Data 101: Mapping and Internship 101: Following Directions

by Alexa Marzina

June 18, 2018

Starting a new job is always scary, especially when you’re thrown into helping with a data mapping class first thing in the morning and set off the emergency exit alarm in the building not once, but twice. The goal: Take the class and write about my personal experience. While I had plenty of anxiety about it, I was ready for the challenge.

The class was the Western Pennsylvania Regional Data Center’s Data 101: Mapping at the Carnegie Library of Pittsburgh in Beechview. Data 101 is a series of classes designed to build people’s skills and confidence in data literacy. The focus was learning about and visualizing different types of data mapping through four hands-on activities, which could later be replicated or expanded upon by using computer mapping software.

I was visibly excited when I found out that this workshop included not only colorful pencils and markers, but fun stickers as well. Obviously, I opted for the sparkly pink unicorn ones, and suddenly, data didn’t seem so scary.

While mapping data doesn’t require magic, having unicorns around always helps.

During the first activity, we learned about choropleth maps — maps that are colored, shaded or patterned based on a variable. In this case, the visualized example was the number of licensed dogs by county. Equipped with a map of eight made-up counties and the number of their licensed dogs, it was up to us to decide the color-coded scale to represent those data on a map.

Mapped data can tell different stories depending on how you represent them. I categorized my dog data into four bins, with no counties falling into the third bin. So, the story of my map was out of those eight counties, none of them had between 15 thousand and 19,999 licensed dogs. It’s not a very long story, but it doesn’t need to be!

I grouped my data using the arbitrary grouping method, as opposed to other more organized methods such as continuous, equal interval or data distribution groupings.

The second activity, though eight steps long and including serious orders from the city’s department of business planning, allowed me to create a beautifully vibrant birds-eye-view street map. The scenario for the workshop was a popular parking garage was closing and the department of business planning needed us to ensure that all the appropriate establishments got notified of the closure, specifically within 250 feet of the garage.

To do so, we used latitude and longitude coordinates — yes, like you learned in middle school social studies — to locate the garage, then used a pushpin and a string as a makeshift compass to draw a scaled 250-foot radius, or buffer, around it. Buffer technology is useful for things like finding the nearest coffee shop near you on your smartphone maps app. Turns out, we needed to notify six business about the garage closure.

Activity three taught us to be careful when using data, and that you can’t only use one type to fully visualize a problem. Given a list of properties and their location on the street, we had to determine which properties were vacant city property and who was eligible to buy them. After looking at the data carefully, we saw that there was one vacant property in the set that was actually owned by a homeowner already, so it was not eligible for sale. We realized that paying close attention to the data is always of the utmost importance to prevent mistakes.

While these don’t look like variables you’d typically need a map for, having a visual can be helpful to solve problems.

The final activity, which required us to fold a paper into a four-by-four grid, or a 16mo if you’ve ever studied bookmaking, to see which areas of a fictional town were at high-risk for spreading cooties. An important lesson about data is that while a lot of it is public knowledge, you never want to release data that could put a person, represented here as a point on a map, at risk for violence or other negative circumstances. In this case, if one of the grid squares had less than three cases of cooties, it couldn’t be represented in the final map to protect the individuals’ privacy.

Each remaining square with three or more cases was color coded based on the number of cases they had, creating a point density map. Though we as the mapmakers had access to each point, or case of cooties, in all areas, users of the final map would only be able to see the color-coded boxes and not each individual case.

Sometimes, grouping your data into different chunks leads to a wonky looking end result, but you can always try to break it up a different way.

While that all sounds like a lot, above all we learned that map making and data mapping is allowed to be fun and doesn’t have to be super serious or boring. Though there are some guidelines with using data, like prioritizing privacy protection and accessibility to those who may be colorblind or have other disabilities, there isn’t only one “right” way to map out a data set.

The next Data 101 is Finding Stories in Data Wednesday, July 25 at 10 a.m. at the Carnegie Library of Pittsburgh in Beechview.