Premium
Information integration for constructing social statistics: history, theory and ideas towards a research programme
Author(s) -
Judson D. H.
Publication year - 2007
Publication title -
journal of the royal statistical society: series a (statistics in society)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.103
H-Index - 84
eISSN - 1467-985X
pISSN - 0964-1998
DOI - 10.1111/j.1467-985x.2007.00472.x
Subject(s) - small area estimation , data science , poverty , data quality , government (linguistics) , unemployment , geographic information system , quality (philosophy) , computer science , geography , business , estimation , economic growth , economics , marketing , metric (unit) , linguistics , philosophy , remote sensing , management , epistemology
Summary. More precise policy making at all levels of government has fuelled tremendous demand for small area data—smaller than ever before. At the same time, there has been an unprecedented accumulation of data in geographic information systems, administrative records databases and more sophisticated survey sampling schemes. Researchers and practitioners have been trying to combine these diverse sources of data. But how should these diverse sources of data be combined in a way that is policy relevant and statistically principled? The paper illustrates these questions with several example applications at the state, county and local level: emerging geographic information systems databases, the need for estimates of small area income, poverty, demographic and uninsurance data by health authorities, and how administrative records databases (such as licensed day care facilities, traffic counts and unemployment insurance records) are being harvested for their information content. Finally, the paper proposes approaches for integrating these diverse sources of data with different error, uncertainty and quality profiles, and surveys persistent challenges in this area.