The Statistical Matching: An Integrated Archive of the Lifestyles of Italian Families

Abstract

The objective of the techniques of integration for data from a number of different sources is to identify records relating to similar or identical units, to estimate the unified distribution of a number of variables observed in different data archives and to merge informative records. The present work will describe a model of data integration through a methodology of Statistical Matching (hot deck distance) for the integration of two surveys (Eu-Silc Istat and Lifestyle Survey University of Bari). The construction of an integrated database on the basis of these two surveys may be useful for the study of consumer behavior in relation to specific groups of commodities, in order to analyse the decisions taken by families with regard to saving, to examine economic and social inequality, and to study the impact of public policies by means of simulations. The coexistence of multiple and differentiated objectives triggers the need to obtain a very general and versatile integrated file, which provides ongoing detailed information on the different types of spending, on levels of saving, on the distribution of incomes, on the occupational conditions of the members of the family unit, etc.

The contribution is the result of joint reflections by the authors, with the following contributions attributed to S. Montrone (chapters 1,2,5) to P. Perchinunno (chapters 4) and to Liliana De Blasi (chapters 3). The data used in this paper are the result of a survey funded by the Cassa di Risparmio di Puglia entitled: “Analisi statistica territoriale della povertà urbana attraverso la costruzione diindicatori di disagio socio-economico”.