To my dismay, even after a good night’s sleep and a fresh perusal of the paper, after reading both of Greg Laden’s thorough articles, Mooney’s latest summary, Orac’s claim that it’s nothing but tailoring your message to your audience, and Nisbet’s roundup of responses, I’m still hopelessly confused. What the heck is this paper telling me to do?
Here is my crude, primitive and confused understanding of frames. If I am an advocate for science, I should avoid saying, “I like science, and I strangle puppies!” I should say instead, “I like science, and I snuggle puppies!” OK so far, I can agree with the general concept, even if it does seem a little obvious … but then, that could be more the fault of my ignorance of the idea than anything else. Unfortunately, I’m not getting much more than that out of the Nisbet/Mooney paper.
I’m also seeing examples of bias in the work. What if my goal is to be an advocate for strangling puppies? Shouldn’t my comment instead be seen as an example of good framing, trying to link my puppy abuse to a positive frame of science? I was a bit put off by the fact that the authors single out religion as something that must be respected—it gives the impression that Nisbet/Mooney consider atheism something akin to puppy strangling, a habit to be practiced in a dark closet and never to be discussed in polite company.
There’s another common scientific practice (not that puppy strangling is common among scientists…) that they tell us to avoid: technical details.
In short, as unnatural as it might feel, in many cases, scientists should strategically avoid emphasizing the technical details of science when trying to defend it.
Oh, no … so we can’t use this webcomic anymore?
Now we all know that we have to dole out the technical details appropriately—I’ve misgauged an audience a few times myself—but our possession of the data is one of our greatest strengths—if we’re going to start equating explaining the evidence to puppy-strangling, we might as well hang it up and go home right now. Rather, I think I’ll argue that Nisbet/Mooney are using poor framing, and what we ought to be doing is hammering away with the wonderful evidence we have, and pointing out the lack of evidence on the other side. Now it’s fair to say we shouldn’t explain PCR and list nucleotide-by-nucleotide differences produced by evolutionary mechanisms, but general principles and explanations of how we know what we know are a core of science; you can’t simply tell us to avoid it. It’s like suggesting that we could do a better job of promoting science if we could only hide that sciencey stuff.
I have the feeling that if I had a Nisbet/Mooney Training Seminar in how to frame science, I’d end up giving fluff talks that play up economic advantages and how evolution contributes to medicine with slides of puppies rather than squid, and I’d never talk about mechanisms and evidence again. That sounds like a formula for disaster to me—it turns scientists into guys with suits who have opinions, and puts us in competition with lawyers and bureaucrats in the media. It’s saying that we should abandon our strengths and adopt the strengths of the other side. Bleh. I think I’ll have to pass.
I have to admit I haven’t had time to read the Nisbet/Mooney paper, but I’m guessing they’re referring to the sort of framing George Lakoff writes about (most accessibly in “Don’t Think of an Elephant!”)
The idea isn’t to throw puppy strangling into the equation at all, but instead to be sure that you aren’t using the terms of your competitors to describe your situation. (For example, gay marriage makes people squeamish because “marriage” has been so successfully campaigned as “between a man and a woman,” so instead you’d frame the argument around whether or not you think it is appropriate for the governmente to tell us who we can and cannot love. None of the facts or evidence change, it’s just a question of how you sell them to the public at large. If this is what Nisbet/Mooney are arguing, I have to admit they’re right on the money with this technique for science, at least in the anti-intellectual contemporary USA.
I think you more or less accurately or at least adequately summarize what is meant by framing in the Science piece. But for me, that thing … that “framing” … is roughly equal to “spinning.” This is probably important.
I feel, as I’ve said in volumes now, that the original concept of a “frame” is actually more useful. So, consider this collection of words:
Fact
Assertion
Assumption
Belief
Observation
Hypothesis
Theory
and you can obviously add more. These words and related words and phrases have different meanings in different frames (that is what frames do) and Science needs to pay attention to the fact that when we blather outside of our frame (actually, when we sit in our frames and blather at others in other frames) we may not actully be communicating properly.
I advocate that scientists learn this and adjust some things, and at the same time the public needs to be “framed up” more than it is.
Consider how you speak to a colleague in a different science. You know that there are words that when you get to them you have to make sure that your colleague knows what you mean, because over there in that discipline they use the same word but with a different meaning.
We need the public to get on board with more of this and at the same time we need to do some of the work ourselves. In the manner of Dale Carnegie, I’ll say it and I’ll say it again and again: Let’s try to drop the “theory” from “evolutionary theory”… this word is doing so much work out there in normal language, it DOES NOT MATTER that scientists think “theory” means a certain thing. It does not. that is not how language works. The minority does not get to define the meaning. So let’s replace our phrases on that one, switch Evo. Theory to Evolutionary Biology, for instance. That’s easy for us to do.
But the word “belief” and it’s derivatives .. Just because the public “believes” stuff does not mean that scientists should go around talking about beliefs. We speak of observations, evidence, even facts, but not belief. The public needs to come towards us on that.
… And damn, PZ, I can’t believe you spilled the beans about the puppies!!!
I tell you what — when the mainstream “legit” free press media (local newspapers included) stop ramming down our throats what a great guy Jesus was and how wonderful it is that he saved us all by his suffering and dying and how marvelous his back to life trick was and how we now have hope because of it, etc. etc. etc. (gag gag barf barf) like all are proven undisputable facts we all NEED to believe in in order to live any kind of a fulfilled and happy life… we – at least me – will become less hostile. However — science is science and we can NEVER “never talk about mechanisms and evidence.” That – dare I say it – would be a MORTAL SIN of a true order!!!
Robin Zebrowski, for the win.
Speaking as a semipro rhetorician, I was into Lakoff before Lakoff was cool. Here is his article Simple Framing, which is kind of a beginner’s look into what weird high-level communicators do and think about.
To recap in brief, every concept we talk about or think about evokes a rhetorical frame, a structure of connotational and denotational baggage that goes with that concept. (I mentioned the difference between connotation and denotation in my comment on the thread about the Nazi anatomy book, incidentally.) Denotation is the literal meaning of the term. Connotation is its symbolic or emotional meaning. So, for example, in the US, the most common term for the legal wedding of two individuals of the same biological sex is “gay marriage,” whereas in Canada, the term “same-sex marriage” is favoured. Denotationally, the terms are equivalent. Connotationally, “gay” carries much more of a negative (pejorative) weight than “same-sex,” which carries a quasi-legalistic connotation.
When scientists — or any debaters — pick up the terminology of their opponents, they have already conceded that connotation set. The difficulty is to find a phraseology that substitutes a completely different frame set for the one you’re trying to replace. As Lakoff points out, if the existing frame lies about a concept, “the frame will stay in place and the truth will dissipate.”
You’re probably just overthinking… :)
More poor framing. We don’t strangle puppies, we sacrifice them, a word that has much more positive connotations to Bible-wallopers.
I’d have to agree with you again, PZ, after having read the Mooney/Nisbet article, and organized my own thoughts a bit overnight. Matt has added (in a comment on several blogs including mine, but responding to Greg Laden’s critique) to the references cited in their paper on framing theory and analysis. While that gives me more to read, it still doesn’t add to this particular paper, nor offer any further help in terms of what we are supposed to do. Perhaps their seminar/roadshow will offer more, but for now they haven’t got me jumping to sign up.
I agree. It is probably only because I am a biologist that I have this opinion or experience, but in my opinion or experience, it seems to me that the biological sciences are the most open to this issue. The issue of having organized groups (mostly religious) claim that we are idiots and our science is all wrong.
In recent months there has been increasing “scientific debate” about global warming. But in that case I don’t see either side claiming that the climate is Intelligently Controlled by the Great Thermostat.
Currently, most people believe that cell phones, computers and most of the other complex science/technology in their daily lives was created by men and women using science, and that at least some engineers still understand how these things work. They don’t try to believe that supernatural powers are involved in, for example, transporting the cell phone signal from the cell phone up to a receiving tower half a mile away.
Although I sometimes get the impression that Steve Wolfram takes more credit that is due, or over-inflates the importance of his work by calling it “A New Kind of Science”, I do think it is an impressive method of demonstrating to laymen how very simple rules can add up to very complex behaviors.
The interaction that most people have had with computers and computer programs so far, has not been with software designed with such very simple rules. I think this is at the heart of why people look at the complex “machines” of living things, and find it so difficult to believe that there is not a very human-like intelligence involved in the design.
The technologies we use today, including cell phones and computers, could not have evolved via Darwinian forces in the past 2,000 years. Human social and technological evolution is super-Lamarkian. We learn not only from our own experiences and the experiences of our parents and grandparents; we also learn from the experiences of thousands of other people. When Johannes Gutenberg helped make printing more affordable, an exponential leap in our power to share ideas ocurred. The Internet is now creating another similar leap.
Until people begin to experience software that has been created by simple Lamarkian and/or Darwinian sources of “intelligence” they will remain skeptical that very simple rules can create complex behaviors.
The work of climatologists and other scientists is beginning to convince people that an intelligent agent is not required to control each puff of wind and the building of each snowflake. It may be several more generations before most people can grasp some of the same ideas about biological phenomena.
An excellent and trenchant point from Greg Laden, too, particularly about “theory.”
“Theory.” What a loaded term. It’s just dripping with all sorts of connotations you don’t want. How many cretinists use the “Evolution is just a theory!” (and you’ve got to love the semantically-squishy “just” there, as well) canard? They’re deliberately confusing a precising definition with the vernacular definition. (I know you know what a precising definition is, so keep up with us, here.) And I don’t mean they’re getting it confused; they’re deliberately muddying the waters.
To amplify on Greg Laden’s point, scientists don’t need to stop talking about theories (or laws, there’s another good one — how many times have you seen the “breaking the law” analogy misused?) amongst yourselves, but, when dealing with the public — who has already been bombarded with this “theory/law” crap (or, should I say, to remove some of the negative connotation by reframing — “who has been exposed to these vernacular-centric uses of language”) — for thirty years now, use another term. It’s really as simple as that.
As it stands right now, you can’t win and you can barely break even, so it’s time to try something else.
As an audience member for “popular science”, I suggest that you give it up. You’re so far out in left field and creating so may straw men by now that I wouldn’t know where to start in trying to give you this audience member’s perspective. Just go do some science …
If I understand the framing concept correctly, you do this by pointing out the amount of dogshit there is on the pavements, and blame the puppies for it all. That way you’ve framed the issue, and the puppies.
Bob
First of all – and please read the entire paragraph before jumping down my throat! – I have the utmost respect for many genuine religious beliefs and practices. That said, that respect is MY opinion and I have zero interest convincing others to agree with it (or disagree with it). Specifically, I see no reason why, if a scientist finds a particular belief wrong, or finds the very concept of religion wrong, they shouldn’t say so, in whatever language they feel is appropriate. Of course, it’s good social hygience not to needlessly make enemies, but it’s equally good social hygiene not to finesse what you believe passionately. In other words, I often disagree strongly with Dawkins, but the last thing I’d do is tell him to shut up.
Secondly, PZ is 100% right: Remove or minimize the data, skip the technical details, and what’s left is not science, but just assertions. It reduces popular science writing to a prime example of the fallacy of arguing from authority. And I’m speaking as a layperson, here. I WANT the data and technical details (explained well goes without saying).
Finally, if we laypeople have problems reading graphs, undertstanding math, and grasping how to infer meaning from data – and many people surely do – that is OUR problem, not yours. The best thing to do is not to drop the technical stuff, but help laypeople understand it better. And understand the importance of it.
Yes, indeed, that requires world-class writing skills. But it doesn’t require re-framing. That requires expertise. And given that most of us have trouble becoming expert at even one thing, that is an awful lot to require of scientists struggling with knotty problems that require deep concentration and focus.
Scott,
I understand your sentiment, and that would be nice, but as long as there are politically powerful and well funded movement to shut us down we can’t fiddle while the town burns. I think I mixed my metaphors, sorry.
Interrobang and others:
One of the points here is to make our rhetoric more digestible. The suggestions was made (in the Science piece, e.g.) to play to the existing biases. We should not do that because the whole point of science is antithetical to this.
But we can certainly USE the existing biases. I have found that one of the best ways to “teach” is to play to the falsehoods … The things people believe that can be used as a “cultural receptor site” for some cool science concept or bit of information. Like the idea that you think your dog can understand your language, or that lightening never strikes twice in the same place, etc. There are a range of ideas from trivial to critical that can be used. The “public” (person) need not personally believe the falsehood …. they just have to know about it, or at least believe you when you tell them it exists.
I’m not sure that this fits, but…
I wouldn’t say give up your strengths, but provide some more digestible tidbits for general consumption.
If I go looking for (or write) a JAVA/C++/C# code sample, I want something readable that shows the basics, not the more thorough code I would actually use in production software.
How do I output a variable to a DOM element javascript?
This optimistic code:
function dump(text, elementName, index)
{
var elements=document.getElementsByTagName(elementName);
elements[index].innerHTML = text;
}
is a lot easier for a beginner to understand than this:
function dump(text, elementName, index)
{
try
{
if((text)&&(elementName)&&(index >=0))
{
var elements = document.getElementsByTagName(elementName);
if(elements.length > index)
elements[index].innerHTML = text;
else
{
//do error handling
return;
}
}
else
{
//do error handling
return;
}
}
catch(error)
{
//do error handling here
return;
}
}
but I would probably use something closer to the second in actual production.
When I read the technical information here, I tend to skim over some of the finer details, hoping to get a general understanding of the process. I don’t know that everyone can do that.
I’ve said this in several different fora, so I’ll keep it short here. Framing is not some alien topic. Science itself is a “frame,” a way of knowing the world that is unique and self-contained. There are some topics that science can’t fully address because of its limits (justice, beauty, the nature of truth, etc.) just as there are some topics that other worldviews or frames can’t address. If we want people to understand the way the world works from a scientific frame, then we are definitely in the realm of science education. But – and here’s where all of these discussions seem to be going off the rails – the best research on science education confirms the idea that understanding and working with framing is essential to being better science educators. Check out How People Learn and see if you agree that this National Research Council book about education can help us understand political framing.
Bob and PZ:
You can SPIN your puppy idea, but you don’t frame it (at least, what you are talking about here is not framing). Framing is about meaning. So you can say “My dawgs sure could use a good stranglin …” and in some parts of Arkansas that may mean that you need a foot massage, in some parts of Oregon that may mean you are about to strangle the puppies. The different meanings come from the same utterances uttered in different frames.
Goffmanian Frames, that is. To put this in the right frame.
RBB:
I basically agree with you but I want to jump in and decouple “Frame” and “World View”
Simply (may be too simply) put, a World View is a frame, and I don’t think you are saying anything here. But if we proceed assuming that W.V. = Frame we get in trouble, because the most interesting and useful stuff about frame analysis is that we are frequently … and critically … shifting frames in day to day live. But we don’t really shift world views (sometimes, but not really) on a regular basis.
One of the most common frame shifts is the suspension of disbelief when we engage in, for example, fiction.
Being “on the same page” is being “in the same frame” … one makes constant adjustments to keep on the same page with whom ever one is communicating, and we have problems, from very funny moments (my freinds mom: “Before your wedding, I think you should have an affair” … gurns out she meant a party, like a shower … ) to the totally disastrous (I suppose a good but too technical example is when NASA used the wrong “frame” … units of measurement that one time and crashed a space ship into some planet. Or, if you’ve read The Sparrow/The Children of God … when the priest says “I am not married because I live to serve others” … if you’ve read the book, you know what I mean … ouch).
I think this is similar to Dawkins vis-a-vis Neil de Grasse Tyson as best illustrated by this short video:
http://onegoodmove.org/1gm/1gmarchive/2006/11/beyond_belief.html
I think both approaches have their place. We need communicators with both styles for differences audiences.
…
…
In the words of SF author Somtow Sucharitkul, the puppies are “compassionately devived.”
In the words of Brent Rasmussen, second-shift assistant supervisor in the Puppy-Grinding division of the Evil Atheist Conspiracy®, “Sure it’s cruel, but think of the jobs!”
…
…
PZ asked:
“I’m still hopelessly confused. What the heck is this paper telling me to do?”
It’s telling you to sell, not just explain. We do it in grant proposals all the time. If amount/intensity of framing increases from left to right (the scale might be exponential), the relative magnitudes of framing one should employ would be:
ms–talk—–most of grant app——–SA page————————laypeople
…with “SA page” meaning the specific aims page of your grant application, usually with a one-paragraph introduction and conclusion flanking the specific aims. Including transgenic line numbers, 30-character genotypes, etc. on your SA page is suicidal.
I think you don’t get this because you aren’t in a situation that involves writing grants.
PZ,
I appreciate Greg Laden’s intense interest in framing. His latest post, however, dismisses literally hundreds of peer-reviewed papers across the fields of communication, political science, and sociology.
Though he references Goffman’s work from 1974, over the past three decades, research in the above fields have developed framing as a theory of media influence. For overviews and applications of the literature see the citations we reference in the Policy Forum article:
Price, V., Nir, L., & Capella, J.N. (2005). Framing public discussion of gay civil unions. Public Opinion Quarterly, 69, (2), 179-212.
Gamson, WA. and Modigliani, A. (1989). Media Discourse and Public Opinion on Nuclear Power: A Constructionist Approach. American Journal of Sociology, 95, 1-37.
See also the latest issue of the Journal of Communication, the flagship journal in the field. It’s a special issue devoted to framing and media influence. See especially the following overview:
Scheufele, D. A., & Tewksbury, D. (2007). Framing, agenda-setting, and priming: The evolution of three media effects models. Journal of Communication, 57(1), 9-20.
See also this earlier article by Scheufele, possibly the most heavily cited article in the field over the past decade:
Scheufele, D.A. (1999). Framing as a Theory of Media Effects. Journal of Communication 49 (4): 103-22.
On your comments about a workshop on framing, I think it would be incredibly useful and interesting to get together on the phone or in person to discuss the topic. Better yet, perhaps we or someone else could organize a mini-conference on the matter, out of which something could be published.
Best,
Matt
I think there is an important point from the framing article that is being missed.
“Frames organize central ideas, defining a controversy to resonate with core values and assumptions. Frames pare down complex issues by giving some aspects greater emphasis. They allow citizens to rapidly identify why an issue matters, who might be responsible, and what should be done.”
This statement doesn’t say to change or dumb-down the data, but instead put it in a context to “resonate with the core values and assumptions” of audience. A good example of this is what has happened with HIV/AIDS in Africa and conservatives over the last few years. At the beginning of the Bush administration conservatives didn’t really care about HIV/AIDS in Africa and nothing was being done about it. Then, someone framed the issue for them that millions of innocent women and children were dying. This resonated with their conservative beliefs that life is sacred. The issue didn’t change at all but it was finally framed properly. The Bush administration and conservatives have subsequently pushed for significant funds to be directed toward research and treatment for HIV/AIDS in Africa.
The key is to “pare down the complex issues” for the intended audience. If the data will confuse the audience explain it with words instead of graphs. Simplify the most important point so it speaks to the audience and doesn’t make their eyes glare over.
RBB: Do you have a blog? You should. Collect your comments and put them all in one place.
Matthew C. Nisbet:
As a card-carrying member of the APS, it is my Official Physicist Bastard duty to point out that hundreds of peer-reviewed papers in sociology are still just hundreds of papers in sociology. This is a special case of the mathematical principle that N times nothing is still nothing.
— Tom Lehrer
I think people who want to engage these belief changing arguments might find this old CSICOP article useful, “Why Bad Beliefs Don’t Die.”
http://www.csicop.org/si/2000-11/beliefs.html
Also, another concept that might be useful would be the “Overton window,” and I’ve got a bit of information on that in one of my blog posts:
http://normdoering.blogspot.com/2007/04/dont-try-this-at-home.html
More seriously, I think that the claim that “framing” is “nothing but tailoring your message to your audience” (to use PZ’s words) misses a big point. To put it more precisely, I think that glossing over subtleties for a freshman biology class or even “spinning” a grant proposal is not the same as doing a more-slick-than-accurate PR job for the general public. As I said at Respectful Insolence,
I couldn’t give a pair of fetid dingo kidneys for a “frame shift”, but changing the feedback mechanisms at work in particular arenas could do a world of good. expanded on this point:
Dang it, I can’t get my HTML right. That second block quote came from Torbjörn Larsson.
Yeah, right. I don’t buy it. You’re not dumb. Stop pretending to be.
Hi Blake,
Our Policy Forum article, with its citations in the paragraph where we define framing is pretty clear as to its anchoring in the literature.
Price, V., Nir, L., & Capella, J.N. (2005). Framing public discussion of gay civil unions. Public Opinion Quarterly, 69, (2), 179-212.
Gamson, WA. and Modigliani, A. (1989). Media Discourse and Public Opinion on Nuclear Power: A Constructionist Approach. American Journal of Sociology, 95, 1-37.
Best,
Matt
Foley wrote: Until people begin to experience software that has been created by simple Lamarkian and/or Darwinian sources of “intelligence” they will remain skeptical
Dunno about s/w – people don’t see it and don’t think much about it … and “any sufficiently complex technology is indestinguishable from magic” as the old saw has it.
Hardware, tho – now that’s something people can appreciate when it’s designed by a non-human process:
http://en.wikipedia.org/wiki/Evolvable_hardware
Their link to NASA’s project using a genetic algorithm to design a novel antenna is especially intuitive.
I’d guess that if nano-technology takes off, it’ll involve evolutionary processes. Hopefully to positive effect.
On a related topic, I’ll note the work of some literary leaders who actually know science, to further an evolutionary and essentially statistical perspective on human lives and narratives, reframing it from the folk-psychology stories we now tell:
Stanislaw Lem’s science fiction
Antonia S. Byatt’s essays, novels and films. Her essays in HISTORIES AND STORIES lay out the perspective explicitly.
#18 (John) got it right. They’re asking us to bring out the Specific Aims (SA) page from our R01s to do the fighting. Take the audience by the hand, tell them why its important to THEM and what you’re going to do about the problem. Do it clearly, simply and concisely. We all know if we cannot do it on the SA page we’re getting triaged. Its simple!
Matthew C. Nisbet:
I’m not disputing the well-anchoredness of your terminology in the sociological literature, nor am I trying to imply that you invented these concepts out of whole cloth. If anything, I am trying to point out (in my fumbling way) that practitioners of the “hard sciences” may have little inclination to take grand sociological notions seriously, particularly when those grand notions sound (or can easily be made to sound) like they conflict with concepts of scientific integrity.
I would also suggest “A Scientist’s Guide to Talking with the Media,” published by the Union of Concerned Scientists.
I would also suggest “A Scientist’s Guide to Talking with the Media,” published by the Union of Concerned Scientists.
PZ, you are a teacher and a scientist. If you find that a student is not responding to the way you first present the science, I’m sure you instinctively seek to present it in a way that is closer to what the student can understand. You don’t have to love the student, feel compassion for them, or even pity them; you just need to get the stuff across in the way that hits the target hardest and sticks the best. I think that’s what they mean by framing.
That said, those of us who are not scientists and who also don’t want to be talked down to desperately need your sort of teaching and your sort of challenging talk. I would personally and selfishly be quite upset if you changed your tone to suit the lowest common denominator. Keep on pitching those fast balls; they’re the most exciting part of the game. It’s our job to learn to catch them.
Ah… coturnix in comment #21 above asks if I’ve got a blog to organize all my comments on the topic. Sadly I don’t, but here is probably my longest and most coherent comment on the topic, if anyone’s interested in reading it (it is similar to one I posted at the Daily Kos version of one of coturnix’s posts).
Shorter RBB: to be really effective science educators we already need to understand our student’s worldviews/conceptual understandings of science. Framing as its discussed in the science article is really no different from that, it is just in the world of politics rather than science education
My best guess (and I can’t read the article till the paper copy arrives) is that scientists frame science issues according to, well, science, and apparently they ought to be framing them like IDists do, according to wishes, moral predispositions, and whatever dishonesty seems to work at a given time.
What else could they mean? It’s not like scientists give speeches on science and emphasize the downside of their work. Now they might be honest enough to mention drawbacks and negative consequences (which is commendable, not execrable), but just like anybody else they try to appeal to the audience to whom they’re speaking. Perhaps they should do better, and even more so, the media ought to do better, however the notion that framing which remains honest to the scientific method doesn’t frequently occur is clearly wrong.
Therefore, the idea must be that we need to find our own versions of hell and heaven for our sticks and carrots, and to utilize whatever lies might work best. Screw science and attempting to raise the public to think like scientists do, we have a propaganda war to win. And if our side wins even without the science, well, we win.
Except that we don’t, because we only took this side because we wanted science itself to improve life. Not only for ourselves, but even to set free the minds of those steeped in superstition.
Glen D
http://tinyurl.com/35s39o
To clarify what PZ said earlier about sacrifice, I believe he was referring to making an offering to Lord Cthulhu?
Theodore wrote:
“#18 (John) got it right. They’re asking us to bring out the Specific Aims (SA) page from our R01s to do the fighting.”
Thanks for the compliment. The other aspect of this that I failed to mention is that grant review is the situation in which scientists act the most like creationists. Most of us can point to cases, even with applications that were well-framed, in which reviewers constructed their own frames and got even the most basic things wrong.
We then have to write a 3-page introduction that politely REframes the whole application, which then becomes more important than the single SA page, and we have to proceed on the assumption that we are likely to get one or more of the same reviewers again. I don’t know about others, but after receiving a review with an unfundable score, I immediately write a perfectly honest, justifiably hostile response, show it to colleages, then delete it, before I write a response that strikes a better tone. Even after that therapy, virtually all of the evolution of this response involves softening the approach.
The irony here is that we are fighting against the human tendency to make snap decisions, brought to us by our good friend, natural selection. Science is a set of rules that helps us avoid these cognitive problems, but the application of those rules is very inconsistent. Grant review is where they are weakest.
IOW, for biologists, the easiest way to achieve a more effective frame than the one PZ uses is to pretend that the lay people are reviewers of our R01 application. Too many of PZ’s posts have the tone of the response (R01 introduction) I write only for personal therapy, and that makes them ineffective for anyone but the choir.
Question for Mooney defenders:
Al Gore recently attempted to “frame” the Global warming reality, by comparing the earth to a baby in a burning crib. A vivid image certainly, but he drew criticism for being too simplistic, and unscientific. I thought it was a brilliant tactic (in some ways). Is this an example of framing, and is this the type of framing that Mooney is proposing?
One of the biggest problems in this whole frame game is that their is a multi generations old body politic( we know who they are) that has had this evolution thing framed in the media (from movies to textbooks to television quips pundits and actual newscasts) as “evolution versus religion” and the public schools if they teach the controvery, still refer to Scopes, etc., rather than the billion generations of “evolution’ since then–not all of which is exclusively controversial to the religious.
So the old debate lingers in the minds of Jan and Joe Q. Public, rather than re-framing( in the mass media sense of framing)the debates and advances into updated models.
In other words: get rid of the notion that the “controversy” exists between religion and science, and re-tool the controversies as debate within science. Controversy sells, but in this case, a little PR is necessary to reframe it.I think the general pereption of science being a no show at the box office is disproven by guys like you, Laden, et. al. who make science an act of every day, rather than a dry boring academic formulaic thing.
Sure you don’t want to look like the proverbial ‘sell out’ but that term has a two edged meaning: one, sure, you like like a science profiteer( and what is wrong with that?) and two, you get peoples heads and minds turned in the right direction, while also being able to present positive face time without the typical right wing sabotage that is already framed in the media. In other words, education campaigns at the level of well crafted press releases TO the media are a good start.
PZ, I agree with the view that framing is not necessarily spinning. There’s ethical framing, too. That happens for example when more than one logical path is available to explain the same problem.
Consider this simple scenario. You’re pouring coffee from a thermos in a cup. Your thermos had less coffee than you thought. You’re (not) able to fill 50% of your cup.
Is your cup half full or half empty?
(I know: old metaphor. But in discourse sometimes you have to choose how will you report facts).
I’m not playing dumb, I really am confused. I’ve got people telling me I already use frames, that I use frames well, that I use them badly, that I’m ignoring frames at my peril, what I’m describing isn’t framing, what I’m describing is framing, that frames are this thing or that thing or this other thing.
I’m getting next to nothing that’s practical. OK, don’t call it “evolutionary theory”, call it “evolutionary biology”. Is that it?
Maybe I do need a course in this.
Here’s a question. The US military hired public relations firms back in ’02 to spin the Iraq War. Does the National Academy of Sciences do its own publicity work? Or does it hire outside firms?
Given that the opponents of science (both on the corporate side and on the religious side) have at their disposal extraordinary resources to push their message (think tanks, rightwing media, rightwing blogs, third-party advocacy techniques, stables of pundits, and all the rest), isnt the main problem the disparity of resources in the framing wars? The lack of progressive infrastructure for framing and publicizing science is part and parcel of the disparity of resources between conservatives and progressives across the board.
In the long run this disparity is going to have to be corrected by the addition of more progressive institutions that counterveil the conservative think tanks. In other words, scientists and scientific orgs need to strategize how to build these institutions effectively. This will eventually have the additional benefit of leaving the science to the scientists and the framing to the framers. Whether or not the NAS picks up on this idea (could be problematic for the government), certainly independent scientific organizations should. Finally, it has to be pointed out that this approach does not buy into the other side’s framing – it only addresses the power disparity on a purely institutional/political level.
For those of us allergic to spinning/doctoring, the main problem sometimes is deciding whether one framing choice is “better” to transmit the message, or -why not- the “winner.” This is illustrated in an paper that you (and your readers here) may have read in the article that first presented it (I read it 25 years ago in med school):
Imagine that the US is preparing for the outbreak of an unusual Asian disease, which is expected to kill 600 people. To combat the disease, two alternative programs are proposed, A or B.
If program A is adopted, 200 people will be saved.
If program B is adopted, there is 1/3 probability that 600 people will be saved, and 2/3 probability that no people will be saved.
Which of the two programs would you favor?
“Frames” and “framing” are fairly old concepts in, say, cultural studies and critical theory. I think the jargon tends to get in the way of understanding what’s being said here.
I am going to restate what I see as the argument of the too-brief Science piece.
1. Society is now faced with a number of policy issues upon which some rather complicated science crucially bears.
1a. Science because it is in possession of applicable technical knowledge on these matters, has a stake in seeing that policy is in consonance with the technical findings.
2. There is no way on god’s green earth that anyone is going to be able to teach the American people enough science to judge any of these matters on their scientific merits on a time scale that will prevent political/policy catastrophe.
3. Presently, people make their judgments on these technical matters using a number of handy “frames” which are essentially heuristics that they employ to make decisions on a range of complex issues in their lives, as well as to make judgments in politics, culture, etc. etc.
4. Science persistently tries to educate through these policy issues in spite of 2 & 3. (Here I disagree, I think science just generally tends to misread what “frames” are being employed and usually goes straight for the “Don’t worry, leave it to the experts” frame, which is rather out of date.
5. Scientists with an interest in policy issues should acquaint themselves with some of the frames folks are using to pass judgment on their work, and start to anticipate which frames their work should be put in and how and to build their messages to facilitate favorable framing.
From a theoretical standpoint: great.
From a practical standpoint, I’ve never seen anyone who has told us what frames folks favor for what that doesn’t sound more stupid than it should.
I think there are some people who have a great sense of how people frame things (your PT Barnum-esque folks–Madonna, Andrew Lloyd Webber, Karl Rove, whoever) but I’ve never seen:
a) anyone spell out how this is done or
b) do it with something where correspondence with reality IS still an important issue (it’s not an issue for art or fashion; and doesn’t register with certain political types, apparently).
How bout if we frame it like this PZ…
“ya can’t teach an old dog new tricks”.
Actually that’s your best bet here, is to duck this one… there are a bunch of folks who are quite vocal (Christians?), who would like to see this who framing thing as putting old PZ back in his place. In my opinion, they are barking up the wrong tree (to keep the dog analogy going), I don’t think the Mooney paper is aimed at PZ. Let it die a peaceful death, or if there is something really useful in it, we will get the picture, even tired angry old PZ. Who knows, the damage has probably already been done, PZ may no longer have the chutzpa to talk about abortion clinic shooters, or the assault on civil rights (same sex marriage ) in South Carolina.
I think I can address this from an advertising standpoint.
I work in pharma marketing/advertising. No I’m not evil.
Say you have a population that are afflicted with a disease with various methods of treatment. One being an oinment that has to be applied twice a day which is smelly and messy, one being a systemic drug (pill) that has to be taken once a day but has some nasty side effects and becomes ineffective after say 8 months, and there’s the drug that you want people to take that can be easily self injected once a week over a long extended period of time and is the most effective of the options.
How do you get people to give themselves a shot?
You have to understand the patient’s needs and desires. How they deal with their illness. You have to answer their questions and prove your drug is worth it.
Once you do, through lots of research it’s much easier to target these people and encourage them to give the drug a try.
The need to believe in something greater, bigger and more powerful?
Much more difficult because people have many many personal reasons for it.
Does science need a warm friendly face? Sure. Does atheism? Sure.
The problem is that no matter how well informed and educated some people may be…
They’re just not going to ever give themselves the shot.
PZ: Yes, you are correct. The only really useful thing anyone has said so far is that one thing I said. (about Evolutionary theo…oh, I mean, Evolutionary Biology).
I really do think that most of this “framing” stuff people are talking about is spinning, and spinning is useful but dangerous and must be better justified than it is right now.
The Goffmanian Frame concept is better than the sociological concept simply because Anthropologists are better at theory and sociologists are better at other stuff. This does not mean that I don’t think most cultural anthropologists shouldn’t be taken out in the woods and strangled with the puppies, but for god sake, save the linguists, they have uses!
Steve:
How do you get people to give themselves a shot?
There are societies in which people are already totally cool with giving themselves shots. This is where your evil pharmaceutical buddies dumb the out of date ampules of stuff.
The point being not to bash the pharmacy… I’d rather stick to the point (ouch)… Many Americans dont’ swallow science, while in other cultures science is broadly accepted.
My proposal all along has been to make some adjustments on the supply side of the information (the scientists) but to also bring the American public in line with much of the rest of the (western?) world through the education process (in all it’s forms, yes yes, even home schooling if you must).
But I’m saying that no matter how well you frame your message and educate your audience some will never accept what you say.
There are people who despite knowing that the injectable drug is better, refuse to do it.
They say… I don’t like putting drugs into my body. Even if they’re taking pills, they see that as different. And some won’t even take pills.
To say we need to frame science for the public… is fine. But rarely are there scientists that can do it. Maybe they do need to be trained. But it’s not even remotely easy to do.
Try listening to a focus group sometime.
Can science be argued in soundbites?
Yes.
Ex: We have the fossils. We win.
Notice that the above also constitutes an example of a soundbite-sized argument in the debate about how to present science.
Now I’M confused… is there some notion here, among actual scientists, that “science a” > “science b” (as long as science b is sociology)? It’s difficult to describe how that makes me feel, as a researcher who is not exclusively focused on physics, chemistry, or biology. If you refuse to take one discipline seriously because of your preconcieved notions about its integrity (ignoring, for example, the data and statistics that demonstrate reliable effects), then you have a COMPLETE and ACCURATE understanding of how creationists view evolution.
Apologies if I’m misunderstanding things here, I suspect we’re having a framing problem of our own.
To say we need to frame science for the public… is fine. But rarely are there scientists that can do it. Maybe they do need to be trained. But it’s not even remotely easy to do.
This smells a little like the original “scientists are crappy communicators” myth, but nonetheless I’m inclined to agree. Which is why we need to bring the public up to speed as well.
I find it interesting that we are having this sudden explosive dialog as though no one has ever thought about this before, but I guarantee that people have. Well, there are scientists, social scientists and philosophers who have, and there are some great citations above.
But it is very important to note that creationists have understood this issue far longer and far better and put it to better use than scientists have. The “theory” issue is a key case in point.
Most teachers in HS and college insist on getting the students to understand that “scientists use the word “theory” differently than the general public does”..
Has anyone had this experience: You cover that, your teaching assistants cover it, the textbook covers it. Then, half way downstream in the semester, a student raises his hand and says “but, this evolution thing is only a theory, right?” And 15% of the other students brighten up and nod along.
Meanwhile read or listen to any creationist rhetoric. The idea that “evolution is only a theory” is number one, number two, number three and number ten on their top ten list.
This smells a little like the original “scientists are crappy communicators” myth
Why is this a myth?
Many Americans dont’ swallow science, while in other cultures science is broadly accepted.
Why is this NOT a myth, at least insofar as you seem to think it reflects a greater understanding of science elsewhere. I don’t think a greater European acceptance of science is evident in, say, the GMO debate.
And is the “greater acceptance” you do see simply respect for social authority rather than an actual understanding?
I think what I’m trying to say is that alot of people already have the notion that science is “hard work” (as Bush might say) and they’ll never understand it, and science is often or unimportant.
Science is hard and people don’t like to be reminded of their ignorance. And I think scientists have a hard time distilling it down into digestable, easy to understand or even pedestrain terms. It goes against alot of what they’ve spent their lives doing… which is gaining more knowledge.
Being extremely smart or knowledgeable (a nerd or geek) isn’t even a positive trait in our society… but if you can sell an idea, no matter how absurd. You’re to be respected.
Scientists are not good bullshitters.
Why doesn’t every grant application and scientific publication require a “lay” abstract? Lay abstracts are a great way to think about framing your work for a general audience.
It’s a great exercise to write a lay abstract and pass it around to your non-scientist friends, then ask them to describe it back to you. You will almost certainly have to make revisions to make it more clear. Would requiring a lay abstract be a useful step to get scientists to spend a little bit more time thinking about the frame of their work?
But a lay abstract requirement would essentially constrain science to those topics that non-experts would understand. There are far too many topics that need to be studied that fall outside of that category, especially since so many studies are intended to complete a small part of a puzzle which itself is part of a larger puzzle, and so on. I don’t think it’s unreasonable for these studies to fall below the “lay person’s” radar.
Maybe this is the problem. Where do we draw the line between research that the public must know about, and research that non-experts ignore? This line seems entirely arbitrary, and an overextension of lay-understandability requirements would provide yet another unnecessary leash for researchers.
Latest to join the swarm is Sean Carroll at Cosmic Variance, who concludes his post in the following way:
Nice suggestion Brian, NIH grants require this already — they are available on CRISP. We could do a better job on them, however. As for manuscripts, why not put the lay abstracts on PubMed Central where the free 6 month NIH funded papers are supposed to be. Append the lay abstract to the paper abstract. Let the authors do it when they submit their papers. I’d be willing to do it for all my manuscripts that are there (I’m in the process of approving them to be hosted on the site for older ones prior to the new policy).
I wonder if your girlfriend would be more interested in an “abstract lay” than a “lay abstract”.
I tried to “frame” the framing debate for the lay person here: http://trinifar.wordpress.com/2007/04/08/195/
Here’s another attempt: if someone doesn’t understand you, do you just talk louder?
Nisbet and Mooney are talking about a real problem at the intersection of science and politics, and I think the issues of global warming and other forms of climate change highlight that problem much better than the creationism/evolution debate. In the latter Jane and Joe Public may just get confused, in the former they may end up dead or, slightly less dramatically, their children may end up living in a horribly transformed world.
Isn’t it obvious we all need to do a better job of influencing journalists and politicians to respond to science?
I write about PZ as “rebel leader” here: http://trinifar.wordpress.com/2007/04/09/the-pzasaurus-pee-za-sore-us/
We need our rebel leaders, but we also need a few Gandhi’s and Sagan’s.
Lay abstracts are a great way to think about framing your work for a general audience.
Yes, except that often that lay info needs framing. IOW, it needs to be put out one way or another. Either or.
How do we accomplish that for the example cited in posting #44? (and we can post other examples, with numbers and proportions to avoid “malicious” framing).
RE 62, Elena
I’ll give it shot: below is an abstract for one of my papers, the lay, or “framed” abstract follows. I’m shooting to make a voter/taxpayer understand the importance of our work.
Role of cation-chloride-cotransporters (CCC) in pain and hyperalgesia.
Price TJ, Cervero F, de Koninck Y.
REAL ABSTRACT: The importance of the GABAergic system in spinal nociceptive processing has long been appreciated but we have only recently begun to understand how this system is modulated by the regulation of anion gradients. In neuronal tissues, cation-chloride cotransporters regulate Cl- homeostasis and the activity and/or expression of these transporters has important implications for the direction and magnitude of anion flow through GABA-A channels. Here we review recent evidence that two cation-chloride cotransporters, NKCC1 and KCC2 are involved in pain and enhanced nociception. On the one hand, NKCC1 activity is upregulated in primary afferents following an inflammatory insult and this produces excessive GABAergic depolarization in primary afferents leading to cross excitation between low and high threshold afferents. On the other hand, KCC2 expression is reduced in dorsal horn neurons following peripheral nerve injury resulting in a loss of GABA-/glycinergic inhibitory tone and, in some cases, inverting its action into net excitation. Pharmacological targeting of these cation chloride cotransporters to restore normal GABA-/glycinergic transmission in the spinal cord represents an entirely novel approach to the development of analgesics.
FRAMED: Pain is the primary reason people seek medical attention and its treatment is hindered by a lack of medicines that can effectively treat pain over a long time course. One of the major problems in patients with chronic pain is touch-evoked pain or allodynia. In this review, we discuss the current state of the art in mechanisms that might underlie allodynia with special attention to new treatment strategies. One promising avenue is a class of proteins (the cation chloride cotransporters) in the spinal cord that affect the way neurons communicate. Under normal circumstances these proteins maintain inhibitor neurotransmission, effectively blocking pain transmission. In pathological pain condiditions, the expression of these genes changes causing inhibitory neuron communication to become excitatory, increasing pain transmission. We discuss evidence that shows that this mechanism leads to touch-evoked pain in pathological pain conditions. Moreover, we show that drugs that can target these proteins might be effective treatments for this persistent type of chronic pain. It is our hope that future research on this project can lead to a new class of analgesics.
I think anyone can understand that while very few would understand the original abstract. I didn’t exagerate, lie or turn it into a sound bite. Moreover, I would happily put it in the paper if journals would provide the space.
Theodore,
I agree. Your lay version is easy to read and understand. The journal version may be too specialized for those not in Neuroscience -if only because there seem to be so many newly described genes and protein (co)transporters that it’s difficult for the rest of the science community to keep up.
Yet I think your example shows one solution to one facet of the “communication” problem attributed to scientists.
Other times, before “lay-translating” we must decide a priori: half full, or half empty?
This is what I was trying to illustrate with the example in post #44 (which is from Science, by the way).
Oran: This smells a little like the original “scientists are crappy communicators” myth
Why is this a myth?
It is a myth because scientists are very often very good at communicating. We need to communicate to colleagues in different disciplines who do not know our lingo, many of us teach hours and hours and hours of intro courses and do well at it, and so on. I feel that this mantra gets said again and again as a form of denial of what is really going on, which is that scientists who deign to involve the public are often shunned by their colleagues. This is not exclusive to science … it is a general thing in academia … and it may in fact be not as bad in science as it is in other fields.
See: http://gregladen.com/wordpress/?p=398
Many Americans dont’ swallow science, while in other cultures science is broadly accepted.
Why is this NOT a myth, at least insofar as you seem to think it reflects a greater understanding of science elsewhere. I don’t think a greater European acceptance of science is evident in, say, the GMO debate.
It is not a myth because it is generally true. This is a phenomenon that has been demonstrated through actual scientific study. Americans are relatively dim when it comes to science, and a large percentage are just plain hostile to it.
And is the “greater acceptance” you do see simply respect for social authority rather than an actual understanding?
Absolutely not. That is not even close to what I’m saying.
Curtis and Brian: About half the proposals I’ve written (in science) required the equivalent of a lay abstract. And no, there is no science that you can’t explain to the general public. The general public can understand anything in science at some level, at some satisfying level. Don’t you think?
Theodore: Why is your lay abstract not sufficient for the science journal? It’s just plain a better abstract. You could throw some of the technical terms into parens and simply have a better abstract.
Seriously.
I second Greg’s last two posts.
Greg’s point about scientists being shunned for involving the public is very important. I also think that most scientists don’t want to have to worry about or have to deal with a public reaction to their work. Regardless of whether this is good or bad, I imagine a simple solution to this problem, leadership.
Carefully chosen leadership figures could speak to the public on behalf of other scientists. Scientific societies could choose the leaders and help frame the work for the public.
Maybe I am alone on this, but I feel that the sciences lack leadership and the lack of leadership is part of the problem we are facing.
Greg, in response to comment #66, I do feel that the public can understand almost all science. I also think that we often make it difficult for them by using jargon and technical terms that they don’t know and then they stop listening.
My five cents here. I have discussed another issue with the current debate on framing on my blog. But now I have a question for all those who are being so hostile to the idea of framing.
Most of you will also say that science communication and framing should be left to science communicators, and let the scientists do the science (and occasionally vent). However, the word is that “alternative careers in science” (because that is what a real science communicator is really going for) are looked down by academics. Not only that. I cannot count anymore the times that people on ScienceBlogs & Friends have been complaining about the “quality of science writing” done by such communicators.
That brings me to a simple question: you do not want to do it yourself, the one the others do is crap, but you want people to understand science – understand it now, not after 50 years of proper science ed. So, how the hell do you want to do it? Seriously, “framing” is a real-life practical suggestion for a short-term solution. Do you have any?
Greg,
To a very large extent I agree, and I’m not trying to be too much of a devil’s advocate here, but I think maybe my problem is in characterizing the actual target audience of a lay-abstract. Are we talking about the American public, or college-educated non-experts? If we’re talking about Americans in general, it’s important to take average literacy in this country into account.
The NAAL (National Assessment of Adult Literacy) from 2003 plops about 25% of Americans (more or less, depending on race) at basic or below-basic document-literacy, and around 50% at basic or below-basic quantitative quantitative literacy. Readers at the basic level demonstrate the ability to read and understand information in short, common sentences, can read simple documents, and can solve simple one-step quantitative problems when the arithmetic operation is specified or easily inferred. In other literacy surveys, this is often listed as something like a 6th grade reading level or lower. With that in mind, would that 25% of the country understand the following sentence:
“In pathological pain conditions, the expression of these genes changes causing inhibitory neuron communication to become excitatory, increasing pain transmission.”
This is not an insignificant portion of the population. Can we simplify all research to the point that the semi-literate quarter of this country will understand it?
Curtis:
Yes, what scientists actually do is not what they can do! On leadership: that may not be the only answer or the best answer, but it is an answer that would work, for sure. And maybe it is the best approach.
Steppen: However, the word is that “alternative careers in science” (because that is what a real science communicator is really going for) are looked down by academics. Not only that. I cannot count anymore the times that people on ScienceBlogs & Friends have been complaining about the “quality of science writing” done by such communicators.
Thanks for making these two points, they are really important.
I totally agree that yes, academics look down on “alternative careers in science” by and large, and that is totally wrong.
I’m the first to stand up and whine about the lousy quality of science writing. But the truth is that a lot of science writing is lousy.
When I was in graduate school, Roger Lewin came and took a class that I was also taking. He was on an MIT scholarship for science writers. He was a pretty good science writer, but taking a couple of classes with Pilbeam and the like turned him into a fantastic science writer. And it only cost about 50,000 dollars to do that. (Is that a lot? A little? I have no idea). Notice, however, that Roger is no longer a science press writer. Too bad…
Science Magazine has somne excellent writers and I’ve worked with some of them. At UMN the press office had a person doing news releases and stuff who was not highly trained in science but who was realy dedicated to doing a good job, and did so.
My point is, there are a lot of good science writers out there, and it is very possible to have more. That does not mean that I’ll stop complaining about the bad reporting when I run into it!
So there is hope… and yes, the short term solution may well be to get more writers trained in science. Also, as was pointed out by the guys who started this whole discussion, news outlets and services are reducing their science staff rather than increasing it (I assume this is true).
I’m always going to be willing to take positive suggestions to improve communication with the public, and I do think that’s important. Everyone should understand, though, that the Science article has precisely two suggestions with respect to the evolution wars:
– Respect religion so you don’t scare people away.
– Talking about data and evidence doesn’t help, so avoid it.
That just means I’m completely out of the picture, and someone else is going to have to do this business. Those are not ideas I can follow.
Brian:
I basically agree with you, and whatever happens, literacy, prior training, etc. needs to be taken into account. There is not “one” audience. My position is to expand the middle … the people who can be reached but are not being reached. At least I think this is a good idea, because that reaches the most people who are not already reached.
By reached, I mean “bring on board” “get with the frame” “joint the club” of people who love science even if they don’t do it. Join us. It’s painless. Really….
Now, lets look at this sentence of yours:
“In pathological pain conditions, the expression of these genes changes causing inhibitory neuron communication to become excitatory, increasing pain transmission.”
is incomprehensible, without a doubt, to 25% of people. So I just called up a biologist I know and read it to her and said “explain this to me”. She said “read it again, go slower.” So I did. So she said “one more time.” So I did. Then she explained it back to me perfectly, and in better English. So a person in the field got this verbally at normal talking speed after three tries (with no context … no frame shifting… had the frames already been shifted, she would have got it first time, of course).
Anyway, the obvious point here is that a lot of work needs to be done on communication.
But you know what? We don’t need trickery! We just need betters skills and some more specialists.
PZ: Right on, bro.
Greg,
I agree 100%. That sentence was cut (out of context) from the lay-abstract a few posts above mine, and just struck me as something that would never be understood by a true layperson.
Expanding the middle MUST happen, and soon.
PZ wrote: Talking about data and evidence doesn’t help, so avoid it.
Or re-frame it, an old article in Science would suggest.
Back to the example on post #44:
To avoid an estimated 600 deaths, two alternative programs are proposed: A or B.
If program A is adopted, 200 people will be saved.
If program B is adopted, there is 1/3 probability that 600 people will be saved, and 2/3 probability that no people will be saved.
Program A and program B are indeed equal. They offer the same. They are only presented differently (“framed”). When people are asked:
Which of the two programs would you favor?, the breakdown of answers is:
– Program A= 72 %
– Program B= 28%
This pattern in answers -the authors explain*- reflect the fact that most people are risk averse. The prospect of saving 200 lives is more attractive than a risky prospect of equal expected value, that is, a one in three chance of saving 600 lives.
The paper then shows that if the same problem is re-formulated with inconsequential changes in numbers and wording, the responses will vary accordingly. There are also examples about framing contingencies, framing outcomes, and other types of framing.
It’s worth emphasizing that framing refers exclusively to the presentation of facts. No misleading, no spinning, no embellishments. But being aware of the framing we’re choosing for our discourse is important because humans (audiences) don’t always stop and reason.
Responses to categorical (yes/no) questions (e.g. evolution vs religion) will vary depending on how the questions (or issues) have been framed. And on this, religious people may have more intuitive training than we do.
But we can train, too.
* Tversky A, Kahneman D. The framing of decisions and the psychology of choice. Science, 1981;211:453-458.
Can we acknowledge that ‘framing’ is just another buzzword and move on with the process of explaining things to people now?
Here is my crude, primitive and confused understanding of frames.
Okay, here’s mine:
Me: I am going to talk to you about X (science, or Bible, or widgets), which is my field of expertise.
Other guy: Okay, but first, if I may: What the hell does it have to do with me? And how will I understand?
Me: Those are fair questions; I am going to try to talk about X in a way that takes them into account.
I am not sure if I properly understand “framing,” but if I am close, then I have to agree that science has shown some failures. Take NASA: all the spinoffs from the space program, especially in the medicine of an aging America, and yet NASA’s sole public selling point seems always to be, “Moon rocks and Tang, pretty cool! Oh, and exploration feels good!”
Caledonian: no, framing is not just another buzzword. The phenomenon has been studied for years. See footnote in post #79. About the biological bases of framing, this was published in Science last year:
Frames, Biases, and Rational Decision-Making in the Human Brain
Human choices are remarkably susceptible to the manner in which options are presented. This so-called “framing effect” represents a striking violation of standard economic accounts of human rationality, although its underlying neurobiology is not understood. We found that the framing effect was specifically associated with amygdala activity, suggesting a key role for an emotional system in mediating decision biases. Moreover, across individuals, orbital and medial prefrontal cortex activity predicted a reduced susceptibility to the framing effect. This finding highlights the importance of incorporating emotional processes within models of human choice and suggests how the brain may modulate the effect of these biasing influences to approximate rationality.
http://www.sciencemag.org/cgi/content/abstract/313/5787/684
Yes, the concept has been studied for years. ‘Framing’ is just a buzzword.
I just realized why the Nisbet-Mooney article irritated me so much.
My very first post, September 23, 2006. This same argument from Mooney about the Ivory Towers right in ‘RWOS’.
Greg:
Good to hear from you again. (congrats on the wedding!).
I would have to say – based upon my personal experience, that scientists who reach out, and make an effort to communicate to the general public, are NOT looked down upon by their peers.
At least, not in the States.
In the past two weeks I’ve given my Physics of Superheroes talk as a physics colloquium at MIT and Illinois – Urbana. In both universities, the audience (ranging from high school students all the way down to… full professors. ahem)got what I was trying to do and appreciated it. In particular, the faculty responded very positively to my efforts to use unconventional examples to illustrate physics principles.
Where did I receive feedback that using superheroes were an inappropriate and demeaning way to teach physics? At the Hahn-Meittner institute in Berlin. (Interestingly enough, the German translation of my book is a big seller for my German publisher – I’m like David Hasselhof! But then, the book is not intended for physicists – who know all of the physics I’m describing).
There are advantages to reaching out this way – the screenwriter of an upcoming superhero film (I’m not supposed to say which one) has contacted me about discussing the physics aspects of the main character’s powers. As a fan – I hope he tells a good story – but if I can help add some real physics, so much the better. He is no doubt interested in talking with me because it helps him with his fiction – if he can keep the action physically correct, within the “miracle exception” constraint, in can help maintain the viewers suspension of disbelief and keep them tied to the story.
No they are not. Distributions with identical expectations are not identical (Here’s a hint: what’s the difference between a stock and a bond?). The element of risk is present in B and not in A, so the survey does not indicate what you claim, but instead indicates people prefer the safer, smaller payoff to the larger, risky option. “Framing” has nothing to do with it.
Well, I don’t think anyone is proposing you as the ideal candidate for the post of Missionary to the Godridden. (Dawkins, on the other hand, does hold some position related to promoting the public understanding of science, IIRC. Fortunately for him he’s in England, where religion is both less widespread and less, um, dedicated in its opposition to facts.)
But I also think your central point is right on: scientists can’t afford to avoid the messy details of science, or it will really be just another competing viewpoint. There’s quite enough people who think that already, without encouraging it by hiding the advantage science has over other “ways of knowing”.
I know I’ve said this already on several related threads (and the point isn’t original to me by any means), but we don’t need more people to believe science, we need more people to understand science. It does no good to say that this is too high a standard – it is the only standard worth having. Without the scientific way of thinking – examining the evidence and scrutinizing both new and old claims – all you have is a new breed of dogma.
For those people who really think that most people are too stupid and lazy to look at evidence before making up their minds, well, at least the Enlightenment is going to go down fighting. (But if that were so, how would it have gotten started in the first place? Such a contemptuous dismissal of the intellect of the human species doesn’t seem to hold water, given our species’ achievements.)
And you’re consulting on a movie?
Why am I having a flashback to the Spongebob Squarepants movie?*
*Which movie, by the way, was actually an allegory for the evolution/creation conflict. I shoulda done a review.
No – Spongebob was a warning about the dangers of sub-prime mortgages – advising people to sell their spilt-levels and start living in a pineapple under the sea!
What I meant was that I’m like David Haselhof in that I’m popular in Germany. And I have a car that talks to me.
And a toaster that talks to me. And a fork.
In fact, I hear voices all the time. But they don’t tell me nothin’ I don’t already know!
#79, #86: If I recall correctly, the interesting result was that when the descriptions of the same two programs were reworded (in such a way that Program A2 really *was* the same as Program A and Program B2 the same as Program B), the majority preference reversed, showing the effect of the wording of the question on the answer.
Still, although the programs are not the same, neither is clearly better than the other. It would be interesting to see how many people could be induced by the “frame preferences” to pick a mathematically worse program just because it sounded better. If Program B became a 50/50 shot, how many would switch to it, and how many would continue to cling to the guaranteed 200 lives saved?
Elena, you have a good point in general, but as Rooney correctly points out, your choice of example does not match it, because the second scenario has an element of risk that the first one does not. I have not kept up with Jason Doctor’s and John Miyamoto’s work on the subject, but they have done some work on whether the cross-over point between scenarios (roughly, how much more potential benefit is necessary to assume the added risk) is rational or not. That is quite different from the point you want to make about framing identical information in varying ways.
Gerd Gigerenzer’s work is, I believe, much more on target for what you are asserting. He looked at how physicians and patients got confused in trying to untangle a series of probabilities, and proposed a more natural way of presenting the information that removed the need to calculate those probabilities.
I think the following is an example of what you are referring to:
Here is an example of how he frames identical information in more and less user-friendly ways:
Jim:
Good to hear from you! Congratulations on the book, by the way.
I’m glad your experiences have not been negative. And I’m not too surprised that physics is an area where there is limited, or at least varying degrees of snobbery. There is the Physics Force, there was Bill Nye, and now there’s Jim Kakalios. (I had suggested that this was a general academic thing and maybe not as bad in the sciences as in other areas. In anthropology, the archaeologists are looked down on for many reasons (some probably deserved) but I think that the tendency for archies to engage with the public is part of it.
In any event, tenure codes rarely include communicating to the public as important. The efforts of NIH and NSF over the last several years to support “broader impacts” which includes education at many levels and public engagement wasn’t because they were sitting around and some guy said “Hey, scientists do this stuff really well, lets start including this as a criterion in the grant process.” I suspect they were fixing something.
In any event, I’m mainly trying to get away from the explanation that “scientists can’t explain things because they are not good at it” because I just don’t believe that. Science IS explaining difficult and mysterious things. We are the ones that are explaining mysteries. We and the rabbis. We have different approaches, of course.
Hey, I’m popular in Japan … maybe we should go on the road.
(You have a fork that talks? Only one?)
Greg:
I totally agree that “scientists are lousy explainers” is a lousy explanation for troubles with public perception. I, and many of our colleagues, believe that you don’t really understand something unless you can explain it at the Freshman level. This is probably a legacy of Feynman, who championed this opinion.
I’m a little leary of road trips. The last one ended with a freeze frame as we drove into the Grand Canyon.
And my fork doesn’t only talk. It also fights crime.
OK – time for Little Nemo to head for bed.
It’s the only standard that will work. Most people already “believe” in science, they just don’t understand how it works. That’s why the ID/creationist and anti-climate change guys always want “scientists” on their teams. And they are pretty good at framing.
They are also good at attacking the over simplified models that you have to use when you reduce a complex model into a bite sized chunk you can communicate in a short newspaper article.
The best thing to ever happen for the climate change crowd was Gore’s movie. That appears to have worked. It changed minds. So, maybe evolution could use a good movie? Remember, “March of the Penguins” made money, so did Gore’s film. I think Dawkins has enough clout to get Hollywood interested in making a buck.
I’m already quoted, so I haven’t much new. Well, unless I repackage and add to some of my comment from Greg’s last post:
I think it is clear that Nisbet & Mooney has failed in efficiently communicating what they want to do. They also seem to have failed to put their message (in the Goffman sense) in the correct frame for scientists.
But N&M larger and fuzzier framework (which seems to be about social construction of meaning) is correct in the sense that this is what we want to do.
Normally people switch between frames all the time, but the problem is to break or modify the bounded rationality of an inappropriate frame.
However, I don’t think it is possible to step into any individuals fundamentalist frame to change it. These frames and most of their individuals are heavily guarded against change and/or switch. (And perhaps even individually indicative of pathology.)
The conversion descriptions from fundamentalist positions one reads is most often triggered by them being presented by material that shows that they have been lied to or at least not gotten all of the available alternatives presented for them.
On the other hand, one could possibly publicly try to affect change in the common frame as it is presented.
Well, in the last century were was a frame called “discussion” or “discourse”. It seems people met to freely exchange and evaluate ideas. I believe the roots goes back to the Greeks and the Socratic method.
But it is all extinct now. In the media world of today we respect all ideas and frames, we just don’t value them.
I’m already quoted, so I haven’t much new. Well, unless I repackage and add to some of my comment from Greg’s last post:
I think it is clear that Nisbet & Mooney has failed in efficiently communicating what they want to do. They also seem to have failed to put their message (in the Goffman sense) in the correct frame for scientists.
But N&M larger and fuzzier framework (which seems to be about social construction of meaning) is correct in the sense that this is what we want to do.
Normally people switch between frames all the time, but the problem is to break or modify the bounded rationality of an inappropriate frame.
However, I don’t think it is possible to step into any individuals fundamentalist frame to change it. These frames and most of their individuals are heavily guarded against change and/or switch. (And perhaps even individually indicative of pathology.)
The conversion descriptions from fundamentalist positions one reads is most often triggered by them being presented by material that shows that they have been lied to or at least not gotten all of the available alternatives presented for them.
On the other hand, one could possibly publicly try to affect change in the common frame as it is presented.
Well, in the last century were was a frame called “discussion” or “discourse”. It seems people met to freely exchange and evaluate ideas. I believe the roots goes back to the Greeks and the Socratic method.
But it is all extinct now. In the media world of today we respect all ideas and frames, we just don’t value them.
Ronnie writes (#86): The element of risk is present in B and not in A, so the survey does not indicate what you claim, but instead indicates people prefer the safer, smaller payoff to the larger, risky option.
The element of risk is indeed present in both problems with equal weight. (Unless someone assumes that the statement “will save 200 people” does not refer to probabilities but to Jesus’ promise?). Problems A and B do describe the same epidemiological contingencies.
“Framing” has nothing to do with it.
It’s a demonstration of framing. May I suggest you read the entire Science article by Tversky and Kahneman.
RavenT (# 91) writes: your choice of example does not match it, because the second scenario has an element of risk that the first one does not.
As I pointed out to Ronnie, these are not my scenarios but those presented by Tversky and Kahneman in their explanation of framing.
I think the following is an example of what you are referring to… That is quite different from the point you want to make about framing identical information in varying ways.
Right. It is a different subject. What you’re illustrating is Bayesian probability. It relies heavily on the prevalence (or pre-test probability) of a disease. (The Bayes’ theorem provide the bases to understand for example why mammograms and biopsies are overdone in the US).
It’s not framing, as you correctly point out, although perhaps it could be considered a specific form of it.
Jim:
I’m a little leary of road trips. The last one ended with a freeze frame as we drove into the Grand Canyon.
I would have figured that when it comes to big canyons, the recommended procedure is UPto, not INto…
I totally agree … Freshman level. In computerland, there is the “telephone test” and I’ve adopted that in training for graduate TA’s … The computer version is take the procedure you are working on and call your mother, who is NOT a computer programmer, and explain it to her on the phone so she basically understands, if perhaps is also bored by, it.
EG. you have to be able to call your mother and explain, say, the concept of “fitness” … have her get it, and not screw up the explanation so she has accidently learned something wrong (like, for instance, that fitness = reproductive success).
Your poor mother! Your TA’s poor mothers!
I tell my students to only imagine explaining it to their mothers. Imaginary mothers don’t have any lives other than tending to their children, so they’re a little more patient.
Elena (Lennie?):
No, you need identical scenarios for framing. Clearly, the two scenarios are different. The variance is zero in the scenario A and 80 000 in scenario B. I’ve read the beginning of the paper and the wikipedia article summarizing it. I don’t know what the 2129 articles that cite it have to say, and I honestly don’t care. The conclusion is simply false. The scenarios differ by more than presentation.
By what definition of risk? In scenario A, you are assured of saving 200 people with probability one. In scenario B, you risk losing even these people (with a large probability). In A, actual outcome=expected outcome. In B, actual outcomes != expected outcome.
You used them and endorsed the conclusions. You said they “were the same”, when in fact they were not.
Well, the first part of this sentence contradicts your response to me, and the latter part contradicts the first. You are not framing “framing” very well.
You miss the point, which is that the doctors misunderstood Bayesian probability, and that proper framing was needed to help them. RavenT is not trying to illustrate Bayesian probability, she is trying to illustrate that better illustration of Bayesian probability is necessary,
I tell my students to only imagine explaining it to their mothers.
… that’s a good idea. I would get a lot fewer irate phone calls from people’s mothers.
Plus, there is a high probability that a mother is an evolutionary biologist. These things sometimes run in families…
That there what I wrote would be what’s called a question. To wit: is European “acceptance” of science a function of greater understanding or is it a function of greater deference to authority?
And if Americans ARE laggard in science knowledge (I’ve never seen a study of general populations that measures scientific knowledge as opposed to attitudes), doesn’t that say something about the state of science communication in this country?
And being better at PR than post-modern literati or philosophers is setting a pretty low bar for science communication, I’d say.
In spite of the fact that many scientists teach, a great many of them seem to have no idea how–“you either get it or you don’t” seems to be the pedagogy often employed–which is bitterly complained about by their students.
I really think science has no room for complacency here.
Another problem is the tendency of scientists who ARE good communicators to use their expertise as a lever to make all kinds of claims that have nothing to do with it or which are, at best, real stretchers (Pinker and Dawkins would be my two big examples of this tendency.)
The idea that scientists claim to know much they do not actually know has a lot more traction than it should in the public realm. Everyday oversell is probably one reason why.
Ronnie writes (#100): Clearly, the two scenarios are different.
No, that’s not a logical inference.
By what definition of risk?
By that which is well explained in a number of books. I can suggest a couple: Biostatistics. The Bare Essentials, by G. Norman and D. Streiner; Mosby-Year Book, and: Designing Clinical Research, by S. Hulley and S. Cummings, Williams and Wilkins.
I’ve read the beginning of the paper
How about reading the entire article?
I honestly don’t care.
It’s a free world. My posting was intended to offer a “media-unpolluted” background on the concept of framing, which is certainly not new.
Through post #82 I also wanted to share here the finding, relatively new (for many readers, at least) that framing appears to have neural bases. See:
http://www.sciencemag.org/cgi/content/abstract/313/5787/684.
This is what I find most fascinating -and worrisome.
Counter-framing
When I was a boy growing up in the 40s and 50s it was obvious to my smart-ass friends and me that the creation story in Genesis was a fairy tale. Growing up civilized us, and taught us a little tact and diplomacy, but I doubt that any of us changed our minds. Many of us became scientists of one sort or another.
Now in my declining years I’m free to embrace my Old Curmudgeon status and suggest that a discussion that begins with “evolution is just a theory” should be met with “‘God’ is just a myth.”
Not God, but ‘God’: the typographical framing does the real work. It’s healthy for the naively religious to be forced to confront the concept of ‘God’.
Elena:
Tone-deaf are we? You’ve egregiously mis-spelled my name three times now, so fuck you.
I am using the financial definition of risk (see wikipedia), which is the probability of the actual outcome being different from the expected outcome, which is clearly zero in A and nonzero in B. This word seems to have a completely different definition in other areas, so I apologize for the confusion. Semantics aside, my point stands. There is an obvious difference in statistical spread in the two options, and you really need to brush up on probability if you think the two are identical.
Why the hell would I read an article that is obviously flawed from the get-go?
Also, your link is broken.
Just take the period off the url.
Rooney, not Ronnie. Normally, I would not bother with correcting small reading errors like that, but because our discussion will hinge on precise reading of discipline-specific definitions and probability, I believe that it is worth pointing out the need to read carefully.
You are conflating several things, among them “risk” and “utility”; riskless gambits and risky gambits, and “framing” and “framing effects”. In fairness, Tversky and Kahneman conflate those first two pairs in their examples as well, but they were pioneers in studying decision-making. A great deal of refinement in basic concepts has taken place since 1981. Part of the later research (new as of 2002-2003, which is the last time I seriously followed the literature) is whether the switch from riskless gambits to risky gambits is on a rational basis or not. It refines the basis of some of scenarios in the Tversky article, and would not claim that certainty versus risk are epistemologically identical.
The conflation between risk and utility/riskless and risky gambits is, as you say, straight out of the Tversky article. But you yourself are conflating “framing effects” (which you cited a neurobiological article for the basis of) with the “framing” that this post discusses.
“Framing effects” are the change in decision taken by the receiver of the information in response to different presentations of information. By contrast, framing is an action by the transmitter of the information about how to effectively transmit exactly the same information to the receiver (Gigerenzer’s example of choosing conditional probabilities versus natural frequencies). Even Tversky and Kahneman stated:
(meaning their goal is not to shift the basic infomation conveyed) and then went on to state:
In other words, Tversky and Kahneman are right on principle, but failed to live up to that principle in their example. From a utility point of view, 200 people living/400 people dying with certainty may be the same as (maybe no one lives/dies OR maybe 600 people live/die); from a risk point of view, they most certainly are not the same, as Rooney has already pointed out. That’s why risky and riskless gambits are called by those names–they are categorically different decision strategies.
I am using Bayesian probability to illustrate framing. The information in both the conditional probabilities framing and the natural frequencies framing are identical; the only difference is the choice of presentation. That is an example of framing, rather than Tversky’s framing effects.
You misunderstand my point. I did not point out that it is not framing; I pointed out that it *is* framing (rather than framing effects), and that Gigerenzer explicitly refers to it as “framing”. The framing effects, then, are that the doctors and patients are all over the map with their estimates based on conditional probabilities, and they are much more reliably correct with their estimates based on natural frequencies.
BTW:
There is also the element of culpability here, and how people process it, even in the face of big outcome differences.
For instance when faced with a game like this:
I have a gun. If you don’t strangle a person I randomly select, I am going to shoot twenty random people. You have absolutely no doubt whatsoever that I will fulfill my threat. You have absolutely no means of turning the tables on me. What do you do?
Most people refuse, in spite of the bad outcome for reasons of (they perceive) avoiding personal culpability.
I think this may be part of what’s driving your Asian disease cure scenario–feelings of culpability in the event(2/3 possibility)option B doesn’t work out.
This may not go to people’s perception of outcome at all.
Rooney writes (# 105): You’ve egregiously mis-spelled my name three times now.
Then I egregiously apologize. Sorry that you have lowered yourself to calling me “Lennie” (which sounded as innuendo: you thought I was a man personifying a woman). Sorry that you’ve smeared PZ’s blog with swearing. But hey, now we all understand better where you’re coming from.
I am using the financial definition of risk (see wikipedia),
Rooney: 200 lives saved in the future divided by 600 people is still one third probability-no matter how you slice it. Unless, again, we’re attributing the life-saving capability to a deity and then is no longer a probability but a certain prophecy.
About framing: it seems as though you are falling precisely into the box that the article describes, i.e., framing. But don’t take my word for it. Why not send Science a letter to the editor?
RavenT writes (# 107): The conflation between risk and utility/riskless and risky gambits is, as you say, straight out of the Tversky article. .
There’s no conflation. The life/death risks and the framing of acts, outcomes, and contingencies are described in the article under separate subheadings. The example I transcribed above is the first of two examples under “The evaluation of prospects.”
Wow. My jaw is currently on the floor. I sincerely hope you’re not employed in a profession that requires statistical reasoning.
I thought no such thing. I tried to mangle yours as you mangled mine. I swore because you didn’t catch on.
It seems to me more that you are proving Gigerenzer’s point about the human mind and its difficulty with statistical reasoning. You have a fundamental misunderstanding of how probability works. In itself, that’s certainly no crime; far worse is your demonstrated refusal to let go of your idée fixe, no matter how much evidence to the contrary is presented to you.
Elena, we all can read; that’s not the issue here. The problem is that you cut and pasted verbatim something that you clearly don’t understand to support your incorrect argument. Then, much as the creationists do, when people who actually know what they’re talking about try to point that out to you, you dismiss their input.
I tried to gently correct your misunderstanding by explaining in more detail, but since you won’t listen either to Rooney’s direct feedback nor my more patient one, then we’re all just wasting our time here.
Rooneym
[quote]Wow. My jaw is currently on the floor.[/quote]
Again, may I suggest you consult a textbook of Statistics. See post #103.
I’d like to see a well-thought rebuttal of the Science article. Do send a letter to the editor (though they may disallow swearing) -then share it with us. I’m genuinely curious.
RE: “OK, don’t call it “evolutionary theory”, call it “evolutionary biology”. Is that it?”
Sure.Drop the theory, and practice it.
And then, get some “hot blonde” quasi intellect, ala Ann Coulter, to write about it, make appearances, add rude dismissive statements about Mary Magdalene( she was just a stupid whore), and God ( he was a prude) etc.
I mean: biology is a ‘physical’ science, not like those icky temporal cultural anthropologists;-)
Rooney,
My jaw is currently on the floor.
Again, may I suggest you consult a textbook of Statistics. See post #103.
You admit to not having read the article. You question (with an inadequate argument) one bit transcribed above. Why not read the article instead and write a reasoned rebuttal to Science? Do send a letter to the editor (though they may disallow swearing), then share it with us.
Elena, this is getting really sad.
I gave you a very gentle and face-saving out from your problem–all you had to do is say, “yes, the Gigerenzer example is a much clearer and unproblematic example of what I meant to say by framing”, and your point would have stood very strongly.
Instead, you dismiss it and insist on demonstrating your basic misunderstanding of statistics, insist that other people do your homework for you, and insist that we write critiques which already appear above.
You are not going to find any statistics book, biostatistics, epidemiology, or otherwise, that will define a probability of 1 as identical to a probability of 2/3 or a probability of 1/3. No one is going to write a letter to the editor of Science critiquing a 26-year-old article, when there has been a ton of research in the interim refining the basic concepts from the original article, and no one is going to do your homework for you. Copying and pasting someone else’s work in an almost-30-year-old article does not take the place of defending your own arguments.
That’s simply not how it works. People are busy, and if you don’t show any inclination to get the basics right, and to learn from feedback and constructive criticism, then they won’t spend their time on a lost cause; they’ll invest it where there is a chance of something productive happening.
Sorry for double-posting.
RavenT:Elena, we all can read; that’s not the issue here.
Is not? Rooney has asked above why he should read the article before arguing that its contents are wrong. And his source for statistical background is apparently Wikipedia.
Maybe you have read the entire article. You question its examples on the premise that risks and probabilities are misused. I truly see Tversky’s point, and I’d also invite you to write to Science with your points of contention and share your letter here.
Regardless of Tversky’s specific examples, though, my aim in bringing them up here was to underscore that given X facts, presenting them as f, a, c, t, or s will make a difference in how X are understood.
About framing effects: what you say is exactly what I understand. The paper on the neural bases of framing shows how framing comes into existence, by showing its effects on the “framee” and his/her decision making.
RavenT
You are not going to find any statistics book, biostatistics, epidemiology, or otherwise, that will define a probability of 1 as identical to a probability of 2/3 or a probability of 1/3.
I agree. That’s not what the example proposes. “Saving 200 lives” (out of 600) by a medical program “N” is always a probability proposition.
Elena, I am a graduate student in mathematics. I’m pretty sure I don’t need to consult a statistics book. I don’t know much about framing or psychology, but I know plenty about probability. I was lurking and I saw an elementary mathematical error, so I spoke up. If your entire field has not caught this mistake, maybe I should write a letter, but I have no idea what else has been written on the subject, and I feel no inclination to sift through all of those citations. That takes time I don’t have.
RavenT, on the other hand, seems to be quite knowledgeable with the literature, and appears to agree with me. If there’s a letter to be written, she seems to be the person most qualified. It would be pretty arrogant of me to swoop into this area and start writing letters regarding 26-year old papers just because of the error of a blog commenter.
Elena, I just took a look at the original article, and I believe you’ve misunderstood what Tversky and Kahneman are saying. They are not claiming that the two scenarios are identical, nor that they are equally risky. In fact, they explicitly attribute the choice of one scenario over the other to the fact that people are risk-takers under some conditions and risk-averse under others. (And, as another math grad, I have to agree that these scenarios aren’t remotely identical from a statistical PoV.)
What T&H are claiming–and Chris explained this well up-thread–is that two differently-worded problems are identical, because they both equate to choosing between these two scenarios. But because one problem describes the scenarios in terms of deaths, and the other one in terms of survivors, the majority of people choose a different scenario in each problem. That’s the framing effect. It’s not about whether they consider the two scenarios as equally preferable or not; it’s about whether they change their preference depending on the scenarios’ description.
118:
Yes, it is *exactly* what the example proposes. Tversky and Kahneman are using an abstraction about utility which they consider makes the scenarios equal. But as I mentioned, a lot of refinement of these principles has been done since the original article, and I don’t agree that their abstraction suffices to make the situation the same, certainly not about risk, which serves to confound their point as a result. But that is not the important point about framing. Their controversial abstraction derails their point about framing, when other examples could reinforce it.
Let’s try an experiment in framing, à la Gigerenzer. Perhaps we can come to agreement in a way which has mathematical rigor *and* makes the point about framing at the same time.
I think we agree on the problem as posed in 79:
The controversy about this example illustrates Gigerenzer’s point about the confusion around information presented in statistical form. Let’s test his hypothesis about natural frequencies by seeing if we can come together better on the same information presented in that format.
Do we agree that this is a fair representation in natural frequencies of the problem?
Now, if these programs offer the same, shouldn’t the outcomes be the same–that stands to reason, doesn’t it?
At no time in these three trials is the outcome the same for each program. So comparing programs A and B is comparing apples and oranges, on the basis of probabilities.
Like Tversky and Kahneman, a policy maker may argue on the basis of utility that since Dr. Raven and Dr. Elena both save 600 people, it does not matter which program they fund. But that is a rarefied point, based on a specific abstraction, and it relies on assumptions about making comparisons across groups, which we have not spelled out and dealt with in our example.
An interesting question is this: if you are a patient, and you are trying to decide with which doctor your chances are better/your risk is lower, do you have a reason to prefer one doctor over the other? If so, why? If you can see why patients should have a clear preference for Dr. Elena over Dr. Raven, then you see the point that Rooney was making, and why the scenarios are not equal in a way that confounds T&K’s point.
To quote T&H:
“The majority choice in this problem [1] is risk averse: the prospect of certainly saving 200 lives is more attractive than a risky prospect of equal expected value, that is, a one-in-three chance of saving 600 lives.”
Later,
“The majority choice in problem 2 is risk taking: the certain death of 400 people is less acceptable than the two-in-three chance that 600 will die.”
They seem quite aware that the scenarios are not equal, except in terms of expected value.
Very good point, Anton, and I thank you and Chris for the correction to my own misreading of the article. Chris, I don’t know how I managed to miss your comment making the same point earlier, but somehow I did. You’re right, and my explanation of why T&K were considering them the same was mistaken.
I actually do know that H and K are not the same letter. It’s just that I’m in math, so I’m not required to apply that knowledge or anything.
It’s too bad his name’s not “Nahneman”; then you could just say you were writing it in Cyrillic :).
I read a murder mystery once which hinged on a Russian woman, Vera Rossakoff, using her handkerchief to frame some guy with the initials “BP” that way.
–Raven, who never manages to spell “Kahneman” twice the same way without looking it up…
Brian Foley: About the “Great Thermostat” – don’t be so sure. There are some who are hostile to the GW issues because they think “god will provide” or that the rapture is happening soon, etc.
Blake Stacey: A lot of crap does get published in sociology, but there is good stuff, too. (Often the better papers I’ve seen don’t seem too useful since they are idea-poor and the idea-rich papers are outlandish, alas.) All these weaknesses call for better scientific study of society, not abandoning such to “cultural studies” in English departments and such.
Chris Hallquist: Doesn’t work. Someone will immediately retort – “No, you don’t! Fossils are faked by Satan!” (or something like that).
Greg Laden: “At some level” is a weasel-phrase. What level is appropriate?
Agatha Christie, right? I love Rossakoff’s romantic tension with Poirot. It’s just so…sordid.
Raven,
To address one of your previous comments, the reason I didn’t reply on Gigerenzer’s paper is because I haven’t read it (and thus can’t pretend that I have).
I understand your example about patients. It’s valid. However this comment So comparing programs A and B is comparing apples and oranges, on the basis of probabilities is not.
As you know, when you toss a coin you have a 50/50 chance to get tails vs heads. But you can toss a coin 10 times and get 10 tails. Does this disprove the 50/50 prediction? (And yes, randomness is a different topic; my point here is only about prediction/ probability).
As you and Anton point out, the article states that A and B have “equal value.” This claim is numerical but devoid of ethical considerations -which, along with the the responders’ personal predispositions to risk, are instead the hinges around which the responders’ answers pivot.
Ah, you understand :).
Thanks to my uncle, who encouraged indiscriminate reading by dumping his old library (age-appropriateness no object!) on me, I devoured those books around 4th grade or so. That instilled a life-long passion that I still indulge once every few years or so.
And Christie still owes me for the 6 weeks I failed spelling for insisting on writing “colour”, “flavour”, “honour”, etc., on all my homework.
Keith Douglas:
I can certainly accept that. Ever higher — never surrender — damn the resonance cannons, full speed ahead! Etc.
Which doesn’t stop me being irritated at a discussion which seems almost consciously to avoid the practical.
if somebody already mentioned it, i apologize, but i’m opposed to [hand-waving]. what we need is a more science-literate public, not further dumbing-down of science.
a good communicator chooses words appropriate to the intended audience, but a great communicator can introduce unfamiliar terms [and concepts] and enlighten [and educate] the audience.
i dunno. i’m skeptical of this whole idea of “framing.”
Hipparchia, while I agree with your comments about what a gifted communicator is capable of, on the subject of framing I believe you’re confusing content with context. I don’t think it’s about dumbing-down. I think it’s about establishing appropriate context and tone.
OK, PZ, everyone, something came in the mail yesterday that should address all concerns about this issue:
50% off all your framing needs. At Michaels. Through April 22, 2007.
No, it states that they have equal expected value. ‘Expected value” as a mathematical term has no more to do with “value” in the ordinary sense than any other numerical value you could extract from a probability distribution–the variance, for instance, or the supremum or infimum of the distributions’ support. One certainly could base one’s preference for a scenario on the expected value of some quantity it determines, but that’s no less arbitrary than any other criterion.
If that were true, their answers wouldn’t change between Problems 1 and 2, since their ethical codes and predispositions to risk presumably didn’t change radically over the course of the survey. But they did change, because their preference is also strongly dependent on the way the scenarios are described. That’s the framing effect.
Elena did make the point that the she interpreted the statement as a probabilistic one, that is to say “will save 200 lives out of six hundred” only means that it will on average save 200 out of 600 lives. This, it seems, is the incorrect way to interpret it from what people seem to be saying and quoting (don’t have time to read the article).
(i’m also a maths postgrad heh)
Does anyone else find it amusing that a group of what are ostensibly scientists are going on gut reactions to tease out meaning, rather than reading about the topic itself? This is not directed at PZ, obviously.
I can’t claim to know Mooney and Nesbit’s full intent, but let me suggest again that a practical place to start is the UCS book “A Scientist’s Guide to Talking with the Media.”
It is all too easy to bog down in the odd nomenclatures and conceptual frameworks of unfamiliar disciplines. Which, of course, is the point – if we as scientists want to communicate with the broader world, we can’t expect the world to travel that distance alone (or indeed, care to), we must learn to meet the world halfway. Science is too important to leave to scientists alone.