CASA Working Paper 201
1 July 2015
A Synthetic Longitudinal Study for the United kingdom
In the United Kingdom, there exist three Census-based longitudinal datasets, known collectively as the Longitudinal Studies. The England and Wales Longitudinal Study (LS) is a 1% sample of the population of England and Wales It started with a sample from the 1971 Census and links the records of individuals (and other members of the sample member’s household) from the last five censuses. There are around 500,000 individuals present at each census, although not all of these will be linked for the full time-series. The LS also records key life events such as births, deaths, marriages and cancer registrations. Similar datasets of linked Census records exist for Scotland and Northern Ireland, although these data have a shorter time-series (currently linked back to 1991) and larger sample fractions (which constitute around 5% and 28% of their populations respectively at each Census). Whilst immensely valuable datasets for demographic research in the UK, all of the Longitudinal Studies are under-used when compared to other Census data products. Part of the reason for this is the restricted access researchers have to the microdata due to the potentially disclosive detail contained within. Consequently, in order to introduce potential researchers to the data and increase the user-base, a synthetic general-use version of the Longitudinal Studies is proposed.
Author: Adam Dennett, Paul Norman, Nicola Shelton and Rachel Stuchbury
Publication Date: 1 July 2015