Preserving Census Microdata and Making Them Useful: IPUMS

1
Preserving Census Microdata and Making Them
Useful IPUMS Sudan
Prof. Awad Hag Ali Central Bureau of Statistics,
SudanProf. Robert McCaaUniversity of Minnesota,
USA

• First Arab Statistical ConferenceAmman,
  JordanNovember 12-13, 2007

2
The Problem 1973 census tapes at risk!
3
A Solution Data recovery
4
IPUMS-International Goals

1. Recover and preserve census microdata (raw
   data) and documentation
2. Integrate anonymized, high precision samples
3. Disseminate extracts of samples to researchers
   (trusted users)

UN Statistical Division Principles and
Recommendations (2007)

• Preserve census microdata and documentation
• Disseminate anonymized microdata
• Note No recommendation regarding prior censuses
• IPUMS pays to recover, document, and archive

5
First Population Census of Sudan 1955/56

• Conducted by the British
• Time span one year and a half
• Used traditional administration (chiefs of the
  tribes).
• Total population 10.1 million, adjusted to 10.3
  for under-count
• A landmark in global census history
• First African census to use key-punch cards.
• Thoroughly modern in design, execution, and
  evaluation

6
First Population Census of Sudan 1955/56
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
Population censuses of Sudan

1. 1956 - 10.3 millions (adjusted from 10.1)
2. 1973 - 14.8 millions (adjusted from 12.3 to
   account for Gezira camps and Nomads)
3. 1983 - 20,594,197
4. 1993 - 24,941,000
5. 2008 - ?? Since 1973 uniform
   methodologieslong short forms urban (long),
   rural (sample long short)

11
The Recovery process

1. Test 3 tapes by data recovery firm (Muller Media)
2. When tests were successful, ship remaining 32
   tapes
3. Due to US economic blockade, tapes were first
   shipped to a Spanish University, re-packaged then
   trans-shipped to Muller Media
4. 8 tapes were readily recovered 3 tapes were
   recovered with hand spooling 4 tapes were
   unrecoverable and 17 tapes were deemed earlier
   versions of recovered data total cost US3,000
5. Construction of database is now underway at the
   Minnesota Population Center

12
Feast in Khartoum
13
1973 Tapes recovered by IPUMS
14
(No Transcript)
15
May IPUMS help recover data for Arab States?
?
16
Beyond data recovery anonymized, integrated
samples for research

• IPUMS the Hubble Telescope project of
  statistics
• From 1999, funded by National Science Foundation
  of the USA, to
• Construct high precision samples
• Anonymize the microdata
• Integrate microdata and metadata
• Disseminate to researchers world-wide at no cost
• 70 official statistical agencies have endorsed
  project Memorandum of Understanding
• Samples for 80 censuses (26 countries), 200
  million person records are now available

17
IPUMS-Sudan

• 1956 complete documentation scanned
• 1973 recovery of 100 data by IPUMS
• 1983 recovery of 100 data by CBS
• 1993 archived on CDs
• Documentation for all censuses is translated
• In 2008 construct samples, anonymize and
  integrate
• In 2009 launch integrated samples for
  International Statistical Institute meeting
  (Durban, South Africa)

18
Recovered
19
Beyond data recovery anonymized, integrated
samples for research

• IPUMS the Hubble Telescope project of
  statistics
• From 1999, funded by National Science Foundation
  of the USA, to
• Construct high precision samples
• Anonymize the microdata
• Integrate microdata and metadata
• Disseminate to researchers world-wide at no cost
• 70 official statistical agencies have endorsed
  project Memorandum of Understanding
• Samples for 80 censuses (26 countries), 200
  million person records are now available

20
IPUMS Best practice Dennis Trewin
(Australian Statistician and chairman of UNECE
committee on Managing Microdata Access)

• Anonymization to protect statistical
  confidentiality
• Integration of microdata and metadata
• Dissemination to researchers fair and free
• --topics for a second presentation

21
Invitation to participation in IPUMSWhat needs
to be done?

• Official formalities
• Endorse memorandum of understanding
• MPC pays US5,000 per census for microdata,
  non-exclusive license, and documentation

• Unofficialeach statistical authority agrees
  to details
• What census microdata are to be recovered 197?,
  198?, 199?
• Calendar begin now? or in 2008?
• Historical censuses do microdata tapes or
  computer cards still exist?
• Can earlier microdata be recovered?
• IPUMS does the work and provides the funding

?
?
22
(No Transcript)
23
The Fifth Sudan Census

• According to article 215(1)
• of the Constitution
• A population census throughout the Sudan
  should be conducted and completed by the end of
  the second year of the interim period.
• Therefore this census is a constitutional one.

24
The Central Bureau of Statistics (CBS) and the
Southern Sudan Commission for Statistics and
Evaluation (SSCSE)

• According to article 215 (2) and based on the
  comprehensive peace agreement (CPA )the Census
  shall be conducted by the Central Bureau of
  Statistic and the Southern Sudan Commission for
  Statistics and Evaluation

25
Monitoring Quality assurance

• A committee for monitoring and quality assurance
  was established by the PCC in March 2006 .
  It consists of
• International Experts.
• National Experts.
• Civil organizations.
• Political parties.
• The committee is chaired by the states council.

26
Achievements -

• (1) A detailed and comprehensive project document
  has been prepared through project appraisal
  meetings attend by (CBS) and (SSCCSE) world
  bank, UNFPA, and other supporting bodies. The
  document consist of the followings -
• A detailed work plan.
• A detailed Budget with activities and time
  frame for implementation for each activity.

27

• (2) Two questionnaires has been designed-
• Short and long
• The short Questionnaire will be used for the
  full coverage of the population with a
  minimum number of questions (11 questions). A
  long questionnaire will be applied for 10
  percent of the population (EAS).
• (3) The Enumerators and supervisors manuals have
  also prepared.

28
(4) Cartographic Work was-

• Started in the early 2006 and continuing up to
  now. GPS is used for the accurate location of
  settlement and delineation of (EAS). It is
  also used for allocating service centers.
  Almost all cartographic work have been finished
  in the North except Darfour which in a half way.
  The south is also in a half way. We except to
  finish mapping by the end of September.

29
Enumeration phase -

• The actual enumeration will start in the
  second of feb.and finish by the15 of feb. 2008.
  (a proximally 2 weeks).
• Defacto method will be applied.

30
Decentralization -

• Decentralized data capturing will be followed
  for the first time on Sudan.
• Final data processing and tabulation will be
  performed centrally.

31
Data dissemination -

• National and regional tabulation will be
  analyzed and published using different data
  dissemination methods such as-
• Printed reports.
• Electronic media (websites, Emails).
• Data archiving.
• Seminars and workshops.

32
Challenges -

• Lack of Proper infrastructure and experienced
  manpower in South Sudan due to the war.
• Mines and insecurity in some parts of
  Sudan.
• Continues population movement to the South
  (returnees) from inside and out side the
  country.
• Insecurity in some areas in Darfur States and
  Equatoria.
• Problems concerning nomadic population and their
  continues movements in huge areas and across the
  boards.
• Lack of continues flows of Funding from both
  government and Donors.