Data Points is a new series where we explore the world of data visualization, information graphics, and cartography.
Numbers themselves don’t lie, but how we represent them can be really misleading. While charts and graphics can help us visualize large amounts of data, they can also be dangerous in the wrong hands.
The good news first: If the chart is not plain lying, just playing with the way we perceive visual information, there are some tips you can use to reveal the truth. The bad news is that you might have to do this analysis quickly if a chart flashes by quickly on a TV screen. Also, let’s be honest: Not everybody is willing to look further into a chart if the result confirms what they want to believe.
If you want to detect a cheating chart, here’s a quick guide to the most common tricks:
1. Broken scales show drama where it doesn't exist.
This is probably the most common way graphics lie, whether intentional or not. Something that changes by 0.1 percent over 10 years and something that changes by 1,000 percent in one year can look exactly the same depending on the scale, or range of values used on the chart.
It gets worse when we compare two different elements. In this case we’re not just adding drama to the change, we're making unfair comparisons.
That doesn’t mean that charts should never break a scale. Maybe we want to show deviation from 50%, and that should be our starting point. But there has to be a good reason to do it. Showing a more dramatic view of the change is not one.
2. Showing data on two different scales can make for an apples-to-oranges comparison.
Is 170 pounds more or less than 5 feet 8 inches? The question has no answer because we’re talking about two unrelated units, although they speak about related fields. Still, many charts draw these kind of correlations.
This lie is usually unintentional. Many people just want to include all their information in less space, but the results still draw conclusions that aren’t anywhere but in these charts.
3. Showing a correlation can imply causation.
Two people doing the same dance moves may not be dancing to the same song. The same way, just because two sets of numbers follow a similar path doesn’t mean there’s a correlation.
There’s even a website devoted to make fun of this non-existing game of cause and effect. Don’t let the form of change look like the reason of change.
4. Ignoring population size makes rates impossible to compare.
There are also many ways to lie with maps, which we won’t dive into in this post. Still, when we’re talking about where people live it's important to have in mind how many people live there.
Which cities have the highest number of murders? Not the most dangerous, but the most populated. More people = more everything.
5. Decoration can be deceiving.
Who doesn't love 3D pie-charts? (Disclosure: many people, including me.) Yet they have a big audience in scientific papers, economic reports, and PowerPoint charts. They look fun and modern, but including volume and perspective in a chart where they don't mean something is a big no-no.
If you’re making a chart to visualize amounts and you distort the way amounts are visualized… what’s the point of making a chart?
Of course, there are many more ways to lie with charts—some subtle, some not—just as there are ways to mislead with words and pictures. The biggest enemies of chart cheats are context, analysis, and common sense. Try to apply those next time you see some "shocking truth" in the form of a chart.