Mind the map
Conversations with Data: #62
Do you want to receive Conversations with Data? Subscribe
Welcome back to the latest edition of our Conversations with Data newsletter.
In this episode of Conversations with Data, we spoke with science journalist and author Betsy Mason about how journalists can accurately use maps to help visualise stories. She also explained how bad design, geographical quirks and perceptual illusions can confuse the public, particularly when it comes to election maps.
What we asked
Tell us about your career. How did you first become interested in maps?
This is a question I often get and it is surprisingly hard to answer because I think I've just always had an affinity for maps. When I was little, I could spend hours poring over road atlas'. Before I became a science journalist, I was a geologist. I got a master's degree in Geology and I ended up working for an oil company in Houston for almost two years. While I was there, I learnt that there was this thing called science writing. So I applied to UC Santa Cruz’s graduate science communication programme and went there and switched careers and have really enjoyed it since.
How did this interest turn into you writing a book?
My obsession with cartography and maps began when I started writing about them at WIRED Magazine as a science editor. I started to blog about maps with my friend and colleague Greg Miller called "Map Lab". We blogged about maps and making maps. The idea was we would learn all about maps and about how to make maps in public and our readers would learn along with us. The more I learnt, the more I loved maps even more. We eventually moved that blog to National Geographic, where it's called All Over the Map, and then did a book together of the same name published in 2018.
Why do maps matter to humankind?
I do think that there's definitely something special about maps that uniquely draws us, humans, in. Maps give us an understanding of our place in the world. I believe having a representation of the world that we can place ourselves in is important. It helps us be less inward thinking, less focussed on ourselves and our immediate surroundings. It's about perspective.
In your book, you explain how our brains are built for maps to absorb visual stories. Tell us more about that.
I think that our brains relate to maps differently than to other data visualisations and that we more readily form emotional connections with maps. This is backed by both science and my experience of seeing how others interpret them. Maps are this perfect combination of constraint and creativity. They're based on a very real framework defined by actual geography, but they leave just enough space for creativity and artistry that hits the right spot for human brains.
You also focus on the history of maps in your book. What do maps tell us about the past and why do they matter?
I'd say maps are an important way to record history because they can be so specific to the time and place that they were made. They can reveal so much about the person who made them, their view of the world, what was important to them, what they left out. So they can record history in a way that I think has these sort of different angles that you don't get in other ways.
You wrote an article days before the US election entitled "Election maps are everywhere. Don't let them fool you". What was your perception of how news organisations used maps to communicate the election results?
Well, I thought that, as expected, the cable news stations that were on for hours and days on end with their magic wall maps -- which are all made with extremely bright reds and blues -- I could see the Chromeo stereotypes. I could see all the problems. We've got a situation where rural, less populated areas tend to vote Republican and therefore are coloured red and concentrated urban centres tend to be more Democratic and are coloured blue. So when you look at the map, it's just a sea of red with dotted blue Islands. But in this election, more people voted Democrat than Republican. That's not the way the map looks though. I think that's particularly problematic right now when we have people who are suggesting that the election was rigged and that there actually is more support for President Trump than there is because it does look like that on the map.
What about election maps published by other news organisations?
Print news organisations like The New York Times and The Washington Post do a lot better. They have different kinds of maps. They use a more even brightness of red and blue. They're not always shading the entire state or county in the colour of the winning party. They also do a thing called normalisation on vote share. So if a county or state is extremely Republican, it'll be a much darker red than if it just barely went Republican. That gives you a much better impression of where the country actually stands.
Those newspapers that I mentioned and some other publications have got really great cartographers on staff, and they're always innovating new ways to try to make more accurate representations of the election. But it's a really interesting and difficult problem. It is also true that when you're mapping by state and it's representing the Electoral College, that sort of unevenness is representative of the Electoral College system. So it's a more accurate representation of this weird way that we elect a president that isn't a direct, popular vote. So it doesn't really matter how Republican Utah or Wyoming are. It just matters that they're Republican and all the electoral votes go that way.
As far as journalists and their mapping skills go, do you ever see a story and wonder why a map wasn't included?
More often than not, I think "this should not have been a map". And that's one of the things that journalists should do is think about whether or not a map is necessary. Generally, there are two kinds of maps when it comes to journalism. There are maps that you use to display information or an idea, and then there are maps that are used basically with data journalism and GIS analysis to reveal the story and to find the patterns. And so obviously those kinds of maps are integral to the story.
What should journalists be mindful of when using maps in their stories?
Throughout history, maps have been used to convince people of things that were true and things that weren't true. And maybe it comes back to that connection with maps that I was talking about earlier. But I think people are very prone to believe things that they see on maps. If it's on a map, it just looks and feels more true. So one thing journalists need to keep in mind is that the same data set can be mapped many different ways that will show or convey a different message. And that's not to say that journalists should just pick the way to map the data that matches the story they're telling. The point is to use maps carefully and responsibly and understand that they're not just illustrations to break up the text.
Can you recommend any useful resources for journalists wanting to learn more?
It's a perfect time to to be a journalist who wants to make maps because it's more possible than it's ever been. For journalists who just want to start incorporating maps or maybe just playing around with maps, a lot of different resources exist out there. I can recommend two free online courses. Dr Anthony Robinson teaches one at Penn State University, and it's called Maps and the Geospatial Revolution. You'll learn about the basic principles of cartography, map design, how to use colour, and how to make maps using the software platform of Esri, which is the biggest mapping software company. There's also Alberto Cairo's course about data visualisation for journalists through the Knight Center. While that course generally focuses on data visualisation, it has a lot about mapping in there, too.
Latest from DataJournalism.com
- Julia Angwin, The Markup
- Matthew Kauffman, Solutions Journalism Network.
- Kiko Llaneras and Mariano Zafra, El Pais
- Karoline Albuquerque, Julliana de Melo Correia e Sá & Ciara Carvalho, Jornal do Commercio
Want to attend? Register here.
Our next conversation
In the next episode of our Conversations with Data podcast, we will speak with Shaida Badiee, the co-founder and executive director of Open Data Watch, a non-profit organisation working at the intersection of open data and official statistics. She will speak to us about the 2020 release of Open Data Inventory, which aims to assess the coverage and openness of official statistics to help identify gaps, promote open data policies, improve access, and encourage dialogue between national statistical offices and data users.
As always, don’t forget to let us know what you’d like us to feature in our future editions. You can also read all of our past editions here.
Tara from the EJC Data team,
P.S. Are you interested in supporting this newsletter as well? Get in touch to discuss sponsorship opportunities.