Thematic Focal Person: Anne Goujon
Overall Vision
The laboratory will aim at collecting data on population by levels of educational attainment and literacy for the 20th century up to now including data mining, adjusting, reconstructing, validating, and analyzing. This will serve two main purposes: The recent estimate for the 21st century will provide the population basis for the WIC human capital projections; the estimates for earlier times will offer scholars a solid, consistent, and detailed base for studying the impact of education and literacy on most of the demographic, economic and development processes of the last century.
Specific work plans for mid-2011 to mid-2013
Estimates of projection base year population for projections to be published between 2012-2013
In 2010, IIASA and VID constructed a first set of population projections by levels of educational attainment (from 2000 to 2050) and reconstruction (from 2000 back to 1970), based on the collection of detailed data on levels of educational attainment by age and sex for 120 countries around the year 2000 (see W. Lutz, A. Goujon, S. K.C., W. Sanderson. 2007. Reconstruction of population by age, sex and level of educational attainment of 120 countries for 1970-2000. Vienna Yearbook of Population Research, vol. 2007, pp 193-235; Samir KC, Bilal Barakat, Anne Goujon, Vegard Skirbekk, Warren Sanderson, Wolfgang Lutz (2010). Projection of populations by level of educational attainment, age, and sex for 120 countries for 2005-2050. Demographic Research 22(15): 383-472. See also this link for detailed results: www.iiasa.ac.at/Research/POP/<wbr></wbr>Edu07FP/).
With the plan to extend these projections to some 180 countries, expand the number of education categories from 4 to 6, and update everything with the most recent census and surveys, the data collection exercise under this project will be very substantial. We will use all available data sources on education level, in particular, censuses provided by national statistical offices or IPUMS (Integrated Public Use Microdata Series), CELADE or UNESCO, and surveys such as Demographic and Health Surveys, Multiple Indicator Cluster Surveys, Labor Force Surveys, and the Living Standards Measurement Study. Activities under this collection exercise also include identifying education reforms and other changes that affected the definition of education categories, as well as validating data with other sources, historical and present.
Understanding the education data problem
Although a key indicator of individual and societal well-being appropriately used in much academic and policy oriented work as a proxy for socio economic development, the data on human capital are far from being consistent. As a derivative of our project listing and analyzing the data on levels of educational attainment coming from many surveys for many countries, we will document and analyze the difficulties in arriving at a clean dataset on levels of education. These difficulties include:
Inconsistencies between datasets for different years, that can be due for example to changes in the length of schooling or simply by the formulation and/or tabulation of the question on the level of education of the respondents in the questionnaire. • Inconsistencies between surveys that are mostly due to sampling errors that can lead to different results between surveys in the same country. • Inconsistencies between countries that are mostly due to the lack of a globally harmonized classification of education systems, despite the existence of country mappings according to ISCED categories.
Documenting these difficulties will demonstrate the importance of reconstructing and projecting levels of educational attainment to arrive at a database where most consistency and accuracy problems have been removed.
Construction of a database on levels of education and literacy
Historical understanding of the diffusion process of education in the population and of the contribution of educational attainment to the process of socio-economic development, including the demographic transition, requires detailed historical data on education. Under this project, we will study the possibility of partially gathering and reconstructing data on levels of educational attainment and literacy skills by age and sex for the 20th century for as many countries as possible. Several sources will be key: UNESCO, which has been the primary international source of data, especially since the 1960s, IPUMS, and other data sets such as population censuses and those assembled by individuals and research institutions,
e.g. Observatoire démographique et statistique de l’espace francophone (ODSEF) at Laval University. We will first check and compile data availability on population structure by levels of education and literacy (by age and sex). We will then decide on the design of the retrospective Human Capital Database which will be partially reconstructed by back¬projection.
