If Mark Twain were around in the digital age, I have no doubt he’d say, “There are lies, damn lies, and then infographics.”
Infographics seem to be everywhere now. What started off as perhaps a unique curiosity of the newspaper biz has now turned into a non-stop cavalcade of “data visualization” by publishers of all types and stripes. There are tens of thousands of graphics being produced daily, quite frankly, most of them are awful.
But even worse, a lot of them are straight out liars. They lie like a rug – and chances are you – you’ll never detect it.
The reality of how we “see” is a complex process between the eyeball and the brain. The brain “looks” at an image in a couple of different places in the brain, before ultimately making an assessment if it should be sent to the cognitive areas of the brain for perception. That process happens as fast as 13 milliseconds. In that time, we generally “believe” what we see because visual content bypasses much of our “giggle test” gate-keepers and get mainlined straight to the cognitive functions of the brain with an “eyeball tested, brain approved!” sticker on it.
This confusion is hardly new. In 350BC, Aristotle noted that “our senses can be trusted but they can be easily fooled”.
He noticed that if you watch a waterfall and shift your gaze to static rocks, the rocks appear to move in the opposite direction of the flow of water, an effect we now call “motion aftereffect” or the waterfall illusion. Tracking the flow of the water seems to “wear out” certain neurons in the brain as they adapt to the motion. When you then shift your gaze to the rocks, other competing neurons over-compensate, causing the illusion of movement in the other direction.
These “infographics” that are afraid of the truth are not about optical illusions. I’m talking about infographics that seem perfectly fine – giving you VOLUMES of information, unfortunately, most of it is absolutely false.
These are the worst infographics I’ve ever seen in telling the truth:
Lying Infographic #1 – NBC Nightly News “Misremembers” Ethnicity Data
This infographic about race and ethnicity in the United States from 1960 to 2060 is transformed, through bizarre visualization, into a map of the United States in which California, living in 1960, is entirely White, and New York, over in 2060, is entirely Hispanic. The NBC Art department must have “misremembered” what the census data actually said.
Obviously the problem is how the infographic itself is designed. When you create an infographic that uses the US as a “texture” – people presume the overlay of the percentage charge applies to the territory and geography. Your brain makes that connection and you conclude that the United States is a) overwhelming white, b) and that the fastest growing segment of non-white people are blacks. Both conclusions are wrong if you look at the underlying data in this research. If it was presented properly as a stacked area chart (absent the texture of the US map) understanding the reality of the Pew Research would be considerably easier.
I can only conclude the deception is intentional in the design of this infographic, given NBC News political agenda. If we hadn’t pointed out to you the problems – you would have likely accepted this graphic without question. Seeing “proportions” is hard to do when you’re comparing graphics and numbers (which I’ll get to below when we talk about nurses).
Lying Infographics #2 – Sometimes I can’t tell the difference between boobs and a heart attack
Back when the ALS Ice bucket Challenge was so popular, people were dumbstruck by how much money it raised, in comparison to how many people are affected by it. To illustrate this point, Vox apparently decided to compare breast cancer fundraising with coronary heart failure.
While I am sure no doubt some men (including myself) do feel an increased heart rate staring at a great pair of boobs – I think it perhaps a bit misleading visually to put breast cancer on top and then heart attacks on the other column. Human nature is to naturally connect the two. For the record, the third little gumball sized blue ball is heart disease (on the left), and the pea sized pink ball (on the right) is breast cancer victims.
Vox I believe is just lazy and that this chart is not intentionally deceptive. I can see a designer going “ok, let’s just order everything by magnitude” and not realize that such an assessment leads to erroneous conclusions visually. Obviously what your brain does is called “connecting”. We see it all the time, and it leads to erroneous conclusions. For example – tell me which box is larger?
I have no doubt you believe the box on the right, surrounded by larger boxes, is smaller than the box on the left, surrounded by smaller boxes. The reality is – they’re the same size. You beleive they are not the same size because your brain (and your eyes) work together to collect information about the visual context in which you evaluate an object. Your brain has a hard time disconnecting the context from the object of study. As a consequence, in the infographic above, you suddenly wind up comparing boobs and heart attacks, and you don’t even know it. Your brain never really sees the legend that would tell you it’s breast cancer v. heart attack. What you saw was – we raise a lot of money in the pink blob compared to lots of dead people in the blue blob. In other words, what this infographic told you is you can’t tell the difference between this…
Because there’s no real advantage in making that correlation – I’m going to just assume Vox was stupid in their infographic design, not intentionally misleading.
Lying Infographic #3 – That pesky decimal point always makes for complicated math.
Ok this one is awesome because it’s one of the oldest chart “cons” in the book. So you look at the chart, but forget the little people icons for the moment. If you compare the trend over time, just looking at the numbers on a piece of paper (which I did), it is only increasing by 7%, or .07. However, if you actually look at the stacked little people, if all four of the columns of icons are equally valued (in other words – 4 people represents 43, 147, but in the last graphic 40 men represent 47,500) the overall size of the chart increases 700 times… or 700.00 or 7000%. Oops! That’s only off by a factor of 5x. Not to mention it’s misleading to have the people units represent the same amount (ostensibly the same – only 300 people different) for two years, and then a 3K increase in the numbers equals a 7 times jump in icons, and then another 1500 people only increases the total icons by roughly a third.
Math – not just for mathematicians anymore.
This is an intentionally screwed up infographic in my view because this is a very very old “con” in terms of manipulating the conclusion. A key rule in data visualization is the baseline of a chart ALWAYS needs to remain consistent. It doesn’t matter if it’s infographics or charts or whatever. However, it’s not just infographic authors that do silly things like this – Fox News does it all the time.
There’s no way those two numbers are baselined to the same zero – and if they are baselined properly there is no way your eyeballs will tell your brain, because you have no way of understanding the context beyond the snippet. Your brain automatically assumes the bar on the bottom is the baseline. Again, because your eyes and your brain work continuously to establish the context of what it sees. If you manipulate that context – you wind up with yet another infographic afraid of the truth.
Lying Infographic #4 – College? I don’t need no stinking College!
This infographic by Business Insider is interesting. The point they’re trying to make is that essentially since 1980, the amount that college education costs a student has made the ROI of that education negative when compared to the earnings of that student. It’s a brilliant point – except for two facts they missed. One – the ability of non-degree holding candidates to get a job is significantly worse, thus, while they pay less only for a high school degree, their ROI is undoubtedly negative (in the aggregate) as they are the greatest percentage of unemployed (thus earning no income). Second, the pay gap between college graduates and everyone else reached a record high last year, according to the new data, which is based on an analysis of Labor Department statistics by the Economic Policy Institute in Washington. Americans with four-year college degrees made 98 percent more an hour on average in 2014 than people without a degree. That’s up from 89 percent five years earlier, 85 percent a decade earlier and 64 percent in the early 1980s. Ooops!
Facts aside, obviously we have a baseline problem – but it’s on both axes – not just the Y axis. While this chart has the same “base number” on the X axis for both the earnings and the college costs, the magnitude of the differential is amplified because the chart dimension increases 171% on the Y axis, but you go only 2% (1964 to 2010 is really an increasing rate of only 2%) on the bottom. As a consequence, the magnitude of the differential is amplified because the Y variable looks like it is increasing faster than the time variable.
To give you a sense of how difficult the manipulations are to detect – here’s some other graphics that use the same underlying data – and tell different stories:
According to the College Board, the net price of college tuition — that’s the figure after grants, tax credits, and discounts are factored in — has indeed risen by roughly 75 percent between 2002 and 2011. But that translates to just about an $1,130 increase — or 2 percent of your median young bachelor’s holder’s annual earnings, at least if they work full time. Now, there are problems with this chart as well – for one thing, the X axis is 1/3 longer than the Y axis (bet you didn’t see that did you), so it undoubtedly flattens out the curves. In this case it’s a mild exaggeration – because if anything, shaving off 2002-2004 would make the chart even less remarkable in terms of variance.
Now why is it the Business Insider chart looks like “Jaws” and this chart looks like a cardiac patient flatlining? Well – indexing for one thing. Notice that BI chart indexes its data (it adjusts the dollars based on an inflation index). That’s why the trends look so dramatic. But when you look at it in terms of actual dollars and cents, the changes seem considerably less pronounced.
I have to conclude that BI’s infographic/chart is intentionally designed to mislead – or perhaps this other chart is lying – but someone isn’t telling the truth.
Lying Infographic #5 – Greater than? Less than? When it comes to Math – stand your ground.
Ok, last time I checked, 721 was less than 873, even with common core mathematics being taught in schools. Now, if you flip this cockamamie chart upside down, and flip it horizontally (so that time actually moves the right direction), you actually get the correct implication of the chart (although all the numbers are backwards):
This one uses all types of tricks. The shaded region misdirects your eyes. The baseline being flipped to the top confuses you (because you normally evaluate integrations visually from zero to infinity as down to up).
When I flip the chart around – the story told is dramatically different than the story Reuters wanted to tell you. This is undoubtedly an intentional manipulation.
Simple Infographic Design Rules that Indicate Lying & Deception:
- Are the baselines consistent? If not – it’s usually being used to either conceal change or magnify it.
- Are the shape sizes relatively proportional? This is tough to know – but don’t believe everything as seen until you really think about what’s depicted.
- Is the legend designed to simplify or obfuscate the results?
- Look for hidden data – indexed numbers? Things rebaselined? Inconsistent in indexing? (These are just a few tricks to make things look a certain way)
- Are unnecessary things in the chart to distract you? (Think of how that NBC chart would have been if it was just a stacked percentage chart.)
While your brain may make a conclusion in those 13 milliseconds, your conscience needs to be more discerning. When designing infographics, be aware of how your brain actually works to “get to the point” quicker. You can use those facts to actually make valid conclusions stand out. Because if I catch you lying with your infographic, you’ll undoubtedly wind up here at some point.