幾十年來，過程寫作常因為『重內容輕形式』和『輕忽寫作成品』而導致批評聲浪不斷。 近年來，部落格網誌的四大特色 — 自我表達、自我反省、互動和發布 — 吸引學者和語言學習者的注意。本文旨在探討部落格融入『過程寫作』之教學如何影響四名台灣高三學生的寫作表現。文獻探討涵蓋過程寫作，部落格和三項共通因素 — 回饋、自我反省日誌和電子檔案評量。 本研究的對象是研究者所教授的過程寫作班的四位高三學生。本研究的程序以寫作循環 (writing cycle) 為主軸。資料包括多次校訂 (multiple revisions)、兩篇草稿 (前後測)、學生寫作態度問卷、教師日誌(teacher’s log)和訪談紀錄。首先，以Yagelski’s coding schemes (1995) 和Johnson’s (1994) 內容和組織指標 (indicators of content and organization) 解讀多次校訂，以了解學生的校訂類型 (types of changes in multiple revisions)。接著，分析前後測分數的差距 (score differences)及文章長度 (the length of essay)、探討多次校訂與前後測草稿的關聯性，再找出前後測草稿中的內容與組織指標，以顯示學生在兩篇草稿的進步程度。訪談資料則用於探討部落格的功能、學生遇到的困難和需要何種教師支援。至於學生的寫作態度，則以前後測問卷的分數差距 (score difference) 為主，再以訪談紀錄作為佐證資料或疑點澄清。最後，與教室情境有關的發現，則以教師日誌和訪談資料為主。 本研究的主要發現如下： 1. 生有自己偏好的校訂類型 (types of changes in multiple revisions)。 2. 作班的四位高三學生在內容和組織方面有顯著進步。 3. 級部落格在過程寫作的每個階段都有不同功能。 4.多次校訂的枯燥須加以克服，以及建議實施老師與學生間的迷你會議。 5.在信心、焦慮、實用性與偏好四方面，學生寫作態度改變。 6.教室情境所衍生的幾項發現。 最後，本研究對過程寫作在實際教學上應用與未來研究方向提供建議。 / Over the past decades of practice in process writing, criticism over “the stress on content over form” (Badger & White, 2000) and the neglect of final writing product (Barnes, 1983) has never ceased. Considering that blog has attracted scholars and learners for its unique features — self-expression, self-reflection, interactivity, and publication, this researcher integrated blog into the instruction of process writing to investigate the writing performance of the four 12th graders. The literature review covered studies on process writing, blog and three elements commonly found in both process writing and blog— feedback, self-reflection journals and e-portfolio. The participants of this study were four 12th graders of Taiwan’s high school in a blog-mediated process writing class taught by the researcher. The procedure revolved around the process writing cycle, in which these four participants repeated the recursive writing process. Data included multiple revisions, two drafts (pre-test and post-test), questionnaires on students’ writing attitude (pre-test and post-test), the teacher’s logs and interviews. First, to obtain what types of changes students made in multiple revisions, the 48 multiple revisions were coded by Yagelski’s coding schemes (1995) and Johnson’s indicators of content and organization (1994). Second, to investigate if the four participants made progress in content and organization, this researcher touched on four aspects of analysis: 1) the score differences between the pre-test and post-test drafts, 2) the length of essay, 3) a connection analysis between the multiple revisions and the two drafts and 4) a draft analysis using Johnson’s indicators of content and organization (1994). Third, the interview data helped to explore how the blog functioned in each step of process writing model, and to present the challenges students encountered and the teacher support they needed. Fourth, to discover if these students experienced any attitude changes, she also calculated the score differences between the pre-test and post-test questionnaires in students’ writing attitude and verified the questionnaire results or clarified the unclear points with interview data. Finally, some other classroom-context-related findings were also revealed in the analysis of the teachers’ logs and the interview data. The major findings were listed as below: 1) students showed preference for making some types of changes in multiple revisions; 2) the four students in the blog-mediated process writing class gained significant progress in content and organization; 3) the class blog functioned differently in each step of process writing model; 4) the boredom of doing multiple revisions and the teacher-student mini conference were the challenges encountered and the teacher support recommended; 5) there were writing attitude changes in the four categories — confidence, anxiety, usefulness and preference; and 6) several findings were also revealed in the context of classroom. Some pedagogical implications and recommendations for future research on process writing were presented at the end of the thesis.