Goodreads: How Charts Lie
Published: October 15, 2019
A leading data visualization expert explores the negative—and positive—influences that charts have on our perception of truth.
We’ve all heard that a picture is worth a thousand words, but what if we don’t understand what we’re looking at? Social media has made charts, infographics, and diagrams ubiquitous—and easier to share than ever. While such visualizations can better inform us, they can also deceive by displaying incomplete or inaccurate data, suggesting misleading patterns—or simply misinform us by being poorly designed, such as the confusing “eye of the storm” maps shown on TV every hurricane season.
Many of us are ill equipped to interpret the visuals that politicians, journalists, advertisers, and even employers present each day, enabling bad actors to easily manipulate visuals to promote their own agendas. Public conversations are increasingly driven by numbers, and to make sense of them we must be able to decode and use visual information. By examining contemporary examples ranging from election-result infographics to global GDP maps and box-office record charts, How Charts Lie teaches us how to do just that.
How Charts Lie is an important and accessible look at how charts can be used to present misinformation or how people can misinterpret charts to come to inaccurate conclusions. In an age where information is becoming increasingly visual, it offers readers a chance to improve their visual literacy and to do their own research before believing everything they read or clicking “share” on a provocative chart they run across on social media.
The chapters are divided into subtopics, addressing things like how badly designed charts can be misleading or how charts can leave out information (intentionally or unintentionally) to make a point that isn’t actually true, and explanations are illustrated with examples made up by the author, examples of bad charts from real publications, and examples of what a correct chart might look like instead. The format is clear and easy-to-follow—although one might find reading a whole book about charts slightly dry, even if one is interested in the topic and recognizes its importance.
One simple example of bad chart design you’ll run across: bad proportions. For instance, I might say, “I read more books in 2019 than in 2018” and illustrate it with these simple bubbles. The relative sizes of the bubbles imply, however, that I read significantly more books in 2019 than in 2018, while if you actually read the data, it says I only read 3 more books. Even if you do read the data carefully—and many people don’t—the visual difference in the bubbles will affect your interpretation subconsciously.
Things get more complex as the book goes along, and one thing that struck me is that sometimes (or often) readers need background information on a topic to understand that a chart is lying, particularly if it’s accompanied with annotations explicitly saying that it shows something—that it actually doesn’t. For instance, you might need to know something about American politics and the electoral college to know that a chart that “proves” Trump won over most of the country in the 2016 election actually shows that he won in a lot of counties over a large geographic area—but he didn’t actually win “the most votes” and cannot claim that “the most people voted for him.”
The book does emphasize the necessity of research—looking at who made the chart, what information they used to create it, what information they excluded, etc.—but I was still struck by the sense that there are cases where you would need to actually know something about the topic at hand to 1) think the chart seems a little off and might need more digging and 2) know that the data used to make the chart is not complete. This isn’t a flaw of the book, of course, but it does illustrate that people do need to do research and may need to defer to experts on certain topics, instead of thinking that understanding a chart and trusting it is a straightforward process.
My one disappointment with the book is that I foresee it being popular mainly as assigning reading in college classes (perhaps in high school, though I wouldn’t expect a teacher there to assign the whole book). It’s a topic that’s important for everyone, but “how to read charts correctly and increase your visual literacy” isn’t a subject that screams “bestseller” material, even with the catchy title about charts telling us lies. I do hope this book finds an audience, however.