ABSTRACT

Indices of lexical diversity have been used to estimate the size of a writer’s vocabulary and/or a writer’s lexical proficiency for some time. One issue with many commonly used indices of lexical diversity (e.g., TTR and index) is that they vary as a function of text length. Accordingly, much research has been devoted to the development of indices that are text length independent. However, very little research has investigated the degree to which indices of lexical diversity are reflective of human ratings of diversity itself. In this study, the relationship between indices related to three dimensions of lexical diversity (abundance, variety, and volume) and human ratings of lexical diversity are explored in a corpus of L1 and L2 argumentative essays. The results indicated that abundance was the strongest predictor of lexical diversity ratings, followed by volume and variety. Multivariate models indicated that 74% of the variance in lexical diversity ratings could be explained by abundance and variety, and that speaker status had a negligible effect. Implications for research, writing instruction, and writers are discussed.

摘要

一段时间以来，词汇多样性指数已被用于估计作家词汇量和/或作家的词汇熟练程度。许多常用的词汇多样性索引（例如TTR和索引）的一个问题是它们随文本长度的变化而变化。因此，许多研究致力于发展与文本长度无关的索引。但是，很少有研究调查词汇多样性指数在多大程度上反映了人类对多样性本身的评价。在这项研究中，在L1和L2论证性文章的语料库中探讨了与词汇多样性的三个维度（丰富度，多样性和数量）相关的指数与人类对词汇多样性的评级之间的关系。结果表明，丰富度是词汇多样性评分的最强预测因子，其次是数量和种类。多变量模型表明，词汇多样性评分中74％的方差可以用丰富度和多样性来解释，说话者地位的影响可忽略不计。讨论了对研究，写作指导和作家的影响。