Some issues on non-sampling errors in EU-SILC

The objective of this paper is to carry out a review of the sources of non-sampling errors and lack of comparability in EU-SILC ( EU Statistics on Income and Living Conditions) data across countries and across time, together with an assessment of their impact on income variables and poverty measures.The paper begins with a broad literature review on different frameworks for classifying types of error encountered in surveys. Such frameworks linking error components to theirs sources can be helpful in identifying and controlling those errors from the data production side. From the side of EU-SILC data use and analysis it is helpful to identify and classify patterns of distortion to which particular results are subject. It is from this perspective that we identify a wide range of potentially useful studies in Section 3. We also note how differences in concepts and procedures can affect comparability. We discuss in details the specific aspect concerning the quality and comparability of EU-SILC data related to the presence of large values and outliers in the measured household disposable income. This is only one example of many more in-depth studies on quality and comparability of EU-SILC data that could be usefully carried out for augmenting the comparability of such a fundamental survey for monitoring the social landscape of European Union.