The Integrated Canadian-American Public Use Microdata Series (ICAPUMS) is an SPSS .sys file which comprises three Canadian and American census data sets. The ICAPUMS was created in 1995 and 1996 at the Minnesota Historical Census Projects by Lisa Y. Dillon as part of her doctoral research. In this file, the 1871 Canadian census public use sample has been integrated with similar samples of the 1850 and 1880 American censuses. The 1871 Canadian PUMS was created by Gordon Darroch and Michael Ornstein at the Institute of Social Research at York University (Toronto, Canada) in 1979. The American PUMS are drawn from the IPUMS files created by the Minnesota Historical Census Projects, University of Minnesota. The ICAPUMS was created to facilitate comparative study of the Canadian and American populations during the late nineteenth century. These data have been fully integrated by uniting the separate data files with a common set of comparably coded variables.

At present, the ICAPUMS is 271,019,808 Mb and contains information on over 724,000 individuals. The Canadian PUMS includes over 24,000 individuals as a base sample, and more if several forms of oversampling are employed in analysis. The American PUMS include almost 200,000 individuals in 1850 and over 500,000 in 1880.

At the moment, Dr. Dillon is updating the ICAPUMS to include the 1901 Canadian census data set and a PUMS of the 1900 U.S. census. She is conducting this work as a post-doctoral fellow with the Canadian Families Project, Department of History, University of Victoria. The 1901 Canadian census consists of about 265,000 individuals, while the 1900 U.S. PUMS contains 100,438 individuals. Eventually, the 1860 and 1870 U.S. PUMS will also be integrated into the ICAPUMS. Once this new phase of census data integration is complete at the end of May, 1998, the ICAPUMS will be available once more. See Dr. Dillon's Current Activities page for further information or contact her at dillon@hist.umn.edu. Further information can also be found in Lisa Y. Dillon, "Integrating Nineteenth-Century Canadian and American Census Data Sets," in Computers and the Humanities.