Empirical investigation of a CATI sampling design for rare populations

(Empirische Untersuchung eines CATI-Stichprobendesigns für seltene Populationen)

Summary

Due to a variety of problems (e.g. data privacy, access problems, federal structure of the official statistics) there barely is register data or official data for social policy-relevant subgroups, such as disabled people, educationally disadvantaged people and persons with migration background. Therefore in all of these cases, the planning, preparation and evaluation of social policies depends on survey data.

For general population surveys, there are well developed and established sampling procedures. These procedures can hardly be used for rare populations, because they are either nearly impossible to apply, they lead to biased estimates or they lead to too little observations with the characteristic of interest.

Thus, in practice often non-random sampling methods are used, whereas quota samples are the most common. As a result of non-random sampling, inference to the target population cannot be realized, which actually is the fundamental objective of data analysis.

The proposed project therefore has great practical and societal relevance. The objective is the implementation and evaluation of a sample design for nationwide telephone surveys of rare populations. The tested sample design provides a cost effective and rapid implementation of surveys of rare populations such as ethnic minorities.

Contact

Volker Stubbra, email
Kjell Sembritzky, email

Funding

The project is funded by the Federal Ministry of Education and Research (BMBF)