- JoNova - https://joannenova.com.au -

Lewandowksy, Oberauer, Gignac – Is the paper bad enough to make history?

Stephan Lewandowsky, Gilles Gignac, Klaus Oberauer

The scathing blog posts are popping up everywhere.
From William Briggs we get a sense of the historical importance of the Lewandowsky et al effort.
One day a terrific psychological study is going to be written on the madness and mass lunacy which arose after climate change swam into the public’s ken…
The cornerstone of this future pathological report may well be the peer-reviewed Psychological Science paper “NASA faked the moon landing—Therefore (Climate) Science is a Hoax: An Anatomy of the Motivated Rejection of Science” by Stephan Lewandowsky, Klaus Oberauer, and Gilles Gignac, perhaps the completest, most representative work of its odd era.
“Everything that could have been done wrong, was done wrong. Every bias that could have been manifested, was manifested. Every fallacy pertinent to the matter at hand was made. The conclusions, regurgitated from unnecessarily complicated statistical procedures, did not follow from the evidence gathered, which itself was suspect. In its way, then, the paper is a jewel, a gift to the future, a fundamental text to how easy it is to fool oneself. “
Steve McIntyre goes through the statistical tests, finds questionable practices, questions he can’t answer, and general failure paired with incompetence. Some people wonder in the comments if there is a point to doing this when the methodology and data are flawed beyond hope. While I doubt this analysis will tell us anything about skeptics, it may reveal something about warmists, and in particular the Dept of Psychology at UWA.

Trying (Unsuccessfully) to Replicate Lewandowsky

I wasn’t able to replicate Lewandowsky’s claim at all. I got explained variance of 43.5% in the first factor(versus Lewandowky’s 86%). I notice that the explained variance for two factors was 86%: maybe Lewandowsky got mixed up between one and two factors. If so, would such an error “matter”? In Team-world, we’ve seen that even using contaminated data upside down is held not to “matter”; perhaps the same holds in Lew-world, where we’ve already seen that use of fake and even fraudulent data is held not to “matter”.

There are several instances of similar outcomes. I won’t repeat them here.  In comments Steve McIntyre finds more, and drops this gem:

Using my present best guess as to his calculation of latent variables, here is his Table 1 and my estimated correlation matrix. The two resemble one another except for conspiracy where the sign is reversed. OLS methods (of which a correlation matrix is an example) are VERY poor methods for this sort of data set. Lewandowsky may set a sort of incompetence landmark in this respect that will take many years to surpass.

It appears for all the world that Lewandowsky has replied, sort of, mentioning the SEM that McIntyre referred too, but without daring to link to him, or even mentioning Steve McIntyre’s name. Though Lewandowsky is too busy to post up all of the data he collected two years ago he has time to craft deep and insightful lines, like  …”it is easily overlooked that data analysis is also a cognitive activity.” …. O’ Really?

He-who-shall-not-be-named has replied in comments on Lewandowsky’s blog.

 stevemcintyre at 02:01 AM on 18 September, 2012

I’ve attempted to replicate the factor analysis results reported in the paper and have not been able to do so based on the information available.

Given the sketchy description of methodology in the paper, I suggest that you place the script for your results online. I’ve regularly done this and found that it both clarifies methodology for readers and adds to their interest.

Your assertion that “SEM permits computation of the error-free associations between constructs,” is a very bold statement in statistical terms and a script implementing that claim would definitely be worth sharing.

Tom Fuller, pro survey writer describes some of the flaws

Among other things Fuller describes the medicalization of dissent, a delicate topic if ever there was one, but so apt. Fuller has done over 1,000 surveys himself, and he lists five flaws, two of which I found particularly interesting:

Toodle, Lew

4. Lewandowsky allowed multiple responses from the same IP address. This means that someone could spam the survey, entering time and again to influence the results. Would they? One of the sites that linked to Lewandowsky’s survey has as part of their secret tribe of activists a person who wrote, “...people like us have to build the greatest guerilla force in human history. Now. Because time is up…Someone needs to convene a council of war of the major environmental movements, blogs, institutes etc. In a smoke filled room (OK, an incense filled room) we need a conspiracy to save humanity” and another who wrote of skeptics, “Sometimes you just want to let loose and scream about how you want to take those motherfucking arseholes, those closed-minded bigotted genocidal pieces of regurgitated dog shit and do unspeakable violence to their bodies and souls for what they are doing to the safety of what and who we all hold dear.” So, yes, they would probably do so in support of their cause.

My thoughts about this are that if you, hypothetically, wanted to find a group of people who would feel motivated to fake up a survey to make skeptics look stupid, where else would you go but Deltoid, Skeptical Science or Tamino? (Not that I’m suggesting that was his aim, I’m just putting a perspective on how poor the choice of sites was.)

5.  Lewandowsky discussed the objectives of the survey while the survey was open for responses, so those who wanted to prejudice the results knew they could do so. This alone amounts to research misconduct and is cause for throwing out the results of the survey as well as the paper based on it.

Lewandowsky’s inability to address any of these issues, despite writing a paper describing it and hyping it on a weblog with 8 blog posts in the past week, is evidence that he cannot address them. He simply decided before his research began that climate skeptics are conspiracy theorists and gamed a survey to produce the results he wanted.

Lewandowsky’s site Shaping Tomorrows World has deleted about 50 of Thomas Fullers comments.  Strange —  since we’re told the skeptics were proving the Lewandowsky hypothesis in droves, you’d think they’d want to leave all the samples of “denier” comments up for show?

Did you see That Survey in 2010? We want to know

Watts Up is looking at participation.  See The Lewandowsky participation question: for everyone who did and didn’t notice the survey two years ago. Please help out with a comment. There are 898 responses, mostly of people who didn’t see the survey.

If you saw the survey two years ago, then please also add a comment at this WATTS UP page here. So far there are only 23 responses.

Replicating the Lewandowsky survey (Your chance to answer these questions)

The survey is titled “Climate Skeptics Views Survey” and it is hosted on SurveyMonkey.com. The data collection period is currently set to run until 5 October 2012 at 3:00PM ET (UTC 20:00). I will use the Oct 6th weekend to examine the initial data and plan to publish preliminary results the following week. Final results will be published sometime thereafter (it really depends on how much free time I have to finish the analysis and paper prep).

Your blog readers may access the survey at http://www.surveymonkey.com/s/TZC6MNS or by clicking on the following to take the Climate Skeptics Views Survey. (UPDATE: Two skeptics had the same idea. This is not the same as the A Scott survey (password REPLICATE), I am talking behind the scenes to try to compile the two duplicate replicates).

You can also Vote on Watts Up: Has Dr. Stephan Lewandowsky usurped Dr. Michael Mann as the most irrationally emotive spokesman for climate alarmism?

Debate is fierce in the thread. Commenters are undecided, but Foxgoose takes a leaf from Shaping Tomorrows World. Those who have read Lewandowsky will appreciate it.

foxgoose says:
September 17, 2012 at 8:26 am

I’ve just performed an SEM latent variable analysis on the results of Anthony’s poll so far, using an advanced technique so obscure and powerful that none of you people of lesser intellect could get anywhere near understanding it.

I cannot get into the details here, but basically SEM permits computation of the error-free associations between constructs, such as one’s attitudes towards Mann and one’s Lewandowsky ideation. It is because measurement error has been reduced or eliminated, that correlations between constructs are higher in magnitude than might be suggested by the pairwise correlations between items.

It clearly shows there is a latent signal that most here believe Professor Lewandowsky to be an intellectual giant and a prince among men.
QED

 

REFERENCE:

Lewandowsky, S., Oberauer, K., & Gignac, C. E. (in press). NASA faked the moon landing—therefore (climate) science is a hoax: An anatomy of the motivated rejection of science.. Psychological Science.
————————————–

My posts on this topic:

PART I  Lewandowsky – Shows “skeptics” are nutters by asking alarmists to fill out survey

PART II  10 conspiracy theorists makes a moon landing paper for Stephan Lewandowsky (Part II) PLUS all 40 questions

PART III here Lewandowsky hopes we meant “Conspiracy” but we mean  “Incompetence”

PART IV  Steve McIntyre finds Lewandowsky’s paper is a “landmark of junk science”

PART V Lewandowsky does “science” by taunts and attempted parody instead of answering questions

PART VI Lewandowsky gets $1.7m of taxpayer funds to denigrate people who disagree with him

Part VII  Lewandowksy, Oberauer, Gignac – Is the paper bad enough to make history?

also UWA sponsors world wide junkets for poor research, inept smears: Oreskes

 

9.6 out of 10 based on 63 ratings