Tuesday, July 29, 2008

Dealing with Uncertainty

The New York Times has an interesting article about communicating science to the public

Though the article is mostly about the presentation of climate change and health issues, the points raised are more generally applicable. Kimberly Thompson, an associate professor of risk analysis and decision science at Harvard, is quoted with
“Words that we as scientists use to express uncertainty routinely get dropped out to make stories have more punch and be stronger,” she said, adding that those words are important to include because “they convey meaning to readers not only in the story at hand, but more generally about science being less precise than is typically conveyed.”

but she also points out that “scientists themselves sometimes fail to carefully discriminate between what is well understood and what remains uncertain,” indicating that the communication problem is two-sided, and that small inaccuracies can have large backlashes because “the flow of scientific findings from laboratory to journal to news report is fraught with 'reinforcing loops' that can amplify small distortions.”

I agree with these assessments. As I have previously said (eg in my post Fact or Fiction?) uncertainties are part of science. Especially if reports are about very recent research, uncertainties can be high. Uncertainties need to be documented accordingly, even if that lowers the entertainment value. There is no place in science for inaccuracies. I understand the need to make popular writing more accessible, but on no account should it be outright wrong. Leaving out details clarifying under which circumstances which conclusion applies to what certainty can kick a statement from being vague to being wrong. And that doesn't even take into account the distortion of media reports in echos of original articles on the internet.

As David Malone wrote very aptly in the New Scientist, Aug 2007
“We are faced with all kinds of questions to which we would like unequivocal answers […] There is a huge pressure on scientists to provide concrete answers […] But the temptation to frame these debates in terms of certainty is fraught with danger. Certainty is an unforgiving taskmaker. […] If we are honest and say the scientists conclusions aren’t certain, we may find this being used as justification for doing nothing, or even to allow wiggle room for the supernatural to creep back in again. If we pretend we’re certain when we are not, we risk being unmasked as liars.”

In a similar spirit, the author of the NYT article, Andrew Revkin, writes on his blog in a posting about Media Mania and Front Page Thoughts:
“[O]ne danger in this kind of coverage — not accounting for the full range of uncertainty or understanding in dealing with very important environmental questions — is that it ends up providing ammunition to critics charging the media with an alarmist bias.”

I am afraid 'alarming' is exactly the reason for such headlines — it's a way to get attention. Revkin ends his article with an optimistic quote by Morris Ward, the editor of yaleclimatemediaforum.org (a forum on Climate Change and the Media with focus on improving media coverage)

“Ward [...] says that it will be up to the public to choose to be better informed on momentous issues that do not fit the normal template for news or clash with their ingrained worldviews. 'At some point,' he said, 'the public at large has to step up to the plate in terms of scientific and policy literacy, in terms of commitment to education and strong and effective political leadership, and in terms of their own general self-improvement.'”

12 comments:

Phil Warnell said...

Hi Bee,

“At some point,' he said, 'the public at large has to step up to the plate in terms of scientific and policy literacy, in terms of commitment to education and strong and effective political leadership, and in terms of their own general self-improvement.'”

If this were related to religion one would be tempted to say amen, yet since this is science I will simply remind and concur that this is extremely non trivial.

Best,

Phil

Plato said...

Better to be left with the definitive term in science in this case.

Malone:If we are honest and say the scientists conclusions aren’t certain, we may find this being used as justification for doing nothing, or even to allow wiggle room for the supernatural to creep back in again. If we pretend we’re certain when we are not, we risk being unmasked as liars.”

Well it doesn't help that when one can identify "them who are in awe of beauty" by name, by others who are educated in science, who find this beauty, ugly. That a scientist can be atheistic declined, because of such a "human feeling." What does this say about "respect for another" who had devoted their life to science?

This sets the "moral tone" on how crackpotism is assigned, and to whom. How it seeps into the discussion. Its rating, and all the while, "a religion was invoked" from a defencive position and a illusive one by the vanguardof science.

Uncertainty, is clear to all scientists?

As if, a scientist by name, can hold a ethic and moral stance above another by name on uncertainty?

So what does this show to the public? What does it show to the students? What does it show to other scientists, who do not know the subject?

Most certainly:), such reviews can be good for what that book/posting/quote is portraying.

Well balanced views, are always appreciated.

Anonymous said...

It seems that scientists, because they are scientists, can have a sense of being a member in a community. But, to regard all who are not scientists as being members of another community (the public) is erroneous. Non-scientists have no sense of being community 'members' because they are 'not-scientists.'

It would be wonderful if this kind of mistake could be eliminated by science, though it is doubtful this will happen as long as scientist remain as human as non-scientists obviously are.

:(

Anonymous said...

Less hostile comment rules:

Please submit only comments that are related to the topic of the post. To do so, you will have to read it. Further, please also read the existing comments in order to avoid redundancies. If you want to introduce a topic different from the one that is posted, please consider starting a blog of your own instead. Also, this is not an open 'ask-the-expert' forum. If you have a question that is not related to the post, please ask it in a forum that can accomodate random questions.

Thanks!

Bee said...

Hi Anonymous:

? Do you think these comment rules are hostile? If so, I am genuinely sorry, that wasn't my intention. Would you let me know what is offensive about them? Best,

B.

Bee said...

Hi again Anonymous,

I've changed the 'comment rules' using your suggestions. I wasn't particularly happy with what I wrote either. Thanks for the feedback!

-B.

Plato said...

I guess people/scientist change by setting the tone.

Non-scientists have no sense of being community 'members' because they are 'not-scientists.'

You may like to think so. And you may like to think "scientists are not human." You might like to think because one is not a scientist, they work only on "feeling and on developing a philosophy" while working on the science. Scientists can hold "dual interests" while working science, just as the public can.

Impeccable-adj 1: without fault or error; "faultless logic"; "speaks impeccable French"; "timing and technique were immaculate"; "an immaculate record" [syn: {faultless}, {immaculate}] 2: not capable of sin.

You will then be our example then of a scientist? Your logic. I am only a ghost. :)

Anyway, back to the posting.

Dialogue "between scientists" would be more the example( you have to witness this in action) then tossing mud like a child, "at" and "over character," while battling the limits of knowledge with your refutations of any position.

Is it fraught with all the failings if it is left alone to speak from only from one side? Parent/child, child/child, parent to adult, child to adult, more like then, the adult to adult. More a Venn Logic integration here?:)

If you have no person who will stand front and centre, and is "impeccable in the avenues of science by example" how is it such illusive theoretical examples about representation of the facts( you know the history had to be progressed from first) can ever be thought relevant? The "limits of knowledge" had to have been reached.

I, am only Joe Public

Best,

bellamy said...

Humans are, by biological 'decree', insecure. Transcend this, embody funxional principles and allow them to emerge. Audience is secondary (I feel even in a quantum sense).

Anonymous said...

"I've changed the 'comment rules' using your suggestions. I wasn't particularly happy with what I wrote either. Thanks for the feedback!"

Gentle, but firm. I think that works.


Plato:

Dialogue "between scientists" would be more the example( you have to witness this in action) then tossing mud like a child, "at" and "over character," while battling the limits of knowledge with your refutations of any position.


Anonymous:

Anthony Blake addresses this subject well, I think.

http://www.duversity.org/TheSupremeArt.html

I haven't yet read this new book, but hope to soon. It seems an important part of the overall effort of science. Of course, this was initiated by Bohm decades ago. Hopefully Blake will be heard now and we can experience a better signal-to-noise ratio.

Thanks, Bee, for your efforts in this!

Neil' said...

So, is Anthropogenic Global Warming "unequivocal" (or deserves to be)? Anyone here have good answers?

cynthia said...

Scientists, in my view, have difficulty communicating science to a general audience because science is largely stochastic in nature and, more importantly, stochastic thinking is counterintuitive for most humans, especially for the subspecies of humans called creationists.;^) In short, most of us, as a rule, like to find a deterministic cause for everything.

Here's a comment from the econ-blogosphere which conveys what I can only dream of conveying:

Humans beings, generally, are not very good at thinking about probabilities, statistics, and, in the immortal Rumsfeld's words, "unknown unknowns".

Humans like simple narrative storytelling, where character confronts a challenge and overcomes, with good moral effect. In fairy tales, life is simple and moral, and power is magic. We like magic, we humans, precisely because it does require us to understand. The rules of magic are sensitive to our intentions, and only fussy about limits when it makes for a good plot turn.

Science is far more demanding. In fairy tale or a comic book, a character may be he granted the ability to fly, but he's not required to work to understand how to fly. He's not constrainted by laws of physics, which are indifferent to moral purpose.

Knowledge, real knowledge has a moving, foggy frontier, where discovery is often perilous. Real power brings responsibility for inintended, unanticipated consequences.

Millions think that believing or not-believing is virtuous. How moral is that?

Plato said...

Anonymous1,

With "Scientific knowledge," better to learn Html language, and to direct perspectives more appropriately.

Maybe an image is even better?

Best,