| Summary: | Nonresponse is one of the main problems for mass surveys. In this paper we analyze the issue of unit nonresponse, when selected respondent doesn’t provide all (or almost all) data required. Literature demonstrates that rate of nonresponse differs across countries, regions and socio-demographic factors. On this ground we suggest that nonresponse should be treated as a socially determined action. This hypothesis has never been tested, partly because it is extremely difficult to collect sufficient information about nonrespondents. In this article I focus on social distance as a determinant of successful communication in surveys, and draw upon the writings of Chicago school of sociology to analyze social distance in big cities. The paper suggests a practical approach to collecting data about all sample units, including nonrespondents. I also discuss the results of empirical approbation of this approach.
|
|---|
