Title: A proposal to preserve, integrate and manage access to anonymized census samples of the Official Statistical Agencies of the Arab States in cooperation with the Arab Institute for Training and Research in Statistics * * * Robert McCaa, Professor of
1A proposal to preserve, integrate and manage
access to anonymized census samples of the
Official Statistical Agencies of the Arab States
in cooperation with the Arab Institute for
Training and Research in Statistics Robert
McCaa, Professor of Population HistoryUniversity
of Minnesota, rmccaa_at_umn.eduhttp//international.
ipums.org additional information
atwww.hist.umn.edu/rmccaa/ipums-global
2Invitation to participate http//international.i
pums.orgA global collaboratory to (1) preserve
microdata and documentation,(2) integrate
microdata and metadata(3) and manage access to
anonymized, integrated samples for researchers
and policy makersregardless of country of birth,
residence or citizenshipwithout cost.Funded by
National Science Foundation National
Institutes of Health (USA)
3IPUMS-International (ISI Lisbon, 2007) dark
green disseminating (26 countries, 80 samples,
200mpr)green harmonizing (32 countries, 94
samples, 185mpr)lightest green negotiating
ISI 2007--21 countries at IPUMS workshop
Mollweide projection
4http//international.ipums.org
Apply for access (see form and conditions of use)
Construct a custom-tailored request select
countries, years, sub-populations, variables
Examine integrated metadata (samples)
Study integrated documentation (variables)
Link to Official Statistical Agency home pages
5Who Uses the Microdata? Examples from China
- Affiliation (Mar. 2007) University professors
and graduate students from - China Center for Economic Research, Peking
University - Chinese Academy of Social Sciences
- Guanghua School of Management, Peking University
- Institute of Policy and Management, Chinese
Academy of Sciences - Institute of Population and Labor Economics,
Chinese Academy of Social Sciences - Jinan University
- Management and Administration Institute
- Nankai University
- Renmin University of China
- Sociology Department, Nanjing University
- The Institute of Population Research, Peking
University - Tsinghua University
- University of International Business and
Economics - University of Science and Technology of China,
etc.
61. Data recovery. Example Bangladesh Bureau of
Statistics (1981 census, 276 tapes to recover)
gt3,000 tapes recovered 1971 Germany1980
Mexico, Mali 76, Sudan 73and many more
72. International and chronological integration of
microdata and metadata methods and procedures
- Integrated microdata
- Challenge retain all significant detail,
integrate everything - Solution composite coding scheme (multiple
digits, each carries meaningthink ISCO) - Use international standards where appropriate
- Integrated metadata (documentation)
- Summarize and describe commonalities
- Explain unique characteristics
- Dynamically generate metadata according to needs
(countries, samples, variables) of individual
researcher using XML database
8Plan international and chronological integration
of census samples
- Preserve and integrate microdata from each census
round 2000, 1990, 1980, and earlier, if extant - The Minnesota Population Center
- licenses microdata for fee of US5,000 per census
- imposes confidentiality protections so that
samples may be supplied to bona-fide researchers
world-wide - integrates both microdata and metadata (2-3 years
of work) - disseminates, on a restricted-access basis, to
users free of charge - advises NSOs on users, uses and publications
- Note Original source files are never accessible
to researchers
93. Managing access to microdataThe
IPUMS-International Case Study
- Why is IPUMS a good practice?
- Transparency of access
- Fully anonymized, thorough documentation
- Integrated data available without charge
- Sanctions for violations of user license
- Cited as good practice by UNECE Task Force on
Microdata Access (2007) - Cited as best practice by Dennis Trewin (former
Australian Statistician and chair UNECE task
force on microdata access)after 3 day on-site
inspection (Nov 07) - What are IPUMS strengths?
- What are IPUMS benefits?
10Researcher Conditions of Use License
11Conditions of Use License (acceptance is
indicated by clicking each individual box)
12Managing access to microdataThe IPUMS case
study strengths
- What are IPUMS strengths?
- Recover historical microdata and documentation
- NSOs retain ownership may recall data if misused
- Metadata and microdata are fully integrated
- Security of data and anonymization measures
- Equitable access, without cost freedom of
inquiry - Sanctions are strong and clearly spelled out
- Sustained funding
- What are IPUMS benefits?
13Managing access to microdataThe IPUMS case
study benefits
- What are IPUMS benefits?
- Greater credibility of census operations
statistics - More stakeholders
- Better use of census data
- Empirically based policies
- Equitable access to census microdata for
researchers regardless of country of residence - Participation in a global research initiative
- License fee of US5,000 per census
14Participating Official Statistical Agencies
Country List (Nov. 2007, total 73 62 of world
pop.)
- Africa (14) Ethiopia, Ghana, Guinea (Conakry),
Kenya, Lesotho, Madagascar, Malawi, Mali,
Mauritius, Rwanda, Sierra Leone, South Africa,
Tanzania, Uganda - America (23) Argentina, Bolivia, Brazil, Canada,
Chile, Colombia, Costa Rica, Dominican Republic,
Ecuador, El Salvador, Guatemala, Haiti, Honduras,
Mexico, Nicaragua, Panama, Paraguay, Peru, Puerto
Rico, Saint Lucia, USA, Uruguay, Venezuela - Arab States (4) Egypt, Iraq, Palestine, Sudan
- Asia (15) Armenia, Bangladesh, Cambodia, China,
Fiji, Indonesia, Israel, Malaysia, Mongolia,
Nepal, Pakistan, Philippines, Thailand,
Turkmenistan, Vietnam - Europe (17) Austria, Belarus, Bulgaria, Czech
Republic, France, Germany, Greece, Hungary,
Italy, Netherlands, Portugal, Romania, Slovenia,
Spain, Switzerland, Turkey, United Kingdom
15Invitation to participate
- 1. Endorse Memorandum of Understanding
- 2. Recover pre-2000 census microdata and
documentation - Shipping and recovery costs at project expense
- 3. Integrate microdata and documentation
- Entrust microdata and documentation (forms,
instructions, codebooks) - Bill for license fee (5,000 per census
including pre-2000) - MPC anonymizes and integrates microdata sample
from each census with corresponding metadata - 4. In 2 years (before 57th ISI), we would propose
to - Recover all data from old tapes (that are
recoverable) - Integrate both microdata and metadata for the
selected samples and subset of variables - Launch integrated samples with 40 other
countries - Benefits of participation many new users more
comparative research much new knowledge
institutional credibility
16IPUMS-International, 57th ISI (Aug. 2009) dark
green disseminating (45 countries, 150 samples,
500mpr)green harmonizing (?? countries, ??
samples, ???mpr)lightest green negotiating
?
ISI 2009--be there with IPUMS!!
Mollweide projection
17(No Transcript)