Validating a test of second language written lexical retrieval: a new measure of fluency in written language production

Abstract

Lexical retrieval is an essential subprocess of language production and crucial in fluent writing and speaking. In this study we discuss a new measure of the speed of written lexical retrieval in a second language, the Written Productive Translation Task (WPTT). In contrast to Picture Naming tasks, the WPTT is not restricted to concrete nouns and verbs that can be depicted by pictures. The WPTT can also test combinations of words. Moreover, the WPTT can be used in a classroom setting because individual testing is not required. Using Messick’s 1989 framework, we investigated whether the WPTT is an appropriate measure of the construct of written lexical retrieval. In a multiple regression approach we examined the relationship between the WPTT and another test of the same construct, a written Picture Naming task. In addition, we looked at the relationship with other related constructs such as lexical access and orthographic encoding. Results suggest that the WPTT is both a reliable and valid measure of written lexical retrieval. When corrected for typing speed, the WPTT is related to the Picture Naming task and Orthographic Encoding. Finally, the results show that the relationship between the WPTT and the Lexical Decision Task is not very strong. This study makes clear that Lexical Decision Tasks do not constitute a valid substitute for measuring written lexical retrieval. At the same time it underlines the advantages of using a test specifically developed for measuring speed of written lexical retrieval like the WPTT.