Home > Publications . Search All . Browse All . Country . Browse PSC Pubs . PSC Report Series

PSC In The News

RSS Feed icon

Smock cited in story on how low marriage rates may exacerbate marriage-status economic inequality

Shapiro says Americans' seemingly volatile spending pattern linked to 'sensible cash management'

Work of Cigolle, Ofstedal et al. cited in Forbes story on frailty risk among the elderly

Highlights

Sarah Burgard and former PSC trainee Jennifer Ailshire win ASA award for paper

James Jackson to be appointed to NSF's National Science Board

ISR's program in Society, Population, and Environment (SPE) focuses on social change and social issues worldwide.

McEniry and Schoeni host Conference on Long-run Impacts of Early Life Events

Next Brown Bag


PSC Brown Bags will return in the fall

George C. Alter photo

Completing Life Histories with Imputed Exit Dates: A Method for Historical Data from Passive Registration Systems

Publication Abstract

Alter, George C., Isabelle Devos, and Alison Kvetko. 2009. "Completing Life Histories with Imputed Exit Dates: A Method for Historical Data from Passive Registration Systems." Population, 64(2): 293-318.

Using family reconstitution methods, historical demographers are able to reconstruct life histories with data from vital records. In countries where they exist and where data are of good quality (notably in the Nordic countries, Belgium and the Netherlands), population registers can be used to describe the fertility and mortality of historical populations. A major problem with these data sources, however, is the under-reporting of information on migration. The presence of individuals remains certain so long as the events concerning them (or their family) are registered, but if no further information is recorded, there is no way of knowing whether they have migrated or not.

2 In this article, George Alter, Isabelle Devosand and Alison Kvetko estimate the migration probability of each subject "lost to observation" on the basis of his or her individual characteristics, then randomly impute an exit date. Bias due to missing data can thus be corrected by assigning an exit date to each subject. Classic event history analysis methods can then be used, as if individual records were complete up to the exit date.

3 Using a nineteenth-century population register of a Belgian municipality in which emigrants are identified and their exit date is known, the authors test their method and confirm its robustness, adding a simple and effective new instrument to the demographer's toolbox.

Country of focus: Belgium.

Browse | Search : All Pubs | Next