Friday, February 14, 2020

'How Charts Lie': a few edits to the first print edition

If you read the first print edition of How Charts Lie you may notice a few printing and layout errors. These should have been corrected in the e-book and the audiobook already. If you detect anything that looks strange other than these, please let me know.

Also, I've printed out cards containing these edits. If you want to receive one and keep it inside your copy of the print book, contact me.

(Update 01/08/2020: you can now download all figures from the book in high resolution and in two different color schemes.)

Here are the corrections:

On page 24 the transparency effects that should emphasize or hide parts of the charts disappeared between the galleys—where the graphic was perfect—and the final printing. Mysteries. Here's how that graphic should look like:

On page 45 the line corresponding to the United States didn't show when printed. Here's the real graphic:

There's a minor issue with the gradient on the second bar of the chart on page 142: it doesn't fade to white for some reason. It should look like this:

A chart on page 116 is slightly misplaced.

On page 92 there's a needless “is” in a sentence that should read “Unless a crime is premeditated...” On page 104 there's an “s” missing at the end of “assess” (this one made me giggle.) At the bottom of page 128 there's an “as” missing before “Assange”. And on page 157 there's a tiny label that should read MS instead of MI. The last label on the Y-scale of the chart on page 172 should read “600” instead of “490”.

On page 44, where it says “people became richer or poorer” or “people in those countries contaminated more or less,” I think I'll add “on average” just in case, as these are per capita numbers.

Thursday, February 13, 2020

Design can destroy the world—but it can also make it better. My opening remarks for the 2020 Data Intersections conference

Data Intersections, the University of Miami's conference about the ethics of data, design, and technology is this same afternoon. Some of you have asked me in private whether we'll record it, and the answer is yes. We'll make all talks available in few days, as soon as the videos are edited.

In case you're interested, here's the draft of the remarks I'll offer at the beginning of the conference (spoiler alert: there'll be a book about some of this in 2021):

Hello, welcome to the Data Intersections conference. First I’d like to thank you for being here this afternoon.  
For those of you who don’t know me, I’m Alberto Cairo, the Knight Chair in Visual Journalism at the School of Communication of the University of Miami. I’m also director of visualization and information design at our Center for Computational Science, CCS, one of the sponsors of Data Intersections. CCS is about to become UM’s Institute for Data Science and Computingor iDSC,—as you’ll learn in a minute from our Provost, Jeff Duerk. I'll be the director of iDSC's Center for Visualization, Data Communication, and Information Design, so that's exciting news...
But before that, let me briefly explain how the 2020 edition of the Data Intersections conference came to be. It all begun in 2019, when Mike Monteiro, one of our speakers today, sent me an early copy of his book Ruined by Design: How Designers Destroyed the World, and What We Can Do to Fix it. 
You'll see, I’m a journalist and also a data visualization and infographics designer, so I was interested in learning about how and why I was destroying the world. 
I liked Mike’s book so much that I ended up writing a blurb for its back cover. I said about Mike’s book, and I quote, that it is “Victor Papanek’s Design for the Real World updated for the 21st Century—and with much more swearing.” 
If you are a designer or technologist, you probably know Victor Papanek’s famous book. If you’ve never heard of it, go get a copy. As with Mike’s book, Papanek's Design for the Real World is a passionate discussion of how design and technology can go wrong, and what we can do to get them right instead.
The reason Mike’s book had such an impact on me is that I’ve always been fascinated by numbers, design, and science. That's why I make data visualizations and infographics, and also teach how to design them. At the same time, Ive always been interested in thinking about how we, the creators of those numbers, designs, and technologies, can make good and informed choices not ignorant or even destructive ones. 
This is related to a third interest: moral philosophy, Since I was in High School, I’ve been reading informally, as a proud amateur, into the literature of ethical thinking, so I'm somewhat familiar with the major debates and schools in the philosophy of ethics—virtue ethics, deontology, consequentialism, and their multiple variants. 
That literature might help us answer questions that we all face: what is the difference between what we can do and what we ought to do? How can we train our moral intuitions, and our sixth sense of what is right or wrong? How can we weigh the possible consequences of our actions and creations? What is it appropriate—or not appropriate—to do with the data and technologies we design? 
Ultimately, I’d argue that the key question we should try to answer is: how can we use data, design, science, and technology to help human beings have better, happier, and wiser lives, and to make our societies flourish—instead of destroying them? 
This is what Data Intersections is about. Our four speakers today, Otávio Bueno, Heather Krause, Yeshi Milner, and Mike Monteiro, will surely inspire us to be more ethical data scientists, designers, journalists, and technologists. Or, in general, better human beings. 
Once again, thanks so much for being here this afternoon. I hope you’ll enjoy our great speakers and the reception at the end of the day. Now, I’d like to introduce the Provost of the University of Miami, Jeff Duerk. Jeff, thanks for being here. The stage is yours.

Wednesday, February 12, 2020

Visualization often puts stories in perspective

Following an election or primary race too closely is entertaining, but also stressing. Media organizations often read too much into individual polls—I try to stick to the mantra “every single poll is noise; what might matter is the weighted average of all polls,”—and pundits love to read tea leaves—aka: spotting and discussing signals in randomness.

The recent Democratic primaries in Iowa and New Hampshire are big news these days, but how much do they really matter? I don't know; I'm no political analyst, but the following visualization by WSJ's Brian McGill puts things in perspective. Maybe those primaries matter less than we want them to, as there's still such a long way to go?

Brian designed that cartogram manually, painstakingly copying and pasting nearly four thousand squares. I sympathize with his effort. I've made cartograms like that myself in the past.

Monday, January 27, 2020

A nice short video about misleading graphs

While reading How Charts Lie, one of my undergraduate students told me that TED-Ed published a short animation (paired with a lesson containing extra materials) about misleading graphs in 2017. The author is Lea Gaslowitz. It's quite nice:

Thursday, January 23, 2020

An example of how to annotate a visualization, by The Financial Times

The Financial Times visualizes how Britons spend their time at weekends vs. week days, a graphic that is part of a story about how weekend working affects families.

The array of graphs is elegantly designed, as it often happens with the FT, which has developed a massive library of high quality visualizations and even has a column about data journalism. That said, what really does the trick for me is the detailed annotations.

Several members of the FT data and graphics team have repeatedly expressed their belief in the power of the annotation layer. John Burn-Murdoch, for instance, has said: “I and my colleagues here at the FT, we really do think one of the most valuable things we can do as data visualization practitioners is add this expert annotation layer.” I agree. Explanatory visualizations shouldn't consist of visualizing data alone, but also of adding words to put the data in context, highlight the most relevant facts, or dispel possible misinterpretations.

Monday, January 13, 2020

Announcing Data Intersections, a free conference about the ethics of data, design, and technology

In the afternoon of February 13 I'm hosting Data Intersections, a free half-day conference about the ethics of data, design, and technology.

Our speakers are Mike Monteiro, author of Ruined by DesignYeshimabeit Milner, founder of Data for Black Lives, Heather Krause, founder of We All Count, and Otávio Bueno, philosopher of science, Math, and technology.

If you wish to attend, sign up for free, and see more information here, including the schedule.

Saturday, January 11, 2020

El País wisely uses animated transitions

Spain's El País recently got a new graphics director and stepped up its visualization game by strengthening its graphics desk with some other new hires. The results are showing already. Their latest data-driven piece is an analysis of the European countries currently governed by multi-parti coalitions. The designers wisely used a restrained color palette and elegant and smooth animated transitions to connect each section of the story to the following one.

The print version of this piece also looks quite great:

Wednesday, January 8, 2020

All graphics from 'How Charts Lie' freely available in two color schemes (for now)

I'm making most figures—not all, due to technical and copyright issues—from How Charts Lie freely available. Find then in this Dropbox folder. I also decided to create different versions using alternative color schemes. The first one is blue and red instead of gray and red, and you can download the graphics here. (Also, don't forget the edits page.)

Here's a comparison between two versions of the same graphic, the one that appears in the book and the one with a red-blue color scheme:

Thursday, December 26, 2019

All The Economist Graphic Detail visualizations in one convenient PDF

Alex Selby-Boothroyd, head of Data Journalism at the Economist, has just announced a special gift for their readers: they've collected more than 60 articles/visualizations from their Graphic Detail section as a PDF. To download it you need to sign up (it's free) or subscribe to the magazine.

The document contains some of the finest, most elegant, and most tightly edited print graphics I've seen this year. You can pair it with this article, which explains what Alex and his team learned by publishing a weekly data-driven article for more than a year. If you teach visualization, both the document and the article are great teaching resources.

Monday, December 16, 2019

Politico visualizes why the Blue Wall crumbled in Pennsylvania, Wisconsin and Michigan

The latest piece I've read in Politico is a multimedia experience about why Pennsylvania, Wisconsin, and Michigan became more Republican in the 2016 election.

The article combines video, audio—and visualizations that are as straightforward as they are thoughtfully designed. I particularly like the heat map at the beginning of the piece, but the line graphs that support the three reasons the authors offer for the shift—Trump-Obama voters in Pennsylvania, young voters in Wisconsin, and African-American voters in Michigan—are also quite nice.

Tuesday, December 3, 2019

"Wow" first, then "me" and details on demand

The New York Times has designed several deceptively simple visualizations depicting air pollution in different places in comparison to New Delhi, the most polluted city on the planet. This project does many things well, I believe.

First, it has a me layer, detecting where you are, or letting you search for it. As I wrote in The Truthful Art and in this blog, I think that letting viewers see themselves in the data greatly increases interest and engagement.

Second, the story opens with an animated dot density chart. The designers may have realized that the strength of their dot graphic isn't its accuracy—you can't compare cities to each other very well, but just get a rough impression of their relative pollution levels. Instead, what makes this visualization a good opener is its power to capture your attention. Later in the piece viewers can get all details through more traditional maps and graphs. The one below, for instance, which breaks the layout in a quite dramatic manner:

(Note: the title of this post is a pun based on Shneiderman's mantra.)

Monday, December 2, 2019

VizUM 2019: Uncertainty in visualization

Every year I organize a small free symposium at the University of Miami called VizUM. The next one is very soon, on December 11 at 4:00-7:00PM. This year, VizUM is paired with our annual conference about Big Data, which is also free. Here's the schedule for the day.

Our guest speaker is Northwestern University's Jessica Hullman; I'll also present after her. If you're interested in attending, see details here, and register for free through this form.

You can download the flyer on the right as a PDF here.

Friday, November 29, 2019

Old dogs, Tom Hanks, and analogies in visualization

Christopher Ingraham writes about a new study comparing the aging of humans and Labrador retrievers. I've been using the 7-year rule—each 7 human years equal to 1 dog year—for ages, although I'd heard of other formulas. One of my dogs is 3, and I always assumed that his goofy and playful behavior was due to being a late teenager in his twenties. It turns out that he may be something closer to a very active 50-year-old person!

The graph in the article is lovely not only because of its clarity, but because of the cute comparison between dogs and Tom Hanks. I wish we could see analogies like this more often in visualization. They were pretty common in news graphics decades ago, and I have the hunch that they aid understanding of otherwise abstract quantities.

Thursday, November 28, 2019

The bar chart race is the fidget spinner of data visualization

Andy Cotgreave has recorded a video humorously critiquing bar chart races—animated ranked bar graphs,—calling them “the fidget spinner of data visualization.” Are bar chart races good or bad? I guess that, like everything else in visualization, it depends on the purpose you have in mind. Decisions in visualization need to be based on justifiable reasons. If you want to wow viewers and capture attention, then go ahead and try to make one. But if you want them to extract meaning from the data, spot trends or patterns and, as a consequence, learn something interesting or useful, in many cases you might want to choose a different visualization type, such as our old friend the time-series line graph:

Tuesday, November 26, 2019

Data visualization fashion

Giorgia Lupi never ceases to amaze me. Her latest project is a fashion collection based on data sets created by three pioneering women scientists, Ada Lovelace, Rachel Carson, and Mae Jemison.

I love seeing data visualization used in unexpected ways: in public spaces, physical objectsfine art, or wearables. I have the hunch that such initiatives help demystify the field and bring more attention to it.

Giorgia's collection is already sold out, which is hardly surprising; I want this sweater; maybe I could pair it with a tie that Michele Banks gave me as a gift a while ago:

You can read more about this project here, here, and here. Side note: A passage from a Vogue piece annoyed me a bit because it repeats tired stereotypes designers and journalists are still too fond of (italics mine):
As an information designer, Lupi is known around the world for her singular, artful approach to data: Instead of relying on hollow charts and graphs, she creates beautiful hand-drawn prints that lend a “human” touch to sterile numbers and statistics.
There's nothing “hollow” about common charts and graphs, and there's nothing intrinsically “sterile” about numbers and statistics. If we perceive them that way it's likely because we don't know how to read them well.

Monday, November 18, 2019


C-SPAN2's BookTV recorded a public How Charts Lie lecture I gave at Northeastern University a few weeks ago. They broadcasted it this past Sunday afternoon and they've just made it available online, too. Here it is.

The slides are also available as a PDF, in case you find the talk useful and want to steal some ideas for your own classes or talks.

Friday, November 15, 2019

Visualization stuff I've seen recently

I guess I should launch a newsletter soon, as I'm seeing plenty of interesting visualizations and visualization-related articles lately—more than usual,—and it's difficult to keep up. Here are a few:

• I'll be on C-Span 2's BookTV on Sunday at 3:30PM EST to talk about How Charts Lie. The program will be available online right after, I believe.

• Kaiser Fung redesigns a chart that was discussed recently in social media.

Betsy Mason explains why scientists should get better at visualization. Her article is a good introduction to some elementary principles. She even has a section about why rainbow color schemes are evil.

Jon Peltier writes about maps (maps!) in Excel.

Nathan Yau talks about a Javascript library that lets you design XKCD-like visualizations.

• Randall Munroe, the creator of XKCD, is in the DataStories podcast, and has many thoughts about science communication.

• Cole Nussbaumer's Storytelling With Data website has a new look.

• There's a new episode of EagerEyesTV, Robert Kosara's video series; it explains data formats.

• Matteo Moretti explains how visualizations can be embedded into larger “informative experiences” through a couple of projects he designed.

Monday, November 11, 2019

Beautiful visualizations that reveal ugly truths: a neofascist party gets 15% of the vote in Spain

El País's coverage of yesterday's general election in Spain was exemplary, and that includes their graphics. El País made significant changes to their data and visualization desks recently, and it shows. They have a detailed map of the country and before the complete results were available they experimented with forecasts and real-time counting.

The bad news of the day was that a neofascist party, Vox, got 15% of the vote. More than 3.6 million people supported it. That should worry anyone who cares about the future of liberal democracy. I miss boring times.

Friday, November 8, 2019

Data and the junk in our lives

Spotting data visualizations and infographics in movies is a little pastime of mine. The latest case appears in the trailer of the upcoming Pixar movie, 'Soul'. A character is sitting in front of a large dashboard chockablock with data visualizations. The narrator says “don't waste your time with all the junk of life,” and the character tosses the screens away. Maybe the message of the movie is that Jazz is preferable to data? Why not both? I'm intrigued. Watch the trailer here; it looks beautiful.

Wednesday, November 6, 2019

A King and a Princess walk into a lecture about charts...

Yesterday I gave a talk about How Charts Lie in Barcelona. It was part of an event celebrating the 10th anniversary of the Princess of Girona Foundation, which supports young Spanish artists, scientists, and entrepreneurs. The King, the Queen, and their two daughters, Leonor and Sofia, attended the opening ceremony on the first day.

Several parallel talks and workshops took place the day after. When I was setting up in the auditorium I was assigned, Alfredo Martínez, the chief of protocol of the Spanish Royal Household, showed up, introduced himself, and casually said—this is not verbatim—“hey, did anybody tell you that the King is coming to your lecture?” Nobody had. “No pressure,” I thought.

The King brought Princess Leonor; they sat in the front row for most of the session. I got encouraged by two facts I observed: first, the King repeatedly whispered in his daughter's ear and pointed at the screen to further explain to her some of the graphics; second—and most importantly—the Princess, who is now 14, didn't fall asleep at any point. Success!

My dear friend Xaquín G.V. took some photos with his cell phone. Here they are: