Small area methods and administrative data integration

The literature of small area estimation (SAE) is dominated by sample-survey-based applications, where administrative register data are used as covariates in the models. Register-based statistics, however, are becoming more and more common, and integration of survey and administrative data can raise many distinct issues. Compared to sample survey data, an important advantage of the register data is that statistics can be produced at much more detailed aggregation levels. For instance, register-based census have been carried out in a number of European countries. Statistical measures of uncertainty, however, are rarely produced for register-based statistics, partly due to a lack of theoretical developments, partly due to the complexity of the errors involved. These issues are particularly relevant also for the production of poverty and well-being indicators. In many countries data coming from large sample surveys, such as the Labour Force survey (LFS) or the EU-SILC (European Union - Statistics in Income and Living Conditions), can be complemented or integrated with data coming from several population registers to obtain either more accurate estimates at the local level, or multidimensional indicators that could not have been produced using each source on its own. We shall characterize the settings of SAE which involve administrative data according to (a) whether there are relevant additional sample survey data present and, if so, (b) how the target measure is related to the available data in the two sources.