose of this research was to investigate the effects of design-based knowledge building on pre-service teachers’ teaching knowledge. This research adopted a case study design. Participants were 44 undergraduate pre-service teachers who engaged in a course pertaining to the design of instructional media. They were divided into 10 learning groups. Data sources included: (1) pre-service teachers’ design of three lesson plans; (2) pre-service teachers’ discussion within groups; and (3) peer feedback after presentation of each group’s lesson plan. The process of data analysis is as follows: First, lesson plans were assessed by an evaluation form, and then analyzed by using nonparametric tests to see if there was any change in the design of lesson plans. Second, the three discussion and peer-feedback activities within each group were analyzed by open coding using two different coding schemes: “Technological Pedagogical Content Knowledge—TPACK” (Mishra & Koehler, 2006), and quality of ideas. Third, using cluster analysis to divide 10 groups into different clusters, this study further compared students’ learning processes and outcomes between groups. The findings were as follows: (1) It was found that the design quality of lesson plans were enhanced, especially in terms of the innovative and creative teaching performance dimension (Z = 2.87, p <.01) and the design framework dimension (Z = 2.55, p <.05); (2) In the course of group discussion, teaching ideas that contained only one type of teaching knowledge (e.g., content knowledge) had decreased significantly (Z=-2.81，P<.01). Teaching feedback (received from other groups) regarding two types of teaching knowledge (e.g., pedagogical content knowledge) increased significantly (Z= 2.19，p < .05). (3) Using cluster analysis, this study classified 10 groups into 2 clusters based on groups’ discussion about teaching ideas. One cluster outperformed in both cognitive and meta-cognitive types of teaching ideas was named high-productive cluster; the other cluster was named less-productive cluster. High-productive cluster was superior to low-productive cluster in terms of the “Innovative and Creative Teaching” and “Instructional Media and Technology Use” dimensions for lesson plans, and in terms of the quality of feedback they provided to their peer groups. The main findings were as follows: (1) Design-based knowledge building helps pre-service teachers’ to design more innovatively their lesson plans and to enhance their teaching knowledge; (2) It was found that the groups who could work more creatively with ideas could also produce more innovative lesson plans, and were also more likely to give away quality feedback to other peer groups. Based on the findings, it is suggested that teachers should not just pay attention to the quantity of discussion posts, but also the quality of idea discussed. It is also important to encourage student to work innovatively and collaboratively with ideas during discussion. Teachers should also try to intervene and help provide necessary scaffolds to the low-productive groups so as to encourage them to continuously improve their teaching ideas.