Statistics Canada
www.statcan.gc.ca
Common menu bar links
How these files were created
Archived Content
Information identified as archived on the Web is for reference, research or recordkeeping purposes. It has not been altered or updated after the date of archiving. Web pages that are archived on the Web are not subject to the Government of Canada Web Standards. As per the Communications Policy of the Government of Canada, you can request alternate formats on the "Contact Us" page.
These microdata files were created from the 1991 Census Public Use Microdata Files (PUMF) for Individuals. This original file consists of 809,654 records representing a 3% sample of persons enumerated in the 1991 Census. Since this massive amount of information is difficult for schools to work with, smaller representative systematic samples were extracted to produce one national and 10 provincial files containing from 800 to 1,000 records. A 12th data file was produced containing data from the Yukon and Northwest Territories.
In order to create smaller files that could be easily handled by statistical software packages in schools, a systematic random sample was drawn. To ensure the resulting files were representative of the populations under study, the original file was sorted by the following fields in sequence: province, Census Metropolitan Area, household size, sex, age and ethnic origin.
For each file, R (the skip value) is calculated based on the original number of records in the PUMF for that area divided by 1,000 and rounded up to the next integer. A random number generator was assigned to select a start point and we selected every Rth record after that, so that we ended up with less than 1,000 records in each file. Since the random number calculation was performed independently for the French and English language versions of the files, some of the English and French language microdata files differ by 1 in size.
