A controlled donor imputation system for a one-number census |
| |
Authors: | Fiona Steele James Brown Ray Chambers |
| |
Institution: | Institute of Education, London, UK;and University of Southampton, UK |
| |
Abstract: | Summary. The 2001 UK census was a one-number census. An integral part of such a process has been the creation of a transparent census database that has been adjusted for the underenumeration in the 2001 census. The methodology for creating this database is based on a controlled donor imputation system that imputes individuals and households estimated to have been missed in the census. This paper describes this methodology and provides results from a statistical assessment of its performance using data that realistically simulate the census process. |
| |
Keywords: | Calibration Census coverage survey Census underenumeration Donor imputation Non-response modelling |
|