The Science Exchange, 55 Exchange Place, Adelaide SA 5000 [View map]

Communicating risk and certainty in science

Communicating risk and certainty in science
When:
Time: to

While in Sydney recently as part of the Australian Science Communicators conference I hosted a discussion on communicating risk and certainty in science. These are potentially the biggest stumbling blocks in getting the messages of science out into the public arena. There is no certainty in science and there is always an element of risk. But how we perceive risk as individuals and how it’s expressed in science are two different beasts.

Richard Dawkins recently came-a-cropper dealing with uncertainty during a radio interview about his views on atheism. Part of the underlying philosophy of science is that you cannot know anything with absolute certainty, there is always a statistical chance that a new piece of information might come along and overturn the understanding that you have of a particular phenomenon. It’s a philosophical loophole that theoretically tightens with every piece of evidence you muster in favour of your hypothesis. But that loophole can never be completely closed.  So, technically, you must never express certainty with any proposition in science.

So how did Prof Dawkins fall foul of the media in following this basic uncertainty principle? In a discussion with a bishop he was asked if he was certain there is no God. As a scientist, he could not be certain about anything but he tried to explain that all the evidence he had examined, supported the proposition that there is no God. He expressed his confidence in that hypothesis as 6.5 out of 7 (a strange metric but then Dawkins is often unconventional!). He was only doing his scientific duty according to the underlying philosophical tenants of science. But how did the media report this? “Dawkins Unsure There Is No God” –  gave the impression that the level of uncertainty was closer to a serious doubt rather than a statistical improbability.

“The Science Is Not Settled” is a catch cry of the climate denial clan, implying that the level of doubt is a confounding challenge to climate science rather than an increasingly small statistical chance that they have it all wrong. Technically the challengers are correct, the science is not settled, because the science is never settled on any subject. But the science is beyond reasonable doubt in this particular instance. If the weight of evidence supporting the proposition that the climate is changing and that human factors are responsible were ever to be tested in a court of law, it’s difficult to imagine that any reasonable jury could not return a guilty verdict.

That’s clearly not how the media deal with the inherent uncertainty of science in this instance. The journalistic credo of balance demands that opposing side are presented with equal merit and gravity. While this is perfectly acceptable, and should be mandatory, when reporting political debates, it utterly fails to convey the true landscape of scientific thought on the issue. The same old anti-climate war horses are wheeled out time and again to spout the same, disproved and irrelevant factoids all in the name of creating ‘balance’ and exploiting the nature of uncertainty. 95% confidence is reported as a one in twenty doubt. Objective analysis of the scientific data flies straight out the window.

Science communicators do have to remain true to the principles of uncertainty and try to convey degrees of risk in a media environment that demands black and white; a terrain that seemingly cannot deal with subtly and complexity. Add to this that the presentation of science in the media is often in the context of debates with non-science based entities who do not want to play by the science rules and the frustration of conveying accuracy almost becomes self-defeating.

So what should we do? Should we keep the uncertainty to the science and present confidence in our communications? Do we have the capacities and abilities to educate the public into understanding the subtleties of certainty as a scientific concept? Do we soldier on and try to convey risk in a more accessible way? As you may now imagine, this became an intense discussion on the night and I don’t think we came up with any simple solutions. You will be able to make up your own mind soon because the whole discussion was captured by the ABC for later broadcast on its Big Ideas program – I’ll keep you posted as to when that goes to air.

By Paul Willis, @Fossilcrox 

4 Responses to “Communicating risk and certainty in science”

  1. Anne Powles says:

    As a the one non scientist from a scientific family of origin I contend Scientists have to be true to their science and allow in doubt always, like Dawkinns, but explanations, such as above must be plentiful, especially to the young. The mathematics of statistics must be taught more comprehensively in schools.

  2. Sarah says:

    Oh Paul,
    As long as you are blinded by certainty you’ll never win this argument because you don’t offer ANY evidence for “the proposition that the climate is changing and that human factors are responsible.” And none of the other ACC proponents see the gaping hole in the public discourse. “We all agree” is not evidence in support of anything otherwise Dawkins would be convinced by theologians who are the only people actually qualified to make statements about God.
    Don’t get lost in the rabbit tracks of denier assertions think about the whole of the proposition: that the Climate is changing – Evidence is plentiful; that human activity is responsible – evidence for this is never broached. I had to look up Tindal myself and he didn’t think the tiny emissions from an infant industrial world were sufficient to tip any balance significantly. Deniers would jump on this. Climate is dynamic as we know from D-O events in every age. I dare not mention Bond events so as not to sidetrack the discussion. Of course what we do impacts the globe. We live here we are part of this.

Leave a Reply


3 + 5 =