許多英語學習者總是將主動與被動任意轉換。這樣錯誤的使用被動式來自結構主義的觀點；相對而言，很少學習者會以功能主義的觀點使用被動式。奠基於篇章功能影響，本研究檢視論文中的文獻回顧章節中被動式用法及訊息分布方式。本研究的資料是來自台灣應用語言學碩士生所寫的碩士論文。有三個研究問題：(1)在文獻回顧章節中被動式出現的頻率為何？(2)與被動式連
用的動詞為何？(3)是否可歸納出語態選擇和使用被動式功能的規則?本研究的結果顯示與被動式連用的動詞多與使用方法及報告結果有關，而非與論文主題有關。有限制、短的被動式用的最頻繁。至於訊息分布上，長／短被動式經常出現在已知／新獲和新獲／新獲訊息環境中，而非限制長／短被動式經常出現在已知／已知和新獲／新獲的訊息環境中。最後，研究者將分析四種被動式的結構與訊息環境。被動式的出現與文類息息相關，並非任意出現在文句中，因為文類的關係被動式的出現會受到訊息分布及篇章結構的影響。

Many EFL learners view active and passive voice as interchangeable in all contexts.This inaccurate perception perhaps results from learners’ understanding of passives only from a structural perspective. Relatively few EFL learners learn passives from a functional perspective. Influenced by the focus on function and purpose in discourse (Swales, 1990; Tarone et al., 1981), this study examines the use of passives and the information status in the literature review section of research writings.
The data collected in this research is a corpus of theses written in English by Taiwanese graduate students in the field of applied linguistics. The three research questions guiding this research are as follows: 1) What is the frequency of occurrence of the passive voice in the literature review section of nonnative learners’ master’s theses? 2) Which verbs are most frequently used in the passive voice? 3) Are there
general principles governing voice choice and functions of the passive voice?The results show that verbs frequently used in the passive deal with the methods examined and findings reported and the functions of the verbs do not vary greatly across different research topics. In addition, finite short passives have the highest frequency of occurrence out of the four passive constructions. As for the information status, finite short/long passives frequently occur in Given-New and New-New patterns while nonfinite short/long passives often appear in Given-Given and
New-New patterns. Lastly, the positions of the four patterns where passive
constructions are embedded are also analyzed in the student writers’ literature reviews.The conclusion is that passives do not occur in text randomly. Under genre constraints, the occurrence of passives is closely related to the information status.Their use is affected by the discourse patterns.