Following the retraction of Lewandowsky’s Fury, the validity of University of Western Australia ethics “investigations” is again in the news. At present, we have negligible information on the University’s investigation into Fury, but we do have considerable (previously unanalysed) information on their earlier and illusory “investigation” into prior complaints about the ethics application for Moon Landing Hoax (“Hoax”).
This earlier “investigation” (recently cited at desmog here and Hot Whopper here) supposedly found that the issues that I had raised in October 2012 were “baseless” and that the research in Hoax was “conducted in compliance with all applicable ethical guidelines”.
However, these conclusions were not written by a university investigation or university official but by Lewandowsky himself and simply transferred to university letterhead by UWA Deputy Vice Chancellor Robyn Owens within minutes after Lewandowsky had sent her language that was acceptable to him.
In today’s post, I’ll set out a detailed chronology of these remarkable events.
Hoax was published under the supposed authority of the University’s ethics permit RA/4/1/4007, a permit which had been originally issued for an entirely unrelated project under which pedestrians in Perth were interviewed about their “understanding of statistical trends in time series data”. The original ethics application included an ethics checklist, which, according to Australian policy, included the following question whether the research involved any deception or concealment: “Does the research involve active concealment of information from participants and/or planned deception of participants”. To which Lewandowsky answered “NO”.
By August 2010, Lewandowsky had become bored with the time series project and instead wanted to show that skeptics were conspiracy theorists. Instead of interviewing pedestrians in downtown Perth about trends, Lewandowsky wanted to do an internet survey about conspiracy theories.
Lewandowsky’s new project was so different from the existing approval that many important sections of the existing application ceased to apply (even the purpose of the study as stated in the original application no longer applied.) But instead of filing a new ethics approval for the entirely different project, Lewandowsky chose instead to pass off the new project as merely an amendment to his existing project, falsely assuring the ethics administrator in an amendment request that the survey would only be “modified slightly”. Attached to the email request were questions used in the Hoax survey. (^^- I’ll add a link to the documents.)
I am writing to seek approval for an amendment to the procedure for RA/4/1/4007.
In a nutshell, I want to administer the survey not in person but via the internet using professional survey software (e.g. http://www.surveymonkey.com or equivalent. As before, completion of the survey will be taken to constitute consent, and as before a variant of the approved infomration sheet will be shown before the survey commences.
The survey will be modified slightly as follows,
(1) The graphical extrapolation task is removed
(2) In addition to the already-approved items, some further questions will be presented that are enclosed in this email [information regarding questions already approved not reproduced here]
(3) In all other respects the approved procedure remains unchanged except that it is administered via internet, with consent again being expressed by completion of the electronic questionnaire.
(4) Participants will be recruited by posting links at relevant websites (e.g. http://www.uwa.edu.au/climatescience or science-oriented “blogs”.).
Misled by Lewandowsky’s false assertion that the survey was only “modified slightly”, ethics officer Kate Kirk approved the changes the next morning, the speed of the change surprising even Lewandowsky. (“wow, thanks for the quick approval.”) Emboldened, Lewandowsky then requested permission for active concealment of his association with the survey (a concealment that Lewandowsky only applied to skeptic blogs):
One question: would it be possible to mention only my assistant’s name, Charles Hanich, on the online survey? The reason for this is that I have been writing on the climate issue in public e.g. [here] and my name alone routinely elicits frothing at the mouth by various people, not to mention the hate mail I receive.
Because I am interested in soliciting opinions also from those folks, I would like to withhold my name from the survey as I fear it might contaminate responding.
Under the Australian National Statement, “active concealment” and/or deception in the course of a research project must be approved by a Human Ethics Research Committee and cannot be delegated to an ethics officer (section 2.3.4). Evaluation of proposed deception/concealment requires that the concealment or deception will not create any “harm” to the participants and that “a full explanation, both of the real aims and/or methods of the research, and also of why the concealment or deception was necessary, will subsequently be made available to participants” (section 2.3.6).
Kirk knew or ought to have known that only a Human Research Ethics Committee could approve deception/active concealment. Nonetheless, only five minutes after Lewandowsky’s email request, Kirk approved Lewandowsky’s deception/concealment request, saying that she “look[ed] forward to receiving the hate mail”:
Yes, fine for you to leave your name off as long as the standard complaints paragraph and contact details are there. I look forward to receiving the hate mail. I’ll let you know if I get any
Lewandowsky himself knew or ought to have known that only a Human Research Ethics Committee could approve deception/active concealment and that this was beyond Kirk’s authority. Nonetheless, Lewandowsky accepted this flawed approval. Lewandowsky even departed from these terms: although he withheld his association from the survey from skeptic blogs (not that any of them had heard of him at the time), he personally promoted the survey with eco-activist blogs.
Lewandowsky’s deception/concealment of his association with the survey later became an issue when, following publication of Hoax in August 2012, people wondered which skeptic blogs had been contacted by Lewandowsky. Because Lewandowsky had originally concealed his association with the survey from skeptic bloggers, searches of 2010 emails for “Lewandowsky” all failed (as I and others soon determined). Rather than explain his earlier concealment, Lewandowsky gleefully sought to embarrass bloggers who were unable to locate a survey invitation and taunted skeptic bloggers in a series of posts on a UWA blog. In one of his taunts, Lewandowsky coyly mentioned his “assistant”; I noticed the sly change from Lewandowsky himself and did a further search on “uwa.edu”, locating an invitation from Hanich. Within a few days, Pielke Jr, Roy Spencer and Marc Morano also located invitations, solving Lewandowsky’s game by September 10. A few hours after Spencer and Morano had identified themselves, Lewandowsky published a blog post purporting to expose the various bloggers (though they had already identified themselves.) Lewandowsky even back-dated his blogpost to seemingly precede the Spencer and Morano self-identifications and, in Fury, claimed to have outed all of the bloggers (even me).
Turnill’s October 2012 Blog Post
Returning to the sequence of events: in September 2012, Simon Turnill had filed an FOI request for the ethics application for Hoax, receiving the dossier on or about October 12, 2012.
Lewandowsky also received a copy of the FOI dossier and brazenly told university administator that the FOI release would show that he and ethics handler Kate Kirk had meticulously administered the amendments:
UWA is about to release a batch of my emails relating to the conspiracy paper to an individual in Sydney. I have enclosed the package, which to untrained eyes might suggest that Kate Kirk and I crossed every t and dotted every I with respect to the ethics application.
However, the reality was entirely different from Lewandowsky’s fantasy. Turnill immediately noticed the cursory approval process, describing the events in detail blog post on October 12. I covered Turnill’s first post at CA here). A couple of days later, Turnill wrote a followup post, in which he contrasted relevant sections of the National Statement (see post here) with the actual events. Turnill sent five ethics questions to UWA (at present, I don’t know whether the UWA responded.)
The October Complaint
After reading Turnill’s account of Lewandowsky’s ethics ‘amendment’, I notified Eric Eich (editor of Psychological Science) of Turnill’s findings, observing that the documents showed that Lewandowsky had not obtained “informed consent” for the Hoax survey, copying DVCR Robyn Owens:
Dear Dr Eich,
I draw your attention to the fact that Lewandowsky did not obtain informed consent from the University of Western Australia Ethics Committee for the survey reported in Lewandowsky et al 2012.
Lewandowsky had obtained consent for a study entitled “Understanding Statistical Trends” in which he proposed personal interviews on how people understood statistical trends. He then requested a “minor amendment” from an administrative officer, which was approved within 24 hours without the administrative officer apparently considering the major changes proposed by Lewandowsky.
Instead of the original proposal about “Understanding Statistical Trends”, Lewandowsky changed to a questionnaire about conspiracies. Instead of interviewing people in Perth, he changed it to an online survey. Lewandowsky said that he would leave his name off the survey request for fear of contaminating results, but the survey posts at several anti-skeptic blogs specifically refer to Lewandowsky. See http://australianclimatemadness.com/2012/10/12/lewandowsky-foi-substantial-last-minute-changes-to-project-waved-through-by-uwa-ethics-committee/
Twenty minutes after receipt of my email, someone (presumably from Owens’ office) forwarded my complaint to Lewandowsky, asking him to have the research ethics committee send Owens’ office a letter saying that “they have no qualms about your study”:
Hi Stephan- [McIntyre] is back. His latest objection is in a different class from the others, as it relates to ethical issues surrounding your survey. Please check with your university’s research ethics committee and ask them to send me a letter to show they have no qualms about your study, the way it was actually done.
Later that morning (Saturday Oct 13 9:23), Lewandowsky replied as follows:
Dear Robyn [Owens] (and Gina [Barron]):
See below for an approach to the ^ of ^ by the same individual [McIntyre] who encouraged people to contact Robyn with allegations of research misconduct when he failed to find the emails he received in his in box.
In my view, the FOI release shows that I dotted every ‘i’ and crossed every ‘t’ with respect to ethics, and I have forwarded the entire tranche to the ^. However, it would be good if he could also receive a direct statement from UWA confirming that the study was done in accordance with HR Ethics stipulations.
My understanding is that you have already responded to similar such allegations directly, so I am wondering if you might have a letter handy that you could send to [Eich]. His address is on the CC line. Alternatively I can contact HR Ethics, please advise what you prefer.
Late in the afternoon (October 13 18:53; FOIT,37), DVCR Owens replied to Lewandowsky (cc Barron; ^; Wilkin; Ferns; Dixon) that she would work with the Ethics Office to get a letter to Eich:
I’ll work with the Ethics Office to get a letter to the editor [Eich] as soon as possible next week.
In the evening (Oct 13 19:20; FOIT,37), Mark Dixon of the Ethics Office undertook to draft a response the following Monday (Oct 15):
Hi Robyn, I’ll compile the documentation on the ethics application, amendment, and approvals on Monday. If you like I can draft a response if you think this warrants a response. I can’t get TRIM from home, but I doubt that a minor amendment would have required a HREC review under the National Statement which we use as our guide for ethical review. Cheers, Mark Dixon
Dixon’s premise was, needless to say, diametrically opposed to the facts that Turnill had uncovered, which had shown both that the amendment was not only not “minor”, but an almost total transformation of the project, which additionally introduced active concealment of Lewandowsky’s association with the survey.
The next day (Sunday Oct 14 10:12; FOIT,36), Lewandowsky wrote to Dixon and Owens, co0unseling them to limit their response to the bare assertion that “the research was conducted in accordance with UWA ethics procedures”, drawing on his supposed experience with “this individual and others of his ilk”:
Hi Mark, just in case it saves you some time, I’ve enclosed the FOI release which contains all ethics correspondence for this study: see folio 13 and then folio 27 onward.
Based on my experience with this individual [McIntyre] and others of his ilk, I would be remiss if I didn’t point out that his actions are not motivated by concerns about research ethics. In particular, my experience compels me to advise against responding to his concerns either directly or indirectly with anything but the briefest note that “the research was conducted in accordance with UWA ethics procedures.” (Which it most definitely was).
I can assure you that anything beyond that (e. g., an explanation of why the amendments were appropriate) would simply provide traction for interminable further rounds of dissections, complaints, and allegations. Please do not assume that those individuals are guided by the same ethical standards to which we subscribe.
l must hasten to add that I have no intention to interfere with your intended actions, so please feel free to dismiss my comments. However, equally, I would feel irresponsible if I didn’t alert you to the nature of the situation and the characters involved as best I can.
That evening (Oct 14 20:12), Dixon wrote to Lewandowsky cc Owens, stating that he had looked at Climate Audit (“the inquirer’s blog”) and determined that I had “a private agenda and distorts and omits material where it suits his needs”:
Thanks Stephan, Much appreciated.
Whatever I produce by way of a compilation or draft will be at Robyn Owen’s discretion to use. I don’t plan on making any public statements on this outside of DVC(R) or HREC Chair. I can see from the inquirer’s blog that he has a private agenda and distorts and omits material where it suits his needs. So I am rather glad that I am not the one standing in the firing line. My congratulations, condolences and encouragement to you in that you are one such.
Cheers, Mark Dixon
Early Monday morning (Oct 15 7:36; FOIT,36), Lewandowsky replied to Dixon:
thanks, mark. I’ve gotten used to being in the firing line, although it does tend to waste time that I could otherwise spend on research. but then, that’s the whole point of their attacks. cheers steve
Dixon sent his promised comment on the ethics protocol to Owens and Lewandowsky on Monday morning (October 15 9:15):
The research protocol Stephan is publishing from, RA/4/1/4007, was assessed as having low risk in accordance with §5.1.18 through §5.1.21 of the National statement on ethical conduct in human research (2007, p. 79) and in accordance with the National Statement’s definition of risk. The National Statement does not require such research to be reviewed by an HREC [Human Research Ethics Committee]. The people assessing the protocol for ethical considerations met all the requirements of §5.1.19 in making that assessment. Similarly amendments to the protocol which did not alter the risk did not require HREC review either.
Sincerely, Dr Mark Dixon
Let’s pause and parse for a moment. The protocol originally approved in RA/4/1/4007 was for the interview of pedestrians in Perth regarding their understanding of trends. Dixon observed that the risk associated with this protocol made it eligible for ethical review by qualified personnel (rather than the full Human Research Ethics Committee) and that the people who had assessed the original protocol had been so qualified. Dixon does not consider whether the (almost total) change in protocol requires a fresh ethics application, rather than a mere amendment. Be that as it may, the salient issue is whether the amended protocol – which now included the deception/active concealment – required HREC review.
An hour later (Oct 15 10:44; FOIT, 11), Owens asked Dixon to draft a letter to (apparently) Eich:
Thanks Mark. The [^ – editor ??] is [^ -Eich ??] (copied into your email), but I think it would be good to draft a letter to him, on my letterhead and one I sign, explaining this in a way that’s a little clearer to a US audience – can you do this please? Or we can talk first?
Dixon replied 15 minutes later (Oct 15 10:59; FOIT, 11) with draft language for Eich saying that the original research and each amendment had been determined to be “low risk” and “approved according to the requirements” of the National Statement. (This was over-egging the pudding, to borrow Briffa’s phrase, since a fresh questionnaire had not been completed for the wholesale changes of August 2010. Nor did Dixon address the need to send the proposed deception/active concealment to HREC.)
RO[wens]: “it would be good to draft a letter to him, on my letterhead and one I sign, explaining this in a way that’s a little clearer to a US audience”. Happy to talk about it, although the following might be enough. I suggest:
The research protocol followed by Professor Lewendowsky and published in Lewandowsky et al 2012 was submitted for review to the University of Western Australia’s Human Research Ethics Office where it was assessed against the principles and definitions of Australia’s national guide to research involving people: The National statement on ethical conduct in human research (2007, updated 2009). Both the original research, and each amendment to it, were determined to be ‘low risk’ and approved according to the requirements of that code of practice.
Later on the same day (Monday Oct 15 14:48; FOIT, 10), Dixon sent Lewandowsky the proposed language of the letter to Eich:
Hi Stephan, I have wording for Robyn Owens to prepare a letter, as follows:
The research protocol followed by Professor Lewendowsky with results published in Lewendowsky et al … was submitted for review to the University of Western Australia’s Human Research Ethics Office where it was assessed against the principles and definitions of Australia’s national guide to research involving people: The National statement on ethical conduct in human research (2007, updated 2009). Both the original research, and each amendment to it, were determined to be “low risk” and approved according to the requirements of that code of practice.
Robyn Owens would like to include a list of those publications, because she understands there may be several and we can cover all of them with the one statement avoiding, "oh, but what about … " inquiries. What do you think? Do you have such a list you can send me?
Cheers, Mark Dixon
Half an hour later (Oct 15 15:18; FOIT, 9), Lewandowsky replied by adding the sentences bolded below, which add the claim that the University had “considered” my claims and found them to be “baseless” and that his research had been “conducted in compliance with all applicable ethical guidelines”:
The research protocol followed by Professor Lewendowsky with results published in Lewandowsky, Oberauer, & Gignac (in press, Psychological Science) was submitted for review to the University of Western Australia’s Human Research Ethics Office where it was assessed against the principles and definitions of Australia’s national guide to research involving people: The National statement on ethical conduct in human research (2007, updated 2009). Both the original research, and each amendment to it, were determined to be “low risk” and approved according to the requirements of that code of practice. We have considered the issues raised by [^- Mr. McIntyre] in his letter to the [^ – editor of Psychological Science] dated 12 October and found them to be baseless. The research reported in the above paper was conducted in compliance with all applicable ethical guidelines.
Lewandowsky added that it was important to clarify that all my objections had been “found to be baseless” as I wasn’t easily fobbed off:
Are you ok with the added sentences above in red ink? The [^ – ??? ] asked for something along those lines, presumably so he can close the door on further approaches by [^ – McIntyre?]. I believe I forwarded that individual’s letter [McIntyre] to you over the weekend, and it would be important to clarify that all his objections were found to be baseless or unsubstantiated. Otherwise he will keep gnawing away at this-as he has on another paper that wasn’t to his liking for 14 (!)years.
The identity of the person who “asked for something along those lines” was redacted. Perhaps it was Eich in an undisclosed email. Perhaps someone else. A couple of hours later in the same afternoon (Oct 15 17:07; FOIT, 9), Dixon forwarded Lewandowsky’s re-draft to Owens:
Please see the following updated text with insertions by Stephan and the rational for those.
Ten minutes later and without any further consideration or reflection (Oct 15 17:16; FOIT, 8), Owens asked her secretary to put Lewandowsky’s revision on her letterhead for signature to be sent to (presumably) Eich:
Can you put the following on my letterhead for signature, to be sent to [^ – Eich ?]. He’s at ^.
On October 16 (FOIT, 4), DVCR Owens sent (presumably) Eich the letter exactly as re-drafted by Lewandowsky.
On October 26, Eich sent me a blow-off letter quoting and highlighting the exact sentences that Lewandowsky himself had written and inserted in Owens’ response.
Dear Mr. McIntyre: I’ve received a letter from Robyn Owens (Deputy Vice-Chancellor (Research) at the University of Western Australia that relates to your most recent email to me (shown below). Dr. Owens states that “We have considered the issues raised by Mr McIntyre in his letter to the Editor of Psychological Science dated 12 October and found them to be baseless. The research reported in the above paper was conducted in compliance with all applicable ethical guidelines”
If you have a complaint about ethics, please address them to UWA. If you have a concern about Dr. Lewandowsky’s paper, please submit a Commentary after it has been published. As noted in my email to you of 25 September, I would be happy to provide more details on the Commentary submission and evaluation processes. In the meantime, I will entertain no further correspondence on this subject.
The blog posts at CA and ACM in mid-October also triggered a couple of other complaints. One correspondent wrote on Oct 16 (FOIT,2):
Professor Lewandowsky has become embroiled in just such a situation. He is accused of what can only be described as gross misconduct. He is alleged to have got approval for a certain research project, then of switching the project to something else, and then of attempting to deliberately mislead in the publication of his results. The allegations can be found well summarised on the “Climate Audit’ web site with substantial supporting information, and on many other places on the web.
I wondered if the University was aware of these allegations and whether the university deemed it necessary to investigate these allegations to either clear Professor Lewandowsky or take the necessary actions if the allegations are found to have merit. To simply leave the allegations unanswered cannot be good for the university. 1 don’t believe that the responses from Professor Lewandowsky have been adequate and that as a result it may now be time for the University to step in.
The draft response to this complaint (the final response is not in the FOI materials) asserted that Lewandowsky had complied with process requirements and falsely stated that his protocols had been “approved by the Human Ethics Committee”:
A comprehensive application process is required for human research projects. This application process includes the proposal and any research instruments, such as a survey, that are to be utilised. The project undertaken by Professor Lewandowsky followed this process and the project, including the survey, was conducted as approved by the Human Ethics Committee.
Recently, in the wake of the retraction of Fury, Graham Redfearn of desmog referred back to this earlier incident, claiming that the university had “investigated” the earlier complaint and determined that the allegations were “baseless” and that the research had been “conducted in compliance with all applicable ethical guidelines” – language which, as shown above, was written by Lewandowsky himself:
FOI documents previously released show the complaints about ethics approvals were “baseless” and that Lewandowsky had carefully followed ethics guidelines. Several sceptics called for for investigations to be carried out and complained that researchers did not have proper “ethics” approvals for the study… UWA investigated and told Psychological Science the “research … was conducted in compliance with all applicable ethical guidelines”, and that complaints on these grounds were baseless.
Hot Whopper linked to Redfearn, falsely claiming that it showed that I had “made up stuff” and making “false accusations”:
Graham also wrote about how Steven McIntyre made up stuff, including falsely accusing the scientists of not get ethics approval (which they did)
Under UWA policies, if they receive an allegation of research misconduct, the “Designated Person” (DVCR Owens) is supposed to “conduct a preliminary assessment of the allegation” to determine whether “the substance of the allegation, if proven, would amount to research misconduct; and whether a prima facie case of research misconduct exists”. The Designated Person is required to maintain records of their investigation.
UWA documents show that no such investigation took place in response to complaints by me and others about Lewandowsky’s failure to properly inform the University about the scope of changes to the Understanding Statistical Trends protocol. These changes included the introduction of deception/active concealment of skeptic bloggers, a change that could not be granted by ethics officer Kirk, but which necessitated approval by the University’s Human Research Ethics Committee.
Most remarkably, the widely-cited key conclusions of the “investigation” – “We have considered the issues raised by Mr McIntyre in his letter to the Editor of Psychological Science dated 12 October and found them to be baseless. The research reported in the above paper was conducted in compliance with all applicable ethical guidelines” – were not written by an investigator or university official but ghostwritten by Lewandowsky himself and signed by DVCR Owens within minutes of receipt from Lewandowsky.
Today’s note pertains only to the ethics approval of Hoax. The circumstances surrounding the ethics application for Fury are much worse and will be discussed separately.