CensusMapper
The tongfen R package is now on CRAN, so it’s time for an overview post. Tongfen has changed a bit since it’s inception and is now a lot more flexible but slightly more abstract to use.
What is tongfen? Tongfen, 通分 in Chinese, generally denotes the process of bringing two fractions onto the least common denominator. This is akin to the problem of making data on different but congruent geographies comparable by finding a least common geography.
Canadian 1996 census data is now avaiable on CensusMapper for anyone to make maps, for API access and via the {cancensus} R package. Yay!
The geographic data is not freely available from Statistics Canada, but can be custom ordered (via a small processing fee). Now the data is freely available on CensusMapper. The geographic data is slightly processed, we clipped out water areas and geographies from CSD upward are slightly simplified for better mapping performance as usual on CensusMapper.
Geographic data often comes on different geographic breakdowns. A prime example is census data, where the underlying census geographies can change from census year to census year. This makes it difficult to compare census data across censuses. But comparing census data across censuses at fine geographies is important for many applications.
There are two main ways how people deal with this problem. 1. Estimate data for one of the two geographies by (usually at some point) relying on area-weighted interpolation.
Recently we added a cross-tabulation of Structural type by Document type to CensusMapper for mapping and API use, including in our {cancensus} R package. Today we added another datasets to this.
Taxfiler data at the census tract level for tax years 2000 through 2017 Both of these datasets come through a project we are currently doing with CMHC, and we are excited to be able to turn these into open data, free for anyone to use.
To help deal with the response effort in the recent Newfoundland and Labrador show storm, StatCan created vulnerability maps of the most affected areas. It’s great to see StatCan putting their data to use. After seeing this fly by on twitter and then flagged by Simon on Linkedin I had some thoughts on this that might be worth a quick blog post.
I am biased in how I think of these kind of disaster response maps via the tools that I have developed for working with StatCan census data, namely CensusMapper interactive mapping platform and the the cancensus R package.
