I recently published a paper on the motivated rejection of science that is forthcoming in Psychological Science. The abstract of the paper is as follows:
Although nearly all domain experts agree that human CO2 emissions are altering the world’s climate, segments of the public remain unconvinced by the scientific evidence. Internet blogs have become a vocal platform for climate denial, and bloggers have taken a prominent and influential role in questioning climate science. We report a survey (N > 1100) of climate blog users to identify the variables underlying acceptance and rejection of climate science. Paralleling previous work, we find that endorsement of a laissez-faire conception of free-market economics predicts rejection of climate science (r @ .80 between latent constructs). Endorsement of the free market also predicted the rejection of other established scientific findings, such as the facts that HIV causes AIDS and that smoking causes lung cancer. We additionally show that endorsement of a cluster of conspiracy theories (e.g., that the CIA killed Martin-Luther King or that NASA faked the moon landing) predicts rejection of climate science as well as the rejection of other scientific findings, above and beyond endorsement of laissez-faire free markets. This provides empirical confirmation of previous suggestions that conspiracist ideation contributes to the rejection of science. Acceptance of science, by contrast, was strongly associated with the perception of a consensus among scientists.
Perhaps unsurprisingly, this paper has caused a considerable media response and a flurry of activity on the internet. I have also received a fair amount of correspondence, so much in fact that I have been unable to keep up with it. I apologize to those who have not received a reply to recent messages, and I hope this post covers some of the issues raised.
In a somewhat ironic twist, given that the paper addressed conspiracist ideation, much attention has focused on the source of participants, which were “Visitors to climate blogs voluntarily completed an online questionnaire between August and October 2010 (N = 1377). Links were posted on 8 blogs (with a pro-science science stance but with a diverse audience); a further 5 “skeptic” (or “skeptic”-leaning) blogs were approached but none posted the link.”
To clarify, this means that participants were recruited from those blogs that posted the link—not those that did not. One might therefore presume that attention would focus on those blogs that provided entry points to the survey, not those that did not, because it is entirely unclear how the latter might contribute to the results of the survey. For example, the website of the British RSPCA also did not post a link to the survey, and neither did the Australian Woolworths website, so how might their non-involvement affect the results? I am keen to hear about potential mechanisms, perhaps we have overlooked something.
However, attention has primarily focused on those non-participating blogs and their identity. I have been inundated with requests to release their identities, and I have thus far declined to comply with those requests because I believe that a presumption of privacy should apply to my correspondence with potential participants in research.
Unlike some of the people who have been emailing me, my work is subject to ethical guidelines and is subject to approval by my University’s ethics committee—as is the work of any other behavioral scientist in Australia and elsewhere. It is therefore not solely my decision whether or not to reveal the identity of people who were approached on the presumption of privacy.
Because this issue is likely subject to different opinions, I have therefore approached the Australian Psychological Society and my University’s Human Research Ethics Committee to provide guidance on this decision.
There is an obvious asymmetry of potential harm here: If I release the names but it turns out to have been unethical, this cannot be undone. If I decline to release the names, as I have done to date, and it turns out that this was unnecessary, then no harm is done if release of the names is delayed by a few days.
I am therefore awaiting guidance on this issue.
In the meantime, I understand that there is a list on the internet of individuals who have declared that they were never contacted. As we are awaiting the decision about release of the names, just a matter of general principle, there can be no harm if those folks were to again check their inboxes (and outboxes) very carefully for correspondence from my assistant at UWA in August and September 2010. I know how difficult it is to locate individual emails among thousands received in a year, and a double check may therefore be quite prudent. (Who knows, it might even prevent some overly trigger-happy and creative people from floating a conspiracy theory about how I just made up the fact of having contacted those blogs, similar to the way NASA faked the moon landing.)
There are other issues that have been raised in connection with the paper, including some interesting points regarding the statistics, and they are worthy of further commentary in the near future. As it happens I am attending a conference at the moment with one of my co-authors, which ties us up for most of the time but which also provides an opportunity for discussion that is likely to lead to further posts in the not-too-distant future.
Here are the27 comments at the original post:
Comments 1 to 27:
1.Barry Woods at 03:57 AM on 4 September, 2012
I notice that no link to the paper is made here, nor the rather contentious title is not spoken..
“NASA faked the moon landing|
Therefore (Climate) Science is a Hoax: An Anatomy of the Motivated Rejection of Science”
http://websites.psychology.uwa.edu.au/labs/cogscience/documents/LskyetalPsychScienceinPressClimateConspiracy.pdf
Given that this was the reason a number of people were concerned, perhaps just a tiny oversight, in attempts to clarify the concerns made by ‘sceptics’ to the readers here?
As Steve Mcintyre – Climate Audit and Condon (The Air Vent are both named in the paper, I do think it appropriate, that the authors confirm thatthese 2 blogs were NOT approached – They both state that they were not.
Secondly, Tom Curtis at the ‘consensus blog’ – Skeptical Science’ one of the so called ‘pro-science’ blogs surveyed’ had this to say about the paper and it’s data. Perhaps a response from the authors is due?
Tom Curtis:
“Given the low number of “skeptical” respondents overall; these two scammed responses significantly affect the results regarding conspiracy theory ideation.
Indeed, given the dubious interpretation of weakly agreed responses (see previous post), this paper has no data worth interpreting with regard to conspiracy theory ideation. It is my strong opinion that the paper should be have its publication delayed while undergoing a substantial rewrite.
The rewrite should indicate explicitly why the responses regarding conspiracy theory ideation are in fact worthless, and concentrate solely on the result regarding free market beliefs (which has a strong enough a response to be salvageable). If this is not possible, it should simply be withdrawn.”
http://www.skepticalscience.com/AGU-Fall-Meeting-sessions-social-media-misinformation-uncertainty.html#84398
2. tlitb1 at 04:35 AM on 4 September, 2012
“In the meantime, I understand that there is a list on the internet of individuals who have declared that they were never contacted.”
You may also be interested to hear of a list on the internet of blog hosts who say you have their permission to reveal any contact you had with them regarding this survey, here.
http://rankexploits.com/musings/2012/tweet-your-permission-for-lewandowsky-to-out-you/
Names include: Anthony Watts, Steve McIntyre, Marc Morano and many, many more.
3. tlitb1 at 04:39 AM on 4 September, 2012
Actually strike Marc Morano, I don’t see his explicit permission there.
4.omnologos at 04:39 AM on 4 September, 2012
There is another asymmetry in your words. If Nasa really had faked the Moon landings, the conspiracy behind it would have been involving thousands of people across many decades (to this day, in fact!).
It is actually simple to retort any conspiracy theory about NASA and the Moon landings by saying that, if they really faked it, THAT would have been by itself an even bigger accomplishment than the landing.
On the other hand, for you to make up having contacted this or that blog, all it takes is for…you (and you alone – your co-authors don’t even have to be involved) to make up having contacted this or that blog.
Not a conspiracy by a long shot, and impossible to dismiss using the “thousands of people across many decades” argument.
Therefore the analogy “….floating a conspiracy theory about how I just made up the fact of having contacted those blogs, similar to the way NASA faked the moon landing” is deeply flawed. Same applies AFAIK to all examples of “conspiracy” described in your paper, all of them extremely unlikely hypotheses based on the nefarious cooperation of untold number of conjurers (BTW: I do not believe in any of them).
None of that is remotely comparable to the idea that reluctance to divulge names may be due to something else than privacy considerations.
5. tlitb1 at 05:09 AM on 4 September, 2012
One might therefore presume that attention would focus on those blogs that provided entry points to the survey, not those that did not, because it is entirely unclear how the latter might contribute to the results of the survey.
You stated the idea of exposing your survey to skeptic blogs in the paper
; a further 5 skeptic” (or skeptic”-leaning) blogs were approached but none posted the link.
Why did you tell us that in the paper, if, as you say “it is entirely unclear how the latter might contribute to the results of the survey”?
Once one reads that isn’t is reasonable to ask: What made you think of exposing your survey to skeptic blogs? What then made you give up after only the first 5 skeptic blogs rebuffed you?
6. oneillpt at 06:30 AM on 4 September, 2012
Entertain us please with an explanation of any possible relevance of the British RSPCA or Australian Woolworths not posting a link to your survey
7. ManicBeancounter at 06:36 AM on 4 September, 2012
Having examined the data in some detail, I am far less concerned about which “skeptical” blogs that were contacted, but the consequences for the results of not having them included. Looking at the four questions on climate science, it possible to score from 4 (complete rejection) to 16 (complete acceptance. 125 scored from 4 to 7 compared with 854 who scored 13 to 16.
On the conspiracy theories (excluding CYNewWorldOrder and CYClimChange) I found an average score of 1.55, with little difference between Pro & anti science groups.
8.Eli Rabett at 07:23 AM on 4 September, 2012
FWIW, the pressure on the usual skeptics to sign up to the tweet going about is likely to make the Human Subjects folk not allow Prof. L. to name them on the grounds that they were subject to group social pressure
9. Foxgoose at 07:55 AM on 4 September, 2012
There are several logical leaps in Eli’s desperate attempt to blow smoke.
The current premise is that there are no “Human Subjects” and as more and more known sceptical blog proprietors add their voices, this will become increasingly apparent.
Unless of course Stephan has approached the proprietors of five “sceptic blogs” which no-one in the sceptic community has ever heard of.
That scenario would carry a quite powerful message of its own.
10. Ben Pile at 09:28 AM on 4 September, 2012
Given that blogs are categorically ‘public’, I don’t see how there can be an expectation or assumption of privacy. The editor of a newspaper or company couldn’t expect any correspondence *to* him to be kept secret.
There are a number of other things about the contact between the researchers and the bloggers that might be useful to know.
1. How were the blogs chosen — what was the criteria?
2. How were the bloggers approached?
3. What was said/written in the request in each case?
4. Was the text of the request the same in each case?
5. Were there any existing relationships/friendships/etc between the researchers and the bloggers?
6. How were these potential contaminating influences anticipated, excluded, or managed by the researchers?
11. Eli Rabett at 10:18 AM on 4 September, 2012
“there are no “Human Subjects”
Well Eli would not have been quite so blunt.
12. Sou at 11:25 AM on 4 September, 2012
I’ve noticed a couple of curious things.
Firstly, blogs that regularly smear scientists are most upset, claiming they were not invited to take part. (This includes blogs that have maligned Prof Lewandowsky in the past.) These bloggers could consider that they are making a post-survey contribution, providing data for further research. The comments made by people on these blogs tend to align very well with the responses to the initial survey so I doubt that it would generate a complete new paper.
Secondly, ‘skeptical’ bloggers appear to confuse crank conspiracy theorists with extremist right wing views. My reading of the paper is that the survey found extremist ideologies (laissez faire variety) were a stronger predictor of the rejection of climate science than holding more generalised conspiracy theories. The findings are very clear in the paper, yet lots of people do not seem to have picked up on this.
Finally, many people have expressed an opinion about the title of the paper. I have some suggestions on my blog that should satisfy any concerns:
1. No Market is 100% Laissez-Faire – Therefore (Climate) Science is a Hoax: An Anatomy of the Motivated Rejection of Science
2. Societies elect Governments – Therefore (Climate) Science is a Hoax: An Anatomy of the Motivated Rejection of Science
Others suggestions eg “I resent paying tax – therefore …”
13. Sou at 11:38 AM on 4 September, 2012
PS I’m told that at least some of these bloggers who ‘denied’ being invited to post a link to the survey have finally found the email requesting they post a link to the survey, which they ignored at the time.
Could you make up this stuff?
14. TonyM at 15:08 PM on 4 September, 2012
The oft quoted adage “there are lies, damned lies and statistics” has been levelled at your paper already so I won’t add to that other than to emphasise that such a survey is wide open to be treated with contempt with some participants deliberately taking the “mickey” out of it depending on mood, predilection and sensing the survey inclination.
Clearly that does mean that the type of site chosen for the survey does affect the results contrary to your bleating protestation by invoking Woolworths irrelevancy.
The other issue is your claim of having asked a number of skeptical sites to participate and they declined. A number of these most popular “skeptical” sites have stated that they were not contacted and have given you public permission to name them.
You seem to duck for cover behind an illusory ethical dilemma which could easily be dispelled if you so choose. Simply contact those five sites and ask for permission to disclose their identity. Then, equally simple, make public those results by naming those which gave you permission. If zero, then state it clearly.
You may indeed be right that some of those sites who claim not to have received the invite may be in error and I am inclined to accept that. But it is hard to see how your approach and excuses help.
15. oneillpt at 17:54 PM on 4 September, 2012
Sou, there is an old Irish saying (non PC perhaps) which translates as “a woman told me that a woman told her”, which indicates possible unreliability of the information being offered.
“I’m told that at least some of these bloggers who ‘denied’ being invited to post a link to the survey have finally found the email …” seems to fit this pattern quite well. If any of the said bloggers have in fact said that they have found the email request, there should be no impediment to providing those specific names. Without such names, to use your own words, you could indeed be making up this stuff.
16. MikeH at 17:58 PM on 4 September, 2012
Steve McIntyre punctures the conspiracy theory about the survey by conceding that yes, he did receive an invitation (twice).
“In Lewandowsky’s post today, he reported that the inquiry was not sent out by him personally but by his research assistant. I searched again this time under the term “uwa.edu.au” and located an email from Charles Hanich on Sep 6, 2010 asking that the survey be posted by Climate Audit and a second request two weeks later.
Like many people, I get lots of emails. I didnt know Hanich and I didn’t pay any attention to the request at the time. I didnt reply.”
Can the readiness of the conspiracy theorists to accept a conspiracy theory about the survey itself lead to an addendum?
17. MikeH at 18:00 PM on 4 September, 2012
oneillpt has a conspiracy theory about the deflating of the conspiracy theory.
18. Ian Woolley at 18:11 PM on 4 September, 2012
Might be interesting if the questionnaire was now distributed via the sceptical blogs. In the interests of the fuller picture. Would this be possible, and would it now be accepted as part of the study?
19. Sou at 18:42 PM on 4 September, 2012
Ian Woolley, no need – seriously. Not only does the analysis of the survey results conform quite well to findings in similar studies, comments on the ‘skeptic’ blogs discussing this topic support the findings very well.
onillpt – you could check it out for yourself. I did. Of course, going by some of the comments I read, some ‘skeptics’ will ‘believe’ it’s all a conspiracy and that someone was impersonating Steve McIntyre when he admitted to finding the two requests after all.
(Another approach would be to ask all the various ‘skeptic’ bloggers to release all the emails they sent and received in 2010. There are lots of ‘skeptic’ bloggers. You never know – they might all be happy to provide all their emails on the internet in the interest of openness and full disclosure.)
20. Ian Woolley at 19:06 PM on 4 September, 2012
Sou – good use of the word ‘conform’. As Geoff Chambers points out at the Bishop Hill site, with Prof Lewandowsky giving a presentation at Monash university a mere 3 days after the second email to Steve McIntyre which included the following slide:
Lewandowsky & Gignac (forthcoming)
•Internet survey (N=1100)
•Endorsement of climate conspiracy (“hoax by scientists to get grants”) linked to endorsement of other conspiracies (“NASA faked moon landing”)
•Conspiracy factor without climate item predicts rejection of climate science
‘conforming’ to previous studies was very much the aim.
21. Sou at 19:20 PM on 4 September, 2012
Ian Woolley – so you are saying that you don’t accept the evidence from this particular survey, is that correct?
Do you also not accept scientific evidence that matches the physics of global warming?
Out of curiosity, if you answered ‘yes’ to both the above, what other conspiracy theories do you ‘believe’?
Also out of curiosity, what are your political leanings? (Whatever your answer to the above questions.)
Evidence suggests that more extreme laissez faire views are a good predictor of climate science rejection. However it doesn’t follow that all people who have such views reject climate science. Nor does it follow that a left-leaning person will accept climate science.
22. Barry Woods at 19:42 PM on 4 September, 2012
Perhaps the authors could adress some of the concerns raised at the blog Skeptical Science. Despite the title, this is a non ‘sceptic’ blog and as Stephan Lewandowsky contributed to this blog, interestingly at the same time as the survey was put on line there. Perhaps these thoughts will recieve due consideration, as they cannot be dismissed as just ‘sceptics’
Tom Curtis (at Skeptical Science – John Cook’s blog):
“The consequence of this lack of a neutral response combined with excluding all results that do not complete all questions is to:
1) Bias the sample by excluding some people who are trying to complete the survey accurately;
2) Force some of those who do complete the survey into more definite responses than they actually hold.
In my wife’s opinion, this flaw alone is enough to make the survey scientifically worthless; and I trust her judgement on this issue.
My wife further said that she would automatically reject a paper with Lewandowsky’s title as being politically motivated. In the social sciences, politically motivated papers are a major problem, and generate an excess of background noise and confusion.
Part of my wife’s response to that is simply to ignore as worthless clearly politically motivated papers. I can see her point, but disagree with the response. Data is data, and so long as you are clear as to how it was obtained, and the results obtained, can be interpreted without consideration of the views expressed in the paper.
What is more, in this case the views expressed in the paper are sober analysis. The title makes the paper seem very much worse than it is.
Never-the-less, given title, and given the (several) methodological flaws discussed in this post and in my post @12, this has confirmed my opinion that this paper is an “own goal” for opponents of “skepticism” about AGW. It contributes nothing of value scientifically to understanding AGW “skepticism”, and its title is a disaster. ”
http://www.skepticalscience.com/AGU-Fall-Meeting-sessions-social-media-misinformation-uncertainty.html#84398
I do think this other comment desreves some response as well from the authors of the paper:
Tom Curtis (Skeptical Science – John Cook’s blog):
“Given the low number of “skeptical” respondents overall; these two scammed responses significantly affect the results regarding conspiracy theory ideation.
Indeed, given the dubious interpretation of weakly agreed responses (see previous post), this paper has no data worth interpreting with regard to conspiracy theory ideation. It is my strong opinion that the paper should be have its publication delayed while undergoing a substantial rewrite.
The rewrite should indicate explicitly why the responses regarding conspiracy theory ideation are in fact worthless, and concentrate solely on the result regarding free market beliefs (which has a strong enough a response to be salvageable).
If this is not possible, it should simply be withdrawn.”
23. Sou at 19:56 PM on 4 September, 2012
I think this spate of comments is what is sometimes disparagingly referred to as ‘blog science’. ‘Skeptics’ come out in force, ironically probably including those who, had they responded to the survey, would have supported the results. They want every blog comment addressed. Par for the course in anything to do with climate science.
There are channels for commenting on published papers, including writing rebuttals. Blog review doesn’t have quite the same status.
(Incidentally, reporting on interim findings of an internet survey can be done very quickly. I can provide interim results within an hour or so of downloading responses to date. Had McIntyre posted a link when he received the first request, N quoted in the Sept presentation and subsequent paper would maybe have been higher. Even had he posted a link when he received the second request a couple of weeks later, N in the presentation and subsequent paper may have been higher. But McIntyre didn’t post the link at any time.)
24. Eli Rabett at 20:38 PM on 4 September, 2012
The whole thing is a great argument for the OSX and Mail, because the search function actually works.
But we are now well into the Gish Gallop phase of the thing. Tally ho!
25. Ian Woolley at 20:49 PM on 4 September, 2012
Sou – what’s your take on Tom Curtis’s opinion as quoted above by Barry Woods? My take, to answer your question, is that the results may well be ‘significantly affected’ given ‘the low number of “skeptical” respondents overall’.
Second question,(physics of global warming): not sure. There seems to be an awful lot of back and forth. But also, how we respond politically is much the more interesting question. And that *is* unsettled.
Conspiracy theories I believe in: I think the coach-ticket booking service for the National Express coaches here in Britain is rigged to ratchet-up the cost of ‘funfare’ tickets if you leave the site (to compare prices with other service providers)and come back even within minutes. I am absolutely convinced this is the case. There’s no way the prices would jump that quickly in that short an amount of time. And also pan-dimensional beings run H&M; (the ‘Swedish’ clothing company).
Political leanings: so tricky these days isn’t it? Left and right are hollow categories. I’m suspicious of anyone who sincerely describes themselves as socialist because I just think you can’t seriously have missed the last 20 years, can you?
Anyway, what do you think of Tom Curtis’s comments? Does he have a point?
26. Sou at 21:03 PM on 4 September, 2012
Ian, thank you for your response. I take it you are not sure about the science and are more interested in the policy responses. Policy responses are becoming critical every year.
Re my take on Tom Curtis’ remarks, if you read my comments in the same thread on skepticalscience.com (and my blog, which should be linked to my name from that same site) you will get a good idea.
27. oneillpt at 23:05 PM on 4 September, 2012
Sou – my apologies for doubting you. I had in fact of course tried to check it out for myself, and found nothing at Climate Audit or at the Blackboard, the two obvious places to look. A comment from Steve McIntyre now appears at the Blackboard confirming what you reported, together with a discussion of WordPress cache problem preventing comments being seen for a period other than by people who posted a comment there themselves.
I would have been less suspicious had you said plainly that “Steve McIntyre comments at the Blackboard …” rather than “I’m told that at least some of …”, not helped by your previous reference to suggestions at your own blog, without providing any link to that blog – I do actually like to check out things for myself, and you are as unknown to me as I probably am to you.
Unfortunately, I’ve had long experience of dubious claims based on dubious (or even not so dubious) surveys. It does lead to a rather suspicious mindset. See below for an example.
Drawing conclusions from a very small number of respondents in a survey can be very unwise, and when latched upon by activist groups downright dangerous.
As a cautionary tale, and draw any parallels you may wish, take one survey I felt compelled to comment on, many years ago, during a particularly divisive divorce referendum (divorce at that time being prohibited under the Irish constitution). One activist claimed that “statistics show that stepfathers are five times more likely to molest their stepdaughters than are natural fathers” and another activist group claimed that “virtually half the girls with stepfathers were victimised”. On going back to the original survey (I do like to check for myself as I said) it turned out that the first claim was based on only two claims of sexual victimisation by a stepfather in 530 responses from female students, and the second claim cleverly omitted the minor detail, on the same page in the source article, that of the 30 respondents with a stepfather, 14 had claimed victimisation by others, and that “paradoxically some of these stepdaughters were victimised prior to meeting their stepfather”.
Both claims however were accurate quotations from the source article. They just omitted qualifications such as “If we were to extrapolate on the basis of this data, something we are not really entitled to do, given the limitations of the sample”. Unfortunately the author, David Finkelhor, in the journal “Child Abuse and Neglect”, contributed to the misuse of his survey by failing to resist the temptation to provide these “useful” quotes even while admitting that they were not justified given the limitations of the data, and in fact warning in his book on the subject that “such small numbers are not a reliable basis on which to calculate prevalence accurately”
In the context of removing a constitutional ban on divorce I did mischievously ask in that newspaper comment whether we should abolish farms as the same survey claimed a victimisation rate of 44% (7 of 16) for those who lived on farms.
MikeH: MikeH has a conspiracy theory about oneillpt having a conspiracy theory about the deflating of the conspiracy theory. Over to you. Recursive conspiracy theories are such fun.