follow CCP

Recent blog entries
popular papers

What Is the "Science of Science Communication"?

Climate-Science Communication and the Measurement Problem

Ideology, Motivated Cognition, and Cognitive Reflection: An Experimental Study

'Ideology' or 'Situation Sense'? An Experimental Investigation of Motivated Reasoning and Professional Judgment

A Risky Science Communication Environment for Vaccines

Motivated Numeracy and Enlightened Self-Government

Ideology, Motivated Cognition, and Cognitive Reflection: An Experimental Study

Making Climate Science Communication Evidence-based—All the Way Down 

Neutral Principles, Motivated Cognition, and Some Problems for Constitutional Law 

Cultural Cognition of Scientific Consensus
 

The Tragedy of the Risk-Perception Commons: Science Literacy and Climate Change

"They Saw a Protest": Cognitive Illiberalism and the Speech-Conduct Distinction 

Geoengineering and the Science Communication Environment: a Cross-Cultural Experiment

Fixing the Communications Failure

Why We Are Poles Apart on Climate Change

The Cognitively Illiberal State 

Who Fears the HPV Vaccine, Who Doesn't, and Why? An Experimental Study

Cultural Cognition of the Risks and Benefits of Nanotechnology

Whose Eyes Are You Going to Believe? An Empirical Examination of Scott v. Harris

Cultural Cognition and Public Policy

Culture, Cognition, and Consent: Who Perceives What, and Why, in "Acquaintance Rape" Cases

Culture and Identity-Protective Cognition: Explaining the White Male Effect

Fear of Democracy: A Cultural Evaluation of Sunstein on Risk

Cultural Cognition as a Conception of the Cultural Theory of Risk

Saturday
Nov262011

Market Consensus vs. Scientific Consensus

Some smart researcher should invent a market measure of belief in climate change

An index could be constructed that reflects things like investements in climate change adaptation, investments in new business opportunities created by change in climate, and offering of & changes in price for insurance against adverse impacts.

As the price of the index rises (or falls!) it would be evidence of market consensus on climate change. Market pricing is relevant to trying to figure out lots of things, obviously (indeed, there is a cottage industry  in academia now to create prediction markets to compete with other types of predictive models). But the value of a market consensus measure here is cultural, too: for some citizens, market consensus will have a positive cultural resonance that scientific consensus (at least in this context) lacks. They could be expected, then, to give information from the market more engaged & open-minded attention.

People culturally disinclined to pay attention to markets as information might pay more attention to this one, too, and thus learn about the value of being more open-minded about information sources.

Last but not least, having a market measure of belief in climate change would be great for people trying to investigate dynamics of science communication -- for all the reasons I just gave.

Wednesday
Nov232011

Politics, Cognition & Pepper Spray

Does “pepper spray” really hurt? The answer probably depends on the relationship between the ideology of the person who was sprayed and the ideology of the person asking/answering the question.

There is an internet buzz emerging over the suggestion by Fox news commentators & equivalent that “pepper” spray (it’s orders of magnitude more irritating than habanero) isn't all that painful. The debate is politically polarized along predictable lines.

If the demonstrators who were sprayed had been protesting abortion rights outside an abortion clinic, would there be an ideological inversion of the perceptions of how much the spray stings?

The answer is that we are unlikely even to get to that point in the discussion before we are already tied in knots over other facts relating to the behavior of the protesters and the police.

 

My colleagues at the Cultural Cognition Project and I did a study in which we instructed subjects to view a videotape of a protest that (we said) was broken up by the police to determnine if the protestors had crossed the line between “speech” & intimidation. Our subjects said "yes" or "no" -- said they saw shoving, blocking or only exhorting, persuading -- depending on the subjects' own values & what we told them the protest was about & where it was taking place: an anti-abortion demonstration outside an abortion clinic; or an anti- don't/ask/don't/tell protest outside a college recruitment center.

This is an example of “cultural cognition,” the tendency of people to conform their view of legally relevant facts to their group values. It’s a big problem for law — not just because these dynamics could affect juries & judges but also because they generate divisive conflict over the political neutrality of the law. I wrote a long law review article about this problem recently but I admit (as I did there) that I don’t think there is any easy solution to it.

But here is one thing concerned citizens might do to try to counteract this dynamic. When they see something unjust like UC Davis incident, try to look & find out if the same injustice has been perpetrated against others whose political views are different from one's own -- & complain about both.

I looked for stories on abortion protesters being "pepper" sprayed. Found some, but not many. Either anti-abortion protesters don't get sprayed as often (in absolute terms) as Occupy Wall Street & anti-war protesters or the spraying doesn't get reported as often, perhaps because of the impact of cultural cognition in reporting of news (the facts that get reported are the ones we are predisposed to believe) . . . .

reposted from Balkinization

Thursday
Nov172011

Two Communication Sciences for Plata's Republic

gave a talk last night at Harvard Law School in connection with the Supreme Court Foreword. Below is an *outline* of points I made. It is *not* text of my talk; I spoke extemporaneously & merely used the outline as something to think about as I thought about what to say in afternoon. (Maybe I'll try to remember what I said--was not nearly so dense as this-- & write it down, but I doubt it!) "Plata's Republic" is play on case Brown v. Plata in which Scalia's dissent looks motivated reasoning in the eye & proclaims it the truth of the role of empirical claims in democratic policy deliberations (I think the most surprising thing I've ever seen in U.S. Reports).

 

            1.  My basic claim is that political conflict over the neutrality of the Supreme Court is generated by psychological dynamics unrelated to whether the Justices are genuinely partisan or whether genuine neutrality is possible. That is, such conflict can be fully explained even assuming that neutrality is meaningful and that the Court is an acceptably neutral decisionmaker.  If such conflict is undesirable—as I submit it is—then we must perfect our understanding of nature of these dynamics and of how to control them.

            2.  We can make sense of these dynamics by considering political conflict over policy-relevant science. Valid science does not publicly certify itself: because citizens are not in a position to reproduce scientific findings on their own, they must necessarily rely on social cues  to certify for them what insights have been genuinely established through the use of valid scientific means. As a result of  motivated reasoning,  diverse groups of citizens will often construe those cues in opposing ways.   When that happens,  there will be political conflict over science notwithstanding its validity and notwithstanding the political impartiality and good faith of scientists. The existence of such conflict, moreover, will impede adoption of policies that effectively promote ends—including public health, national security, and economic propserity—that diverse citizens agree are the appropriate objects of law.

             3.  The dynamics that generate political conflict over the Supreme Court’s constituitional decisionmaking are exactly the same ones that generate political conflict over policy-relevant science. Just as they cannot verify the validity of science on their own, so citizens cannot verify the neutrality of constitutional decisionmaking on their own; they must rely on social cues to certify the validity of such decisionmaking. In this context, too, motivated reasoning will often drive citizens of diverse values to diverge in their assessments of what those cues mean.  Politically diverese citizens will disagree about the neutrality of constitutional decisionmaking in such circumstances despite the impartial application of valid doctrinal rules for enforcing the state’s obligation to be neutral in the manner that citizens of diverse values agree it should be. Such disagreement, moreover, will itself vititiate the value of the impartial application of those doctrines insofar as the benefit of neutrality consists largely in public confidence that the law is not imposing on them obligations incompatible with respect for the freedom of diverse citizens to pursue happiness on terms of their own choosing.

            4.  Both of these problems—political conflict over policy-relevant science and political conflict over constitutional law—reflect communication deficits.  The impediment that political conflict poses to the adoption of informed policies is the price we pay for failing to recognize that d oing valid science and communicating the validity of science are entirely different things.  Likewise, some portion of the toll that political conflict over Supreme Court neutrality exacts from our experience of liberty—likely a very large portion of it—reflects our failure to recognize that doing netural decisionmaking and communicating it are entirely different things too.  How to shield public policy deliberations from the recurring influences—accidental and strategic—that trigger culturally motivated reasoning with respect to both policy-relevant science and constitutional neutrality are both matters that admit of and demand scientific investigation in their own rights.

            5.  Developing these sciences—fixing the communication failures of Plata’s Republic—is a mission that lawyers, and the institutions that train them, are ideally situtated to address.  It is a central part of the lawyer’s craft to match the content of information with the cultural cues (the social meanings) that enable its comprehension and that vouch for its credibility.  Our experience with and sensitivity to this dimension of effective communication can thus help to remedy the sad and costly inattention to it reflected in public policy discourse.   Moreover, because a training in law always has been and continues to be a form of preparation for the exercise of significant civic responsibility—we educate Presidents, after all, as well as Supreme Court Justices and Supreme Court advocates—it is perfectly natural that law schools should play a role in perfecting the science of science communication.  It is all the more obvious that they are the natural location to address the judiciary’s own peculiar and ironic neglect of the fit between its professional conventions for doing neutral law and the cues that communicate constitutional neutrality.  Not only are we ideally positioned to promote scientific inquiry into what effective neutrality communication demands; we are uniquely empowered and responsible for implementing what such investigation can teach us through the self-conscious and enlightened cultivation of our profession’s norms.

Wednesday
Nov162011

Cognitive Illiberalism & Neutral Principles of Constitutional Law

Harvard Foreword on motivated cogniton & constitutional law is now published. Basic argument is that the same interplay of cognitive & political dynamics that polarize Americans over climate change & other risk issues polarize them over the neutrality of the Supreme Court. Judges need help from communication science just as much as scientists do (although at least some Justices bear more responsibility for the communication problem in law than any scientist I can think of does for the one in public deliberations over risk regulation). There are two very thoughtful replies, one by Mark Tushnet & the other by Suzzana Sherry. I'll have to think their arguments over & see whether & how my position changes.

Tuesday
Nov152011

Saw a Protest Slide Show

My slides from CELS 2011 presentation of Saw a Protest. Best ones are definitely ## 42-46!

Saturday
Nov122011

Profiles of Risk Perception: Normality & Pathology

A friend asked me if I could supply him with graphic representations of data that illustrate the bimodal-- i.e., culturally polarized -- state of risk perceptions over climate change & contrast that distribution with a "normal" -- nonpolarized -- one on some other risk or issue. So I put together this:  

The bottom histogram is the bimodal cultural distribution for perceptions of climate change risks. The top histogram is the normal distribution for nanotechnology risk perceptions.  I selected nanotechnology as the comparison case not only because perceptions of its risk are not polarized but also because there is nothing that guarantees that they will stay that way. Indeed, in our study Kahan, D.M., Braman, D., Slovic, P., Gastil, J. & Cohen, G. Cultural Cognition of the Risks and Benefits of Nanotechnology. Nature Nanotechnology 4, 87-91 (2009), we used nanotechnology risk perceptions to test the hypothesis that  that cultural predispositions can induce biased assimilation & polarization when people are exposed to information about a novel risk, one about which they had little if any prior knowledge and on which they were not polarized prior to information exposure:

 

In sum:

(1) the top histogram is picture of a (deliberatively) "healthy" distribution of risk perceptions;

(2) the bottom histogram is a picture of a "pathological" one; and

(3) among the goals of the science of science communication should be to learn to identify risk  sources that are vulnerable to becoming infected with this pathology -- as nanotechnology  evidently is -- and to perfect techniques for building up their resistance to it (techniques for  treating pathologies is critical too-- but it is a lot harder, I think, to change polarizing meanings  than it is to stifle their formation). 

 

 

Friday
Jul222011

Science communication isn't soulcraft (or shouldn't be)

President Obama has recently been taking heat from environmentalists, most conspicuously Al Gore in a recent Rolling Stone essay, for not using his “bully pulpit” to force the public to attend to the threat posed by climate change. “By excising ‘climate change’ from his vocabulary,” said one critic, “the president has surrendered the power that only he has to explain challenging issues and advance complex solutions for our country.


I definitely agree that President Obama should be taking the lead to improve public comprehension of climate change science. But I suspect I have a very different opinion on what the President should be trying to communicatealso how and when. What the public needs, in my view, is not more information about climate change, but a new, more inclusive set of cultural idioms for discussing this issue.

My argument will be easier to understand if I start by describing a national public opinion study conducted by the Cultural Cognition Project, a research consortium of which I am a member. There were two principal findings.
  • First, public controversy is strongly associated with differences in cultural or group values. People who subscribe to an individualistic, pro-market worldview tend to see climate change risks as small, while people who subscribe to an egalitarian, wealth-redistributive worldview tend to see them as large.
  • Second, differences in science literacy (how knowledgeable people are about basic science) and numeracy (a measure of their facility with quantitative, technical reasoning) magnify cultural polarization. As egalitarians become more scientifically literate and numerate, their concerns grow even larger; as individualists become more scientifically literate and numerate, their concerns diminish all the more. (For this reason, levels of science literacy and numeracy have essentially no meaningful impact overall).

These data suggest that conflict over climate change, far from reflecting a deficit in public comprehension of scientific information, demonstrates how adept people are in forming beliefs that express their group commitments. Should that surprise anyone? Right or wrong, the risk perceptions of an ordinary individual won’t actually affect the climate: the contribution an individual makes to carbon emission levels by her personal behavior as a consumer, or to climate change policymaking by her personal behavior as a voter, is just too small to matter. If, however, an individual (whether a university professor in Massachusetts or an oil-rig worker in Oklahoma) forms a belief about climate change that is heretical within her community, she might well forfeit the friendship and respect of people she depends on most for support in her everyday life.

Because it’s in the rational interests of ordinary people to conform their beliefs to those that predominate in their cultural groups, it’s also not surprising that science literacy and numeracy magnify cultural polarization. People who know more about science and have a greater facility with technical reasoning can use those skills to find even more evidence that supports their culturally congenial beliefs.

Of course, if we all follow this strategy of belief formation simultaneously, the collective outcome could be a disaster. I’m not hurt when I adopt a belief that “fits” my values but that is wrong, as a matter of scientific fact; but I and many others might well suffer harm if society adopts policies that don’t reflect the best available science about consequential societal risks. Because we live in a democracy, moreover, the risk that society will fail to adopt scientifically enlightened policies goes up as individuals of diverse cultural affiliations form the impression that it is in their expressive interest to adopt beliefs that affirm their groups’ values over their rivals’.

So back to President Obama and his role in the climate change debate. I think it is one of his
Administration’s responsibilities to foster a science communication environment that spares us from these sorts of tragic conflicts between individual expressive interests and collective welfare ones.

When our leaders talk about risk, they convey information not only about what the scientific facts are but also what it means, culturally, to take stances on those facts. They must therefore take the utmost care to avoid framing issues in a manner that creates the sort of toxic deliberative environment in which citizens perceive that the positions they adopt are tests of loyalty to one or another side in a contest for cultural dominance.

Where, as is true in the global warming debate, citizens find themselves choking in a climate already polluted with such resonances, then leaders and public spirited citizens must strive to clean things up—by creating an alternative set of cultural meanings that don’t variously affirm and threaten different groups’ identities.

 In that sort of environment, we can rely on the trust in science and scientists common to the overwhelming majority of cultural communities in our society to guide citizens toward acceptance of the best available science—much as it has on myriad other issues so numerous, so mundane (“take penicillin for strep throat”; “use a GPS system to keep from getting lost”) that they are essentially taken for granted.

 In his Rolling Stone essay, Al Gore calls the debate over climate change a “a struggle for the soul of America.” He’s right; but that’s exactly the problem. In “battles” over “souls,” citizens of a diverse, pluralistic society will naturally disagree—intensely. We’d all be better off if the issue had never come to bear connotations so fraught. Obama’s primary science communication task now is to lower the stakes.

 It won’t be easy. But any progress will depend indispensably on respecting the separation of science communication from soulcraft.

 President Obama, at least, seems to actually get that.

Tuesday
May242011

Prison Overcrowding, Recidivism & Crime

I was one of many, many experts contributing to briefs to the Supreme Court on this case.  In a 5-4 decision, the Court upheld a decision to require California to reduce the number of prisoners to a number that the state itself deemed safe for inmates.  Part of the Supreme Court's calculus involved weighing potential risks and benefits to public safety involved.  The majority cited expert testimony (based on numerous studies) that lowering prison populations may, on net, enhance public safety.   

Monday
May032010

More Immigrants, Less Crime? 

Like seemingly every other major cultural flashpoint (guns, the death penalty, and even abortion), both sides of the immigration debate have seized on anti-crime arguments.  No one in the mainstream debate disputes that immigrants, on average, are less likely to commit crimes than native-born citizens, but I doubt that is very convincing to supporters of the new immigration law.  There have also been several high-profile crimes committed by immigrants in Arizona, though I doubt those have swayed opponents of the new law.  I suspect that, as with other debates about the sources of crime, the evidence is culturally loaded enough to make it hard for anyone who feels passionately about the issue to process contrary information.  On the bright side -- and unlike gun control, capital punishment and abortion law -- nearly everyone agrees that immigration reform is needed.  There also used to be a number of Republicans like McCain who campaigned on the issues.  There's no predicting how the issue will play out this round, but I doubt that arguments about crime are unlikely to be decisive.  It does, however, provide a rich field for anyone interested in doing empirical research into the way cultural cognition shapes receptivity to arguments and information about immigration! 

Monday
May032010

Cultural Cognition and the Supreme Court

The NYT has an interesting op ed by Charles M. Blow today.  What I find most interesting isn't the notion that opposition to abortion is waxing, but the way this appears to be tied to attitudes about the Supreme Court.  Here's a little clip from the side graph to the article.  

Basically public perception appears to have reversed course after Obama was elected, with more Americans thinking that the Court is more liberal now that Obama has been elected and Sotomayor appointed.  While there are some interesting theories about justices trending liberal over their tenures, I suspect that more obsessive SCOTUS watchers would, whether they are happy or upset by it, say that the Court has either maintained its ideological balance or trended conservative in recent years.  

Why does public perception data seems to trend the other way?  It's a small change, to be sure, but I wonder if perceptions about the Court aren't the product cultural cognition.  If so, then it would make sense that people who think of the country as a whole as becoming more liberal under Obama as thinking that the Court, too, has become more liberal.  As a cultural touchpoint, it would be disconcerting to people at both ends of the ideological spectrum to think that Obama has had no impact on the ideology of the Court or -- even more disconcerting for ideologues on both sides -- that the Court may trend conservative in his administration.  Just a theory for now -- we'd need more data to test it. 

Tuesday
Feb232010

NPR Story on Climate Change

Christopher Joyce has a nice story on how cultural cognition shapes perceptions of climate change:   

"Basically the reason that people react in a close-minded way to information is that the implications of it threaten their values," says Dan Kahan, a law professor at Yale University and a member of The Cultural Cognition Project. 

Have a listen!

Monday
Dec142009

NYT Sunday Magazine Calls for End to Cognitive Illiberalism

We were delighted to discover that the CCP's study of the Supreme Court's decision in Scott v. Harris made it into New York Times Sunday Magazines Ninth Annual Year in Ideas (standard for selection: "the most clever, important, silly and just plain weird innovations..."). It was especially fitting to share that honor with the Ruppy, the glow-in-the-dark dog, public fears of whom are being investigated in CCP's synthetic biology risk perception project.

Friday
Oct092009

Scalia and the Cross

The New York Times is reporting on a Supreme Court case about a cross erected in the Mojave National Preserve in 1943.  While most of the Court seemed focused on whether the attempt to transfer the land to a private party (and thus avoid establishment issues) was proper, Justice Scalia went right for the establishment question: 

The question of the meaning of a cross in the context of a war memorial did give rise to one heated exchange, between Justice Scalia and Peter J. Eliasberg, a lawyer for Mr. Buono with the American Civil Liberties Union Foundation of Southern California.

Mr. Eliasberg said many Jewish war veterans would not wish to be honored by “the predominant symbol of Christianity,” one that “signifies that Jesus is the son of God and died to redeem mankind for our sins.”

Justice Scalia disagreed, saying, “The cross is the most common symbol of the resting place of the dead.”

“What would you have them erect?” Justice Scalia asked. “Some conglomerate of a cross, a Star of David and, you know, a Muslim half moon and star?”

Mr. Eliasberg said he had visited Jewish cemeteries. “There is never a cross on the tombstone of a Jew,” he said, to laughter in the courtroom.

Justice Scalia grew visibly angry. “I don’t think you can leap from that to the conclusion that the only war dead that that cross honors are the Christian war dead,” he said. “I think that’s an outrageous conclusion.”

Stephen Burbank (in an email) points out that this has all the markers of cognitive illiberalism as described by our article in the Harvard Law Review on the Supreme Court's decision in Scott v. Harris:

Because they are not generally aware of their own disposition to form factual beliefs that cohere with their cultural commitments [judges] manifest little uncertainty about their answers to [policy questions turning on issues of disputed fact]. But much worse, because they can see full well the influence that cultural predispositions have on those who disagree with them, participants in policy debates often adopt a dismissive and even contemptuous posture towards their opponents' beliefs....

It may be cognitively difficult for someone with the cultural commitments of Justice Scalia to understand the cross as anything other than a universal symbol profound respect, and to struggle with evidence to the contrary.  But struggling with cultural blindspots is something we expect judges to do, particularly in cases involving questions about the establishment clause.  

Wednesday
Sep302009

Supreme Court to Examine Chicago Gun Ban

The Chicago Sun-Times and just about every other news source in the country is reporting the Supreme Court decision to hear a challenge to the city of Chicago's ordinance barring handgun ownership (McDonald v. Chicago, No. 08-1521).  The debate over the ordinance and the case is ostensibly one about rights, but those rights are, as the majority opinion in Heller indicated, to be balanced with concerns about public safety.  Just what public safety requires, though, is precisely what cultural cognition predicts people will disagree over.  And, sure enough, as the headline in the Chicago Sun Times (surely intended to generate outrage and rejoicing in different communities) states: "Gun advocates predict drop in crime if gun ban is lifted."  McDonald, Heller, and their progeny may strike a compromise that appeals to a broad spectrum of the American public, or they may inflame cultural passions further.  Only time will tell.  But in the meantime, you can read up on the debate and the role cultural cognition plays in it here: 

 

Thursday
Sep102009

Kobe Bryant, Anita Bryant, and Big Ben

It’s not clear that the case will ever make it to trial, but if it does, what sort of person would make the best juror for Pittsburgh Steelers quarterback “Big Ben” Roethlisberger in his defense to the civil sexual assault case filed against him? The answer might come as a surprise -- maybe not to Roethlisberger’s lawyers, but probably to many commentators involved in the debate over law and date rape. A study founded on the theory of cultural cognition suggests that Big Ben would likely be judged much more sympathetically by a jury dominated by women who subscribe to highly traditional gender norms than he would by one consisting literally of his “peers.”

Click to read more ...

Friday
Aug072009

Easterbrook on Climate Change

Steve Easterbrook has a thoughtful post on his blog about cultural cognition and climate change.  In the comments on his post, one of his readers describe a common problem that lay citizens often ascribe to experts: sometimes experts "lie for hire."  It's true that members of the public may worry about that, but it's also true that individuals selectively attend to this particular risk when evaluating expert opinion.  Dan is in the middle of a study in which he demonstrates precisely this phenomenon, and I'm sure he'll be posting about it here soon!   

Wednesday
Jul292009

Robots That DO NOT Eat Humans

Dan's post suggests to me that cultural cognition might influence how reassuring individuals find a recent press release by Robotic Technology, Inc., setting the record straight. While the company's Energetically Autonomous Tactical Robot ("EATR" for short), "can find, ingest, and extract energy from biomass in the environment (and other organically-based energy sources)," RTI wants you to know that, "[d]espite the far-reaching reports that this includes “human bodies,” the public can be assured that the engine Cyclone (Cyclone Power Technologies Inc.) has developed to power the EATR runs on fuel no scarier than twigs, grass clippings and wood chips -- small, plant-based items for which RTI’s robotic technology is designed to forage. Desecration of the dead is a war crime under Article 15 of the Geneva Conventions, and is certainly not something sanctioned by DARPA, Cyclone or RTI."  

(Thanks to Sarah Lawsky for the link.) 

Sunday
Jul262009

The Next Frontier of Risk Perception: AI

Story today in NY Times on growing concern about the risks posed by artificial intelligence and in particular the possibility that artificially intelligent systems (including ones designed to kill people) will become autonomous. Interesting to consider how this one might play out in cultural terms. Individualism should incline people toward low risk perception, of course. But hierarchy & egalitarianism could go either way, depending on the meanings that AI becomes invested with: if applications are primarily commercial and defense-related and the technology gets lumped in w/ nanotechnology, nuclear, etc., then egalitarians will likely be fearful, and hierarchs not; if AI starts to look like "creation of life" -- akin to synbio -- then expect hierarchs to resist, particularly ones who are highly religious in nature.  Wisely, AI stakeholders -- like nanotech & synbio ones-- recognize that the time is *now* to sort out what the likely risk perceptions will be so that they can be managed and steered in way that doesn't distort informed public deliberation:

 

The meeting on artificial intelligence could be pivotal to the future of the field. Paul Berg, who was the organizer of the 1975 Asilomar meeting and received a Nobel Prize for chemistry in 1980, said it was important for scientific communities to engage the public before alarm and opposition becomes unshakable.

"If you wait too long and the sides become entrenched like with G.M.O.," he said, referring to genetically modified foods, 'then it is very difficult. It’s too complex, and people talk right past each other."

This is a topic ripe for investigation by cultural theorists of risk. 

Saturday
Jul252009

yellow statistics

With apologies to Coldplay, here's a lament for all the "stargazers" out there:

Click to read more ...

Wednesday
Jul152009

Combining Likert Categories and Embracing a Simulation-Based Mindset

Quite often we'll be developing simulations based on models in which the DV is a 6-point Likert-style response scale. (Usually it's somethign like: strongly disagree / disagree / mildly disagree / mildly agree / agree / strongly agree.) For presentation purposes, it's often useful to reduce this down into two categories: any form of agreement / any form of disagreement. In particular, when graphing, it is much easier to show one cut with confidence intervals than to show five cuts with confidence intervals. 

In the past we've done this by converting the DV into a binary variable, then running a logistic regression. But this has numerous drawbacks. First and foremost, it simply throws away all the information about how strongly a person agrees or disagrees. As a result, errors tend to be larger than necessary. Second, and relatedly, the results often aren't as similar to the ologit regressions run against the more information-rich likert DV as one would like. And third, if we want to report both kinds of findings -- binary and likert-style -- this means reporting two separate models that don't always give the same results. In short, it's been a mess, and we've usually just chosen one or the other. But when we've gone with a logit regression, this seems like sad choice to make just to achieve greater simplicity of presentation.

Recently, though, I had coffee with Jeff Lax-- of state-level policy analysis & Gelman Blog fame -- and he suggested something that, in retrospect, reveals that I'm still often trapped in a non-simulation mindset. In essence, he suggested this: "Run simulations on your ologit model & combine the simulations for the agree levels and again for the disagree levels; then take your confidence intervals from those combined simulations." In retrospect, that is so clearly the correct approach that the question is why I didn't see it myself. The answer, I think, is that I was still thinking in terms of the regression model rather than the simulations.