Language dominance in Turkish-German bilinguals: methodological aspects of measurements in structurally different languages

Abstract

The purpose of this study is to establish measures of language dominance in bilinguals who speak structurally different languages, in our case German and Turkish, with tools that are based on fluency and oral proficiency. A 'balanced' bilingual with equal proficiency in two (or more) languages is hardly ever found (e.g. Grosjean, 1982; Olsson, & Sullivan, 2005) but the identification of the dominant language is a huge methodological problem, especially in studies of structurally different languages (see Daller, van Hout, & Treffers-Daller, 2003). The participants in the present study are a group of Turkish-German bilinguals who grew up in Germany and returned to Turkey during their school career, the so-called 'returnees' (n = 60), and a group of Turkish secondary school students who grew up in Turkey and learned German as an L2, the so-called control group (n = 55). We firstly establish the language dominance of the two groups with a C-test. We then use oral picture descriptions in both languages to measure a variety of fluency measures, both manually and using scripts written in 'Praat' (Boersma & Weenink, 2007). On the basis of these scores, we are able to develop measures of fluency that correlate highly with the C-test scores and have a highly predictive value in a logistic regression in the prediction of group membership (returnee or member of the control group). We conclude that this corroborates the validity of the measures. Overall we conclude that it is possible to develop measures of language dominance based on fluency and overall oral proficiency.