Census - Data Sources

Data Source

• The Leeds, Grenville and Lanark District Health Unit received the 1981 and 1986 Census from the Public Health Branch, Ministry of Health, through its Health Planning System (HELPS) initiative and the 1991 and 1996 Census data from the Information, Planning and Evaluation Branch of the Ministry of Health.
• The census is conducted once every 5 years to gather information about the social and economic situation of Canada.

Data Limitations

• Because a census must include all people in Canada as opposed to a selected sample, coverage errors may exist which affect the accuracy of the census counts.
• Undercoverage occurs when a person is not enumerated. Overcoverage occurs when persons are counted more than once or those not in the census population are counted (i.e., foreign visitors).
• Comparability of workplace location data (commuting) between censuses is affected by changes in geography and census subdivision, census division, census metropolitan area, census agglomeration and census tract boundaries. Therefore, caution should be used when comparing workplace location data between censuses.
• It is believed that censuses prior to 1996 have undercounted the number of persons with "no fixed workplace address". In 1996, the term "no fixed workplace address" replaced "no usual place of work".
• Evaluations of the 1991 Census data showed that respondents who did not go to a regular address to work, such as construction workers, were unclear how to answer the question.
• Comparability of ethnic origin data between the 1996 Census and previous censuses has been affected by several factors including changes in the questions format, working examples, instructions and data processing. The change in format to an open-ended question in 1996 is likely to affect response patterns. In addition, the presence of examples such as "Canadian", which were not included in previous censuses, may also affect response patterns.
• The number of multiple ethnic origin responses has grown with each census, which affects overall data comparability.