Various Samples of Anonymised Records (SARs) have been extracted from the 1991 and 2001 Censuses and these have provided an invaluable research resource. Three levels of microdata are planned to be extracted from the 2011 Census: a Teaching file; Safeguarded files and Virtual Microdata Laboratory (VML) files available in an ONS safe setting. The SARs complement census area data: by allowing the user to derive custom variables; to create versatile crosstabulations; and to use statistical models which, for example, help avoid the ecological fallacy (Norman & Boyle 2010). Whilst large and complex files, the SARs microdata are not so much of a challenge to the user as the Office for National Statistics (ONS) Longitudinal Study for England & Wales (and the Scotland and Northern Ireland equivalents) and allow a study coverage of more than just England & Wales, or Scotland or Northern Ireland. Other datasets which a user might access in census years and during inter- and post-censal years are area data from administrative records and large scale government surveys such as the Labour Force Survey (LFS). The SARs though are a much larger sample than a source such as the LFS and cover both household and institutional residents. The SARs have underpinned a wide range of research which includes: Boyle et al. (2002), Norman & Purdam (2013) Norman (2008) and Norman et al. (2014). The Office for National Statistics (ONS) has devised a 2011 Census Microdata Teaching File to help users learn how to use microdata and prepare initial analyses for the use of the other SAR releases. This document first steps through the use of a ‘customised’ version of the 2011 Census Microdata Teaching File and provides SPSS syntax to first read in the original file to SPSS and label the variables and then to revise some of the variables to improve their utility in the context here.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)