There's an old saying in journalism: "Dog bites man, who cares? Man bites dog, now that's news."
The aphorism captures the idea that only a sliver of what actually happens constitutes real news. It's a journalist's job to find that sliver. If you ran a story every time a dog bit someone, you'd have a lot of work to do. And no one would want to read it. But if you hear of a case where a man bit a dog, that's likely to pique some reader's curiosity. That's the kind of scoop every journalist wants.
It's perfectly obvious that news producers have a clear incentive to discriminate among possible stories. Not everything should be publicized. Not everything is news.
But in my conversations about fake news with people, I've come to realize that most people have a naïve view about the nature of these incentives. Many people think that producers' incentives are innocuous: they choose to run some stories and not others because they want their readers or viewers to be informed. Running a story when a dog bites someone won't teach readers anything new. They already know that happens. But running a story where a man bites a dog is something else--here's a chance to introduce your readership to a novel and interesting bit of the world.
The truth is both more complicated and less beneficent. Yes, of course, most producers are interested in informing their audiences. But producers also have incentives to tell the stories we want to hear, regardless of whether those stories make us more informed about the world. That's because at the end of the day, their job is not to improve the accuracy of our worldview. It's to sell advertising. Publish the wrong stuff, and you'll go out of business.
And, in fact, focusing on what's new (or catchy or exciting or upsetting) is positively harmful for developing an accurate view of the world. That's because it distorts our sense of how often things happen and how serious they are. Think of it this way: if you only read stories about men biting dogs and not vice-versa, you'll get the mistaken impression that people bite a lot of pets. The reality, of course, is just the opposite.
A recent Tweet from Alec Stapp at the Institute for Progress provides a quick visualization of this distortion. Have a look at this first image.
This is a chart of actual deaths in the United States over a recent period of time. It's based on data compiled by the experts at the CDC, so it's trustworthy.
The chart divides causes of death into some broad camps: heart disease, cancer, car accidents, strokes, etc. With just a glance, it's pretty obvious that something like 85% of deaths in the country are caused by disease or other medical conditions. People are dying because they are old or sick.
In particular, they are manifestly not dying because of violence. Yes, there are car accidents and suicides, but other violent causes of death are so few as to be negligible in a chart of this size. So, if you get your information about deaths in America from the CDC, you'll get the impression that the country is relatively safe and that most people are living long lives.
Now compare that chart with the following one, this time a chart of the mentions of causes of death in the headlines of the New York Times and The Guardian.
The comparison between the two charts is startling. Heart disease accounts for almost a third of all actual deaths, and yet it's rarely mentioned in the news. Reporting on cancer, on the other hand, is pretty accurate. But violent deaths are grossly overrepresented in the media.
The largest category of actual, violent deaths is car accidents (see chart above). That category doesn't even show up in the headlines of these two newspapers. Worse, homicides and terrorism make up over half of the reported causes of death, yet they are so infrequent as to not even register on the chart of actual deaths. So, if you get your information about deaths in America from the NYT, you'll get the impression that the country is relatively dangerous and that fully two-thirds of all people will die by suicide or at the hands of a terrorist or homicidal criminal.
This is not to say that the media is making things up. I'm sure that the stories about homicides and terrorism are largely true. But the emphasis on these deaths--to the exclusion of others--gives the mistaken impression that they are more common than they really are. A lot of fake news is technically true (that's because there are four types of news).
We get a mismatch between actual deaths and reported deaths because the incentives of news purveyors are at odds with the truth. What sells is one thing. What's true is (often) another. People don't want to read stories about old folks dying of strokes; those are "dog bites man" stories. We want to read stories about masked gunmen shooting innocents in the street; those are "man bites dog" stories. So, that's what successful news producers give us.
The punchline is that it's a mistake to blindly rely on the media to inform your worldview. Next time someone cautions you about the risks of violent death in the USA, ask them what news channel they watch and then send them to the CDC website.
*The image for this blog post was built by the artificial intelligence called DALL-E. If you're not following what AI is able to do these days, sit up and pay attention. From creating images to writing papers, the stuff at Open AI is amazing and terrifying at the same time.
Comments