Author:

Abstract:

Since 2010, several KU Leuven faculties and colleges participate in a large-scale, low-stakes language test of academic Dutch at the beginning of the first year. The test, taken by 11.144 students between 2010 and 2014 is constructed to detect possible at-risk students and give them an early warning signal. It consists of 25 selected-response vocabulary and reading items and lasts maximum 30 minutes (De Wachter & Heeren 2013). A predictive validity study with 2660 university students shows a significant positive correlation with study success (De Wachter et al. 2013). This presentation focuses on the interaction between the language students speak at home and their language of pre-university instruction in relation to their academic language test scores. After filtering out incomplete data, the dataset consisted of the language test scores of 11.144 students in four academic years between 2010 and 2014, of which 3578 college students and 7566 university students. In the data analysis, a linear regression was run with the test score (SCORE) as response variable and as predictors their home language (HOME), whether there was a difference in home and schooling language (DIFF), their sex (SEX) and the year in which the test was taken (YEAR). The most important finding of the analysis is a significant interaction between home language and the language of pre-university schooling. When Dutch speaking students had a language of schooling different from their home language, their language test results were lower than when their home and schooling language were the same. However, when students indicated that they rarely spoke Dutch at home the effect was reversed, meaning that a difference in home and schooling language went together with higher language test scores. Finally, when students indicated that they mainly spoke Dutch at home and sometimes another language the effect was similar, but less strong. Of these two groups 92.5% indicated Dutch as their pre-university school language. That seems to indicate that the Dutch academic language proficiency of students in a multilingual home situation benefits from a pre-university education in Dutch.