1 / 16

Administrative data a dangerous gold mine in official statistics

Administrative data a dangerous gold mine in official statistics. Anton Färnström anton.farnstrom@bra.se The Swedish National Council for Crime Prevention 2014-06-03. Administrative data. Administrative authority. Statistical producer. The public.

byrd
Download Presentation

Administrative data a dangerous gold mine in official statistics

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. Administrative dataa dangerousgoldmine in official statistics Anton Färnström anton.farnstrom@bra.se The Swedish National Council for Crime Prevention 2014-06-03

  2. Administrative data Administrative authority Statistical producer The public • Data is not collected by the statistical producer • Data is generated for other purposes than statistics

  3. Swedish crime statistics • The Police decision codes: • Code 1: No investigation lead • Code 2: Nu result from investigation • … • Code 13: Free-text • …

  4. Swedish crime statistics Cleared offences in 2011

  5. Swedish crime statistics Cleared offences in 2011

  6. Swedish crime statistics Cleared offences

  7. Swedish crime statistics Cleared offences in 2011

  8. Swedish crime statistics Cleared offences in 2011

  9. Swedish crime statistics Cleared offences in 2011

  10. Swedish crime statistics • Other findings: • Different opinions on how to use the codes • Needs of registering personnel conflicting with statistical purpose • Vague classification • Insufficient instructions

  11. Swedish crime statistics Main results: Coding errors have resulted in large flaws in the downstream statistics The level of detail used in the statistics was not adapted to the quality of the raw data Results were similar to a previous quality study regarding reported offences

  12. Quality in register-based statistics Not enough to monitor indata in regular controls Not enough with close collaboration More in-depth quality controls are needed

  13. A simple model for assesingquality Management related indicators Operations related indicators

  14. Swedish crime statistics • Quality indicators – management • The quality of the classifications • The quality of the registration instructions • Routines to inform the statistical producer on changes • Common work groups with the statistical producer • Qualityindicators – operations • Variations between administrative units • Variations over time • Special attention to free-text options • Differing interpretations among registering personnel

  15. Thankyou! anton.farnstrom@bra.se

More Related