Print
PDF

Design


In each of the four countries we started with a sample of 14 year-old children in 2010. This sample comprises immigrant children of different generations as well as a native reference group. The former group was oversampled in order to increase the effective sample size for minority-majority group comparisons.

The CILS4EU study applied school-based sampling. In 2010, we sampled almost 500 schools (net) and interviewed more than 18.000 pupils in the four countries. In order to ensure that the sample contains a sufficient number of children of immigrants, schools with high immigrant proportions in this age group were oversampled. This sampling scheme does not only allow us to technically overrepresent immigrant children but it also represents a powerful design for studying the impact of ethnic concentration and inter-ethnic contacts on key integration outcomes.

For more detailed information about sampling procedures, weighting, development of the instruments and the actual fieldwork process have a look at our technical reports here.



Wave 1 (2010/11)

In the first wave, the school interviews consisted of a 20–30-minute written test in basic cognitive and language abilities, as well as a 45-minute self-completion questionnaire (incl. questions about students' ego-centered networks and classroom networks).

Additionally, teachers were asked to fill out a supplementary questionnaire to measure relevant characteristics at the classroom and school level. Apart from that, 30-minute telephone interviews were conducted with the parents of all children selected for the core sample.


Wave 2 (2011/12)

In the second wave, most of the respondents were again approached via their schools. For this group of respondents we collected again detailed information on their classroom networks.

Only a small part of school leavers and changers were contacted individually via telephone and mail, thus being interviewed in their home context.


Wave 3 (2012/13)

Between the second and the third wave of the study, most respondents either left school or made the transition to a higher level of education. Except for the Dutch part of the study, interviews within the third wave were therefore conducted mainly via telephone, but also as online or postal interviews. In this wave we also repeatedly collected information on the ego-centered networks of the respondents.