在使用中文「一＋量詞」詞組時，不同母語的中文學習者呈現出不同的偏誤規律。本文檢視並分析中級的韓國學習者與美國英語學習者使用中文「一＋量詞」的普遍偏誤。美國學習者傾向多用中文的「一＋量詞」，而韓國學習者與其相反，呈現少用「一＋量詞」的規律。本文認為學習者的偏誤之所以形成，在於數量詞的教學並未觸及如何判斷名詞組在篇章中的所指，而名詞組的指稱性質決定「一＋量詞」的使用。值此之故，本研究在中介語錯誤分析的基礎上，建構一個中文「一＋量詞」教學的流程圖。該流程圖適用於對中級的美國英語學習者與韓國學習者進行教學，透過循序漸進的引導，訓練學習者判斷篇章中的名詞所指，以輸出正確的語言形式，而無需向學習者傳授任何關於指稱的語言學知識。本文也提出未來更進一步的研究方向。

This paper examines the common errors produced by native Korean and American English speakers who are in the process of learning to use yi 'one'+classifier at the intermediate level of Mandarin Chinese. This paper argues that such errors tend to occur when learners are not trained to observe the discursive referentiality patterns of nominals, which determine when a noun should or can be modified by yi 'one'+classifier. The paper establishes a pedagogical sequence to provide such training and tailors it to native speakers of American English and Korean in intermediate Chinese classes. The pedagogical sequence takes the form of a decision tree. The paper also makes suggestions for further research on pedagogy of numeral classifiers at a higher level.