Linked Historical Censuses
The historical microdata permit linking of individuals between census years for longitudinal analysis. Currently, seven U.S. census samples are linked to the full-count 1880 dataset; the Norway 1865, 1875, and 1900 censuses are linked to each other; the Sweden 1880, 1890, and 1900 censuses are linked to each other; and census files from 1851 and 1881 Great Britain are linked. The U.S. data include separate samples that link males, females, and couples between pairs of years. The Norway data link males and couples. The Sweden data link males and females. The Great Britain data link males. Each linked sample contains data on all linked individuals and records for other members of the linked person's household from both years. Currently, the linked datasets contain information on more than 7 million people at two points in time (9,000 in Great Britain, 180,000 in the U.S., 400,000 in Norway, and 7.1 million in Sweden).
Procedures for linking individuals in the different North Atlantic countries are similar. The basic methods were developed by linking the U.S. complete count 1880 census to the sample datasets from 1850 to 1930. These methods were then applied to other countries, adapting the approach to address differences in the information available for linking. The primary goal in constructing these longitudinal databases has been to minimize selection bias and maximize representativeness of the linked cases. This approach requires a very conservative linking strategy; consequently, the samples include only a fraction of the total population in any census year.
The detailed linking procedure discusses the method for creating links for the U.S. and the modifications necessary to create linked data for Norway and Sweden.