Benjamin Schulz
Non?response of Immigrant Parents in School?Based Surveys: Results from the National Educational Panel Study in Germany

5th Conference of the European Survey Research Association, University of Ljubljana, July 15th to July 19th, 2013

Immigrants often show particularly low response rates. Consequently, they are under-represented and object to response bias. Since many surveys lack comprehensive information on non-participants in general and on their immigration background in particular, detailed analyses of immigrant non-response are still scarce. Here, data from the National Educational Panel Study (NEPS) in Germany offer a promising chance for more detailed analyses. Among others, NEPS provides a school-based sample of German 9th-graders. In addition to PAPI assessments of students, teachers and headmasters CATI interviews with parents were conducted. However, for 44.2 percent of the participating students parental interviews could not be realized. While this generally is an undesirable fact, it allows us to analyze parental non-response using data gathered from students and schools as contextual information to predict parental cooperation and response. Thereby, we seek to identify more precisely how non-participating immigrant parents are characterized. To this end, I apply logistic random intercept models to estimate probabilities of cooperation and probabilities of non-response for immigrant parents from Poland, Turkey, the former Soviet Union, former Yugoslavia and Southern Europe. With 42.8 percent the overall response rate of immigrant parents is 17 percentage points lower than for native parents. Furthermore, cooperation and response rates differ strongly between immigrant groups. Context information like seniority of interviewers, students’ language skills, parental education, school resources and school composition are used to explain initial differences. However, even controlling for all these aspects substantive net effects of “country-of-origin” remain. In line with theoretical considerations and earlier research, language proficiency turns out to be the single most important factor for parental response. Results may help to tailor future studies to better meet the particular requirements of immigrants and thus to reduce response bias.