漢語考試大綱詞表和語料庫詞頻的反向檢視研究

國際漢語教材的生詞經過專家編訂及實際教學錘煉，可反向檢視漢語考試大綱詞彙分級的合宜性以及語料庫詞頻的對應性。本研究從九套具有代表性的初級國際漢語教材選出生詞樣本，分別依據中國大陸及臺灣的漢語考試分級詞表及大型漢語語料庫，統計樣本詞的等級及詞頻分佈。發現初級教材生詞竟普遍包括考試大綱的高級詞及超綱詞，顯示大綱的詞量及收詞有待商榷。而其詞頻分佈也普遍有高比例的低頻詞，可見語料庫的詞頻難以作為教材實際編寫的選詞依憑。

The vocabulary of many international Chinese textbooks has been systematically edited by experts and repeatedly tested in class to meet the communication needs of second-language learners. The appropriateness of the vocabulary classification of standardized Chinese test frameworks and the correspondence of corpus word frequency can be examined in reverse. This research selects samples of new words from nine notable beginning-level Chinese textbooks. The distribution of these sample words is analyzed based on the vocabulary level lists in Taiwan’s Test of Chinese as a Foreign Language (TOCFL) and mainland China’s Xin Hanyu Shuiping Kaoshi, or New HSK. We also review the sample words’ frequencies by referencing Mainland and Taiwan corpora to examine whether their frequency distributions match up to frequency-based principles. We found that the vocabulary in the beginning-level textbooks generally includes high-level words and even a sizable number of words from outside of the frameworks. The number of words in these frameworks and the words they include are open to question, and the word frequency distribution in the textbooks generally shows a high proportion of low-frequency words. Therefore, in terms of the actual writing of textbooks, it is difficult to rely on corpus word frequency as the basis for vocabulary selection.