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

« Back in the US ... back in the US ... back in the US of Societal Risk conflict | Main | Build it & they will model ... the CCP data playground concept »

MAPKIA! Episode #73 Results: Stunning lack of any meaningful relationship between vaccine- and GM-food-risk perceptions earns @Mw record-breaking 5th straight MAPKIA! title!

@Mw nuzzles her prize: a baby giganto-technology e. coli -- it's not disgusting!So the results are in!

@Mw has won her Fifth  "MAPKIA!"!, earning her the appellation of MAPKIA “Lance Armstrong”!

Because she already owns 4 I ♥ Popper “Yellow" Jerseys from her previous victories, she selected a giganto-technology genetically engineered e. coli for her prize.  It was the last one in stock—lucky her!

Remember, the question was

What sorts of individual characteristics or predispositions, if any, account for the observed relationship between vaccine- and GM-food-risk perceptions and what, if anything, can we learn about risk perceptions generally from this relationship?  

The “observed relationship” in question was the one in this graphic,

which I constructed in response to a Twitter exchange, which itself was inspired by blog post I wrote in response to a question posed by a “politics & science” webinar member, who . . . Oh, who cares.

Anyway, there were, in effect, two main hypotheses.

@Mw’s was essentially “there isn’t any meaningful relationship between vaccine-risk and GM-Food-risk perceptions in particular—it’s just a weak measure of some indicator of generalized worry about risks.”

That was pretty much my thought, too. I know from lots of previous examinations that general population survey measures are not suited for generating any meaningful insight into either of these risk perceptions. 

Reactions to GM foods are pure static—uninformed noise from survey respondents the vast majority of whom have no idea what they are being asked about.

On vaccines, the vast majority of the US population has extremely positive affective reactions to them, and the small minority that doesn’t has views that are unrepresentative of any of the sorts of cultural or like affinity groups in which clusters of societal risk perceptions tend to form.

If the two risk perceptions are basically just sports, why expect something meaningful to come from the intersection of them?

But resisting this view, @ScottClif & @DaneGWendell, on twitter, seconded more or less by @Cortlandt in comment thread, proposed a “disgust sensibility” link.

Essentially, people who get grossed out easily will be anxious about the effect of ingesting laboratory synthesized variants of food stuffs & being injected with chemical concoctions like vaccines.

Disgust for sure is assigned a risk-detection role, so this is a perfectly plausible conjecture, too, I agree.

But I think at least the data I was able to pull together for testing these competing hypotheses pretty strongly favors @Mw.

A proviso is in order, however. 

Obviously, everything one learns from data, even when the data bear a valid inferential connection to the question at hand, is provisional.  Empirical proof doesn’t “prove” propositions (other than the most trivial ones, I suppose) with probability 1.0; it supplies evidence (again if valid) that gives us more reason or less to believe that one conjecture or another is true.

Accordingly, we have to think about how much more reason we have to believe one thing or another—that is, how much weight the evidence has.  And we have to maintain a permanent state of amenability to adjusting our resulting assessment of the balance of the evidence for or against various hypotheses in light of whatever additional valid evidence might later be adduced.

I’m pointing out these admittedly super obvious things because in fact @ScottClif & @DaneGWendell report that they have collected their own data on disgust sensibilities and vaccine- and GM food-risk perception and believe that theirs do show a connection.

For sure, I’m not saying that what I’m producing here means their conclusions must be “wrong”! I haven’t even seen their study. 

But more importantly, as I just said, it’s not in the nature to treat any valid evidence—assuming that this is; people should weigh in, as it were on that, too—as dispositively resolving an issue.  That's not how empirical proof works!

Obviously, when I do get to see their evidence, I’ll take it into account along with the data I’m about to present and adjust my assessment of the truth about the underlying connections, dynamics, and mechanisms accordingly. 

Indeed, because (I gather) they were setting out to examine exactly this question—whether “disgust” shapes vaccine- and GM food-risk perceptions—I am sure the employed measures that were very well calibrated to testing this hypothesis.  I’m using ones that weren’t designed specifically for that task but that I have reason to think ought to support valid inferences on it.  But maybe the difference in the precision in our respective measurement strategies will make a huge difference.

Or maybe they’ll point out something else about their data that shows how it clears the barriers that I think mine throw down in the inferential path toward the conclusion that disgust sensibilities link vaccine-risk and GM food-risk perceptions.

We’ll see!

And hopefully their observing some evidence that seems to me to be pretty strongly inconsistent with their surmise will help them to sharpen my and others' apprehension of what's even more compelling about their data.

Okay, then. . . back to the “MAPKIA”!

Basically, @Mw proposed a “falsification” strategy: any theory that "explains" the “observed relationship” between vaccine and GM food risk perceptions (which is pretty modest in any case) on the basis of some distinctive affinity between those two risk perceptions loses plausibility if it turns out the same relationship exists between either of them and various other, disparate forms of risk perception.

When we run that test, that’s exactly what we see.

Here is the relationship (in the N ≈ 1800, nationally representative sample featured in Kahan, Climate-Science Communication and the Measurement Problem, Advances in Pol. Psych 36, 1-43 (2015)) between concerns about vaccines and a pile of additional putative risk sources (click for more detail) :

Well, these all look pretty much the same as relationship between vaccine-risk and GM food risk perceptions.  

In all cases, we see simply a very modest positive relationship, which is consistent with the not particularly interesting or surprising inference (one surmised by @Mw) that people who tend to worry about one thing also worry about another (although not very much; the vaccine risk concern level is deemed “low” for those most concerned for each of these risks).

The uniformity of these correlations also seems to tell against the hypothesis that vaccine risk perceptions are related to “disgust sensibilities.”  We can see very modest correlations between the perceived risk of childhood vaccinations and perceptions of the danger of putative risk sources that we might expect to evoke disgust, including pornography, and the legalization of prostitution and marijuana (Brenner & Inbar 2014; McCoun 2012; Gutierrez & Giner-Sorrolla 2007).

But we can see the same very modest correllations betweeen concern over vaccines and concern over over high-voltage residential power lines, private operation of drones, and nuclear power--none of which seems to defile "purity," flout conventional sexual morality, compromise bodily integrity, etc. 

Definitely not what one would expect to see, I'd say, if disgust sensibilities were truly driving vaccine risk perceptions.

Okay. Now consider the same test as applied to GM food risk perceptions.

The correlation between self-reported concern w/ GM foods and the disgust trio—porn, legalization of prostitution, and legalization of weed—is, if anything, weaker than were the (already very modest) correlations between concerns with vaccines and the disgust-eliciting risk sources.

What’s more, the correlations between GM food risk perceptions and the eclectic trio of non-disgust risks is noticeably higher.

I don’t think that’s what one would expect to see if GM food risk perceptions were a consequence of disgust sensitivity.

I did one more test to help sort out affinities between GM food risk perceptions, vaccine risk perceptions, and concerns about various other risk sources: I tossed in responses to a whole bunch of “industrial strength risk perception measure”items into a factor analysis. 

This sort of analysis should be handled with a lot more care and judgment than one typically sees when researchers use it (it’s definitely in the “what button do I push” tool kit), but basically, factor analysis uses the covariance matrix to try to identify how many latent or unobserved variables have to be posited to explain variance in the observed items and how strong the relationship is (as reflected in the factor loading coefficients) between the individual items and those various latent variables.

Here’s what we see:

Basically, the analysis is telling us that we can reasonably make sense of the pattern of responses to all of these ISRPMs by positing three unobserved risk predispositions (because positing any more than that adds too little explanatory value).

It’s pretty obvious what the second "factor" or unobserved latent variable is getting at: the perceived riskiness of socially deviant behavior that, in people who fear them at least, evoke disgust (Gutierrez & Giner-Sorrolla 2007).  In cultural cognition terms, these are the things that divide hierarch communitarians and egalitarian individualists.

I have a pretty good idea what the last one is measuring, too!  The sorts of risk perceptions that provoke conflict between hierarch individualists (particularly white males) and egalitarian communitarians.

The first, then, is just an odd bunch of environmental risks that in fact don't get people very worked up in the US. So I guess they are picking up on some general scaredy-cat disposition.

here are the cool ISRPMs that appear in the factor analysisNotice, that’s where GM Foods (“GMFRISK”) is ending up: connected to neither set of “culturally contested” risk ensembles but rather to the residual “I’m worried about technology, help me!” one, where actually there’s not much political contestation (or even generalized public concern) at all.

That would be in line w/ one of @Mw’s hypotheses, too—that people who are scared of both vaccines and GM foods are probably just scared of everything.

Except that it turns out that vaccine risk perceptions don’t meaningfully “load” on any of these latent risk predisposition variables (in fact, it had anemic loadings of 0.33 on the first 2 factors, and -0.10 on the third).

That is, none of these latent risk predispositions alerts to, or explains variance in, vaccine risks.

Not surprising, given how overwhelmingly positive the general population feels about vaccines and how unconnected those who worry about them are to any recognizable cultural group in the US.

Anyway, that’s how I see it!

Feel free to file a protest of this determination, & I will duly forward it to the Head of Gaming Commission, who rules on all MAPKIA appeals.


Brenner, C.J. & Inbar, Y. Disgust sensitivity predicts political ideology and policy attitudes in the Netherlands. European Journal of Social Psychology 45, 27-38 (2015).

Gutierrez, R. & Giner-Sorolla, R. Anger, disgust, and presumption of harm as reactions to taboo-breaking Behaviors. Emotion 7, 853-868 (2007).

MacCoun, R. Moral Outrage and Opposition to Harm Reduction. Criminal Law, Philosophy 7, 83-98 (2013).

PrintView Printer Friendly Version

EmailEmail Article to Friend

Reader Comments (5)

As you suggest, a common "risk" dimension plausibly underlies survey questions posing different substantive "dangers" (GMOs, drones, nukes). Similar arguments could support a common "trust scientists" dimension across 5 substantive topics in a survey we just completed here, and this dimension does have strong social correlates.

May 27, 2015 | Unregistered CommenterL Hamilton


Are the results written up?

Note here, of course, that your specific examples of GMOs, drones & Nukes aren't linked to any of the latent variables suggested by the factor analysis.

Nuclear is part of a cluster that fore sure will be characterized by cultural identities for which race, gender, and cultural outlooks (or political ideologies) will be indicators.

Those sorts of characterstics, in my expreience, explain only a tiny fraction of the variance on GM food risks, likely b/c what is being modeled is not a genuine social phenomeon; people don't know what GM foods are or care about them, so surveys create spurious (largely uninterpretable) pictures of who worries about them.

I have no idea who worries about drones or why. But I have noticed, casually, that it seems to bug tea party types, who say they find them ... disgusting!

May 27, 2015 | Registered CommenterDan Kahan

Are the results written up?

Not yet, just got the data & will take a couple months to sort out. Have been focused on another project, "Polar facts in the age of polarization," that should hit the street next month.

In my nukes, climate, evolution etc. questions I've asked about trust in scientists for information, not about risks. Although there is some element of domain response (e.g., if respondent does not believe climate's changing) there seems to be a stronger signal beneath that, propensity to trust scientists on controversial topics, that is less domain-dependent.

I have no idea who worries about drones or why.

For a different project I joined forest biometricians on a field trip where they used drones to survey trees. In our discussions (also with local residents) it seemed everyone had a clear about who might be worried.

May 30, 2015 | Unregistered CommenterL Hamilton

I would like to appeal to the herd of gaming commissioners. I think my suggestions came closer to the results you listed than you give me credit for. The opposite of the nuance of the 2nd theory I proposed is captured fairly well by your statement:

Notice, that’s where GM Foods (“GMFRISK”) is ending up: connected to neither set of “culturally contested” risk ensembles but rather to the residual “I’m worried about technology, help me!”

The concern about technology is the flip side of a belief that nature is something beautiful, spiritual, to be protected from human interference. Ditto the ideas of the naturalness of nature, pure, organic, etc. But, contrary to your analysis, I do think that electromagnetic exposure to power lines, cell phones, sacharine, food color risks, beef (if the concern is about growth hormones) and even fracking appear to me to "load onto" this factor.

I wish I could think of a way to transform this notion into a survey question. I am impressed by research designs that transform such notions into industrial strength survey questions.

How important/significant is any of that? Shrug. Who cares anyway. I just want my t-shirt.

Credit for the "nature is something beautiful" concept to this:

... two sentiments which are very much last century. One is the idea that modernization is a project leading to progress, with an endpoint that can be clearly defined ... The other is that nature is something beautiful, spiritual, to be protected from human interference. This sounds a bit like the belief of deep ecologists.

June 2, 2015 | Unregistered CommenterCortlandt

On vaccines, the vast majority of the US population has extremely positive affective reactions to them, and the small minority that doesn’t has views that are unrepresentative of any of the sorts of cultural or like affinity groups in which clusters of societal risk perceptions tend to form.

Do geographic enclaves (e.g., Marin County with respect to vaccine risk?) affect this conclusion?

June 5, 2015 | Unregistered CommenterWade

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>