Immigrant children’s poor school results are often linked to their linguistic background. More traditional measurements which compare groups of non-native and native speakers pose serious limitations and result in too much statistical noise. By developing a mixed method design, the project seeks to develop a more refined instrument for investigating the dynamic interaction between multilingual proficiencies, users’ dispositions and perceptions, and their effects on school results. We make use of a (quantitative) instrument which complements proficiency measurement with an extensive set of questions about multilingual repertoires. In addition, beliefs, perceptions and multilingual practices are charted using qualitative techniques: interviews and focus groups.