An Evaluation of the Age and Sex Data from the Census Population of Canada, Provinces and Territories, 1971 to 2001 BY Shirley Loh, Ravi Verma and Margaret Michalowski Prepared for presentation at the Canadian Population Society Annual Meetings, Winnipeg, Manitoba, June This paper reports the results of research and analysis undertaken by Demography Division, Statistics Canada staff. It reflects the opinions of the authors and, not necessarily, those of Statistics Canada and is released to inform and to encourage discussion.
Outline of Presentation Objectives & Review of literature Methodology: Pyramid, Digit Preference Index, Age-Sex Accuracy index Source of Data Results Conclusion and Discussion
Objectives and Review of literature Objectives: To investigate if age heaping is present in single year age data on population To examine the overall accuracy of population data by age and sex To compare the level of accuracy of age-sex population with some countries Review of Literature: Norland, Litven and Cottrell (1977) Bourbeau and Lebel (2000) Poston and associates (2003)
Methodology Graphical presentation of population by age and sex Computing Myers’ indices for detecting digit preference in single year age reporting Computing age-sex accuracy index (U.N. Method) based on 5 year age group
Myers’ Index (unadjusted)
U.N. Method for Age-Sex Accuracy (Unadjusted, Canada, 1971) Criteria of the U.N. INDEX Under 20 → Accurate 20 to 40 → Inaccurate Over 40 → Highly Inaccurate
Source of Data Census population by age and sex, adjusted for net census undercount,
Results: Myers' Index of Age Heaping by Sex, Unadjusted Census Population, Canada (Age Interval 20-69)
Results (cont’d): Myers’ Index of Age Heaping, Canada, Provinces and Territories Age Interval: Canada Newfoundland Prince Edward Island Nova Scotia New Brunswick Quebec Ontario Manitoba Saskatchewan Alberta British Columbia Yukon Northwest Territories
Results (cont’d): United Nations Age-sex Accuracy Index, Canada, Provinces and Territories Canada N.L P. E. I N. S N. B Que Ont Man Sask Alta B. C Yukon N. W. T : Accurate : Inaccurate : Highly Inaccurate
Limitations of the UN Age-sex Accuracy Index (Siegel and Swanson, 2004) Central age group not included in the calculation of the age ratio Expected decline in sex ratio with increasing age not considered Considerable weight is given to the sex ratio component in the formula
Results (cont’d): Revised Age-Sex Accuracy Index, Canada, Provinces and Territories Canada N.L P. E. I N. S N. B Que Ont Man Sask Alta B. C Yukon N. W. T : Accurate : Inaccurate : Highly Inaccurate
Results (cont’d): Age-Sex Accuracy Index for Canada in 2001 and for 42 Countries of the World in 2000 * based on unadjusted population
Conclusion Myers’ Indices: Single Year Age Data No preference or avoidance of any year of birth Age-sex Accuracy Indices: Five Year Age Group Data Accurate age-sex population data for Canada and all provinces except Prince Edward Island, Yukon and Northwest Territories The quality of Canadian age-sex population data is highly comparable to Sweden and Finland Unadjusted Census Population – Myers’ indices and age-sex accuracy indices slightly higher than those based on adjusted census population
Discussion No explanation provided by the United Nations for the classification of the three categories Did the introduction of question on date of birth in 1971 improve the quality of age-heaping?