The quantitative stage of the project consists of a standardized computer-assisted personal survey of German citizens of Russian and Turkish descent, respectively. 500 interviews were conducted within each migrant group.
Contrary to the German Longitudinal Election Study (GLES), we focused exclusively on German citizens of migration descent so that the case numbers will be high enough to perform multivariate analyses.
Our sampling strategy consisted of a two-step sampling procedure. First, we drew German municipalities (“Gemeinden”) as Primary Sample Units (PSU), using a Probability Proportional to Size (PPS) procedure that weighs the likelihood of a PSU to be chosen by the number of residents of a municipality. Second, personal addresses were drawn from the residents’ registration offices registers, and onomastic procedures were used to determine the origin of the first and last name.
The survey lasted about 60 minutes and was conducted synchronously to the GLES post-election survey in November 2017. Available survey languages were German, Russian or Turkish. The questionnaire consisted of two parts. The first part reflected established items for electoral research that were also used in the GLES post-election survey to ensure the comparability of our sample with the group of autochthonous citizens. The second part encompassed items that capture migrant-specific characteristics (e.g. citizenship, membership in ethnic organisations, and ethnic identity).
We also asked for the participants’ willingness to participate in further surveys so we will be able to use the sample to conduct a second survey.