Today I am happy to announce the results from the R Shapefile Contest.
The contest was an incredible success – there were 19 entries that covered a range of topics. Each entry was well thought out, and I encourage you to read each of them.
Here are the entries, in order of submission:
[content_upgrade cu_id=”3205″]Bonus: Get all the entries as a PDF![content_upgrade_button]Click Here[/content_upgrade_button][/content_upgrade]
Please join me in thanking each of the entrants!
Goals of the Contest
As a reminder, the goal of the contest was to “do something in R, with a shapefile, that does something other than make a choropleth map”. This goal was entirely selfish: I have spent years analyzing data using choropleth maps. But as I don’t have a background in geospatial statistics, I am really not aware of what other analytical techniques I can be using. I hoped that by running a contest I could learn some more useful techniques that I could then apply to my own work.
And the winner is …
There are actually two winners to the contest. They both provided concise explanations, and real-world demonstrations, of geospatial concepts that I was simply not aware of.
- Spatial neighbors in R – an interactive illustration by Kyle Walker. Kyle is a geography professor. This might have allowed him to intuitively understand the types of analyses that I was looking for. His entry demonstrates different definitions of “neighbor” in spatial statistics, and how those definitions can effect interpretations of the data.
- London Crime Analysis by Henry Partridge goes a step further. Henry developed an application to map different types of crime in London. He then used Moran’s I to calculate spatial autocorrelation. There were actually several entries that deal with mapping crime, but only Henry’s entry introduced this extra step beyond a choropleth maps.
It’s worth pointing out that both of the winning entries used RStudio’s Shiny framework.
Honorable Mention
Several entries besides the winners stood out as teaching me something new in the area of R and shapefiles in a concise, enjoyable way:
- Hong Kong Population Center of Gravity (COG) by Fung Yip taught me about the concept of a Center of Population.
- In Working with Shapefiles Dennis Chandler used shapefiles to explore US historical boundaries, from 1629 to 2000. I did not know that such data existed!
- In Washington, DC Parking Violations Andrew Breza contrasts several different visualization techniques for analyzing parking violation data in Washington, DC. Before reading this I would have used just a choropleth to visualize the data. I learned a lot of new techniques from this!
- In Marine Boundaries in R: Reading EEZ Shapefiles Daniel M. Palacios give a thorough treatment of a real-world issue involving geography, marine data and national borders. He is clearly an expert in this field, and I enjoyed learning about his speciality.
Prizes
As a reminder, both of the winners will get two prizes:
- A free copy of my course Mapmaking in R with Choroplethr ($99 value) and
- A free copy of my course Shapefiles for R Programmers ($99 value).
I will be in touch with the winners today about how to get their copies of the courses.
[content_upgrade cu_id=”3205″]Bonus: Get all the entries as a PDF![content_upgrade_button]Click Here[/content_upgrade_button][/content_upgrade]