These are synthetically generated unit and area level population and sample data that can be used for testing model-based unit-level small area methods. To prevent disclosure issues the datasets have been generated by repeated (Monte-Carlo) sampling of real EU-SILC (Survey of Income and Living Conditions) data in Austria. The data include geographical identifies and can be used for fitting unit-level (Battese-Harter and Fuller type) models and area level models (Fay-Herriott- type) models. The datasets are part of the R package emdi. Examples of the use of the data can be found in the emdi manual available via https://cran.r-project.org/web/packages/emdi/emdi.pdf and in Kreutzmann et al. (2019) Kreutzmann, A. K., Pannier, S., Rojas-Perilla, N., Schmid, T., Templ, M., & Tzavidis, N. (2019). The R package emdi for the estimation and mapping of regional disaggregated indicators. Journal of Statistical Software, 91(7). https://doi.org/10.18637/jss.v091.i07Reliable statistics are crucial for policy relevant research. Small Area Estimation (SAE) methods generate robust reliable and consistent statistics at geographical scales for which survey data are either non-existent or too sparse to provide direct estimates of acceptable accuracy. The last decade has seen a rapid increase in the use of SAE. Statistical agencies and Governmental organisations are actively developing their own suites of estimates. In the UK the Office for National Statistics (ONS) has responded to user demands by producing estimates of average household income for wards and using SAE to answer queries from local authorities, policy advisers and government departments. The Welsh Assembly Government (WAG) is actively seeking to develop capacity for SAE. Public Health England produces SAEs of adolescent smoking and chronic kidney disease. Initial demands for small area statistics are now shifting to requirements for more complex statistics that extend beyond averages and proportions to encompass estimates of statistical distributions, multidimensional indicators (e.g. inequality and deprivation indicators) and methods for replacing the Census and adjusting Census results for undercount. These developing requirements pose significant methodological and applied real-world challenges. These challenges are deepened by different methodological approaches to SAE remaining largely unconnected, locked in disciplinary silos. The technical presentation of SAE also impedes more widespread uptake by social scientists and understanding by users. The proposed programme of work aims to (a) develop novel SAE methodologies to better serve the needs of users and producers of SAE (b) bridge different methodological approaches to SAE, (c) apply SAE for answering substantive questions in the social sciences and (d) 'Mainstream' SAE within the quantitative social sciences through the creation of methodologically comprehensive and accessible resources. The project comprises three work packages of methodological innovative research designed to deepen the understanding of SAE and achieve the aforementioned aims. The project will capitalise on a cross-disciplinary research team drawn together through an NCRM methodological network and reflecting a large part of the SAE expertise in the UK. Through long-standing collaborations with national and international agencies in the UK, Mexico and Brazil, which are placed at the centre of the project, we enjoy access to individual level secondary survey and Census data. Collaboration with key SAE users will ensure that the project remains relevant to user needs and that methodologies are used for expanding the set of small area statistics currently available. The involvement of international experts ensures the quality and relevance of the research. Substantive outputs will include SAEs of attributes of interest to users, including income, inequality, deprivation, health, ethnicity and a realistic pseudo-Census dataset for use by other researchers. The project will advance knowledge across disciplines in the social sciences including social statistics, applied economics, human geography and sociology. It will additionally impact on the production of official and Census statistics. The project is committed to adding value to NCRM's training and capacity building activities by developing new resources.

The data are synthetically generated unit and area (district) level population and sample data. The use of synthetic data is for preventing disclosure issues with the real datasets. No survey or interviews are used in this case. Instead, data have been generated by repeated (Monte-Carlo) sampling of real EU-SILC (Survey of Income and Living Conditions) data in Austria to create a synthetic population of Austria. A sample is then selected from the population by using stratified simple random sampling within the Austrian districts.