Infographics have been around for many years and recently the increase of a number of easy-to-use, free tools have made the creation of infographics available to a large segment of the population. Social media sites such as Facebook and Twitter have also allowed for individual infographics to be spread among many people around the world. Infographics are widely used in the age of short attention span. In newspapers, infographics are commonly used to show the weather, as well as maps, site plans, and graphs for summaries of data. Some books are almost entirely made up of information graphics, such as David Macaulay's The Way Things Work. The Snapshots in USA Today are also an example of simple infographics used to convey news and current events. Modern maps, especially route maps for transit systems, also use infographic techniques to integrate a variety of information. The question is: Can infographis be trusted in the hands of individuals who are intent on promoting questionable agendas?
I'm seeing more and more graphs these days in print media and online communications to illustrate, or support, points of view or contentions raised in so-called news stories. Truthfully, I rarely see one that I can fully wrap my mind around, more or less trusting that the graphic illustration accurately portrays the truth of the subject matter.
In a pre-publication question-and-answer interview the professor was asked why he felt the need to write a book for the general public about charts and other methods of data visualization? The short answer was that charts are quickly becoming widespread, so we all need to become better chart readers.
"I've been designing charts—graphs, maps, infographics—and teaching how to make them for more than two decades. In the past few years they've become much more common in the news and in social media," he explained "Just think about the barrage of graphs and maps that are thrown at us whenever there's an election or a public discussion about a relevant topic. Charts can illuminate complex problems and reveal trends and patterns in data. They can make us smarter."
"However, charts can also obfuscate and mislead. Many people believe that charts are like illustrations or pictures—in the sense that they can be understood just by quickly looking at them—and that "a picture is worth a thousand words". Most of the time, that's not the case."
Social media platforms are great to disseminate both good and bad information; both kinds have exploded, so we readers should learn how to separate wheat from chaff. Because they are so attractive and clear, charts have become popular with good actors—statisticians, scientists, educators, journalists— but also with bad ones.
The latter are aware that charts can be very persuasive but that, at the same time, a substantial portion of the public isn't well equipped to deal with them. Therefore, they use charts to push agendas and spread deceptive messages.
The good news, though, is that following a few easy-to-understand principles explained in 'How Charts Lie', we can all become better readers, taking advantage of charts that are well designed, ignoring or denouncing those created to deceive us, and identifying those that, even if they aren't intentionally deceptive, still mislead us because we don't read them correctly.
Why do people trust infographics more than they should? Cairo said that because many of us have internalized that "a picture is worth a thousand words", that charts are intuitive and fun, and that you're supposed to understand them in the blink of an eye, without thinking much about them. These myths are dangerous. Charts can indeed be worth a thousand words, intuitive, fun, and quickly understandable—but only if you are a good chart reader already! That's a pre-condition.
What kind of cognitive biases and traits do we bring to reading visual data? Sometimes a chart will mislead us not because it's badly designed, but because we're all prone to lying to ourselves. Psychologists often talk about phenomena such as 'motivated reasoning' and 'the confirmation bias', which mean that we all prefer information that confirms what we already believe, and that we tend to project our beliefs onto whatever we hear or see, including charts.
"Imagine that I design a line chart showing that in the past fifty years obesity rates—the percentage of people who are obese—and life expectancy have increased in most countries at a similar pace. If you're overweight you may be inclined to read that chart as 'proof' that one of those variables is causing the other, that more obesity leads to higher life expectancies."
"But that may not be true at all! It may be that the negative consequences that higher obesity rates may have (shorter lives because of more diabetes or heart disease) are balanced out by other factors: longer lives because of declining poverty rates, medical science making tons of progress, and health care improving many places," he added.
"This is similar to many examples that appear in 'How Charts Lie' which I use to explain a few key principles of chart readership: first, that a chart shows only what it shows, and nothing else. In this example, my chart shows that both obesity and life expectancy have increased, period. Everything else we may see in the chart, such as possible causal links between obesity and life expectancy, is not in the chart at all; it happens inside our brain."
"Second, that we should try not to read too much into any single chart; a chart alone rarely 'proves' anything. If we want to learn about how obesity and life expectancy relate to each other, my imaginary chart is not enough at all. We need more evidence."
The professor was asked for a recent example of a misleading chart doing real-world damage? "I've seen people in the anti-vaccine movement use charts that show that there has been an increase in autism prevalence that is parallel to an increase in vaccination rates and to the introduction of new vaccines. This is yet another example of a correlation or covariation that shouldn't lead us to establish a causal connection.
"The prevalence of autism has certainly increased everywhere, but scientists explain that it's due to reasons that have nothing to do with vaccines. For instance, the public is much more aware of autism today than it was decades ago, so we may be better prepared to detect it. Second, we have children later in life, on average, and having older parents is a risk factor. Third, the technical definition of what constitutes autism has broaden throughout the years, which necessarily leads to an increase in diagnosed cases. Fourth, diagnoses have become more accurate: in the past, people with autism were often misdiagnosed. Finally, genetic factors likely play a key role."
Also, charts that are related to highly charged partisan topics, such as gun policy, abortion, or immigration, are nowadays thrown around in debates to 'prove' this or that point. As previously explained, a chart rarely 'proves', anything on its own. It can be a powerful instrument in any discussion, but it can't be the only element in that discussion.
What are some traits of a chart that should set off alarm bells in a reader? Cairo said that the main one is if a chart is presented devoid of proper context, or if it doesn't mention the source of the data it displays. If you can't verify whether the source of a chart is trustworthy, whether the numbers themselves are reliable, and whether those numbers are really measuring what the chart says they are measuring, distrust the chart.
What's the best tip you can offer when reading graphs? He stressed: "Pay attention!" It all begins with that. "Whenever encountering a chart, don't just look at it quickly and move on. If you do so, you'll likely misinterpret it. Rather, stop for a minute and try to understand what the chart depicts; check where the numbers represented in it come from, and what they are supposed to measure; pay attention to the patterns or trends the chart reveals. And don't read too much into any chart. Remember that we all tend to project what we already believe—or what we want to believe—onto the charts we see every day."
Today, anyone with a computer or smartphone can create infographics, line charts, maps, diagrams, and various other graphics that appear authoritative, whether true or not. And social media has made them ubiquitous, with no editorial gatekeepers in sight. In his new book HOW CHARTS LIE GETTING SMARTER ABOUT VISUAL INFORMATION (W. W. Norton & Company; October 15, 2019), University professor Alberto Cairo offers a powerful and much-needed tool for cutting through the post-truth in infographics.
Cairo outlines what we need to look for when reading a chart, just how high the stakes are, and what kind of tricks politicians, advertisers, social media influencers, journalists, governments and plain-old liars can use to deceive us. He also explains how to become better chart readers and how to use them to our advantage.
In a pre-publication question-and-answer interview the professor was asked why he felt the need to write a book for the general public about charts and other methods of data visualization? The short answer was that charts are quickly becoming widespread, so we all need to become better chart readers.
"I've been designing charts—graphs, maps, infographics—and teaching how to make them for more than two decades. In the past few years they've become much more common in the news and in social media," he explained "Just think about the barrage of graphs and maps that are thrown at us whenever there's an election or a public discussion about a relevant topic. Charts can illuminate complex problems and reveal trends and patterns in data. They can make us smarter."
"However, charts can also obfuscate and mislead. Many people believe that charts are like illustrations or pictures—in the sense that they can be understood just by quickly looking at them—and that "a picture is worth a thousand words". Most of the time, that's not the case."
Social media platforms are great to disseminate both good and bad information; both kinds have exploded, so we readers should learn how to separate wheat from chaff. Because they are so attractive and clear, charts have become popular with good actors—statisticians, scientists, educators, journalists— but also with bad ones.
The latter are aware that charts can be very persuasive but that, at the same time, a substantial portion of the public isn't well equipped to deal with them. Therefore, they use charts to push agendas and spread deceptive messages.
The good news, though, is that following a few easy-to-understand principles explained in 'How Charts Lie', we can all become better readers, taking advantage of charts that are well designed, ignoring or denouncing those created to deceive us, and identifying those that, even if they aren't intentionally deceptive, still mislead us because we don't read them correctly.
Why do people trust infographics more than they should? Cairo said that because many of us have internalized that "a picture is worth a thousand words", that charts are intuitive and fun, and that you're supposed to understand them in the blink of an eye, without thinking much about them. These myths are dangerous. Charts can indeed be worth a thousand words, intuitive, fun, and quickly understandable—but only if you are a good chart reader already! That's a pre-condition.
What kind of cognitive biases and traits do we bring to reading visual data? Sometimes a chart will mislead us not because it's badly designed, but because we're all prone to lying to ourselves. Psychologists often talk about phenomena such as 'motivated reasoning' and 'the confirmation bias', which mean that we all prefer information that confirms what we already believe, and that we tend to project our beliefs onto whatever we hear or see, including charts.
"Imagine that I design a line chart showing that in the past fifty years obesity rates—the percentage of people who are obese—and life expectancy have increased in most countries at a similar pace. If you're overweight you may be inclined to read that chart as 'proof' that one of those variables is causing the other, that more obesity leads to higher life expectancies."
"But that may not be true at all! It may be that the negative consequences that higher obesity rates may have (shorter lives because of more diabetes or heart disease) are balanced out by other factors: longer lives because of declining poverty rates, medical science making tons of progress, and health care improving many places," he added.
"This is similar to many examples that appear in 'How Charts Lie' which I use to explain a few key principles of chart readership: first, that a chart shows only what it shows, and nothing else. In this example, my chart shows that both obesity and life expectancy have increased, period. Everything else we may see in the chart, such as possible causal links between obesity and life expectancy, is not in the chart at all; it happens inside our brain."
"Second, that we should try not to read too much into any single chart; a chart alone rarely 'proves' anything. If we want to learn about how obesity and life expectancy relate to each other, my imaginary chart is not enough at all. We need more evidence."
The professor was asked for a recent example of a misleading chart doing real-world damage? "I've seen people in the anti-vaccine movement use charts that show that there has been an increase in autism prevalence that is parallel to an increase in vaccination rates and to the introduction of new vaccines. This is yet another example of a correlation or covariation that shouldn't lead us to establish a causal connection.
"The prevalence of autism has certainly increased everywhere, but scientists explain that it's due to reasons that have nothing to do with vaccines. For instance, the public is much more aware of autism today than it was decades ago, so we may be better prepared to detect it. Second, we have children later in life, on average, and having older parents is a risk factor. Third, the technical definition of what constitutes autism has broaden throughout the years, which necessarily leads to an increase in diagnosed cases. Fourth, diagnoses have become more accurate: in the past, people with autism were often misdiagnosed. Finally, genetic factors likely play a key role."
Also, charts that are related to highly charged partisan topics, such as gun policy, abortion, or immigration, are nowadays thrown around in debates to 'prove' this or that point. As previously explained, a chart rarely 'proves', anything on its own. It can be a powerful instrument in any discussion, but it can't be the only element in that discussion.
What are some traits of a chart that should set off alarm bells in a reader? Cairo said that the main one is if a chart is presented devoid of proper context, or if it doesn't mention the source of the data it displays. If you can't verify whether the source of a chart is trustworthy, whether the numbers themselves are reliable, and whether those numbers are really measuring what the chart says they are measuring, distrust the chart.
What's the best tip you can offer when reading graphs? He stressed: "Pay attention!" It all begins with that. "Whenever encountering a chart, don't just look at it quickly and move on. If you do so, you'll likely misinterpret it. Rather, stop for a minute and try to understand what the chart depicts; check where the numbers represented in it come from, and what they are supposed to measure; pay attention to the patterns or trends the chart reveals. And don't read too much into any chart. Remember that we all tend to project what we already believe—or what we want to believe—onto the charts we see every day."
No comments:
Post a Comment