Johan SURKYN and
Patrick DEBOOSERE
2002-2005
Abstract:
This project deals with the future of the national census, as the classical questionnaire-based census formula is currently under review in Belgium. Interface Demography is investigating the prospects of consulting administrative records as an alternative data source. In a broad interuniversity consortium, Interface Demography is project-coordinator and takes responsibility for the data on household dwelling characteristics. Starting from the results of the 2001 Belgian socio-economic survey (census), a relevant, consistent and up-to-date database 'household dwellings' is designed to be implemented at Statistics Belgium. In the first stage alternative data sources (administrative registers, surveys,...) containing relevant information are tracked down. Then, problems in using these alternative sources are investigated. Consequently a quality and feasibility study is undertaken. Finally, in close cooperation with Statistics Belgium, the final database is being conceptualized. Based on the research on the prospects of consulting administrative records as an alternative source of data for the population census, two lines of conclusions can be drawn. First, most classic census domains can potentially be covered using administrative data stemming from the respective departments of education, economy, social security, finance etc. Technically, the main challenge is to combine the plethora of rough information originating from different sources with large variations in scope, coverage, definition of concepts, timing etc. into a consistent and coherent statistical picture. However, it has become clear that the transition to using administrative data largely exceeds technical issues related to availability of data and problems of linking data-sources for statistical purposes. Institutional limitations related to the position of Statistics Belgium, the distribution of competences in the Belgian federal structure of the state and the need for a modernized privacy legislation are all obstacles in the way of a successful transition. Furthermore, international comparison indicates that an administrative census system is usually predicated on and located in a strong structure of administrative registers, such as population registers, social security registers, land and dwellings registers and registers of enterprises. Registers provide the clues for identifying census subjects and linking information stemming from different sources. In particular, the lack of a Belgian “Buildings and Dwellings Register” complicates proper identification and linkage of administrative records, hampering the use of administrative sources as an alternative for traditional census questionnaires on housing and living conditions.
< BACK