Chinese translation norms for 1,429 English words

Yun Wen,Walter J B Van Heuven

doi:10.3758/s13428-016-0761-x

Abstract

We present Chinese translation norms for 1,429 English words. Chinese-English bilinguals (N = 28) were asked to provide the first Chinese translation that came to mind for 1,429 English words. The results revealed that 71 % of the English words received more than one correct translation indicating the large amount of translation ambiguity when translating from English to Chinese. The relationship between translation ambiguity and word frequency, concreteness and language proficiency was investigated. Although the significant correlations were not strong, results revealed that English word frequency was positively correlated with the number of alternative translations, whereas English word concreteness was negatively correlated with the number of translations. Importantly, regression analyses showed that the number of Chinese translations was predicted by word frequency and concreteness. Furthermore, an interaction between these predictors revealed that the number of translations was more affected by word frequency for more concrete words than for less concrete words. In addition, mixed-effects modelling showed that word frequency, concreteness and English language proficiency were all significant predictors of whether or not a dominant translation was provided. Finally, correlations between the word frequencies of English words and their Chinese dominant translations were higher for translation-unambiguous pairs than for translation-ambiguous pairs. The translation norms are made available in a database together with lexical information about the words, which will be a useful resource for researchers investigating Chinese-English bilingual language processing.

Highlights

Translation equivalents have been used extensively to investigate bilingual language processing
In addition to collecting translation norms, the present study investigated the relationship between translation ambiguity and word frequency, concreteness and language proficiency
The present translation norming study revealed that 71 % of the English words were translated into more than one correct translation, and that more frequent words and less concrete words have more alternative translations as revealed by a small but significant correlation

Summary

Introduction

Translation equivalents have been used extensively to investigate bilingual language processing. Even though the masked prime is not visible to bilinguals, the recognition of a target word is facilitated by its translation-equivalent non-cognate prime (e.g., Duñabeitia, Dimitropoulou, Uribe-Etxebarria, Laka, & Carreiras, 2010; Duñabeitia, Perea, & Carreiras, 2010; Geyer, Holcomb, Midgley, & Grainger, 2011; Grainger & Frenck-Mestre, 1998; Midgley, Holcomb, & Grainger, 2009). This priming effect occurs when the two languages do not share the same writing system (e.g., HebrewEnglish: Gollan, Forster, & Frost, 1997; Japanese-English: Hoshino, Midgley, Holcomb, & Grainger, 2010; ChineseEnglish: Wang & Forster, 2010). Additional evidence for non-selective lexical access comes from recent studies using a hidden translation repetition priming paradigm. Thierry and Wu (2007) used this paradigm for the first time to investigate whether translation equivalents are accessed in a

Objectives

Methods

Results

Conclusion