Comparative EU Statistics on Income and Living Conditions: Issues and Challenges

This paper develops and discusses a framework for the assessment of statistical quality in EU-SILC, with focus on comparability as a central dimension of quality. We view data quality as a multidimensional concept, covering not only statistical accuracy but also the relevance, timeliness, comprehensiveness, etc., of the data. There is a broad agreement on what dimensions make up the overall quality of statistical data, and we briefly review these dimensions, noting some relationships between them. All counties participating in EU-SILC produce national Quality Reports. We review them briefly since they provide valuable information on data quality, including comparability. Comparability, as one of the dimensions of data quality, is a particularly important dimension in the context of a EU-wide undertaking such as EU-SILC. In ECHP, for instance, comparability was achieved through a standardised design and common technical and implementation procedures, with centralised support of the national surveys by Eurostat. The survey structure and implementation arrangements are more diverse in EU-SILC; we identify how this diversity makes the problem of comparability more complex and acute in EU-SILC. We begin by clarifying what "comparability" means and how it may be achieved in practice, and identify the basic characteristics and requirements of EU-SILC in their effect on comparability. Comparability means the extent to which the results for different countries can be put together, compared, and interpreted in relation to each other and against common standards. An assessment of how far such comparability has been achieved in practice requires us to examine the data and procedures both from the “input” and the “output” sides. The former involves an analysis of the methodology and implementation of the process of production: how the data were collected, statistically treated, processed, and analysed. The latter involves a comparison of the substantive results actually obtained with appropriate standards such as alternative data sources, prior knowledge, and logical expectations. Both these aspects in the assessment of comparability are important. On the basis of this framework, the paper proceeds to identify a number of specific aspects where problems of comparability are likely to arise in EU-SILC, and elaborates methodologies for the study of some of the more important ones among them. 1 Dipartimento di Metodi Quantitativi, Università degli Studi di Siena, P.za S.Francesco 8, 53100 Siena, [email protected].