1 / 16

Confidentiality and statistics on grids

Confidentiality and statistics on grids. A proposal on common rules for handling confidentiality to the Board of Confidentiality at Statistics Norway. The European Forum for Geostatistics workshop in Haag, Netherlands 5th – 7th of October 2009

Download Presentation

Confidentiality and statistics on grids

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Confidentiality and statistics on grids A proposal on common rules for handling confidentiality to the Board of Confidentiality at Statistics Norway The European Forum for Geostatistics workshop in Haag, Netherlands 5th – 7th of October 2009 “bridge the gap” between theory and practice in GeoStatistics Session 1: Small area statistics Bjørn Thorsdalen Population Statistics Statistics Norway Otervegen 23 N - 2225 Kongsvinger Tel : ++47 / 62 88 50 62 Fax : ++47 / 62 88 50 97 E-mail : vvh@ssb.no Vilni Verner Holst Bloch MSc. landscape ecology and natural resources Statistics Norway Otervegen 23 N - 2225 Kongsvinger Tel : ++47 / 62 88 50 62 Fax : ++47 / 62 88 50 97 E-mail : vvh@ssb.no

  2. Overview of the presentation • Background • System of grids for national statistics • Examples on confidentiality issues • Different confidentiality rules • Examples on use of todays confidentiality rules • Guidelines for grid statistics • Further work

  3. Background • A) Requests from users(insurance companies, scientists, companies with localisation or marketing issues, general public, education puposes) • B) Internal drive within Statistics Norway (coming GIS and censuses) • C) Partnership in National INSPIRE Forum (obligations) • D) New possibilities(better presentation of spatial statistics, spatial analysis etc) The more users need, and we produce, the more crucial common rules for confidentiality becomes

  4. External WMS/WFS providers ”Wall of confidentiality” WMS ssb.no F I R E W A L L Geodatabase ArcGIS coverages, shape files etc. Statistics Norway Norwegian Mapping and Cadastre Authority Local Copy Statbank Statistical base registers Statistics

  5. Statistical grids for Norway • Grid name Cell size Number of cells • SSB100m (1) 0.01 km2 35 000 000 cells • SSB125m (1) 0.01 km2 20 000 000 cells • SSB250m (1) 0.0625 km2 5 600 000 cells • SSB500m (1) 0.25 km2 1 400 000 cells • SSB1km 1 km2 350 000 cells • SSB5km 25 km2 15 000 cells • SSB10km 100 km2 5 000 cells • SSB25km (2) 625 km2 500 cells • SSB50km (2) 2 500 km2 150 cells • SSB100km (2) 10 000 km2 40 cells • SSB250km (2) 62 500 km2 10 cells • SSB500km (2) 250 000 km2 4 cells • Because of limitations in many software packages and for practical use, • these grids are recommended as grids with a county coverage. • (2) These grids might also cover sea territories. Number of cells refers to coverage of Norwegian mainland. • One has however to be aware of deviations in grid cell areas for regions remote from the Norwegian mainland and Svalbard. http://www.ssb.no/english/subjects/01/90/doc_200909_en/doc_200909_en.pdf

  6. Confidentiality examples Number of farms. 1x1km. 1999 1 – 3 farms4 or more farms

  7. Confidentiality examples Building stock. 100x100m. 2007. 1 – 3 buildings4 or more buildings

  8. Confidentiality examples Night time population. 1x1km. Year 2000 over 2008. 1 – 9 persons 2000 10 or more 2000 1 – 9 persons 2008 (new settlements) 10 or more 2008 (new settlements)

  9. Confidentiality examples Number of enterprises. 1x1km. 2008 1 – 3 enterprises 4 or more enterprises

  10. Confidentiality examples Leisure homes. 1x1km. 2008. 1 – 3 leisure homes4 or more leisure homes

  11. Confidentiality examples Night time population. 1x1km. Year 2008 over 2000. 1 – 9 persons 2000(abandond cell) 10 or more persons 2000(abandond cell)

  12. Confidentiality examples Number of grid cells and inhabitants by grid cell sizes. Per cent. Share of grid cells with less than N persons N > 101 N > 51 N > 31 N > 11 N > 4 N > 101 Share of persons in grid cells with less than N persons N > 51 N > 31 N > 11 N > 4

  13. Confidentiality examples Frequency of agricultural enterprises by 1x1 km grid cells.2008 Number of agricultural enterprises Grid cells by number of agricultural enterprises

  14. Recommondation given to the Board of Confidentiality at Statistics Norway • Previous and existing rules for handling confidentiality on grids are not adequate. • Confidentiality rules should be handled at lowest reasonable geographical level. • Official statistics should not be given at all geographical levels/grid sizes. • One should have a set of limits/treshold values dependent of the sensibility of the topics for statistics or quality of sources for statistics.

  15. Recommondation given to the Board of Confidentiality at Statistics Norway • The following has been recommended to the Board • Total figures (persons, enterprises, buildings, dwellings) and non-sensitive variables (age, sex, building type, NACE code) do not need to be anonymised. • Statistics on sensitive variables can be given if total figures exceed threshold values. Threshold value is to be set by responsible department for each statistics, dependending on quality, sensitivity and details. Threshold values are fixed to total figures of 10, 30 or 50. No further anonymisation is done. • Grid sizes of 125mx125m and 500mx500m shall not be used for official statistics.

  16. Further work • Work within the Geostat to make guidelines for handling confidentiality issues • Adoptation of European rules for handling confidentiality in grid statistics ? Thank you for your attention

More Related