By David Poulson
The post-election analysis of the U.S. presidential race contains excellent lessons for communicating research and science.
The question many people are now puzzling over is how could all those highly-educated, highly-paid statisticians and pollsters get the election so wrong. Seemingly no one projected a Trump victory.
And now the science of polling is taking a beating. It may never recover.
Perhaps it never should.
The pushback from the research community is that it didn’t get it wrong. The numbers were good. But the media either failed to report the margin of error, or if it did report it, it failed to explain the concept of margin of error.
Reporters also get nailed for not explaining the challenges of getting a representative sample or the moving target of attitudes as the election neared. Instead, the media went not with the fuzzy truth, but with the story it wanted to tell in a typically superficial and sensational manner, say the critics.
The researchers’ advice: Every journalist needs to take a stats class.
That’s one view. But there is another one, and it is one that is relevant to all researchers.
I’ve certainly made similar assertions about statistics to my students. Our faculty have even discussed if statistics coursework should be a requirement of a journalism major.
If not an entire class, journalists need a basic enough understanding of statistics to ask the right questions. And certainly many very good journalists have more than a basic understanding.
But there are two sides to a communication transaction. The election projection misfire isn’t just on the journalists.
As a former environmental reporter, I cannot count the times a scientist has patronizingly told me that I really needed a degree – let alone a single course – in her or his field before I could adequately write a story about it.
How can you write about climate change without a grounding in atmospheric chemistry?
How can you write about soil fertility without studying agronomy?
How can you report about the Hadron Collider without a degree in physics?
How can you explain endocrine disrupters without a course in environmental toxicology?
Remarkably, many journalists do such things. And they do it well – just before tackling another equally complex but entirely different field quickly and with limited background.
Which is not to say that the point is not well-taken. But it’s just not going to happen. Even when relatively flush, news organizations could never afford the tuition to give a journalist coursework for every field of study she or he might have to cover. You may think it unfair, but they simply cannot support enough reporters to specialize in every conceivable niche.
Even if they could, the journalists would never have the time to report. They’d be in class all the time.
Here’s another view: Every scientist and researcher needs to take a basic course in communications.
Wouldn’t that be far more efficient? Who better than you understands your field?
That said, it wouldn’t be easy. You’d lack the easy advantages of writing for an academic journal with an informed audience and no real requirement to engage it. You’d have to make hard decisions on what not to tell rather than reiterate every piece of research leading up to yours.
You’d need to link nouns and verbs into coherent sentences – a surprisingly unused skill among many of us.
And you’d have to practice talking to the media if you want to reach outside of your lab.
Now the easy response is, “That’s not my job.” But I argue that it is precisely your job if you want your work to inform critical decisions. Don’t complain about the absence of science in policy if you’re not willing to take this on. Who better than a scientist to be involved in civic discourse?
Or, like the pollsters, you can dump information on the media with little context and then complain that reporters don’t know enough about your field to adequately represent your work.
But that doesn’t make you a responsible contributor to the public weal. That makes you little more than an unthinking calculator. You’ve got brains, perspective, knowledge of what is at stake.
Despite the obstacles faced by the grammatically challenged, learning to communicate well is certainly possible.
Ole J. Forsberg, an assistant professor of mathematics at Knox College, recently did an excellent job of defending polling in The Conversation, an independent source of news and views from the academic and research community that targets the public. Here is some of what he wrote:
“Nate Silver’s model gave Trump a 29 percent chance of winning the presidency. My model gave him a 20 percent chance. What do those probabilities actually mean? Flip a coin twice. If it comes up heads both times, you just elected President Trump – two coin tosses in a row coming up heads has the same probability of happening that many of these polls gave for Trump moving into the White House.”
Or, as Forsberg told his class on election day: “Flip a coin twice. If it comes up heads both times, say ‘President Trump.’”
Hmmm…not bad for a mathematician.
David Poulson is the senior associate director of the Knight Center for Environmental Journalism and director of the translational scholars program for the Global Center for Food Systems Innovation, both at Michigan State University. A similar version of this post appeared on The Food Fix.
By David Poulson