Presentazione standard di PowerPoint

1
International Seminar on Population and Housing
Censuses Beyond the 2010 Round
5 Marzo 2007
Emerging methodologies OF Continuous use of
registers and geocoded databases In The Italian
population and housing census Fabio
Crescenzi, Methods and Techniques for
Censuses Istat (National Institute of
Statistics), Italy
Seoul, 27-29 November 2012
2
Pressure for a greater use of administrative
sources

• the need of frequently updated and georeferenced
  data
• Operational complexity of decennial census
  
• Pressure for a greater use of administrative
  sources

Combined census
Register based census
3
The 2011 Combined Census

• mail out using Municipal Population Registers
  (LAC),
• multi-mode data collection,
• field correction of coverage errors,
• short form/long form strategy in Municipalities
  with more than 20.000 inhabitants.

4
Multi-mode data collection

• Online To fill the questionnaire online, using
  the provided password (33,4)
• Post office to Return the paper questionnaire to
  any post office in Italy (22,6)
• Municipal Centre to return the paper
  questionnaire to one of the Municipal Collection
  Centres (31,7)
• Enumerator to return the paper questionnaire to
  enumerator (12,3 )

5
Web Monitoring and Enumerators field recovery

• Web census management system.
• A web management system was devoted to the census
  field work (SGR - survey management system).
• Enumerators recovery
• non-responders
• unregistered individuals
• The number of enumerators dropped from 108.000
  in 2001 to 68.000 in 2011 (-37).

6
Maximum use of administrative and statistical
data sources Rolling Sample Surveys to correct
coverage errors and to supplement
data Georeferenced census data every year
7
The quality of the sources

• Quality errors (Accuracy, Completeness, Update
  status, Coherence among sources, Accuracy of
  Geocoding)
• are a burden
• For the use
• of administrative sourcesfor census purposes

8
The equilibrium of Archimedes
9
ARCHIMEDE, Archive of Economic and Demo-social
Micro-data

• Infrastructure that will provide anonymised micro
  data to Public Administrations, researchers and
  users.Geo-referenced data up to the enumeration
  area level Longitudinal data to study the
  evolution over time

10
The force generated by the care of errors
(Archimede Burning Mirrors)
Sources on individuals and households
Integrated System of microdata
strength of care max
C-sample and D-sample survey
Georeferencing
errors
11
(No Transcript)
12
Give me a fulcrum and I will move the world ...
(Archimede)
13
The leverage effect of continuous census

• The cumulation of data collected by
  administrative sources and cared by sample
  annual survey
• can
• Produce data capable of satisfying the needs of a
  decennial census and the annual needs of users

14
Searching the best leverage

• Maximising the use of local administrative
  sources
• Increasing the care of errors
• Looking for the best design of sample surveys

15
In the rotation movement the revolution
16
C-sample and D-sample to correct coverage errors
and to supplement data
D-sample
C-sample

• The waves of two different sample surveys (the
  C-sample and the D-sample survey) will be used in
  crucial stages of a continuous process designed
  to achieve, separately, the two main goals of
  census operations
• i) correct the counting of usual residents to
  produce the key data on demographic structure of
  population and households
• ii) supplement socio economic data to produce
  census data hypercubes .

17
Georeferencing and mapping

• A key role is given to the ANSC, the
  georeferenced national register of streets and
  addresses
• Census mapping has to be made consistent with the
  requirements of a continuous census

18
C-sample

Starting date 2016
Questionnaire Short only data to count persons and households
Tecnique Paperless CAPI
Sampling design Capture Recapture on about 650.000 households per year
Monitoring WEB (SGR)
19
D-sample

Sampling design Two stage About 1.500.000 households per year
Questionnaire Long
Technique Paperless CAWI and CAPI
Monitoring WEB (SGR)
20
D-sample. Plan of data accumulations by output
area size
Output area size (inhab.) Number of years Pooling Interval Central year
gt100.000 1 t t
35.000-100.000 3 t, (t-1), (t-2) (t-1)
lt35.000 5 t, (t-1), (t-2), (t-3), (t-4) (t-2)
Depend on OUTPUT AREA SIZE

• Reference year of estimates

21
D-sample. Dissemination Starting Plan
Year of dissemination (31th December) Year of reference (1th of January) by size of the output area Year of reference (1th of January) by size of the output area Year of reference (1th of January) by size of the output area
Year of dissemination (31th December) gt100.000 35.000-100.000 lt35.000
2016 2016
2017 2017
2018 2018 2017
2019 2019 2018
2020 2020 2019 2018
2021 2021 2020 2019
2022 2022 2021 2020
2023 2023 2022 2021
2024 2024 2023 2022