臺灣國小國語教科書的常用句型分析──語料庫為本的研究

Corpus-Based Examination of Sentence Structures in Taiwanese Elementary Mandarin Textbooks: Implications for Language Education

句型教學是語文教育的重要環節，然而現今缺少以語料庫實證分析建構句型難易標準的相關文獻。有鑑於此，本研究探討國小國語教科書中的單句在類別、使用率的表現，以此歸納國小學童理解課文所需的句型知識。本文以電腦程式與人工校正從康軒、南一、翰林教科書整理出18,081個單句，再歸納成40個句型。在使用率上，前三高的句型S＋V＋O、S＋V＋O＋V、S＋V＋C使用率涵蓋語料庫約五成的句子結構，4至10名分別是S＋V、「是」字句、S＋A、存在句、T＋S＋V＋O、S＋P＋V＋O、「把」字句，這10個已可涵蓋語料庫八成五的句子結構。本研究也統計句型在不同年級的使用率，發現年級越高句子結構也越趨複雜，大致遵循由簡入繁的原則。在版本對比上，句型在3個版本的類型及比例相當接近，顯示單句較無跨版本差異。本文最後根據使用率提出國小的單句學習建議表，期盼能作為編排教學內容或制訂評量標準的參照。

Sentence structure is crucial in language education. However, corpus-based research categorizing and quantifying sentence structures in Mandarin elementary textbooks are insufficient. To fill this gap, this study examined 18,081 simple sentences extracted from textbooks by three different publishers, categorizing them into 40 unique sentence patterns. The most prevalent patterns (S + V + O, S + V + O + V, and S + V + C) accounted for nearly 50% of the sentence structures in the corpus. The ensuing seven categories (S + V,“Shi”sentences, S + A, existence sentences, T + S + V + O, S + P + V + O, and“Ba”sentences) constituted approximately 85% of the total structures. These findings indicate a clear progression in sentence complexity with advancing grade levels, reflecting a pedagogical trajectory from simplicity to complexity. Despite examining various textbook versions, the study found minimal variation in sentence patterns across the three publishers. This research culminates with a proposed sentence learning table, intended to enhance language education.