Explorations of multi-level methods & ecological inference techniques in the analysis of “Life Courses in Context” Peter Doorn & Luuk Schreven

Slides:



Advertisements
Similar presentations
An Introduction to the UK Data Archive and the Economic and Social Data Service November 2007 Jack Kneeshaw, UKDA.
Advertisements

Transitions from independent to supported environments in England and Wales: examining trends and differentials using the ONS Longitudinal Study Emily.
Will 2011 be the last Census of its kind in England and Wales? Roma Chappell, Programme Director Beyond 2011 Office for National Statistics, July 2011.
The Samples of Anonymised Records: Understanding Individual differences Mark Brown.
The Census Area Statistics Myles Gould Understanding area-level inequality & change.
MICHAEL J DENIS, PO BOX 125, PARKSVILLE, KY Kentucky Vital Records.
A comparison of the characteristics of childless women and mothers in the ONS Longitudinal Study Simon Whitworth Martina Portanti Office for National Statistics.
Kees Mandemakers Les Grandes Bases de Données et l’Histoire Sociale des Populations Bordeaux, Université Michel de Montagne, 7-9 th of February 2008 Development,
Chris Dibben University of Edinburgh Linking historical administrative data.
Bosna i Hercegovina Agencija za statistiku Bosne i Hercegovine Bosna i Hercegovina Agencija za statistiku Bosne i Hercegovine Post-enumeration Survey-A.
Counting the Dutch, The Future of the Virtual Census in the Netherlands Presentation at the seminar Counting the 7 Billion 24 February 2012 * Geert Bruinooge.
REPUBLIC OF TURKEY TURKISH STATISTICAL INSTITUTE TurkStat Population and Demography Statistics Department Population and Migration Statistics Team
Alternative Ways of Presenting Historical Census Data Luuk Schreven & Anouk de Rijk &
Historical Censuses; Numbers from the Dutch Providing access to the Dutch Population census of 1971 drs. L.J.G. Schreven.
The Dutch Censuses of 1960, 1971 and 2001 Producing public use files in the IPUMS project Wijnand Advokaat Statistics Netherlands Division Social and Spatial.
REPUBLIC OF RWANDA National Institute of Statistics Prepared by Emmanuel GATERA National Institute of Statistics of Rwanda Management Information Systems.
Searching the University of Alberta Library’s Statistics Canada-based Websites 2001 Census of Canada Canadian Centre for Justice Statistics Canadian Business.
Competency with the Census E. Turner CSU Northridge FOR MORE INFO...
The ONS Longitudinal Study. © London School of Hygiene and Tropical Medicine The Office for National Statistics Longitudinal Study (LS) o What is it o.
Census 2001 Your window to Census information. What is a Census? The Census of population and housing is undertaken every 5 years by the ABS. It aims.
Digitizing Dutch Censuses Preliminary results & work in progress Luuk Schreven Netherlands Institute for Scientific.
Census Bureau – Fernando Casimiro, Coordinator Lisboa IPUMS - Portugal Country Report.
The Northern Ireland Longitudinal Study: An Introduction.
U.S. Census Bureau Demographic Census 2000 July 8, 2003.
Nigel James Bodleian Library The Census Accessing and mapping British Census Data.
11 The American Community Survey Steve Murdock, Ph.D. Director, Hobby Center for the Study of Texas Rice University.
Liesl Eathington Iowa Community Indicators Program Iowa State University October 2014.
U.S. Decennial Census Finding and Accessing Data Summer Durrant October 20, 2014 Data & Geographical Information Librarian Research Data Services
Resources for International Comparative Analysis: The European Social Survey ESRC Research Methods Festival, St Catherine's College, Oxford, 02 July 2008.
Dutch Virtual Census Presentation at the International Seminar on Population and Housing Censuses; Beyond the 2010 Round November, 2012 Egon Gerards,
Mapping South 4 th Street to Fabric Row Unit #2 - Urbanization.
The ACS and the 2010 Census Richard Lycan and Charles Rynerson Population Research Center Portland State University GIS in Action March, 2011.
Using IPUMS.org Katie Genadek Minnesota Population Center University of Minnesota The IPUMS projects are funded by the National Science.
General Register Office for S C O T L A N D information about Scotland's people General Register Office for Scotland “Information about Scotland’s people”
Record matching for census purposes in the Netherlands Eric Schulte Nordholt Senior researcher and project leader of the Census Statistics Netherlands.
1 Sources of gender statistics Angela Me UNECE Statistics Division.
POPULATION AND HOUSING CENSUSES IN SLOVAKIA ON THE WEBSITE Miroslav Hudec Pavol Büchler INFOSTAT – Bratislava MSIS Geneva
Population census micro data for research: the case of Slovenia Danilo Dolenc Statistical Office of the Republic of Slovenia Ljubljana, First Regional.
The Dutch Virtual Census based on registers and already existing surveys Eric Schulte Nordholt Senior researcher and project leader of the Census Statistics.
Simfirms Leo van Wissen University Groningen & Corina Huisman Netherlands Interdisciplinary Demographic Institute NIDI The Hague.
The Dutch Virtual Census of 2001 A New Approach by Combining Different Sources Eric Schulte Nordholt ECE Census meetings Geneva, November 2004.
United Nations Regional Seminar on Census Data Dissemination and Spatial Analysis Amman - Jordan 16 – 19 May 2011 Determination of the scope and form of.
The Role of Metadata in Census Data Dissemination Presented By Mrs. Shirley Christian-Maharaj Assistant Director of Statistics CSO Trinidad &Tobago.
Statistical data confidentiality and micro data in Albania
The availability of Dutch census microdata Eric Schulte Nordholt Senior researcher and project leader of the Census Statistics Netherlands Division Social.
1 Dissemination Michael J. Levin Harvard Center for Population and Development Studies
Creating Open Data whilst maintaining confidentiality Philip Lowthian, Caroline Tudor Office for National Statistics 1.
Access to microdata in the Netherlands: from a cold war to co-operation projects Eric Schulte Nordholt Senior researcher and project leader of the Census.
The Integrated Public Use Microdata Series database IPUMSwww.ipums.org Lab 1 Background on the IPUMS and SPSS.
UN Workshop for South Asian Countries on Collection and Dissemination of Socio-economic Data from Population and Housing Censuses May 2012 Rudra.
United Nations Workshop on Principles and Recommendations for a Vital Statistics System, Revision 3, for African English-speaking countries Addis Ababa,
Workshop on Collection and Dissemination of Socio-economic Data from Population and Housing Censuses New Delhi, India, May 2012 United Nations Demographic.
Demographic models Lecture 2. Stages and steps of modeling. Demographic groups, processes, structures, states. Processes: fertility, mortality, marriages,
REPUBLIC OF TURKEY TURKISH STATISTICAL INSTITUTE TurkStat Demography Statistics Department Population and Migration Statistics Group EXPERIENCES.
Determination of Census topics and tabulations Census of Population and Housing 2012 Sri Lanka.
Integrated Public Use Microdata Series IPUMSwww.ipums.org.
Methodology of estimating the annual number of usual resident population in Latvia Baiba Zukula Deputy Director of Social Statistics Department Central.
Samples of Anonymised Records from the U.K. Census 1991 and 2001 Integrating Census Microdata Workshop Barcelona th July 2005 Dr. Ed Fieldhouse Cathie.
The Population and Housing Census in Switzerland Marcel Heiniger, Swiss Federal Statistical Office IPUMS/IECM Workshop “Integrating European Census Microdata.
Census developments in the Netherlands
Statistics Netherlands Division Social and Spatial Statistics
Population and Housing Topics -2021
The European Statistical Training Programme (ESTP)
Working Group on Population and Housing Censuses
A review of the 2011 census round in the EU, including the successful implementation of a detailed European legal base First meeting of the Technical Coordination.
Plans for the 2021 Population and Housing Census
Key Considerations for Planning and Management of Census Operations
Technical Coordination Group, Zagreb, Croatia, 26 January 2018
Chapter 5: The analysis of nonresponse
Key Considerations for Planning and Management of Census Operations
Presentation transcript:

Explorations of multi-level methods & ecological inference techniques in the analysis of “Life Courses in Context” Peter Doorn & Luuk Schreven & Data Archiving & Networked Services (DANS) Netherlands Institute for Scientific Information Services (NIWI)

Structure of presentation 1.Introduction to “Life Courses in Context”-project –Life Courses: Historical Sample of the Netherlands (HSN) –Context: Census digitization project 2.Exploration of multi-level methods & ecological inference techniques

Introduction to Life Courses in Context project Two separate components: –Life Courses: Historical Sample of the Netherlands (HSN) –Context: Digitization of (aggregate) Census data 1795 – 1971 One combined grant application to Netherlands Organisation for Scientific Research (+ € 3.6 mln funding)

Aim of Life Courses in Context project ‘…to develop a collaboratory for the study of 19 th and 20 th century population history.’ By combining the HSN and Census data sources: –HSN: Micro data + 40,000 individual life courses –Census: aggregate data from published census tables

Life Courses: Historical Sample of the Netherlands ‘…to construct life courses as completely as possible for a representative portion of the 19 th and 20 th century population in the Netherlands.’ A sample has been drawn from the birth registers of all persons born in the Netherlands between 1812 and 1922 (sample size = 77,000 persons) Data gathering by International Institute of Social History (IISH-IISG) since 1991

Sources of HSN (already mentioned) Birth registers basis of sample Person born, names, addresses, ages and occupations of parents (literacy of father) Death certificates Place of residence, age and occupation of deceased, information on his/her spouse. In case of child occupation and literacy of father Marriage certificates Occupations, place of residence and literacy of couple, parents and witnesses Dynamic population registration system (in use since 1850) & personal record cards (later stage) Family structure, pattern of migration Land registers & tax records (later stage) Occupational history and wealth of subject

Use of HSN 1.Basic resource for historical research in demography, sociology, epidemiology, socio-economics and social geography 2.Control database to compare research data 3.Foundation for the collection of new data 4.Source of expertise on data collecting Questions? Contact Kees Mandemakers

Context: Census Digitization ‘…to digitize all published (aggregate) census data from Dutch population, housing and occupational censuses between 1795 to 1971’ National population censuses are one of the fundamental sources of information on conditions in a country, used in historical and social science research Information on population size and structural characteristics: age, gender, marital status, religion, household status, occupational activity and nationality

Main objectives of Dutch censuses 1.To determine the size of the population on a fixed point in time 2.To probe and improve the reliability of the Dutch population registers 3.To examine the demographic and social-economic characteristics of the population 4.To provide data to facilitate domestic policy making

Census Digitization projects: 1997 – present : –Scanning 200 books, pages –Data-entry census March 2004: –Validation and correction of census data and 1930 –Digital archiving census 1960 and 1971 March 2003 – December 2005: –Life Courses in Context (see: –Data-entry of census data –Documentation, harmonization, access and research

What has been realized? New website up and running –Only in Dutch!  –Some 40,000 pages of tabular (aggregate) census data downloadable from website –Documentation is available –Validation and correction are partially complete –Harmonization schemes for certain census variables (restricted) Access to original micro data files for 1960 and 1971 census –Van Tubergen & Maas (2005)

Still to do… Finishing validation and correction Building harmonization schemes for census variables: –HISCO for harmonization of occupations –Standardizing sub municipal divisions –Harmonizing other variables and categories Better access to the data –Data not only as Excel spreadsheets –StatLine or Nesstar? Or other publication tool? Translation of the website to English!

Combining HSN & Census datasets Census covers whole population; check on data collected in sample Data sets are complementary; more data will be available HSN data are longitudinal; census data are cross-sectional snapshots Census data provide more regional detail Combining data can result in identification of individuals (privacy issues!)

Comparison of variables HSN micro data (birth, death and marriage registers) Census aggregate data Date (and hour) of birthAge (groups) Place of birth (municipality of birth certificate)Municipality of birth (nationality, ethnicity) Sex Date of marriageNA Place of marriageNA Marital status Occupational titleOccupation (-al group, sector) NAReligion AddressNeighborhood/municipality NACharacteristics of dwelling (housing censuses) Age of parents at birthNA Signature (proxy for illiteracy)Educational attainment Relationships to family members and witnessesPosition in Household Date (and hour) of deathNA

Combining data across levels of aggregation Historians have rarely tried to combine data from sources of unequal levels of aggregation Three approaches to combine data from the HSN and Censuses: 1.Aggregating individual data 2.Multi-level or cross level analysis 3.Disaggregating aggregate data

Aggregating individual data Most straightforward way of combining two sources Details of the individual will be lost Aggregating HSN data for cross- sections at census years is no easy task Censuses are not perfect; statistical deviations found can either be caused by HSN or by census

Multi-level analysis No actual linkage of records; in multi-level analysis the objective is to statistically explain a phenomenon in which higher levels of scale are included in the analysis Censuses provide background variables not available in HSN; whereas HSN contains individual detail not found in census tables In analyses at the individual level, ecological effects of higher levels may be taken in consideration

Reconstruction of individual records from aggregate tables Statistical Disclosure Control & synthetic estimation methods –Prevent identification of individual entities from aggregate data –Synthetic estimation methods can be used to reconstruct synthetic individual records from detailed census tables Ecological inference –‘…is the process of extracting clues about individual behaviour from information reported at the group or aggregate level’ –Difficult technique, it remains a challenge to apply it to the Dutch censuses Questions? Contact Peter Doorn

Conclusion and directions for future research This paper makes a plea for more interest by historians for the linkage of data from different levels of aggregation The next step is to elaborate on the approaches described in this paper empirically Data and techniques are available, we need a researcher who wants to take on the challenge

Contact Information dr. Peter Doorn Director Data Archiving & Networked Services drs. Luuk Schreven project coordinator Census digitization Paper available in electronic form from website

Population per municipality in 1795 Source: In 1795 Amsterdam is the biggest city with “souls” Klein-Waspik is the smallest hamlet with 3 inhabitants; a total of 1807 municipalities are mentioned in the census.

Boonstra’s NLKaart Dr. Boonstra’s NLKaart; –mid 1980’s onwards –first Historical GIS? –municipal boundaries between –first SAS/Graph based, later MapInfo

HGIN; a Historical Geographic Information System for the Netherlands Project goals: –Converting and correcting Boonstra’s NLKaart –digitizing maps with sub municipal boundaries ‘wijken’ (neighbourhoods) and ‘buurten’ (blocks) (1920 – 1971) –Setting up a gazetteer of historic places –Making the everything available on the web

HGIN details: technical stuff Scalable Vector Graphics Geoserver as basic geographical data server (OpenGIS) User friendly interface in NIWI’s Content Management Software: i- Tor see: or for working preview of the GIS.

HGIN details: results so far Testversion of mapping application is running at NIDI’s website ( and 1971 sub municipal maps available 1930, 1947 and 1956 maps are being digitized (outsourced) NLKaart converted to ArcGIS Work on gazetteer started

HGIN details: religion 1971 (provincial) Percentage Roman Catholic by province, Census 1971

HGIN details: religion 1971 (municipal) Percentage Roman Catholic by municipality, Census 1971

HGIN details: religion 1971 (submunicipal) Percentage Roman Catholic by block / neighbourhood, Census 1971