Likely. It’s a common enough word. Affirmative, yet suggesting some wiggle room for vagaries.
It hardly seems decisive enough a word to determine the fate of nations—or the course of the natural world—but it is precisely its imprecision that gives “likely” its power.
Few people understand its nuances as well as APA Fellow David Budescu, PhD, professor of psychometrics and quantitative psychology at Fordham University, and a leading expert in decision-making behavior.
He has worked for decades to understand how uncertainty is communicated through natural language, most recently in discussion of climate change.
It turns out that peoples’ concepts of what “likely”—or “unlikely”—mean may hold more sway with public opinion than scientific data compiled by 2,000 leading climate scientists and other experts.
“How do people think in the presence of uncertainty or estimate it and communicate it to others?” says Budescu. “I’m interested in behavioral decision theory and how uncertainty affects groups that have to make joint decisions.”
Over a career spanning several decades, he has conducted extensive studies about the most effective ways to communicate imprecise, or uncertain, information.
He has plumbed topics including group decision making, linguistic uncertainty and test-taking behavior, but it is Budescu’s empirically based studies on the effectiveness of using more natural, colloquial words to describe science-based predictions about climate change that have drawn national media attention.
“If you were going to describe a situation that is the prototype for the highest and most complex type of uncertainty, you couldn’t come up with anything more appropriate than climate change,” Budescu says.
There are unknowns about the natural processes that drive climate, and uncertainties about the actual climate science. There are sociological uncertainties, including how fast the population will grow, and the environmental, geopolitical and economic consequences of that growth.
And perhaps most susceptible to a loose interpretation of “likely” and “unlikely” are the uncertainties posed by some people who question the existence and causes of climate change altogether.
“In terms of climate change … there are people who are using this notion of ‘uncertainty’ strategically,” he says. “Basically, as a justification to do nothing. To me that is the equivalent of, ‘We don’t fully understand cancer, therefore, let’s not treat anyone until we do.’”
The Intergovernmental Panel on Climate Change (IPCC), the leading scientific body that assesses and communicates about climate change, has been wrestling with this problem of getting the public to understand the scope, causes and projections of global warming for more than a decade.
In a recent version of their report (2014), they tried to lessen public confusion/denial about climate change precisely by framing their forecasts and conclusions in more accessible, natural “likely” language.
They worked out a table of parameters to communicate their projections—“very likely” signified greater than 90 percent, “likely” was greater than 66 percent, and “very unlikely” was less than 10 percent.
It was an effort to communicate uncertainty in less uncertain terms—using a common language and percentages agreed upon by the differing scientists, economists, demographers and others contributing to the report.
Budescu and his research partners anticipated flaws in this approach, however. “There is a lot of variability in the way in which different people understand and use these terms,” he said.
“We wondered … if you read a report and see ‘it’s likely that X will happen,’ do people necessarily attribute the same values that are in line with the expectation of the IPCC?”
So Budescu used all the empirical tools in his arsenal to find out. He led an extensive study of more than 13,000 people who read the report—sampled across 24 countries—to find out just how accurately understood these colloquial terms were.
Even he was surprised by the results.
We administered the survey to 27 samples in 25 countries in 18.
Across nationalities, respondents grossly underestimated the extreme possibilities implied by the terms “highly likely” and “unlikely,” judging both to be much closer to 50 percent. However, when those terms were coupled and presented together with their numeric percentages, respondents understood them much closer to the communicators’ intentions.
This may seem like a small ripple in the ocean of climate-change communication issues—and, in fact, the IPCC has not incorporated Budescu’s recommendation to include numerical values when using these terms—but it underscores part of the difficulty in getting people to take action on something that is perceived and understood differently.
Other fields also base their communications on “likely” terminology that should require a studied examination of language, he notes: “They are using similar language in the intelligence community, for instance. Another case where people are using words of this type is for pharmaceutical communication, where they tell you about possible side effects of medication.”
Budescu says that some people might find his granular forays into language an unusual choice for a mathematical psychologist.
“People can joke and say, ‘So, you spent so many years of your career trying to figure out what ‘likely’ means?’” he acknowledges. “But there are precise contexts where how you describe and quantify this uncertainty can drive real-life decisions and can cause people to make important policy decisions.”
Budescu became interested in psychology through an unusual route. The son of Romanian chemists, he grew up in Israel, where army service is compulsory.
“In the army, you get to spend a lot of time with people from diverse socioeconomic and cultural backgrounds that you usually don’t get to interact with,” he recalls. “And I got this sense of how much more variability there is in peoples’ attitudes, in their intelligence, and behavior.”
This widened experience of humanity led him to want to study the differences among people—and to do so scientifically.
“My thinking has always been more systematic,” he says, “so things of that type are more natural to me than the clinical and psychotherapeutic experience, which is more open-ended and subjective. I like the idea of being able to say, ‘Here’s a problem, let me write a model, and here’s a reasonable solution.’”
But there is no universal algorithm for curbing climate change. And Budescu acknowledges that much of the work he has done to try to help climate scientists be more effective communicators has been complicated by the gathering strength of skeptics.
He is not alone in this—there is heated discussion among scientists and policymakers about how to respond when “deniers” or skeptics refute scientific evidence of climate change.
How likely does Budescu think it is that his studies of uncertainty and language might help in this fight?
“Well, I honestly don’t think that the kind of work we are doing is really driving much of the political conversation,” he says, adding: “My personal perception is that changes in response to climate change are more likely to be driven by industry … there is not one insurance or large agricultural company in the world that denies climate change and doesn’t take it into account.
“And people often get their information from the here-and-now,” he adds. “I look out my window today and it’s difficult to get a sense of the global temperature going in one direction. It goes against our intuitions. But I think the moment that people are experiencing this, it has a much more lasting impact.”
APA Fellow David Budescu, PhD, is a professor of psychometrics and quantitative psychology at Fordham University, and a leading expert in decision making behavior.
