從語料庫分析的角度看高中生作文字彙的使用

A Corpus-Based Analysis of Vocabulary Use in Compositions of Senior High School Students

本研究主要目的在於探索不同成就之高中生其作文字彙使用上的差異。本研究包含了在六個熟悉主題之下，六所高中學生所寫的一千一百七十篇作文。研究資料的蒐集歷時一個學期。為幫助分析及比對，本研究使用了文字處理程式（CCLang）及二個對比語料庫（The Brown Corpus 及The LOB Corpus）。研究結果如下。（一）就概念字、字形字、文章長度及句子長度而論，高成就學生的表現比中、低成就的學生來的佳。（二）就字彙密度而言，高成就學生的表現亦比中、低成就的學生來的好。（三）就字彙在對比語料庫中的分配比例而言，中成就學生的表現，比起高、低成就的學生，在各個頻率區間中的分配比率均較高。透過對英文作文的語料分析，教師可以更瞭解學生的詞彙使用情形，並透過字彙學習技巧進而針對學生較弱的字彙段進行增強。

This study was a corpus-based analysis of vocabulary use in English compositions written by senior high school students with three different achievement levels (high-, mid- and low-achievement; HA, MA and LA). It considered whether students with different achievement levels differ in their vocabulary use in English compositions. This study covered, 0 students' compositions under six familiar topics. The data collection lasted one semester. A word-processing program, CCLang, and two comparative corpora, the Brown Corpus and the LOB Corpus, were used in this study. The results were stated as follows. First, HA students performed better than MA and LA ones in concept and graphic word types and in the average length of sentences and compositions. Second, as far as lexical density was concerned, the percentage for HA students was also higher than those for MA and LA students. This implied more lexical variation in HA students' compositions and more challenges in reading their compositions. Third, with reference to the distribution percentage in the Brown Corpus and the LOB Corpus, MA students had a higher distribution of word percentages in each frequency range than HA and LA students. By corpus analysis of students' compositions, teachers would understand students' vocabulary use better and know how to strengthen students' vocabulary by teaching them vocabulary-learning strategies.