My last post examined a tool for exploring current Census data and exporting it in an easy to use format. Now what about historical Census data? Not the data from a few decades ago – we’re talking about the really old stuff. Finding this type of historical Census data is notoriously difficult, more so than finding new data. Sifting through the Decennial Censuses that have been digitized is overwhelming for your average library user. Propriety services that offer access to some historical census data with added value, such as GeoLytics, are typically expensive and not always chronologically comprehensive. Fortunately for us, as is often the case, libraries fill the void between the unpolished raw data and the propriety systems that add costly value to this data.
The University of Virginia Library’s Historical Census Browser offers a way to interact with data from the U.S. Decennial Census 1790 to 1960. The original source of the data is the U.S. Census Bureau and was then compiled in an electronic format by the Inter-University Consortium for Political and Social Research (ICPSR), available here. The Historical Census Browser provides a nice way for your average library user to interact with this data. Users can build queries and generate tables to answer specific questions. The tool even provides a basic mapping function for one variable at the state level. Unfortunately, the tool does not provide a way to export the data. For users keen to manipulate this data, there is an extension for Google Chrome, DataMiner, which allows content to be scraped from web pages and converted into spreadsheets.
It would be great to see more libraries offering services similar to the Historical Census Browswer, built on publicly-available data and shared widely. After all, if the data are freely available, why should libraries pay gobs of money for proprietary services that add minimal value?