Non-sampling error can seriously influence statistical estimates based on survey data. Almost any stage of the survey process can give rise to such statistical error, from initial decisions about the concepts to be measured by the survey through to the final stages of data editing. Two aspects of the implementation of data collection are particularly important: survey participation (or its counterpart, non-response) and survey measurement (the validity and accuracy of the answers provided by respondents). Data collection modes play an important role in determining the influence of these aspects. This book attempts to map out the influence of all possible types of non-sampling error on the European Union Statistics on Income and Living Conditions (EU-SILC) data, and to identify ways in which the error could be reduced. The major¬ity of the chapters report research that formed part of the activities of the Third Network for the Analysis of EU-SILC (Net-SILC3), although there are also some guest chapters. The many practical conclusions include suggestions for improvements to documentation of procedures, improvements to guidance on survey procedures, capacity building in methods for dealing with error sources, and methodological studies, es¬pecially cross-national studies.
