Stichprobenverzerrung durch Item-Nonresponse in der international vergleichenden Politikwissenschaft
The respondents’ susceptibility to item nonresponse is an important determinant for the quality of political science survey data. In particular, one has to expect biased samples when groups of respondents systematically differ in their nonresponse rates. Our study tests firstly whether and how strong four different aspects of political involvement which are frequently used in political studies affect the respondents’ susceptibility to nonresponse. Secondly, we tested the con-sequences of these differences in item nonresponse for the presence of sample-selection bias in general, and in particular for the comparability of the samples in international comparative political science research. The database of our study are Eurobarometer studies in the 10 new member states of the European Union. According to our first result, politically involved respondents are substantially less likely to refuse to answer questions and in particular less prone to “don’t know” responses. Our second result indicates that this causes a systematic over representation of involved respondents. The strength of this sample selection bias differs substantially between the analyzed countries.
In order to set up a list of libraries that you have access to,
you must first login
or sign up.
Then set up a personal list of libraries from your profile page by
clicking on your user name at the top right of any screen.