The phenomenon of declining response rates is a major challenge for empirical social research. Should the loss of response units be non-random, population estimates may become biased. A rising share of the “unreachable” may lead to an increased probability of non-randomness of the loss. Exploring the process is therefore crucial to understanding what can be expected of our achieved samples. In a recent study, we investigated patterns of response unit loss by conducting a lagged recontact survey based on the European Social Survey (ESS) Round 8. We found that before one arrives at a premature verdict of unreachability, it is …

The inferential quality of an available data set, be it from a probability sample or a nonprobability sample, is discussed under the standard of the representativeness of a sample with regard to interesting characteristics, which implicitly includes the consideration of the total survey error. The paper focuses on the assumptions that are made when calculating an estimator of a certain population characteristic using a specific sampling method, and on the model-based repair methods, which can be applied in the case of deviations from these assumptions. The different implicit assumptions regarding operationalization, frame, selection method, nonresponse, measurement, and data processing are …