本研究旨在比較國小二年級典型發展及語言障礙兒童，在敘事「巨結構」及 「微結構」各項指標的表現是否呈現顯著差異，並探討各項指標的效果量，找出 區辨兩組兒童最具敏感度的指標。本研究對象為60 名國小二年級學童，語言障礙 與典型發展兒童各30 名，兩組兒童均以無字圖畫書Frog, where are you? 為材料， 透過故事重述的方式蒐集敘事樣本。本研究結果如下：一、巨結構：兩組兒童在 故事文法要素中的「背景、起始事件、嘗試、結果、內在反應」及插曲完整度等 指標有顯著差異。二、微結構：兩組兒童在「語意」部分的「總詞彙數、相異詞 彙數、詞彙錯用率」、「語法」部分的「平均C 單位長度、複雜句使用率、文法 錯誤句出現率」，以及「凝聚結」部分的「前後關係照應凝聚完整率、關聯詞凝 聚完整率、凝聚結凝聚完整率」等指標有顯著差異。三、比較各項敘事評量指標 的效果量，最能區辨兩組兒童的前五名指標依次為平均C單位長度、詞彙錯用率、 嘗試、文法錯誤句出現率及複雜句使用率。在評量國小二年級兒童的敘事能力時， 可特別針對以下的幾個能力進行探討：插曲完整度、詞彙使用能力（包括總詞彙 數、相異詞彙數及詞彙錯用情形）、語句使用的能力（包括複雜句使用率、平均 語句長度、文法錯誤句出現率）及前後關係照應的使用情形。

Purpose: Children’s narrative performance plays a pivotal role in developing their abilities of conversation, literacy, and socialization. To tell a complete and coherent story, a child must possess various skills, including the use of a specific structure to compose a story, the ability to choose appropriate vocabulary and sentence structures to covey ideas, and the use of proper references and connectives to unite the story. Therefore, it is crucial to evaluate children’s narrative abilities in clinical settings. This study investigated whether second-graders with language disorders and their typically developing peers demonstrated significantly different performances in the various areas of macrostructure and microstructure narrative. The effectiveness of these indicators was also investigated to identify the most sensitive indicators for distinguishing these two groups of children. The results can serve as a reference for relevant personnel in clinical practice. Method: A total of 60 children were examined, namely 30 with language disorders and 30 with typical language development. The two groups were asked to retell a wordless picture book called Frog, Where Are You? in order to assess their narrative abilities. Results/Findings: The two groups had significantly different indicator scores for narrative macrostructure elements such as background, initiating event, attempt, consequence, internal response, and number of complete episodes. Regarding the microstructure elements, the two groups demonstrated significantly different indicator scores in semantics (i.e., total number of words, number of different words, and percentage of incorrect vocabulary), syntax (i.e.,mean length of C-unit, subordination index, and percentage of grammatically incorrect sentences), and cohesive ties (i.e., ratios of complete reference cohesive ties, complete connective cohesive ties, and total complete ties). Concerning the effect size of these narrative indicators, the top five indicators were as follows: mean length of C-unit, percentage of incorrect vocabulary, number of attempts, percentage of grammatically incorrect sentences, and subordination index. Conclusions/Implications: The narrative ability of second-grade students should be thoroughly assessed in the following aspects: the number of complete episodes, vocabulary usage (total number of words, number of different words, and percentage of incorrect vocabulary), sentence usage (subordinate index, mean length of sentences, and percentage of grammatically incorrect sentences), and reference cohesive tie usage. For clinical practitioners, narrative assessment can serve as a supplementary tool for determining children’s language performance in addition to norm-referenced assessment batteries, and is an effective method of monitoring children’s progress over time.