Imputation of Demographic Variables from the 2001 Canadian Census of Population

• Activity Number: 305
• Type: Topic Contributed
• Date/Time: Wednesday, August 14, 2002 : 10:30 AM to 12:20 PM
• Sponsor: Section on Survey Research Methods*
• Author(s): Patrick Mason*+ and Mike Bankier and Paul Poirier
• Affiliation(s): Statistics Canada and Statistics Canada and Statistics Canada
• Keywords: CANCEIS ; Edit and Imputation ; Demographic variables ; Canadian Census

Abstract

CANCEIS (CANadian Census Edit and Imputation System) is a generalized Edit and Imputation (E&I) system that will be used early in 2002 to process the demographic variables from the 2001 Census. Later in 2002, it will perform E&I on the labour, mobility, place of work, and mode of transport variables as well. It performs minimum change nearest neighbour imputation on a mixture of qualitative and quantitative variables. In our paper, the quality of the imputation of the demographic variables will be discussed. The advantages of processing on PCs (which is planned for the Census short forms), versus the mainframe (which is planned for the Census long forms), will be examined. The impact of modifying certain parameters in the distance measures that determine what are the optimal imputation actions will be reviewed. Finally, the operational feasibility and the organizational issues associated with the implementation of CANCEIS will be considered.