1 / 35

Imputation in the 2001 Census

Imputation in the 2001 Census. Robert Beatty NILS User Forum 11 December 2009. Coverage. How Census deals with Missing households Missing people within households Incomplete returns. Coverage. Census is statutory Census Act (Northern Ireland) 1969 Penalties for non-compliance

Download Presentation

Imputation in the 2001 Census

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. Imputation in the 2001 Census Robert Beatty NILS User Forum 11 December 2009

  2. Coverage • How Census deals with • Missing households • Missing people within households • Incomplete returns

  3. Coverage • Census is statutory • Census Act (Northern Ireland) 1969 • Penalties for non-compliance • Therefore counts everyone • Doesn’t it?

  4. Coverage

  5. Coverage

  6. Coverage - international • Australia 2006 – 96% coverage • Don’t impute but adjust MYEs • New Zealand 2006 – 95% response rate • NZ imputed for non-response, but only on 4 key variables • Canada ‘adjust for non-responding households’ – need to know about occupied households

  7. Adjustment issues • 1991 coverage – 98% • But inference about population? • Non-response not homogeneous • Young adults • Lower social class • Deprived areas

  8. Coverage - 2001 • Acknowledge under-enumeration • 1991 Census 1,578k MYE 1,607k • Decision to adjust Census 2001 database • Objective – all Census outputs to fully reflect whole population • ‘One Number Census’ • Census = MYE

  9. Coverage

  10. Coverage - 2001 • ‘One Number Census’ method • Basic principle to use a large-scale Census Coverage Survey (CCS) to estimate under-enumeration in sampled areas • Apply survey estimates elsewhere

  11. Census Coverage Survey • UK split into about 100 Estimation Areas (each about 0.5m population) • Three in Northern Ireland • About 200 postcodes / 3,000 households per Estimation Area • Three socio-economic strata within EA • Separate analysis in each strata within EA

  12. Census Coverage Survey • Fieldwork about 3 weeks after Census day • Face to face interviews • Trained interviewers • Given map of postcode boundary • Asked to re-enumerate the postcode • Short questionnaire - coverage

  13. Matching • Forms scanned into system • Special matching software developed • Database retrieval system • CCS returns carefully matched with Census returns – error rate estimated to be under 0.1 per cent

  14. Dual System Estimator (DSE) • Use matched Census and CCS data • DSE estimates adjustment for those missed in both Census and CCS Counted By CCS Yes No Counted Yes n11 n10 n1+ By Census No n01n00n0+ n+1n+0n++ DSE estimate for the area (under certain assumptions): n++= n1+ n+1  n11

  15. DSE : Simple ExampleFish pond • Day 1: Catch 950 fish, mark with a red dot. • Day 2: Catch 900 fish, mark with a blue dot. • Matched: 855 had blue and red dots. • Question – how many fish in the pond?

  16. Dual System Estimator (DSE) Counted Day 2 Yes No Counted Yes 855 95 950 Day 1 No 45n00n0+ 900n+0n++ DSE estimate of the actual number of fish: n++= 950  900  855 = 1,000

  17. Analysis • Separately for each age-sex group, within each stratum, within each EA • Apply DSE method to each sampling point (postcodes) within CCS area • Estimate function DSE = f(observed count) • Apply to all other sampling points within stratum (within EA), and aggregate

  18. Ratio Estimation • Regression-type estimator • Each dot represents a CCS area • Use Census figure to estimate “true” figure

  19. The One Number Census process

  20. Imputing households • Use dummy forms as location • Use dummy forms as ‘constraint’? • Dependence on enumerators • Ireland 2006 – 15% of properties vacant

  21. One Number Census outcome • 2001 Census response rate of 95% • 4.3% in wholly imputed households (mostly linked to dummy forms(3.0%)) • 0.4% additional people in already enumerated households • Imputed 80,000 people

  22. Coverage

  23. Response rates by age

  24. Quality of returns • So far, considered non-respondents • Person & Household imputation • What about quality of returns actually made? • Decision taken to go for ‘complete’ returns • Item imputation

  25. Edit and Impute - Edit • Limited number of ‘hard’ edits – can’t be married if aged under 16 • Larger number of ‘soft’ edits - quality

  26. Edit and Impute - Impute • General principle of ‘complete’ data set • No ‘Not stated’ entries in outputs • Item imputation used • Donor imputation system • No different in principle to systems used in sample surveys

  27. Edit and Impute - Impute • Level of item imputation differed by variable • Not applied to religion

  28. Summary • Objective in 2001 that Census outputs should reflect whole population • Person and household imputation • 5% of persons imputed • Complete records generated for all returns through ‘item’ imputation

  29. I told them in 1951 it was just you, me and the dog, but they keep coming back every 10 years to check.

  30. Looking forward • Date for your diaries … • 27 March 2011

  31. Any questions?

  32. Usual residence definition • Historical – present on night • Most countries now ‘usually resident’ • Definitions do exist (UN) • 2001 – self-assessed • 2011 – instructions • ‘Intention to stay’

More Related