Canadian Census Data Project

Chuck Humphrey (humphrey@datalib.library.ualberta.ca)
Data Library, Rutherford North

More data has been released from the 1996 Canadian Census than from any previous census. However, this exciting news was accompanied with an announcement that the 1996 Census will result in fewer paper publications than earlier censuses. At first this may seem to be a contradiction. How can there be more information and yet fewer printed pages? The reason is that Statistics Canada made a major effort to provide census results in digital format.

Using a proprietary table format, Statistics Canada has generated hundreds of tables that can be displayed in a PC-based utility known as the Beyond 20/20 Data Browser. In conjunction with a web browser, the Beyond 20/20 software can be used as a helper application to view census tables over the campus network.

In a large and diverse institution such as the University of Alberta, the abundance of 1996 digital census data creates several challenges related to access. One challenge is to provide access to census data using regional criteria. This will entail provision of a user interface that permits the easy identification of desired census results with an easy method of retrieval. Another challenge is to deliver this information to the desktops of researchers, teachers, and students.

The new Sun SITE agreement between the University and Sun Microsystems will help solve access problems by providing a server capable of delivering census tables that range from a couple hundred kilobytes to eighty megabytes in size. The initial interface to this large collection of data will be a searchable title list on the web. By January 1999, all of the 1996 Census tables distributed by Statistics Canada through a post-secondary data subscription service known as DLI will be available to the University of Alberta. This same service provides a licensed copy of the Beyond 20/20 software for University members.

By the fall of 1999, a spatial interface will be introduced to allow searching for data at specific levels of geography. Clients will, for example, be able to identify quickly all of the census information for Edmonton and subsequently select one or more tables with Beyond 20/20 for use on their PC workstations.