本研究調查糾正回饋對於華語學習者目標語法結構習得之效應，所探討的回饋類別涵蓋口語糾正回饋與口語結合書寫糾正回饋，口語糾正回饋包含重鑄與提示，口語結合書寫糾正回饋則包含重鑄結合書寫回饋，以及提示結合書寫回饋。研究對象包含48名來自不同語言背景的華語學習者，以及四位華語老師。研究以不同類別的糾正回饋處理方式，重鑄、提示、重鑄結合書寫回饋、以及提示結合書寫回饋四個組別，調查華語語法補語結構的學習成效。研究結果顯示，在接受不同類別糾正回饋的組別中，重鑄結合書寫回饋提高學習者結果補語結構的語法準確度，在立即性後測呈現其顯著成效；提示結合書寫回饋則在延後性後測呈現顯著性的優勢。本研究證實聚焦型書寫回饋的綜合語言知識可提升華語學習者的語法習得成效，同時闡明華語課室教學性質糾正回饋類別的優勢效應。

This study investigates the effect of corrective feedback (CF), including recasts, prompts, recasts plus written corrective feedback (WCF), and prompts plus WCF, on Chinese-as-a-second-language (L2 Chinese) learners’ acquisition of target grammatical structures. Forty-eight learners from diverse language backgrounds and four experienced teachers participated in the study. Drawing upon four CF treatments (Recast, Prompt, Recast + WCF, and Prompt + WCF) with a pretest, immediate posttest, and delayed posttest design, grammar gains of the learners were compared and measured. The results showed that recasts plus WCF facilitated learners’ accurate use of resultative complements in the immediate posttest, and prompts plus WCF retained the long-term grammatical gains, as evidenced by significant outperformance in the delayed posttest. The findings suggest that metalinguistic knowledge in focused WCF is crucial in improving L2 Chinese learners’ grammatical competence and shed light on the advantages of instructional CF practices in classroom contexts.