Sememe richness: a supplementary indicator of lexical richness in Chinese second-language writing

Abstract

<p>This article uses the vocabulary measurement indicators of lexical diversity, lexical sophistication, lexical density, and lexical accuracy for second-language writing levels to systematically measure the lexical richness of essays written by Chinese as a second language (CSL) learners. It employs the methodology of building a corpus based on the collected compositions for data analysis. It shows that the Uber index of lexical diversity and the lexical density of CSL learners does not always show a positive correlation with increased test scores. Lexical sophistication usually positively correlates with test scores, but the growth trend tends to be flat from the 81–90 score range. Lexical accuracy below 80 points is positively correlated with writing scores; however, the data above 91 points is slightly lower than that of 81–90 points. This article further hypothesizes that the above measurement indicators fail to accurately measure CSL students’ Chinese lexical ability because those indicators do not include the sememe factors of Chinese words and thus rely too much on word form. This article further puts forward the specific manifestations (polysemes, homographs, and multi-category words) of sememe richness and proposes specific solutions to add sememe richness to the four original formulas through the use of the four above-mentioned measurement indicators in CSL.</p>

本文運用第二語言寫作水平詞彙多樣性、詞彙複雜度、詞彙密度、詞彙準確率等指標, 系統地衡量漢語作為第二語言學習者寫作的詞彙豐富度。 研究方法主要為採用構建作文語料庫的方法進行數據分析。 結果表明, 漢語二語學習者的詞彙多樣性和詞彙密度的 Uber 指數並不總與考試成績呈正相關。 詞彙複雜度通常與考試成績呈正相關, 但增長趨勢往往在 81–90 分範圍內持平。 低於 80 分的詞彙準確率與寫作分數呈正相關; 但 91 分以上的詞彙準確率數據略低於 81–90 分。 本文進一步假設, 上述測量指標無法準確衡量漢語二語學習者的漢語詞彙能力, 因為這些指標不包括漢語詞的語義因素並過於依賴詞形。 本文進一步提出了語義豐富度的具體表現形式 (多義詞、同形異義詞、兼類詞) 與運用上述指標增加語義豐富度的具體解決方案。