本研究的主要目的在探討「同儕互評」以及研究者所發展之「同儕互評討論模式」對國小六年級學童之說明文寫作表現，以及對寫作與課程之看法的影響，並根據研究結果提出具體建議，以提供未來實施同儕互評於寫作教學上之參考。 研究者所提出之「同儕互評討論模式」主要整合寫作認知歷程理論、同儕互評與同儕討論之相關研究為核心所建構而成，並結合行動學習輔具教室之優勢，以期對同儕互評與討論歷程提供最有利的協助。 本研究以台北市某國小六年級兩個班級為研究對象，採準實驗研究設計中的不等組前後測設計，將參與研究的兩個班級，區分成互評組和互評討論組，分別實施「同儕互評」以及「同儕互評討論模式」，以了解在國小階段之寫作課程實施同儕互評與討論，對學童寫作能力之影響。 前後共實施10週的實驗教學，資料蒐集與分析兼採質化與量化之方式，分別簡述如下： 1.應用「說明文評分標準」進行寫作能力前後測之測量，以前測為共變數，將前測差異影響予以排除，進行單因子獨立樣本共變數分析，以了解兩組受試寫作能力之差異情形。 2.應用研究者自編之課程意見調查表，以單因子獨立樣本t考驗分析課程回饋問卷結果，了解兩組受試者看法上的差異情形。 3.以言談分析方法分析同儕討論之言談內容，以了解同儕討論對寫作互評歷程之促進效果。 本研究之結果如下： 1.互評討論組之整體寫作表現顯著優於互評組。 2.互評討論組之中低分群學生，在文字修辭之寫作表現，顯著優於互評組之中低分群學生。 3.互評討論組之高低分群學生，在組織結構之寫作表現，顯著優於互評組之中低分群學生。 4.互評討論組之高低分群學生，在內容思想之寫作表現，顯著優於互評組之中低分群學生。 5.互評討論組對寫作與課程之看法，顯著優於互評組 研究結果顯示，在國小階段之寫作課程單獨實施同儕互評，在應用評分標準方面有部分的效果，但加入互評後之同儕討論，可有助於學生對回饋與評分標準的理解，增進同儕互評的成效，促使本研究所發展之同儕互評討論模式能運作得更好，進而提升小六學童之說明文寫作能力。 The main purpose of this study was to explore the effects, on writing performance and attitude, of “peer assessment” and “the model of peer assessment with discussion” developed by the researcher. According to the results, the research provided some suggestions and future researches about the implementation of peer assessment during writing instruction. The subjects were 63 six grade students of two classes from wan-feng elementary school in Taipei City. The research was pretest-posttest design with twoquivalent groups of quasi-experimental design. The subjects were divided into two groups. The first experimental group（N=33）accepted the instruction of “peer assessment”, while the second experimental group（N=30） accepted the instruction of “The Model of Peer Assessment and Discussion”. The experiment continued with ten weeks, the data were analyzed as follow： 1.The criterion of writing performance for expository articles was used to measure the writing performance during pretesting and posttesting , and applying twelve one-way ANCOVA statistics to test the divergence of two groups on posttesting. 2.The questionnaire of curriculum was used to measure the attitude of writing and curriculum, and applying t-test to test the divergence of two experimental groups. 3.Applying “the method of discourse analysis” to analysis the content of peer discussion. The findings were as follows: 1.The whole writing performances of the second experimental group were better than the first experimental group. 2.The middle grade and low grade students of second experimental group were better than the middle grade and low grade students of first experimental group on the performance of rhetoric. 3.The high grade and low grade students of second experimental group were better than the high grade and low grade students of first experimental group on the performance of structure. 4.The high grade and low grade students of second experimental group were better than the high grade and low grade students of first experimental group on the performance of content. 5.The second experimental group were better than the first experimental group on attitude of writing and curriculum. The result above illustrated some meaning as follow: The peer discussion after peer assessment promote the effect of “The Model of Peer Assessment and Discussion” developed by the researcher, also help the development of the writing performance of students. According the results of the research, the researcher also proceed the discussion, and provide some recommendations for instruction and research in the future.
