To end this public health crisis, journalists need to humble themselves
It's a tough truth: most people won't read your article, journalists. You still have a responsibility to them.
Back in December, I wrote an article about a disturbing trend I noticed in media coverage about the newly available COVID-19 vaccines. As with virtually every medical intervention, a handful of people had adverse reactions to the vaccine.
Several mainstream news outlets jumped on these stories, which is fair enough. It’s important that info about those reactions be made public, if only for the sake of transparency. But these outlets — The New York Times, CNN, CBS, Politico, and others — were hyping up the adverse reactions to a disproportionate degree.
To borrow from a point made by Carlos Maza during our conversation last week, the responsibility of news organizations to inform the public is often at odds with news as a business.
This coverage triggers a sort of availability heuristic-type reaction. What’s even worse: the press is doing it again.
To quickly summarize, the availability heuristic is a concept by which the probability of an event or the general correctness of ideas are judged by how easy it is to recall examples that fit the scenarios.
It’s a form of unconscious bias that is heavily shaped by what the press chooses to cover (and those choices are often extremely understandable). Things that make news tend to make the news because they aren’t part of an average day. If a flight from Chicago to New York crashes soon after takeoff, you’ll probably hear about it on the evening news broadcast; if it completes its trip safely, it won’t make the news. This may result in people overestimating the risk of flying.
The Present Age is a reader-supported newsletter. While a free version of the newsletter exists, paid subscriptions make this work possible.
Another example is lottery tickets. You don’t hear about losing lotto tickets, but if your corner store sells a ticket that wins the multi-million dollar prize, you may see the story in the news or in casual conversation. The existence of a winning ticket that you heard about could lead you to overestimate the odds of winning.
This is exactly the problem that came out of the vaccine side effects stories, and it’s happening again with recent reporting on so-called “breakthrough cases” of vaccinated individuals getting infected with COVID-19. Take a look at some of this recent coverage and see if you can spot the problem:
The Washington Post’s July 30 article about a COVID-19 outbreak in Provincetown, Mass., was far less than ideal. “CDC study shows three-fourths of people infected in Massachusetts coronavirus outbreak were vaccinated” is an alarming line to read. “But few required hospitalization” makes it slightly better, though it still leaves readers (or, more accurately, as this group makes up a much larger share of people who will encounter the article: people who see the headline on social media but don’t click through) with the idea that perhaps vaccines aren’t working as well as we’ve been told.
It’s unfortunate headlines like this that fuel availability heuristic thinking when it comes to vaccine efficacy and the likelihood of a vaccinated person contracting the virus.
A day earlier, the Post published an article that warned, “The delta variant of the coronavirus appears to cause more severe illness than earlier variants and spreads as easily as chickenpox,” and also that “vaccinated people infected with delta have measurable viral loads similar to those who are unvaccinated and infected with the variant.”
While the article aims to shine a light on the Center for Disease Control and Prevention’s decision to update its mask guidance last week, a quick look at the quote tweets on the Post’s original article shows what the takeaway of the more conspiratorial and anti-vax segment of the population was:
The Washington Post @washingtonpost
Delta variant infections are probably more severe, CDC document says, citing data showing vaccinated people can spread coronavirus https://t.co/PRJiMxRnpz
The Washington Post @washingtonpost
Delta variant infections are probably more severe, CDC document says, citing data showing vaccinated people can spread coronavirus https://t.co/PRJiMxRnpz
The Washington Post @washingtonpost
Delta variant infections are probably more severe, CDC document says, citing data showing vaccinated people can spread coronavirus https://t.co/PRJiMxRnpzYes, vaccines work.
Yes, the vaccines are still extremely effective. No, the vaccines are not “destroying your immune system.” No, the vaccines are not the delta variant. But that’s what people took away from the article — and why wouldn’t they? It was presented in the most over-the-top, hyperbolic way possible. Rather than taking away the correct message (that unvaccinated people need to get vaccinated as soon as possible), the unvaccinated are seeing this as a win, as a validation of their preexisting beliefs.
Here’s another example of terrible messaging via terrible headlines, from NBC News:
125,000 vaccinated people have tested positive!!! That’s alarming!!! Or is it?
That 125,000 number was out of a total of more than 164,200,000 people who were vaccinated. In other words, less than 0.08% of vaccinated people (roughly 1 in 1,300) caught COVID-19 in that time. The takeaway here is that the vaccines are extremely effective at both preventing infection and even more effective at keeping people out of the hospital.
As people pointed out how misleading the headline was, the author of the story, NBC’s Laura Strickler, came to its defense by noting that the context was included in the story itself. I tried to engage with her point, but she never responded.
She went on to respond to criticism of the headline by copying and pasting a response containing context.
She followed this up with some additional context that would have made much better headlines. Unfortunately, these were in the article itself, which the overwhelming majority of people did not read.
Journalists need to understand where the public is at when it comes to how information gets presented.
A 2014 American Press Institute study (yes, I know that 2014 is ancient by today’s standards, but this at least gives us an idea of how people behave on the internet) found that 60% of Americans read only headlines in the week before they were surveyed. Think about that. It’s not that 60% of people didn’t click on a specific article, it’s that 60% of people didn’t click on anything at all. Whatever is in the headline or the tweet is all the majority of Americans will get.
In 2016, researchers found that 59% of articles retweeted on Twitter were shared without being read by the people sharing them. In 2020, Twitter itself acknowledged this problem and began prompting people to click the articles themselves before retweeting.
There are two routes that journalists and media outlets can take with the understanding that the majority of the public isn’t at all interested in reading articles themselves and instead base opinions solely on headlines (and tweets):
Begin putting the most important information in the headlines themselves so that people just seeing it at a glance will come away more informed even if they don’t click through.
Change nothing and instead insist that it’s readers’ fault if they don’t actually read the articles, even if it means that the majority of people who encounter the article in passing will come away with the wrong impression (which can be deadly when it comes to public health).
Those are the two options. One requires humility, the other arrogance. Too often, journalists and the outlets they work for choose arrogance. The press needs to reflect on the role it plays in shaping public opinion. If 90% of the people who see a story that highlights the effectiveness of vaccines come away from it feeling that vaccines are less effective, are you really going to just shrug and say that it’s “their problem” because they didn’t read it? Or are you going to consider that if something you put into the world is having the opposite of the intended effect, that it’s a problem of your own?
This is a problem I’ve written about for a long, long time. In 2019, I tried to guide newsrooms on this, urging them to “write your tweets and headlines as though they’re the only thing people will see” (because, for the most part, they’re the only thing people will see).
One outlet that nailed the framing here (which was surprising to me, as I’m usually pretty critical of it) was Axios. Good on you, Axios.
Thank you for this--I get you've been saying this for a while, but this is still a powerful and necessary case you're making. The statistics about headlines are eye-popping.
I have heard of a proposal that social media outlets require a "three questions"-type of approach that requires the user answer three content-related questions before reposting/sharing an article, to demonstrate that the article has been at least cursorily read and understood. While I recognize that this is an imperfect one, I think the idea leads to some interesting possibilities for redirecting the onus of information-sharing and for slowing down the process altogether so that we can actually take in what we read rather than eating information like Doritos.