The importance of metadata in the use of administrative sources Workshop on statistical metadata (4-6 July Vienna) An Taelemans & Mieke Booghmans
Outline of the presentation I.Datawarehouse Labour Market & Social Protection II.Setting III.Development of metadata strategy IV.Benefits of metadata V.Conclusion
I.Datawarehouse LM & SP Collect and store administrative data of social security institutions Population and coverage rate Advantages: - Cost - Wide-ranging and detailled statistics - Linking administrative data
I.Datawarehouse LM & SP Derived variables Socio-economic position 1.Employed 2.Jobseeker 3.Professionally inactive 4.Other
Outline of the presentation I.Datawarehouse Labour Market & Social Protection II.Setting III.Development of metadata strategy IV.Benefits of metadata V.Conclusion
II.Setting Agora research project (Belgian Science Policy) Under the authority of Federal Public Service Social Security and Crossroads Bank for Social Security Carried out by Centre for Sociological Research and Policy Research Centre Work and Social Economy (K.U. Leuven)
II.Setting Problems of fragmented, inconsistent and insufficient metadata Create comprehensive and high quality metadata Three movements: - descriptive task - analysing task - task of evaluation
Outline of the presentation I.Datawarehouse Labour Market & Social Protection II.Setting III.Development of metadata strategy IV.Benefits of metadata V.Conclusion
III.Development Why are metadata important? What do we expect of these metadata? How should metadata be collected and stored?
Why? Administrative data were not originally compiled for statistical purposes Interpretability Data quality Harmonisation and comparability
What? Flexible Comprehensive and consistent Up-to-date Accessible
How? Need for efficient navigation and search Benchmarking exercise Variable centred metadata model
How? DWH GENERAL VARIABLES CLASSIFICATIONS DATASETS
How? Metadata items variables Name Removal date Abbreviation Schaansc Theme Self-employment Definition/description The date on which the association of a self-employed with the National Institute for the Social Security of the Self-Employed is ended. Derived variable? No Source (+ link to metadata dataset) National Institute for the Social Security of the Self-Employed (NISSE) Validity 1997-now Measuring level (+ link to codelist) Ordinal (dd.mm.yy) Specifications/remarks The deletion date is not always registered on time. As a consequence a number of persons are wrongly registered as self employed. Yearly this affects persons, or 2% of the total population of self-employed. Date last update
Outline of the presentation I.Datawarehouse Labour Market & Social Protection II.Setting III.Development of metadata strategy IV.Benefits of metadata V.Conclusion
IV.Benefits of metadata Average daily wage: complex and country specific Removal date: interpretation problems Job mobility: derived variable
Outline of the presentation I.Datawarehouse Labour Market & Social Protection II.Setting III.Development of metadata strategy IV.Benefits of metadata V.Conclusion
Metadata are indispensable for correct use and interpretation of administrative data But: they cannot ensure that administrative data are correctly used and interpreted Extensive (inter)national cooperation and exchange of best practises
Further information? An Taelemans Mieke Booghmans