97%gate and UQ

Okay – so what is going on at the University of Queensland? Not only are they threatening to sue a blogger, they’re threatening to sue if he publishes the letter threatening to sue. The letter is available online hosted on US servers.

When I first saw this story I thought this was another Lewandowsky style scandal, but a story in the Weekend Australian sets out the facts – to be fair to UQ the headline is a bit misleading.

UQ withheld only data that could identify research participants who took part in the ­research on condition of anonymity. Such conditions are not uncommon in academic ­research, and any breach of confidentiality could deter people from participating in valuable research in the future.

That is a stock standard approach to doing that sort of research – information that can be used to identify specific respondents is usually suppressed when data has been collected under the promise of anonymity. UQ is entirely correct in trying to ensure that its obligation (moral and, more importantly legal) to maintain the anonymity of survey respondents is met.

Having said that, however, UQ has a very serious problem on its hands. It looks to me that the University has lost control over important and confidential information. From UQ Policies and Procedures:

6.1 The University will make commercially reasonable efforts to protect Primary Materials and Research Data against theft, misuse, damage or loss. Researchers must take reasonable steps to keep Primary Materials and Research Data secure. Research Data must be held in appropriate facilities and adequately recorded in a register or index of research metadata to enable access to be managed as required.

That’s fair enough. Then from their own Ethics Application form:

97gate

The answers to questions 9a and 9b are particularly important. From the UQ letter:

97gate 2

Confidential information was maintained on a non-UQ server? Located in the United States? What is the answer to question 9b in the Ethics Application?

I suspect that there has been a very serious breach of UQ operating procedures around data management and maintenance of privacy. We shouldn’t be focussing so much on the lawyers letter to a blogger, but rather on what is going on inside UQ.

The other thing that interested me is the claim that the confidential information (that I agree should remain confidential) is copyright. Really? I would have thought more like trade secret. The obligation with trade secrecy, of course, is to keep your secrets secret.

The important thing is this: This is not a cover up. UQ is acting reasonably to protect the identity of survey respondents research participants. They have both a moral and legal obligation to do so. The real story here is the privacy breach that has occurred at UQ. Tough questions need to be asked and answered.

This entry was posted in Education, Global warming and climate change policy. Bookmark the permalink.

36 Responses to 97%gate and UQ

  1. Todd Myers

    Sinclair, they aren’t actually acting to supress the survey respondents (there wasn’t a survey as such). They’re actually acting to suppress the names of those who helped with the research (ie. trawling through the abstracts to determine whether a paper is supportive of AGW or not). Names that are actually thanked in the paper. I would have thought it is important to know who is making these determinations.

    As a UQ grad I’m particularly disgusted.

  2. The original stories are on WUWT.

    Shollenberger claims no ‘hacking’ was required to retrieve the data.

  3. Free Advice

    What is not reasonable is the lady lawyer claiming that her lawyers threatening letter is subject to copyright and cannot be published, and that they will sue if published.
    What nonsense! And she calls herself a lawyer.

  4. Sinclair Davidson

    Todd – that isn’t what either the Australian (linked above) or UQ has said:

    Only information that might be used to identify the individual research participants was withheld.

    This was in accordance with University ethical approval specifying that the identity of participants should remain confidential.

    That is not unreasonable. Acting in accordance with the research ethics approval forms is always good form. Losing control over the data, not so much.

  5. On this later post, Shollenberger asserts that the data, along with other graphics noted in the post, has been stored on a server with known to be publicly accessible for more than two years.

  6. Todd Myers

    That is my point, Sinc. UQ is using a straw man argument that they need to suppress the names of the participants but those same participants are mentioned in the acknowledgments of the paper itself. The follow-up post at WUWT shows pretty clearly that suppressing the names of these individuals is only suddenly of concern now, it clearly wasn’t an issue for them when the paper was published.

  7. Chris

    That is not unreasonable. Acting in accordance with the research ethics approval forms is always good form. Losing control over the data, not so much.

    It is from what I’ve heard probably a fairly common problem, it just doesn’t get exploited. Researchers in one field aren’t experts in computer/network security and they don’t have sufficient IT support. Also rather than use and maintain their own infrastructure universities are increasingly oustsourcing storage/email functionality to overseas controlled cloud based systems and there is a lot less control over who access to the data or where it (and backups) reside. It’s cheaper though!

  8. Sinclair Davidson

    Todd – you may be right, and I’m sure some clarity on that point will emerge. At the minute what UQ is doing relative to its stated position isn’t unreasonable.

  9. Washout

    Gee i wonder how the learned lawyer would consider Snowden’s actions then?

  10. Bruce of Newcastle

    I suspect that there has been a very serious breach of UQ operating procedures around data management

    The important thing is this: This is not a cover up.

    Oh yes, it is a cover up. UQ have branded themselves as überclimateer, being home of Prof Ove Hoegh-Guldberg’s manic warmists and having given SkS blogger John Cook (note not Dr John Cook) a job.

    As for the breach of procedures, the reason why Shollenberger had a look for the data online is probably because he knew these people are incompetent, having left wide open their cogitations previously. And yes, UQ/Cook was wide open again.

    Of course the contents of such things can be interesting. SkS, which is Cook’s organ, surveyed themselves as to political leanings expecting they would be a diverse bunch. They turned out to be as diverse as a pack of Socialist Advance warriors – yes, they were all leftist activists by their own admission.

    So here we have UQ defending a lefty activist, whose paper is complete crap, using methodology invented by Stalin. I think UQ could do with some humility, or purging, which given their politics seems the more appropriate.

  11. Tel

    The important thing is this: This is not a cover up. UQ is acting reasonably to protect the identity of research participants. They have both a moral and legal obligation to do so.

    The “researchers” made not the slightest effort to keep participant’s identities secret, and the people reviewing the papers happily chatted amongst themselves on blogs open to the general public. Even after they knew their material was being accessed by all and sundry, they made no effort to introduce the slightest protection.

    Turning around years later and pretending it’s all a big deal with the hush, hush and the l33t hax0r crap, is total buffoonery.

  12. Adam D

    (moral and, more importantly legal)

    There is a lot wrong with that statement

  13. Sinclair Davidson

    missing comma?

  14. Mique

    Those who have for years followed this and the similar/related Lewandowsky affairs (and to some extent even the Gleick affair) in forums like WUWT, Climate Audit and Bishop Hill, are well aware that John Cook’s SkS lacks rigour in all kinds of things. That Cook, his associates and his blog might play fast and loose with UQ’s research data security would hardly come as a surprise to us.

    The real surprise is that UQ would stoop to host, let alone defend, such dubious activities.

  15. Kaboom

    Is not the issue of the identity of the “raters” (i.e. those who perused the abstracts of the published papers, and categorised them as being supportive of CAGW theory or not?)

    Clearly, if there was statistically evident bias in the ranks of the “raters”, then this would lead to troubling concerns as to the statistical conclusions of the “raters” as a whole.

    In other words, if too many published papers were assessed by the “raters” as being supportive of CAGW theory (when in fact they were not supportive or even ambivalent), then the 97% claim turns to ice-melt and slush.

    The “raters” were all named and acknowledged by the Cook paper.

  16. For Gods sake, can someone explain this in plain English, please?
    And DoomLord, I found you’re comma. But then I hid it again.

  17. Sinclair Davidson

    can someone explain this in plain English

    The issue here isn’t the data – its to lack of proper process in maintaining data privacy.

    For everyone else – being able to link the names back their responses is the violation of the ethics approval. So the responses being in the public domain and the names being in the public domain itself isn’t necessarily a problem. Being able to connect those two separate pieces of information is the problem.

  18. Dan

    I host my confidential information in google servers which are fully encrypted and maintained 24/7 by the smartest people in the world. The idea that information is more secure in a server in the corner of the office, security patched every now and then, is quite incorrect. The spate of attacks where patient information has been stolen, encrypted and a random demanded doesn’t seem to have convinced many doctors that even offshore servers can be more secure.

  19. Kaboom

    Sinc, I think the issue is the identity of the “raters”, rather than the “responders”, or “research participants”.

    The “raters” were clearly identified by Cook et al. 2012. It is just the meta data as to the number of “rates” per “rater” that was missing, and subsequently discovered through a security lapse.

  20. Andrew

    Todd – you may be right, and I’m sure some clarity on that point will emerge. At the minute what UQ is doing relative to its stated position isn’t unreasonable.

    So basically UQ is acting reasonably, if measured against their stated position and on the assumption that everything they’ve said is true – including the stuff specifically refuted by Shollenberger (eg that he has no. Interest in the identities). Alternatively, since they’re covering up a grubby fraud who dresses in a Nazi / SS uniform to smear anyone who criticises his rubbish (including professors of the stature of Lndzen) and have been universally dishonest in every instance on this topic, I’ve chosen not to believe them.

  21. Robert Blair

    Them blokes up Norf is all a bit thick mate … Not like what us suvverners is!

  22. cynical1

    You didn’t think John Cook would not be involved did you?

    His nickname should be “John Cockup”.

    The last unsecured server hack featured photos of John dressed as a Nazi.

    Lew was probably dressed as Eva Braun…

  23. HillsHoist

    Todd and Driftforge’s posts should be heeded.

    The published paper acknowledges almost all of the raters, by name or handle, anyway. Then there’s the fact it was published under the Creative Commons Attribution 3.0 license.

    Under those two points alone UQ’s legal threats are pissing in the wind. And let’s not forget Brandon Shollenberger points out that in numerous emails Cook never once said no.

    Please read all of the WUWT, Shollenberger and McIntyre’s (Jo Nova’s post is more a summation and an opinion piece) blog posts, and comments in reply, to get a more complete picture of what’s going on. It looks like UQ are trying to feel ‘the vibe’ in this case.

  24. Joe Goodacre

    Todd appears correct – the reason UQ has given doesn’t match up with the fact that the raters identities were already exposed in the paper. From the acknowledgements of the paper…

    Acknowledgments

    Thanks to James Powell for his invaluable contribution to this analysis, Stephan Lewandowsky for his comments and to those who assisted with collecting email addresses and rating abstracts: Ari Jokimäki, Riccardo Reitano, Rob Honeycutt, Wendy Cook, Phil Scadden, Glenn Tamblyn, Anne-Marie Blackburn, John Hartz, Steve Brown, George Morrison, Alexander C Coulter, Martin B Stolpe (to name just those who are not listed as (co-)author to this paper).

    Shollenberger athis blog has indicated that he was originally conversing with Cook to find out if any copyright claims were being made by Cook. Cook then said that a member from UQ would be in contact – next day the letter came.

    McIntyre has also asked why does UQ have any claim at all given that UQ were not mentioned at all in the study (I’ve read the study myself and confirmed this) and the work was conducted by SkS volunteers, see the SkS summary below which listed that there were a dozen…

    The Team

    A team of Skeptical Science volunteers proceeded to categorize the 12,000 abstracts – the most comprehensive survey of its kind to date. Each paper was rated independently at least twice, with the identity of the other co-rater not known. A dozen team members completed most of the 24,000+ ratings. There was no funding provided for this project; all the work was performed on a purely voluntary basis.

    UQ looks grubby. I wrote an email to Jane Malloch (UQ’s legal representative who signed the letter) two days ago saying the following…

    Jane,

    It is disappointing to read about the University of Queensland threatening Brandon Shollenberger to prevent his release of analysis of data that is relevant to assessing the accuracy of claims as to a 97% consensus amongst scientists on global warming.

    David Appell is a respected ‘warmist’ commentator and if even those with a warmist persusaion conclude that the ‘legalistic heavy-handedness demonstrated by the University of Queensland to Brandon Shollenberger is…unnecessary and troubling’ (http://davidappell.blogspot.com.au/2014/05/the-university-of-queensland-letter.html) then we can be sure that your approach is way off the reservation.

    To the ordinary layman, your letter falls into the category of obstructionism and damages the public’s confidence that global warming research is being conducted without bias. The University should not be surprised to hear calls for taxpayer assistance to be directed elsewhere when this is the example of the University’s ‘objective’ approach to one of the most serious questions of our time.

    Regards,
    Joe

    Her email is below if anyone else feels like this isn’t a stellar example of Australian academia…

    [email protected]

  25. Joe Goodacre

    So to state the obvious – does UQ actually have a claim on data collected by volunteers from multiple countries, hosted off UQ servers, when no funding was provided for the project?

    Or – is Cook just hiding behind UQ’s legal skirts when he’s realised that data he didn’t want released is now out. Particularly as analysis by Shollenberger to date indicates that bias in the raters was not removed.

    UQ should be distancing itself from shoddy whatever you call this study (it isn’t science) – not clamouring for the spotlight!

  26. Sinclair Davidson

    Joe – being able to link people to their opinions and views in the data is the problem. UQ does have a problem as that is a breach of research ethics and protocol. Irrespective what anyone thinks of the research itself.

  27. Joe Goodacre

    Sinclair – understood and agreed and I retract to a limited extent my disdain of the approach that UQ have taken.

    I’m yet to see why the data would be UQ’s property given that they are given no recognition in the study, it’s published under creative commons, overseas and they provided no funding. If a claim were established, I would agree with UQ seeking the non-disclosure of raters names in doing the analysis of whether they were biased.

    This turn of events has only come around because data that should have been made available to determine bias between raters (that could have been made available without disclosing which rater was who) was not made available.

  28. Sinclair Davidson

    I’m not sure of the precise details of the paper – but UQ have provided funding to the extent that they pay salaries to researchers and provide time to do research and then provide facilities and resources to undertake that research. The research project itself would have to comply with UQ research ethics policy and procedures (some of that would be a legal obligation).

    Some of the data has been released into the public domain and some of it hasn’t. As I indicated I can’t see how its copyright – but it is meant to be private and confidential. So at least a trade secret.

  29. Joe Goodacre

    It’s only a short study so it’s easy to read yet I wouldn’t bother. The whole purpose of the study was to join the politics, which they sucessfully did when it was tweeted by Obama’s twitter handle shortly after its release.

    I wonder if failure to release data relevant to establishing the robustness of the study is in breach of the uni’s own policies.

    The SkS team themselves indicated that the study was not funded. Within the article under methodology the authors claim…

    This letter was conceived as a ‘citizen science’ project by volunteers contributing to the Skeptical Science website (www.skepticalscience.com).

    The UQ’s claim to the data actually makes that out to be a lie if Cook undertook the study as a project within his employment at the university.

    Not sure why the data would be private and confidential if the association between how the raters rated, and their name was not disclosed.

  30. JohnA

    Sinclair Davidson #1314566, posted on May 21, 2014 at 10:31 pm

    Joe – being able to link people to their opinions and views in the data is the problem. UQ does have a problem as that is a breach of research ethics and protocol. Irrespective what anyone thinks of the research itself.

    Sorry Sinc, but I followed the thread at WUWT and Mr Schollenberger was trying very hard to confirm the status of the data he had – Was it confidential? Was it subject to copyright? precisely because it was publicly available at the server to which he went. Note that he has not published the data which he has obtained. He began all this because he wanted to conduct the normal scientific exercise of replicating the research and comparing his results to the original research results.

    It seems clear that UQ procedures and internal controls have been weak, and they are scrambling to recover after the exposure by trying to bully the respondent with a super injunction – that is he can be sued if he so much as shows anyone else the legal letter he received. Theoretically he can’t even get legal advice on that letter, for fear of being sued.

    And of course UQ is asserting its copyright over the data. But that would have been fine if they had simply said “Yes we assert copyright. What do you want to do with the data? Further research – OK that’s specifically exempted in the Copyright Act”.

    So UQ and the original researchers have made a dog’s breakfast of both the storage of the original data, and of the business of allowing others to confirm their results. And now they are trying to hide the mess.

    I understand your point about the reasonableness of UQ’s actual position in regard to the data, the research and the fact that the data has escaped, but I have no sympathy for their letter, which I believe portrays internal panic.

  31. JohnA

    For “asserting copyright ” please include ‘privacy and confidentiality’ as well – it’s too cumbersome to keep writing both arms of the argument.

  32. Cato the Elder

    which I believe portrays internal panic

    Say rather profound stupidity. When the UQ letter claims copyright in the letter itself, it moves from the facile to the fatuous. Claims such as that are counter-productive – rather than anxiety or fear they provoke (and deserve) derisory laughter.

    I am delighted to see that Shollenberger has apparently called the bluff and published the letter.

  33. Sinclair Davidson

    From the methodology (emphasis added):

    Each abstract was categorized by two independent, anonymized raters.

  34. karabar

    Sinc
    The anonymity of the raters, the cover-up, and the stupid letter are all red herrings.
    At the root of the issue is the nature of the nonsense portrayed on a personal blog which clearly identifies with the University of Queensland, and now that connection is acknowledged by UQ.
    Any rational analysis of the ignorance and incompetence, the innuendo and logical fallacies portrayed on the blog, as well as the shoddy statistics associated with this paper, not to mention it irrelevance, should be an embarrassment to any institution that pretends to be one of higher learning.

  35. Nanuestalker

    Not only are they threatening to sue a blogger, they’re threatening to sue if he publishes the letter threatening to sue. The letter is available online hosted on US servers.

    That letter has all the marking of a 1st year law student not a commercial lawyer with 20yrs experience. Have you confirmed authenticity? I suggest you do Doomlord (j.malloch AT research.uq.edu.au).

  36. Boris

    precisely because it was publicly available at the server to which he went

    Which was not publicly advertised, and which Shollenberger has already unwisely described in detail that plainly reveals any reasonable person would have realized he wasn’t authorized to access. He’s wide open to a civil CFAA claim.

    There seems to be a certain amount of confusion about what “publicly available” means, especially on his part. Walking down the street checking car doors to see if they’re locked and finding one open does not make the contents “publicly available.” This is more or less exactly what he did by methodically iterating through referrer links.

    Combined with the excerpts in the C&D, I can only conclude that he’s possessed of a degree of boneheadedness that makes him his own worst enemy. He’s surprised that he heard back from an attorney informing him otherwise when he sends an E-mail stating “no obligations have been placed upon me regarding any of this material” and he’s just going to do whatever he feels like with it? He’d do better to worry about more than the foolish language claiming that the C&D was magically privileged.

Comments are closed.