通用学术词表的对比研究——选词、评估和分级标准

A Comparative Study of General Academic Word Lists:Criteria of Word Selection,List Evaluation and Rank Ordering

本文对三个通用学术词表进行对比分析,探究其在选词标准、评估标准和分级标准上的不足。研究发现:1)以频数、分布和离散度等为选词标准,仍存在统计指标和阀值设定上的疏漏之处;2)以覆盖率为评估标准,无法揭示学术词表遗漏学术词汇,却收录专业词汇和低频词汇等缺陷;3)以词族频数为分级标准存在信度和效度问题。学术词表不但在共有词族上的分级值并不一致,而且同族词汇中含有不同频数和分布特征的词目。通过对词汇分类、学术词汇共核和词频概貌等理论的探讨,本文拟构建更合理的选词、评估和分级标准。

This paper compares three general academic word lists （GAWLs） to discover their defects in the criteria of word selection, list evaluation and rank ordering. The findings are that 1 ） the criteria of word selection involve the frequency, range and dispersion of vocabulary, but there is room for improvement in the statistical measures and threshold settings; 2） coverage is commonly used as the criteria of list evalua- tion, which, however, fails to reveal the fact that GAWLs omit some academic words, and include techni- cal words and low-frequency words; 3） word families in GAWLs are rank-ordered by frequency, but this practice is flawed in that the ranks are inconsistent among GAWLs, and members of the same word family may differ considerably in the distribution pattern. Meanwhile, we discuss theoretical issues such as vo- cabulary classification, Common Core of Academic Vocabulary and Lexical Frequency Profile, in order to develop more reasonable criteria of word selection, list evaluation and rank ordering.