We collected number-of-translation norms on 562 Dutch-English translation pairs from several previous studies of cross-language processing. Participants were highly proficient Dutch-English bilinguals. Form and semantic similarity ratings were collected on the 1,003 possible translation pairs. Approximately 40{\%} of the translations were rated as being similar across languages with respect to spelling/sound (i.e., they were cognates). Approximately 45{\%} of the translations were rated as being highly semantically similar across languages. At least 25{\%} of the words in each direction of translation had more than one translation. The form similarity ratings were found to be highly reliable even when obtained with different bilinguals and modified rating procedures. Number of translations and meaning factors significantly predicted the semantic similarity of translation pairs. In future research, these norms may be used to determine the number of translations of words to control for or study this factor. These norms are available at http://www.talkbank.org/norms/tokowicz/.
Participants are given a definition, such as an animal that barks, and respond by naming the object to which the definition refers (in this example, dog). In the translation production task, a word is presented in one language and its translation equivalent must be spoken in the other language.