Are Climate Change “Skeptics” More Likely to be Conspiracy Theorists?

September 25, 2012

Are Climate Change “Skeptics” More Likely to be Conspiracy Theorists?

Posted by faustusnotes under off-topic ranting, Science | Tags: Arguing for the sheer buggery of it, controversy, factor analysis, global warming, Statistics |
[89] Comments

Introduction

In this post I will examine the data on beliefs about climate change, conspiracy theories and free market ideology collected by Stephan Lewandowsky. Lewandowsky conducted an exploratory factor analysis (EFA) followed by Structural Equation Modeling (SEM) on the assumption that established theories of conspiracy theory ideation obtained from prior research could be applied to research in online communities of skeptics and warmists. Lewandowsky collected data from those who accept the science of anthropogenic global warming (AGW), hereafter referred to as warmists, and those who reject it, hereafter referred to as skeptics. Lewandowsky constructed exploratory factors by applying factor analysis to separate sets of variables; I think exploratory factor analysis should be more data driven and, since a dataset of individuals who are active on online communities cannot be assumed to follow the same cognitive patterns as communities from which prior results on cognitive models and conspiracy theories were derived, the data should be examined without constraints on which groups of variables should be associated with factors.

In this post I will show that this simple assumption leads to significant differences in outcome, to very different results about the cognitive framework of skeptics vs. warmists, and to different conclusions about the type of communication strategies that warmists should use to persuade skeptics to change their minds.

This is a long post, but if you skip to the “conclusion” section you will also be able to read a BMJ-style “what is known already” and “what this study adds” section that may help to digest the essential points of the post (and the conclusion should be meaningful to non-statsy people).

Methods

(You can skip this if factor analysis makes your eyes bleed).

The data set was obtained using code available from Steve Mcintyre at Climate Audit. Principal Components Analysis (PCA) was used to extract eigenvectors and eigenvalues from the correlation matrix for all 32 variables, and the values of the first eigenvector (loadings) were examined for information about possible relationships between the variables on this variable. The Kaiser Criterion was applied to eigenvalues extracted from PCA to determine how many factors to retain in factor analysis. Factor analysis was conducted using a varimax rotation (the default in R) with the number of retained factors determined by this Kaiser Criterion. To be clear, this means the core analysis proceeds according to the following stages:

Use PCA to extract eigenvectors and eigenvalues
Examine the loadings of the first eigenvector for descriptive purposes, because they are usually informative
Apply the Kaiser criterion to the eigenvalues obtained from PCA: that is, the number of eigenvalues >1 will determine the number of factors to be extracted from the data
Use factor analysis to extract this number of factors, based on maximum likelihood estimation with a varimax rotation

A variable was considered to load onto a given factor in an explanatory sense if the absolute value of its loading was greater than 0.4. That is, if a variable j loads onto factor r with absolute value <0.4, that variable is not considered to be associated with that factor. The actual values of the factors (the so-called scores) were calculated based on actual loadings, so would include linear combinations of the variables not considered to load onto the factor in an explanatory sense. Some factor analysis techniques reduce the final values of the factors to a straight sum of only those variables that loaded onto the given factor, but for this article I have chosen to use the full linear combination of all variables as identified in the factor analysis. I think this is consistent with Lewandowsky’s approach to calculating factors.

Subjects were defined as warmist or skeptic on the basis of their responses to variables 7 to 10, the global warming questions. Those individuals who scored greater than or equal to 12 on the sum of these questions were considered warmist. That is, skeptics were those who refused to agree with all of questions 7 to 10. Obtained factors were then regressed against this variable to see the relationship between the factors and the AGW allegiance of the respondent.

Factors were interpreted based on their variable loadings, and further exploratory analyses conducted as necessary to explore the difference between factors obtained using this method and those of Lewandowsky.

For sensitivity analysis, factor analysis was repeated based on two possible results of visual inspection of a scree plot of eigenvalues (not shown). Differences between the values of the loadings on factor 1 were checked for all three methods (the Kaiser method and the two possible results of the visual inspection).

All analysis was conducted in R. Code with some descriptive information is linked in the appendix.

Results

(If you don’t understand factor analysis, you can skip reading most of this section. I have tried to include a layperson’s explanation, but it’s very difficult to give a layperson’s explanation of factor analysis so it may not be adequate).

(If you do want to skip the minutiae of the results, there is also a section here comparing Lewandowsky’s factors and mine, which is potentially informative).

There were 32 variables in the dataset and 1145 observations. PCA identified five eigenvalues with values greater than one, and the associated eigenvectors explained 60% of the variance in the data (which is not really very good for physical sciences, but pretty good for a psychological survey). The eigenvector of the first principal component contained large positive values (ranging from about 0.1 to 0.3) for the global warming and science variables, and negative values (ranging from about -0.1 to about -0.3) for the free market variables. That is, the first principal component measures a contrast between endorsement of global warming and other science-related variables, and endorsement of free market variables.

Based on these results, factor analysis was conducted with five factors and a varimax rotation. The values of the loadings for each variable on each factor are shown in table 1. Variables whose loading has an absolute value larger than 0.4 are shown in bold. Variables with negligible loadings after rotation are given a blank value in the table, consistent with output from R.

Table 1: Loadings for a five factor exploratory factor analysis of the Lewandowsky data

Variable	Loadings
Variable	Factor 1	Factor 2	Factor 3	Factor 4	Factor 5
FMUnresBest	0.67				0.37
FMNotEvQual	0.24	0.10
FMLimitSocial	0.54		-0.13		0.19
FMMoreImp	0.68				0.34
FMThreatEnv	0.74		-0.11		0.42
FMUnsustain	0.74		-0.10		0.39
CO2TempUp	-0.86		0.21		0.14
CO2AtmosUp	-0.92		0.16		0.14
CO2WillNegChange	-0.93		0.11
CO2HasNegChange	-0.87
CFCNowOK	0.41
AcidRainNowOK	0.49
CYNewWorldOrder	0.35	0.51	-0.19
CYSARS		0.59		0.13
CYPearlHarbor		0.63
CYAIDS		0.64	-0.22
CYMLK		0.73
CYMoon		0.42
CYArea51		0.53	-0.14	0.67
CYJFK		0.68
CY911		0.73	-0.10
CYRoswell		0.54	-0.13	0.73
CYDiana		0.65	-0.11	0.14
CYOkla	-0.13	0.41
CYClimChange	0.85	0.11	-0.14
CYCoke		0.51
CauseHIV	-0.26	-0.27	0.54
CauseSmoke	-0.24	-0.18	0.55
CauseCO2	-0.86		0.27		0.10
ConsensHIV	-0.12		0.55
ConsensSmoke			0.55
ConsensCO2	-0.69		0.27		0.11

The factors can be interpreted approximately as follows, based on those variables that load onto a factor with absolute value greater than 0.4:

Factor 1 (Free Market-AGW axis): measures conflict between endorsement of global warming and endorsement of free market ideas. Those who agree with one disagree with the other. Note that this factor has a very strong loading from the climate change conspiracy theory variable CYClimChange – this is a crucial point. Those who endorse the free market questions generally disagree with the climate change questions and see AGW theory as a conspiracy, but they do not endorse any other conspiracy theories.
Factor 2 (conspiracy theory): a factor measuring endorsement of conspiracy theories, that loads strongly onto all conspiracy theory variables except the climate change conspiracy.
Factor 3 (Cause and consensus): a variable measuring agreement with measures of cause and consensus on smoking and HIV. This measures the strength of subjects’ beliefs about HIV and smoking issues, and could probably be seen as an endorsement of broader scientific ideals. Note it is uncorrelated with both the AGW/free market axis and the conspiracy theory factor, but explains only about 4% of the variance in the data
Factor 4 (Space Aliens!): this variable measures an additional dimension of conspiracy theory, and marks out those individuals who have a really strong belief in alien conspiracy theories
Factor 5 (meaningless): this factor is a meaningless factor with no interpretation, which does not need to be included in the model but was kept because the Kaiser criterion tends to retain too many factors. We ignore it.

The variance explained by each factor is shown in table 2.

Table 2: Variance Explained by Factors

Factor	Variance Explained	Cumulative Variance
Free market- AGW axis	0.26	0.26
Conspiracy theory	0.15	0.40
Cause and Consensus	0.05	0.46
Space Aliens!	0.04	0.49
Meaningless	0.02	0.51

Comparison of Factors with Lewandowsky’s Constructs

Note that the free market-AGW axis (factor 1) essentially contrasts Lewandowsky’s free market factor and his (separately generated) climate change factor. However, it also includes the climate change conspiracy theory variable, which tends to be endorsed by those who oppose climate change theory and support free market ideals. This climate change conspiracy theory is approximately the fourth most popular (endorsed by 134 individuals) and its inclusion in the free market-AGW axis factor is important. The combination of Lewandowsky’s free market and AGW factors into one factor is important too – in Lewandowsky’s model they were correlated with each other, whereas in this model they form a single factor.

The conspiracy theory factor (factor 2) measures the strength of a respondent’s beliefs about a range of conspiracy theories except the global warming conspiracy, which does not load strongly on this factor. Note that, by design in factor analysis, this factor is uncorrelated with the free market-AGW axis factor. That is, it is unrelated to the factor that measures conflict between free market ideals and global warming theory.

Layperson’s Explanation

When you allow factor analysis to apply to all the variables, rather than applying it to predetermined groups of variables, you get a different relationship between factors to that obtained by Lewandowsky. One factor measures conflict between free market ideology and AGW theory, and one factor measures conspiracies, though just as with Lewandowsky’s work the conspiracy factor does not include AGW conspiracy theory. In fact, those who reject AGW are only more likely to endorse a single conspiracy theory: the AGW one. They are not otherwise more likely to be conspiracy theorists. Note that a simple logistic regression of “AGW rejection” against the various conspiracy theory variables might have identified this.

However, because Lewandowsky has separated the free market and AGW-rejection factors, and hasn’t included the AGW conspiracy theory variable in either of them, both of these factors are now likely to be correlated with the conspiracy theory variable. There is a correlation of -0.15 between the AGW factor and the conspiracy factor in Lewandowsky’s model, and of -0.77 between the free market ideals and AGW factors. I think these correlations work together in the SEM to give Lewandowsky’s results.

Further minor implications of the factor structure

It also appears that HIV denialism (HIVCause) and the cancer/smoking relationship (SmokeCause) are more likely to be endorsed by warmists, but only very weakly (they have weak negative loadings on factor 1). It also appears that those who endorse free market ideals are associated with the new world order conspiracy (it has a loading of 0.35 on factor 1). This is consistent with my personal experience of hanging out with anarchists (who commenter Paul tells me were defined as “left wing” at his university and who did include a few HIV denialists and had kooky ideas about smoking) and also of reading WUWT, where I have read quite a few people endorsing the new world order conspiracy. I’ve never seen a moon landing denialist, but the new world order and AGW denial conspiracies do get an airing over there. I did once see a guy deny special relativity because “you can’t square a speed, man” but I guess he was just commenting while very, very stoned.

Nonetheless, while these implications might be fun for poking fun at our political enemies, they don’t meet our loading criteria (0.4 or above) so we don’t include them in our interpretation of the final factors.

Further exploratory analysis

In this sample, 18% of respondents were found to be skeptics on the basis of their responses to questions 7 to 10. Factor 2 measures conspiracy theories. A linear regression model of factor 2 by whether subjects were skeptics found no relationship between skeptic/warmist position and the conspiracy factor. However, a linear regression of responses to the climate change conspiracy question was highly significant, with those who were warmists having an average value of 1.04 for this variable, while on average the skeptics’ value was higher by 2.76 (t statistic 45.66, p<0.001). Thus, skeptics were highly likely to endorse this conspiracy theory, but showed no significant difference on the conspiracy theory factor.

The two main factors were largely unchanged if factor analysis was conducted with only two or three factors instead of five.

Conclusion

When factor analysis was applied to the entire Lewandowsky dataset without prior assumption about the structure of the underlying constructs, only two important factors were identified. The most important factor measured the contrast between free market ideals and AGW endorsement, and the second measured endorsement of conspiracy theories in general. A fourth factor recorded space alien conspiracy theories. Thus factor 1 represented a contrast between two factors identified as separate by Lewandowsky.

Skeptics are no more likely to endorse conspiracy theories than warmists, except for the AGW conspiracy theory, which they were highly likely to endorse. Because this conspiracy theory is common, if it is included in a factor that measures conspiracy theories it will completely change the factor, and this factor will become statistically significantly different between warmists and skeptics.

Lewandowsky’s exploratory factor analysis assumed three separate factors measuring, separately, AGW endorsement, free market ideology, and conspiracy theories. This has two important implications:

It forces factors to be generated according to pre-existing conceptions of which variables load onto which factors
It does not require factors to be uncorrelated with each other

However, a more data-driven form of exploratory factor analysis that does not make prior assumptions about the structure of the data does not force this association, and leads to a completely different set of conclusions about conspiracy theories and AGW rejection. Specifically, that they are unrelated.

Factor 1, which measures the Free market-AGW axis, provides interesting confirmation of what we all know about the history and state of play of the debate over AGW. We all know that rejection of AGW theory has been driven primarily by some elements of the Republican party and free market think tanks and institutions (like Heartland). It’s therefore not unreasonable to expect that this historical development of the debate has constructed the present context, and leads to a division between free market believers and AGW believers. It’s possible to imagine an alternative universe in which AGW was rejected by unions and socialists on the grounds that it was a capitalist plot to undermine the development of the poor, and in this case we would see the opposite relationship, with those who reject free market ideals also rejecting AGW. Factor 1 in this data is a measurement of the state of play in the current debate, and gives statistical confirmation of what we all know: that those who reject AGW tend to come from a free market perspective (see e.g. Tony Watts on PBS recently whinging about taxes and regulation).

It is also unsurprising that the AGW conspiracy theory is endorsed by those who reject AGW theory. They know a lot of people and scientists agree with the theory, so obviously in rejecting it they will be likely to see it as a conspiracy.

This data also provides information for warmists who want to convince skeptics that they are wrong. Skeptic beliefs on AGW are closely linked to free market ideology, which is currently suffering stresses under the collapse of the neo-liberal global order after the global financial collapse. If warmists want to change skeptics’ minds, they need to target this nexus of free market ideology and AGW rejection, find models of response to AGW that can use free market ideals (such as carbon pricing mechanisms), more closely attack the failings of the free market ideology, point to past successes of free markets in dealing with crises, and separate the issue of the economic response from the science of the problem.

The factor analysis presented here suggests that AGW rejection is associated with a specific ideological landscape, and contrary to Lewandowsky’s findings, that it has no particular relationship with any conspiracy theory except the obvious: those respondents who thought AGW was not true were very likely to believe the science of AGW was part of a conspiracy theory or a hoax. They did not see this conspiracy theory in the same way as other conspiracy theories. Lewandowsky’s findings of a free market/ AGW rejection effect are robust, but his conspiracy findings only arise from the placement of restrictions on the initial factor selection method, and do not represent the true subtlety of skeptic ideology, which can be strongly pro-science in other areas but strongly paranoid about AGW. It is better to attack this specific thinking, which is often ideologically driven by factors external to the science, than to view skeptics as having a conspiratorial mindset in general.

What is already known on this topic

Lewandowsky’s research has shown an association between rejection of AGW theory and free market ideals. His finding that AGW skeptics also endorse conspiracy theories (or have a conspiratorial mindset) is controversial, and has been disputed on the basis of his data selection methods and data sources. His analysis has also been questioned because it assumes prior theory about cognitive models in exploratory factor analysis.

What this analysis adds

This analysis shows that making assumptions about which factors to construct generates a spurious association between conspiracy theory ideation and AGW rejection. This analysis shows that a more data-driven exploratory factor analysis does not associate conspiracy theory ideation with AGW rejection, but does confirm that those who reject AGW are likely to see AGW science as a conspiracy theory.

A note on comments about the analysis

I don’t believe that Lewandowsky’s decision to use pre-determined variables to construct factors was a deliberate attempt to smear skeptics. I think it’s a defensible decision to construct a factor set based on previous research and theory. I’m also not an expert on the underlying theory and philosophy of factor analysis, but I think the online skeptic/warmist community should be seen as sufficiently unique that it deserves its own, data-driven exploratory factor analysis. In this sense I think Lewandowsky made a mistake but I don’t think this should be read as meaning I am competent and he is not (or vice versa). It simply represents a disagreement about the framework from which to approach factor analysis. I think that this situation could potentially make a nice teaching example of the effect of forcing data to fit a pre-existing model when it is actually from a sample with a different underlying structural relationship between variables. Unfortunately, factor analysis includes a lot of art and these decisions can never be said to be wrong or right in any strict sense. So in comments here I am not interested in entertaining accusations of incompetence or deliberate manipulation of the analysis, either about me or Lewandowsky.

Appendix: Code

I don’t usually make my programming available when I report analyses on this blog, but in this case I will. It should not be necessary to read the code to reconstruct the methods as described in this post, but in case readers want to the code is available here. The code is divided into sections, which can be selected by changing the value of the variable contr.var. Set this variable to 1 to import the data, 2 to do basic pca, 3 to repeat pca on centred variables (this is completely irrelevant), 4 to do factor analysis with 2, 3 or 5 retained factors, and 5 to explore the differences between Lewandowsky’s factors and mine. In asking questions about code, please put the code indented in a comment, but first please try and read around it and read the comments – I have commented the code extensively so hopefully you can figure it out if you read and infer.

Most of this analysis was done while offline and I couldn’t check other people’s methods or results, so if you see obvious mistakes about variable choices or definitions, please tell me.

Update

It has been pointed out to me in comments that Lewandowsky excluded the AIDS and AGW conspiracy theories from his conspiracy theory factor. I’ve changed the post to reflect that, and also added a brief comparison of the correlations between his three factors and mine. It doesn’t change the overall story, but IMO makes his results more robust.

89 Responses to “Are Climate Change “Skeptics” More Likely to be Conspiracy Theorists?”

Carrie Says:

September 25, 2012 at 6:32 pm
Faustusnotes,

Thank you. This is generous and comprehensive, very accessible. Thank you for the time you have put into this..

Now to find the “translator” 🙂 for the factor analysis, she’s around here somewhere …

thanks again “a non-statsy person 🙂
faustusnotes Says:

September 25, 2012 at 8:58 pm
Carrie, I don’t think translation is easy, but try this…

If you look at the first column of table 1, you see a number next to each variable. These give you a formula for calculating the value of Factor 1 (the free market-AGW axis), which will be: 0.67*(value of FMUnresBest)+0.24*(value of FMNotEvQual)+…-0.69*(value of ConsensusCO2) for each of the 1145 subjects. Thus a subject who answers a 4 to FMUnresBest will add 0.67*4 to their score for factor 1, and so on. We do this for each person and get a total value for their score on Factor 1. Looking at table 1, you can see that Factor 1 is then composed of the positive sum of responses to free market variables, and the negative sum of responses to CO2 variables. So someone who endorses free market positions will be net positive, and who endorses CO2 positions will be negative; your putative libertarian warmist will be net 0.

The values by which each variable is multiplied are called the loadings and the resultant sum for each individual is called the score. We can interpret a factor as representing those variables that load heavily on it (I have chosen 0.4 as the cut off) and convention tells us that if a group of variables load positively and another group negatively, the factor captures a contrast.

Factor analysis provides a statistically efficient and accurate way of estimating the loadings, and by extension interpreting the factors. Debate about the best way of estimating the loadings is common. In general any method uses information on correlations between variables. If two variables are strongly positively correlated they will get a similar loading and the sign (positive or negative) of the loading will be the same. Variables with negative correlations get opposite signs. So for Factor 1 we have positive loadings on free market variables and negative loadings on CO2 variables, which tell us that the CO2 variables are correlated with each other (a person who endorses one will likely endorse the others) and negatively correlated with the free market variables (people who endorse C02 variables will refuse to endorse Free Market variables). The interesting point here is that all the conspiracy variables are correlated with each other (factor 2) but NOT with the global warming conspiracy variable. That one difference pushes the global warming conspiracy into factor 1, where it behaves similarly to the free market variables.

So then, once we have calculated a factor, we can look at its relationship with other aspects of the data. A classic technique, for example, is to see how factors differ by age, sex etc. In this case I have compared the factor by skeptic/warmist values, in order to see whether the conspiracy factor differs between those who are clearly skeptic and those who are clearly warmist. It doesn’t, but if you include the global warming conspiracy (which differs very strongly by warmist/skeptic), it does.

What Lewandowsky did was generate a single factor from all the conspiracy variables. He forced every conspiracy variable to load on that factor, and this included the climate change factor – in his model it has a loading of 0.25, while in mine it is 0.11. Thus his conspiracy variable contains a factor that is closely correlated with free market ideals and negatively correlated with rejection of AGW. Because this is a commonly endorsed variable amongst skeptics but not warmists, it will add about 0.25 to the total value of the factor for warmists but 0.75 or 1 to the value of the factor for skeptics (compared to 0.11 vs 0.33 or 0.44 for my model). The result is a large difference in the overall conspiracy factor for skeptics compared to warmists. This leads one to think that skeptics are conspiracy theorist tinfoil-hatters, but in fact it is only this one variable that is preferentially endorsed by skeptics[1].

The final question concerns how many factors to choose – Steve Mcintyre made a big fuss about this but the method Lewandowsky used, and I used, is pretty standard (you use a related method called PCA) and as he claimed for his model and I claimed for mine, the number of factors does not change the result.

Lewandowsky’s method derives from a different set of assumptions about how closely the mental models of online communities should be related to previous research. I am being data-driven, he is being theory-driven. I think that creates the whole difference …
—
fn1: actually the New world order conspiracy appears to have quite strong endorsement, but this is offset by the smoking/HIV variables amongst warmists; so it’s only the AGW conspiracy that is clearly important.
Carrie Says:

September 25, 2012 at 9:57 pm
Oh, kay, Wow! This practice of Statistics really is an “Art”?

I am grateful to you for taking the time to explain the reasoning behind the process. And to explain in such a dispassionately clear manner. Will need to reread (x10!), best done in daylight hours I think. 🙂 slainte!
Steve McIntyre Says:

September 25, 2012 at 10:28 pm
FN, if you look at Lewandowsky Table 2 footnote, he says that he excluded CYClimateChange from his conspiracy factor. Thus, I don’t think that your analysis on this issue is on point.

Second, the BIG issue in the Lewandowsky survey is fraudulent responses – an issue that you don’t touch here. In my opinion, there are a number of lines of evidence that numerous Lewandowsky responses were fraudulent. I’ve discussed some of these at CA, but there are others that I’m getting too. One of the main lines of fraud detection is that numerous Lewandowsky respondents evidence an extreme caricature of WUWT opinion (sort of like Gleick’s forgery characterized by Megan McArdle of Atlantic as a Bat Cave villain) that is not actually present or very rare in the recent WUWT survey – not just in the conspiracy items, but in the FM items.

Third, many correlations within the correlation matrix are entirely or substantially impacted by severe outlier (fake) responses – a point that I analysed closely at CA for one important case. You don’t touch on this.

Fourth, in my comments on principal components and factor analysis, far from making a “big fuss” about them, I indicated that I didn’t think that this issue mattered for Lewandowsky’s downstream analysis. It was Lewandowsky who made a “big fuss” about a minor point in my commentary on an issue that was very secondary to the larger question of fraudulent responses.

Fifth, although I haven’t touched on this point in commentary at CA yet, while PCA/factor analyses are popular techniques, both assume gaussianity/normality. The Lewandowsky data severely violates this assumption. For example, a number of the more wacky conspiracies and the CauseHIV/CauseSmoke items only have a few dissenting respondents. IMO, these respondents are nearly all fake. But separately, the distribution is almost nilpotent. If one were to apply PCA to this sort of distribution, one would have to experiment with versions of sparse non-negative PCA with binomial or poisson distributions. I don’t know what such techniques would yield – but they don’t deal with the basic problem: fraudulent responses.
Carrick Says:

September 25, 2012 at 11:20 pm
Steve McIntyre, almsot any data set has outliers. The question is what influence these have on the measurement. I agree it’s possible that all skeptical responses were “fakes”, but that seems unlikely.

I think an interesting thing to do would be to see faustus’ method applied to the WUWT poll.

Sou:

I prefer not to comment on the same site as McI. I do not want to be associated with his baseless and unethical allegations of professional incompetence.

Hm…

You seem to have no problems posting on Lewandowsky’s blog where similar accusations of McIntyre are made. You also frequent SkS, where they do nice little running commentaries such as “Christy’s Crocks’.

Perhaps you’d care to rephrase that so you don’t look completely hypocritical?
faustusnotes Says:

September 25, 2012 at 11:36 pm
Steve, if he excluded CYClimateChange but didn’t include it in the other two factors that could have the same effect. In my FA it’s included in the Free market-AGW axis – he doesn’t have this factor at all.

I don’t touch on the fraudulent responses issue because I think it’s just not true. There are 3 responses in this dataset who are strong skeptics but endorse the AGW conspiracy theory – they could be interpreted as fake but they’re not, they’re just dumb respondents. The reason you don’t see the same effect in the WUWT survey is because a) the presence of a neutral option skews the answers and b) people who responded to that survey were very carefully watching their responses to make them as moderate as possible.

When dealing with psychological data it’s extremely unwise to infer fake data from outrageous responses. I’ve done surveys of transgender sex workers, 5 of 72 claimed they had been gang-raped by police; I’ve seen drug users report use of enough heroin to kill a horse, but their self-report has been confirmed repeatedly by observation; I’ve seen medical records as thick as a dictionary, full of the kinds of insane tales of illness, disaster and chaos that would make you think someone was writing a novel. The world is full of people with freakish views, and the fact that someone endorses an extreme ideology doesn’t make them a faker, it just makes them part of the rich tapestry. How would the “legitimate rape” Republicans respond in a questionnaire about feminism? Like a caricature of themselves, I don’t doubt. My two favourite journal articles tell tales of human strangeness that would, if manifested on a survey being “audited” by you, appear to be fake data. Who fucks a tractor? That must be fake!

Yes, the Gaussian data issue is a problem, and one I alluded to in comments here before. Bear in mind though that with large samples a lot of these problems go away. I don’t know if you’ve ever tried factor analysis on binary data – I don’t recommend it, but if you have a large sample and you do factor analysis with tetrachoric correlations, you’ll end up getting similar answers to an assumption of multivariate normality. I think these data should be measured as points on a line, assigned freely by the respondent, but psychological studies usually use Liekert scales. In large enough samples, that tends to wash out. There is a bigger problem (imo) of non-linearity that no amount of data will get rid of, but if you want you could exclude the extrema and repeat my analysis, see what you get. I’m wiling to bet you a groat that your results don’t change at all.

I thin you will achieve more by redoing the analysis the way you think is most appropriate than by making claims of fraud and fakery. They can’t be proven, and they just lead to the impression that you’re trying to poison the well.
Leo G Says:

September 25, 2012 at 11:39 pm
Faust, I don’t think that there is a conspiracy, I just think that things are not going to go to hell in a handbasket. Is there a a way in the survey that would show this, or would I be lumped into the “Climate Science is a conspiracy” group?
faustusnotes Says:

September 25, 2012 at 11:49 pm
Leo G: quite a few people in the survey endorsed “moderate” views of AGW skepticism/warmism, and didn’t endorse the conspiracy view – I think only about 2/3 of those I classified as skeptics endorsed the conspiracy view and only half of them gave the extreme version. So Lewandowsky’s survey would have allowed for even someone as nuanced as you!

Carrick, sou doesn’t want to be commenting on a thread where Mcintyre also comments, which is not the same as being seen on a thread about Mcintyre.

I don’t think these data points are necessarily outliers, and they aren’t necessarily scammed. Francisco was posting on my last thread with a fairly detailed and extravagant climate change conspiracy, and I’ve seen quite a few new world order conspiracy theorists over at WUWT (as well, as I mentioned, as one guy who thought special relativity was a crock). I remember years ago a guy regularly writing letters to a local newspaper in my town, claiming he was being refused a position in the physics dept of our uni because they didn’t believe his theories of perpetual motion. And if you want to see the full whack of conspiracies in one over-the-top, thread-derailing package, google “Graeme Bird,” a name so infamous in the Australian blogosphere that it is rumoured to be dangerous to the cohesion of a thread just to write it (and yes, he thinks AGW is a scam and he doesn’t believe in evolution). You can rest assured that if he answered Lewandowsky’s questionnaire he would be all 1s and 4s, and he definitely cruises skeptic blogs. You can find a whole range of those people haunting deltoid – last month’s thread got up to 700 comments due to the efforts of a single outrageous commenter. Deltoid used to have threads devoted to a single commenter to ensure that he didn’t derail the rest – and he had a wide and impressive range of kooky theories. Check out Pandagon for an excellent array of right-wing conspiracy theorists, or Counterpunch for the equivalent on the left. The anti-animal research movement has some fruity folks who would score a lot of 4s on Lewandowsky’s survey, but they don’t use the internet…
faustusnotes Says:

September 25, 2012 at 11:52 pm
I have a cold and it’s late here, so I’m off to bed … I’ll endeavour to reply to any further comments in the morning …
Leo G Says:

September 26, 2012 at 12:19 am
Thanx Faust. Amazing to me that 2/3 of respondants think that AGW is a conspiracy! But to be honest, in my work-a-day life, very few of the people that I work for/with even mention AGW. In my world view, it is really not that big of a thought for most people. Yes they will speak to it in generalized “read” somewhere terms, etc.

I think a very good survey would be to go through say 10 warmist and 10 non-warmist blogs and calculate how many of the same people contribute comments, compared to the general population. Pretty sure that the fraction would have to many zeroes on the right side of the point to be of any help.

To me, this whole blog AGW thing is a very small closed population that keep the wheels spinning. If all the blogs were to cease tomorrow, I believe that it wouldn’t really be registered at all in the main population.

But it is a guilty pleasure of mine for sure!
Paul Matthews Says:

September 26, 2012 at 12:29 am
Thanks for this – I think most skeptics would be quite happy to accept those conclusions.

I don’t think you’ve really answered the question about faked responses. The most obviously faked ones are those that tick strongly agree with virtually all the conspiracy theories, no matter how ridiculous, and identify as skeptics. These are lines 861, 890 and probably 964 in the spreadsheet (removing the header, these are responses number 860, 889 and 963). Do you really think these are genuine? What would change if you took those out?

Similarly Steve’s first point needs a bit more consideration. You seem to say in the Layperson bit and the conclusion that Lewandowsky lumped the AGW conspiracy with the others, but he says he didn’t.
Carrick Says:

September 26, 2012 at 12:35 am
faustus:

Carrick, sou doesn’t want to be commenting on a thread where Mcintyre also comments, which is not the same as being seen on a thread about Mcintyre.

Well I can understand that part (people get to select who they hang with). It’s the I do not want to be associated with his baseless and unethical allegations of professional incompetence that was raising a red flag. I guess if you parse that correctly he doesn’t mind being associated with “baseless and unethical allegations of professional incompetence”, just not ones from McIntyre. 😉 That’s a matter of taste.

[That said, McIntyre did post on threads on STW that sou also posted on . People are strange.]

Anyway to substance, thank you for your lucid description of the methodology. I hope that short-circuits some of the non-constructive back and forth.

The FM part isn’t a surprise to me nor the “AGW is a hoax” part on some skeptics. I think this part of Lewandowsky’s paper is publishable work. That’s a comment I’ve made before seeing this reworking of his paper, though it seems like what you’ve done is an improvement over his initial work.

It would have been interesting to see more questions that explored warmer viewpoints, such as “there is a fossil fuel conspiracy to suppress the science of global warming” and stuff like that. I’m sure you’d find a nice symmetry there in belief structures.

It appears that most people pick “truth” based on which group they want to be associated with, and that drives which ideological constructs they accept as fact. I’m pretty sure that’s not a new thought either.
None Says:

September 26, 2012 at 1:33 am
Pretty much confirming prior “skeptics” analysis i’d say.

“We all know that rejection of AGW theory has been driven primarily by some elements of the Republican party and free market think tanks and institutions (like Heartland)”

I think you think you know that, but that its a convenient opinion for your political tendencies rather than a fact. I’d say that rejection of AGW (or at least as most skeptics would say, rejection of the studies forecasting runaway extreme and hazardous global warming, and a relatively stable temperature over the past 1-2k years) has been driven primarily by an outing of the appaulingly bad quality of scientific papers upon which the constructions and forecasts have been made.
Steve McIntyre Says:

September 26, 2012 at 1:55 am
“Yes, the Gaussian data issue is a problem, and one I alluded to in comments here before. Bear in mind though that with large samples a lot of these problems go away.”

But not in this case. In the controversial items, CYMoon etc., there are not “large samples” of people holding the views. There are only a few. The major component of the negative correlation between CYMoon and CauseHIV, which Lewandowsky asserted to be strongly significant, were the few questionable outliers, especially the zealot two.

Because there are so few respondents who dissent from the obvious position in these extreme conspiracies, the correlations end up being dominated by the ones who do. It is a recipe for non-robust responses.

In statistical terms, it’s actually a bit interesting because the data departs so profoundly from the assumptions of PCA or factor analysis.

I took a quick look at variants of sparse and robust PCA. I had trouble getting some robust methods off the ground since a robust standard deviation failed on questions like CYMoon because nearly everyone had the same answer. You really need to think hard about what PCA actually means in such circumstances.
Steven Mosher Says:

September 26, 2012 at 2:59 am
excellant. and code too. superb.
On a side note my sense is the internet seems to atract
and retain conspiracy minded folks.
present company excepted.
A.Scott Says:

September 26, 2012 at 6:36 am
I also offer an “excellent” comment on this. While I don’t agree with faustusnotes point of view on the fake responses, that he undertook this effort is a good contribution to the overall discussion.

As to the false, or falsified, responses, again I disagree – after a lot of review of the Lew. and other data, these fake replies standout like a beacon. They even standout against the respondents who answer extremely that you can tell actually seem to believe their answers. In my opinion there are; enough strong “tell” questions, and clear “patterns” which strongly identify the clearly false responses from even the outlying actual believers.

Regardless – even if you include the false data you have a total of 6 Strongly Agree and 4 Agree responses to the CYMoon question. Ten total responses against N=1145 total, and N=>200 “skeptic” responses. I cannot see how anyone can make any meaningful statistical analysis, let alone find legit correlation, with only 10 responses, especially considering those 10 responses are far outliers, whether false or not. 1,135 respondents answered opposite these 10, and most strongly opposite.

To try and draw any meaningful inference from these 10 scammers or outright kooks – let alone the sensationalized inference in the title – is simply ridiculous.

With the authors strong advocacy positions, and other work on things like how to “debias” people from the “misinformation” regarding climate change, it is pretty clear this is advocacy unsupported by science.
DGH Says:

September 26, 2012 at 8:39 am
faustusnotes

Very impressed by your conversion from flame thrower to devil’s advocate. I suspect you’ve gained some respect for SM (and he for you). I hope to read more of your comments on his blog moving forward.

You’ve made it clear that you don’t want to host a conversation of motives, etc, so I will tread very lightly. I won’t complain if you use the snip. But I have questions about the methodology that I hope you’ll entertain…

It’s difficult to understand your belief that the extremist answers were authentic.

When you conducted the survey of transgender sex workers did you do that at a fundamentalist Christian church? If you had would you have expected reliable results? Do you think you might have gotten more authentic self identified transgender responses or more folks posing as transgender?

I know that there are many flaws in that analogy. But I’m sure you get the point. And I would appreciate a further comment on how the decision to collect data about skeptic attitudes on science from non-skeptic blogs might influence the results.

It would also be interesting to hear how you feel about the fact that the authors have not published or even discussed the data that they collected from websites that were not climate related blogs. It is known that the survey was posted at 3 other sites. One of those was connected to UWA – clearly a problem to put that in the general data. But the other two were professional psychology websites which were apparently included at the outset but later discarded when the focus became blogs.

Thanks.
Lady in Red Says:

September 26, 2012 at 9:34 am
I am impressed.

It is interesting the communication that can take place when the mask is dropped, the sarcasm eschewed.

I’ve read the comments at Lewandowsky’s site, the hurt, the bitterness, the sense that Steve McIntyre seeks to destroy. Nah. SM just looks for good stuff, to get it right…. to archive the data and allow others to poke at it.

The mainstream climate “science” community has become so infiltrated with “noble cause corruption” and, sadly, too many too ill-educated to be in the positions they hold, one needs the likes of Watts and McIntyre (and Jeff Id and the guy at the furniture factory in Michigan who took out Eric Steig…) because truth and science matter.

The dog and cat may, still, be circling each other here. But, there is a respect I’ve not detected before.

Thank you, Faustus, for retracting your claws. The cat touch paw may still be tentative, but there is outreach. Congratulations.
….Lady in Red
faustusnotes Says:

September 26, 2012 at 10:37 am
Thanks for your comments everyone. I can’t agree that this is an improvement over Lewandowsky’s work (though of course I did it so I want to think that), or that I’ve switched to devil’s advocacy, or that this implies I think any more highly of Steve’s project, my opinions of which are abundantly clear.

This evening when I have time I will amend some parts of the post to reflect that Lewandowsky didn’t lump the AGW conspiracy in with the rest.

Regarding fake responses, I think Steve is showing a touching naivete about the political and economic opinions of the majority of his fellow travellers. This survey was hosted on Deltoid, right? I know of at least two conspiracy theorists who have frequented Deltoid (the aforementioned and infamous Graeme Bird and another, I think JC, who was regularly banned). Lewandowsky’s blog linked to an AGW skeptic who was a conspiracy theorist extraordinaire – that’s three individuals who, if they responded to the survey, would definitely be considered “fake” responses. As for the free market dimension – the Australian blog Catallaxy is stuffed to the gunnels with AGW skeptics who would easily answer 4s on all the economic variables. If you read the Australian Libertarian party policy positions, they want to abolish all speed limits in school zones and privatize the issuing of money. The same is true of the lunar crazies over at Reason. I think it’s quite easy for that mob – who are almost all AGW skeptics of one form or another – to stack the deck of any survey they answer. There are critics of AGW in the UK Tory party who want to abolish all government controls in the economy and return to the gold standard. I don’t think you have to scratch the political right – and especially the internet political right – very deeply at all to find a very ruthless free market ideology. So no, I don’t think those responses prove by their own extremeness that they are faked.

Addressing more specific points…

Steve, I’ll come back to the four point scale thing later. It’s a big problem I have with psychological research in general, but there’s this overall problem in science: someone designs a bad method, people adopt it early, it becomes the accepted practice, after that you can’t easily get your ideas published without using it. I think Liekert scales need to be revised, but psychologists don’t seem to agree with me. Thus, until they do, we have to accept some degree of weakness in the instrument. I think a lot of the problems you and others are identifying arise from being used to working in a field where you can control the sampling process, in general you can get any data you need through the right sample design, and responses follow strict physical processes which mean that outliers can be confidently claimed to be machine errors or fraud. Sadly, we can’t take ice cores of people’s minds and if we did we really wouldn’t like what we saw. As a result psychological data is difficult to measure and extremely unreliable in general. Just consider the different meaning of “strongly agree.” e.g. if your girlfriend/boyfriend says “I think we should have pasta for lunch” and you endorse “strongly agree,” this can have a very different meaning to endorsing “strongly agree” that Hitler did bad things. Yet they are worded the same way in a pscyhology questionnaire. What can you do to handle this? People are weird, and assigning measurements to their opinions is a difficult and ultimately not particularly fruitful exercise, but people who’re used to thermometers and ice cores sometimes forget this. I think when Lewandowsky wrote his paper he was probably bearing all these issues in mind, as would be the people reading it – but for outsiders used to another field of science it appears a lot more inflammatory because they don’t have those unwritten caveats.

I’ll also address the online survey thing in a post tonight. This is another tree I think is not for barking up.

I don’t want to comment about Lewandowsky’s use of UWA-connected data, DGH, because any such comment inevitably requires supposition about motives (one way or another) and that I am not going to do.

LeoG, I think you’re suggesting a capture-recapture analysis to estimate the size of this community. Excellent plan.

A. Scott, your analogous response to my transgender example doesn’t work because warmist sites aren’t a fundamentalist church – lots of atheists wander in and yell at the preacher on a regular basis (to extend your analogy). They even get to dispute scriptures and read their own. So if one were to put an anonymous survey on the church wall, some atheists would no doubt fill it in. It would be nice if more skeptics had answered the survey but you really really can’t blame Lewandowsky for that, or for pursuing his analysis despite their non-cooperation. And I think my analysis shows that the data can be interesting regardless, and ultimately doesn’t disagree markedly with Lewandowsky’s.

Steve Mosher I don’t usually show code for my analyses – I think one should be able to reconstruct everything from the methods section. I did this only to forestall the inevitable demands for it (and I’m sorry it can only go up in R format on this blog). I have done other analyses of military casualties in Afghanistan, the cost-effectiveness of using magical healing to reduce infant mortality, Australian home fires after insulation installation and, of course, statistical proof Game of Thrones is misogynist where I don’t show code. I assume it will be a one-off. However, in this case as well you will note htat my post is already up to about 3100 words without a proper introduction. Were I to try and get this into a major journal I would need to shave off probably 500 – 1000 words, and it’s likely the easiest way to do that would be to put the majority of methods into an online supplement, and/or assume the reader can infer much of my methods from domain-specific knowledge. Steve Mcintyre’s lack of the latter and impatience about the former has tripped him up a few times, and so I do think it’s best to wait, and not to overreact to a single psychology article that suggests some of us are batshit crazy.

It’s the internet man, … we’re all batshit crazy!
Kevin R. Lohse Says:

September 26, 2012 at 2:32 pm
What a heroic effort to stick to the science and ignore the politics! Lots of information with a minimum of snark made it a good read, espec. the layman’s section. This sceptic will be back.
Paul Says:

September 26, 2012 at 4:14 pm
This is my second favourite mathematical analysis paper ever. Coming right after your studies of Dungeons and Dragon’s fighter stat distribution on PC survival rates. Four and a half stars. 🙂

Also, I strongly endorse every mathematical factor analysis containing the disclaimer ”(You can skip this if factor analysis makes your eyes bleed).”

Bonus thought: For those who feel that some of the data is made up, how much stronger a result do you want than “Skeptics don’t tend to be conspiracy theorists (AGW conspiracy aside)”? It’s not like deleting those three extreme samples are going to get a result other than “Still not correlated”.
faustusnotes Says:

September 26, 2012 at 4:21 pm
Typical, judging the value of the paper from its topic rather than the quality of the methods used! Just because D&D fighter stats are a more important issue doesn’t mean you should downgrade the rating of the more technically advanced paper!
DaveA Says:

September 26, 2012 at 10:20 pm
Another thanks for the effort, it’s well done – even if I don’t understand all of it.

And a comment. Faustus you make a point of how freaky some people are, and relate this to extreme responses as an argument against fakes.

You’re willing to accept that some people (or one) will romance a tractor, but you’re reluctant to admit that out of 1000+ anonymous respondents to an online survey, that some will be fake. Even when the audience offered the survey consists of people we would expect might be tempted to offer fake responses (making skeptics look nutty).

Who’s more freaky, the tractor guy or the fake survey respondent guy?
faustusnotes Says:

September 26, 2012 at 10:45 pm
Fake survey guy, obviously. Who hasn’t fucked a tractor?

More seriously, though, the faking is likely to end up just noise, and it won’t be that common – people will fake in both directions, so it’ll wash out. Also, people fake their own responses too. This is part of the sick game that is psychology research, and there’s not much that we can do about it.

Also, what’s more common – a libertarian anti-AGW bastard who is desperate to have his voice heard by researchers, or a sick internet puppy who wants to fake being a libertarian anti-AGW bastard for shits and giggles? I think we can only answer this question through a very large community survey funded by the government.

Also, if there’s a part of the analysis you want explained, please do ask.
Carrick Says:

September 26, 2012 at 10:50 pm
Just a few comments, then I’m out of here. First on the topic of statistics in RPGs, Roughly 1:10 in this video is one of my favorite statistical analyses (it obviously didn’t include outliers like stoner heads).

Regarding sampling…it seems to me you’re pretty cavalier about the abnormal group within your population, and its ramifications on the use and interpretation of PCA. If the samples don’t belong to the population, including them is an error, and you would have to verify that including them did not substantially distort your results.

If they belong to a “real” second population, then you still have real a problem, because as you admit yourself they are not typical samples of one of the populations that the survey intended to study (skeptics). Either way distortion of the resultss by outlying responses is a big problem here.

Next, while I agree in principle with your comments about being able to reproduce the results using the descriptions in the method section, that assumes your code is not bug free. It also assumes the person who wrote that section understands what the person actually did who performed the analysis (they are not always the same person and the Leroy Jenkins effect definitely applies to grad students).

The recent debacle of the withdrawn Gergis paper is an example of that. Had the code been included, it’s at least possible that the error would have been caught in the review process. If a paper is important it will get vetted after the fact, and Gergis should be happy that the error wasn’t reported to her in a very highly visible venue like AGU. Sometimes it’s politically hard to recover from mistakes like that, especially as a young professor.

Simply describing what you set out to do in principle isn’t going to protect you in the event of an error. So it’s in your best interest to include the code you actually used to analyze the data. In 99% of cases, nobody outside of the reviewers would look at it, in about 10% of the cases a reviewer would.

Finally validation issues…

By using PCA to pull patterns from data that you then use as factors for a factor analysis, because of the aberrant nature of the dataset you have, you are in danger of (metaphorically) looking at patterns in clouds and divining meaning from them. While Kaiser test is supposed to help prevent this, given the methodology of this survey, I suspect it is not a particularly robust criterion here (the interpretation of the values of the eigenvalues, even possibly the ranking of the eigenvalues can get mixed up by outliers). As you are probably aware Fabrigar 1999 heavily criticizes this criterion and argues it shouldn’t even be used (see text page 278 second column).

Just because the Kaiser test “blesses” a particular PCA doesn’t automatically mean it is really a sensible choice for your basis set. You might criticize McIntyre over his language (in which case your criticism is very asymmetrically applied, but no matter), but at least in his area, if he makes a mistake, he’ll soon find out the hard way (there is no gold in them there hills). How do you drill for oil here? How do you know when you’ve hit a dry well?
Carrick Says:

September 26, 2012 at 10:54 pm
Arg … Make that “Just because the Kaiser test ‘blesses’ a particular PC” .
DGH Says:

September 26, 2012 at 11:00 pm
Faustusnotes-

You’ve taken a fair amount of flak on this thing – but then again you put yourself in the line of fire…

Is it fair to point out that you assume a great deal about the skeptic side of the debate based on the extreme opinions of a minority. Then you refuse to discuss the merits of Dr. L’s paper because at the outset and the end you agree with the results. I think fellows in your field have a term for that. Confirmation bias much?

IMO This is the same mistake that got Dr. Lewandowsky into this mess of a survey. And by the way the same mistake (to a different degree) that got Heartland in trouble with their stupid Unabomber billboard.

If you allow extremist behavior to color your opinion of the majority of the participants on the other side of the debate you’ll necessarily end up talking past them and they past you. Welcome to the climate blog wars.

As an example, you clearly heard the part of Anthony Watts interview where he objected to taxes. DId you hear the part where he talked about his solar panels? DId you know he drives an electric car? Did you notice that he isn’t living in an Exxon Mobil funded palace?

My intent isn’t to defend everything about WUWT. He’s got some wacky stuff on there and he makes his share of mistakes. That said, his surfacestations project was interesting and made a contribution – even if his belief that bad siting had a significant positive influence on the U.S. temperature record doesn’t ever pan out.
faustusnotes Says:

September 26, 2012 at 11:14 pm
Carrick, I’m assuming that Lewandowsky’s purpose is to analyze online skeptics (see my next post), i.e. not everyone with a skeptical frame of mind. That’s an important difference, and one you might not agree with. If you think his purpose was to give a general overview of cognitive processes among all skeptical people, then you’ll obviously have objections to an internet survey; me, I’m not so sure that is what he was trying to do. I’m happy to differ about that, because this is a single paper in a psych journal.

re: code and the like, I think many journals don’t even provide an avenue for publishing code. Make of that what you will. As a result, it’s better if your work can be reproduced through the methods. Also, stats is an art, not a science, and different statisticians will do things differently anyway.

I don’t think the Kaiser criterion is best: I used it here because it was simple. Also, I did sensitivity analysis, which I think is important for this stuff. I think Lewandowsky may have too, though it’s in his online supplement or he may, like me, just say “our results didn’t change when…”

DGH, I’m happy to discuss the merits of Lewandowsky’s paper but I refuse to ascribe motives to him or to his research subjects. The latter is unethical and the former is unprofessional. I think if Mcintyre had eschewed both of those practices he wouldn’t have tripped up in his factor analyses because Lewandowsky would have been more helpful. That is also why my language at Mcintyre’s place was aggressive – I don’t think his project of auditing is helpful.

I was aware of Watts’s solar panels thing, &c, and I think it’s irrelevant (I always think people’s personal lives are irrelevant). The important thing is his manipulation of the scientific debate, and yes I agree that his surface stations project has been probably an interesting (if minor) addition to the scientific literature (though I know very little about it). However, he started doing his science late in the game and is one of very few who have done so, and much of it is disingenuous. It’s too little too late to look like anything except a grand exercise in concern trolling. But shall we try and keep this comment thread on topic, lest it become just a monumental troll-fest?
Lady in Red Says:

September 27, 2012 at 1:00 am
Well, something, Mr. Faust, that someone is going to have to deal with at some point is the corruption of “climate” “science.”

You write:

“Also, what’s more common – a libertarian anti-AGW bastard who is desperate to have his voice heard by researchers, or a sick internet puppy who wants to fake being a libertarian anti-AGW bastard for shits and giggles?”

I just want science, plain ole archived data, repeatable science. We can decide about the future of the world when we get that right.

I think that’s all McIntyre and Watts and Jeff Id and Judith Curry and all the rest are about, as well. Science.

Sadly, “climate” “science” is not attracting the best and the brightest these days, but, rather, the most gullible. Noble cause corruption tree-huggers and lovers of polar bear cubs. It’s cute, but it ain’t science.

These losers would have trouble changing a tire or getting a job as a waiter. They slept through most of Statistics 101 and they know nothing about physics. But, they believe!

If *you* could help get science back on track, you would see all the skeptics and “deniers” fade into the woodwork.

Sadly (but fun for me to watch), is that the really smart folk are outside the lock-step, “science is settled” compound. And, those left inside — like Michael Mann — are soooo dumb it’s kinda like shooting fish in a barrel.

(….which would be sad, I think, were it not for the overwhelming arrogance and pretension.)

Have you seen Stephan Lewandowsky’s YouTube videos? They are/should be parodies. They are an idiot professor. Instead of embarrassing shame, Lewandowsky is proud of them. He hasn’t a clue what a caricature he is.

Didn’t he have any professional sense to know that, if it made sense to do this survey at all, that he was far too steeped in prior bias to conduct it with any degree of objectivity?

So. If the folk who believe in science come out and stand up and we can revert back to some simple rules which should dictate in all science, you will find the skeptics fading.

As it is, they are merely the smart ones, the independent thinkers.

…Lady in Red
Paul Matthews Says:

September 27, 2012 at 2:05 am
“I’m assuming that Lewandowsky’s purpose is to analyze online skeptics”

That’s a dodgy assumption. He sent his survey to 8 of his skeptic-abusing pals (“pro-science blogs” on planet Lewandowsky). Then several days later he got his helper to email a different version of the survey to 5 other people. One of these (not a climate skeptic) wrote back questioning the validity of the survey. If Lewandowsky had genuinely wanted to analyze online skeptics, he could have written to more skeptic bloggers, or he could have just posted the link to the survey as a blog comment. I won’t spell out my assumption about Lewandowsky’s purpose as it might break your blog policies.

Anyway, kudos points to you for
(a) writing a clear and fair post
(b) modifying it after comments
(c) saying “It’s the internet man, … we’re all batshit crazy!”
Carrick Says:

September 27, 2012 at 3:22 am
faustus: <blockquoteI think many journals don’t even provide an avenue for publishing code.

That must be a difference in fields. Every journal I publish in makes provisions for an SI, as I understand it, even the one Lewandowsky is publishing in does.
Jean Demesure Says:

September 27, 2012 at 5:47 am
“Then several days later he got his helper to email a different version of the survey to 5 other people.”

To be fair to Lewandowsky, you should have said :
“then several weeks later, after he had the conclusion he wanted for the next day seminar, he got his helper email a different survey to 5 other people.”
A.Scott Says:

September 27, 2012 at 7:38 am
More seriously, though, the faking is likely to end up just noise, and it won’t be that common – people will fake in both directions, so it’ll wash out. Also, people fake their own responses too. This is part of the sick game that is psychology research, and there’s not much that we can do about it.

Accurate as to overall conclusions and correlations.

Highly inaccurate when looking at individual claims. When you only have 10 affirmative responses (6 Strongly Agree and 4 Agree) as with Moon Landing is a Hoax, then even a couple fake responses, are not noises, and clearly have a huge effect.

Lewandowsky tries to play the old “misdirection” magicians trick. Instead of answering the direct question about scientific support for the ridiculous and sensationalized title claim, he does what you do here – he says its just noise, even if we take it out the bigger conclusions still hold (which is another issue altogether).

It is an accurate response – and one I agree with – but it isn’t the answer to the specific question core to all this. His silly claim attempt to smear skeptics by associating them with the Moon Landing Hoax – which was wholly unsupported by any credible or legitimate scientific analysis.
mrsean2k Says:

September 27, 2012 at 7:47 am
The idea that a description of your method without the code used to realise that method is sufficient may or may not be standard practice is irrelevant to it actually being sufficient.

If I “replicate” based on your method only, and get different results, how is that resolved? How *should* it be resolved?

There may be situations where it is onerous or impractical to provide source; if that’s the case it should be justified as an exception, not hidden behind as a rule. If you’re embarrassed about the quality of code, or concerned its not ready for scrutiny, it isn’t reliable enough for you to base your conclusions on.

Your code is not a minor adjunct to your method and conclusions; the description of your method is a cross check for the intention expressed in your code; the results of your code are the realisation of, and sanity check for your method.

They are indivisible and should be available for inspection.
Paul Says:

September 27, 2012 at 8:17 am
@Lady in Red:
“I just want science, plain ole archived data, repeatable science. We can decide about the future of the world when we get that right.”
This is a fair request when discussing physical sciences, but the survey under discussion is a psychological one so (without any expert knowledge on my part) it’d seem reasonable to expect thinks like different opinions to have an unavoidable impact.

As you noted in your earlier post, a respectful approach and clear statement of biases (which Faustus is pretty clear with here [1]) is the best antidote to the problem. We seem to be moving past that and into a space where Faustus is pressured to align with the skeptic line that the original survey is biased and wrong rather than just something that could be performed differently and get different results if you have a different set of assumptions.

On a side note to all, if we want evidence that the survey has flaws in it I’d point to the fact there isn’t a perfect correlation between being a skeptic and thinking there is a conspiracy on AGW. Every comment here that says the science is biased (or “corrupted”) is an implicit statement that there is a conspiracy (of incompetence and ass/funding covering at the very least, if not deliberately mendacious).

In light of that discrepancy, 3 potentially false samples, that result in a conclusion that you agree with (i.e. that skeptics aren’t tin-hat wearing conspiracy believers) seems to be unimportant. I mean, Faustus’s results above actually demonstrate that if someone did try to distort the results they failed to do succeed.

At this point it’s probably best to push Faustus to publish this analysis, get it mentioned in mainstream papers and post gloating comments that if anyone were attempting to distort the sample they failed. You could even go so far as to say that Lewandowsky’s assumptions are questionable and lead to distorted results that can be improved upon by improved analysis [3] [4].

[1] Unlike in Tolkien discussions. He really needs to accept that the Compromise and Conceit setting is never going to replace Middle Earth [2] 🙂
[2] And if it did, some blogger would take it as evidence of Faustus anti-religious biases regardless of how much personal evidence of reasonableness was presented in his defence 😛
[3] As I understand it, this is basically the polite journal way of saying “Ha ha!”
[4] Faustus – I concede I didn’t bother re-finding your original analysis conclusion, so this may be overly harsh.
Paul Says:

September 27, 2012 at 8:30 am
@A. Scott
I can understand that your ultimate aim is to have the claim that skeptics are tin-hat wearing conspiracy theorists either totally debunked or withdrawn. Given how it reflects on you, your cohorts and your attempts to influence debate on the topic it’s not unreasonable.

But Faustus has already put in a good piece of analysis that does debunk the claim. To insist on holding open a “second front” that the survey data is also spurious won’t make the skeptic = tinhat claim weaker. It’s just nitpicking because you don’t like some data that’s been collected.

If you want to refute the data then a larger survey that is published on both skeptic and non-skeptic blogs is required (to capture a comparison of skeptics versus non-skeptics). Because its open to anyone (even if it’s once per IP) you’re going to face a risk of false reporting. If you only survey skeptics then you can’t say “skeptics are no nuttier than the general populace” because you don’t have a general populace survey. If you try for an absolute statement of “skeptics are not nutty at all” I would find any survey purporting to support such a result as untrue as its a simple fact that some of the general populace is nuts and that should be replicated in the skeptic community.

I’ll say again that if you want to debunk Lewandowsky then asking Faustus to publicise these results is the best you can hope for. And he’s always going to attach his notes on how these can be used to change your mind (I assume) because he’s allow to have an opinion of his own (as are you) and surveys like this should be used for a purpose (even if it’s a purpose you dislike, like changing your mind through reasonable fact based discussion)
faustusnotes Says:

September 27, 2012 at 10:12 am
Thanks for your patience with sleeping faustus, everyone. Scroll through this comment to see your particular opinion addressed …

A. Scott, I agree with Paul here[1], this data don’t show anything particularly bad about skeptics and are broadly consistent with what we know: lots of (online) skeptics think AGW is a conspiracy and lots of (online) skeptics have free market ideas. I think you’ll find the same thing if you do any robust survey of skeptics. Complaining further about the data just looks petulant. You’ve collected your own data and I have no doubt that Mcintyre will analyze it, and I suspect he’ll find the same thing, though for all the reasons I stated I think the repeat data set is going to be hugely contaminated. Also as Paul says, if the warmists were aiming to scam this data they did a bad job, and furthermore if they can’t scam a simple psychological survey, how have they managed to scam the rest of the science for 30 years?

Paul, I’m not going to publish this analysis – reanalyses don’t usually get published, psychology research isn’t my focus, and getting an article published takes a lot of time and will get my real name associated with global warming research, which I don’t want. The journal will undoubtedly have a mechanism for people to write a letter opposing the methods or findings of the article, and although accusations of fake data will not get published, criticism of the methods probably will. I think Mcintyre could team up with his commenter HAS (who seems to have similar views to me) to write a letter to the journal – this will be published along with a response by Lewandowsky (probably). It shouldn’t take long – the word limit will be 500 at the most, so it’s an afternoon’s work to whip it out. That creates a legitimate debate in the literature about whether Lewandowsky’s findings are true, and will force others following in his footsteps to do more exploratory analyses before they can go on.

Mrsean2k, you’re confusing “replication” with “reproduction.” Journals publish articles on the assumption that the author knows what they were doing and did it right, and demand a clear statement of methods over code and data for lots of reasons:
1) the article has to stand the test of time, but data and code won’t, so a method that can be replicated is more important than code that can be reproduced
2) the scientific method survives by replicability, not reproducibility. It’s more important that you be able to repeat the experiment on different data and get the same results than that you slavishly reproduce it on the same data
3) peer reviewers don’t have time to redo the whole analysis, and if you expected them to you would never get a peer reviewer
4) reproducing material doesn’t catch theoretical errors: this article is a case in point
5) science journals work on the assumption of good faith on the part of the authors, and assuming that they are potentially lying and you need to double-check their work does not create that atmosphere
6) usually the data cannot be shared, and if you collected data on sensitive matters from human subjects you would have to tell them “this data will be made available to anyone who asks for it.” Good luck getting an unbiased sample with a consent form containing those words…

The main reasons that someone’s research is wrong are usually not buried in their code, but are in their broader methodology: in sample collection, statistical design, power analysis or conclusions. You can check this by careful reading of a well-written methods section. As an example: would the Andrew Wakefield scandal have been prevented by access to his code and data? No, because he faked the data and collected it unethically.

A. Scott:

When you only have 10 affirmative responses (6 Strongly Agree and 4 Agree) as with Moon Landing is a Hoax, then even a couple fake responses, are not noises, and clearly have a huge effect.

You’re right, that’s not noise. The sample contains 10 moon landing conspiracy theorists, but they don’t change anything and outliers (and how to manage them) are a common problem in stats, usually easily handled. The noise arises in the bulk of the data that drives the correlations. Incidentally, of those moon landing conspiracists, 3 were skeptics, right? That means 7 weren’t, which is a sure sign that the survey wasn’t scammed by warmists.

Jean and Paul Mathews, I’m not going to get into the timeline of the controversy or any of that stuff – that horse has been done to death, and there are other blogs where you can flog its corpse.

Carrick, a Supplemental Information file doesn’t necessarily contain code, but is for extra data and the like. For example, as far as I can tell the Lancet info for authors contains no suggestion that you should supply code, and no advice on how to do so. I’ve never been asked to supply code by any journal or peer reviewer, but I’ve had my methods questioned, flaws identified and corrected, and articles rejected[2] all without reference to code. As I said to Mrsean2k, focusing on reproducibility doesn’t avoid huge problems in the theoretical underpinnings of the model. I think Mcintyre’s insistence on “auditing” these things rather than just contesting the methods is counter-productive for these reasons. It is a narrow and mean-spirited approach that doesn’t uncover methodological flaws, but includes an implicit assumption that the scientists whose results he doesn’t like might be lying or fraudulent or staggeringly incompetent. Science did not advance over the last 300 years on that basis.

Lady in Red, I don’t want this blog to turn into a debating house on AGW – it’s not my interest. But I will say that I have no problem with the science of AGW, I note healthy debate within the field, and I see nothing wrong with the stats (mostly) used. When I compare it with the stats-free zone that is WUWT, or the narrow focus on fraud and reproducibility at Climate Audit, I know which I would rather trust. I would further say that I don’t believe you are being spoken to truthfully by the leading lights of the skeptic scene, and would encourage everyone to be as skeptical about what they read there as they think they’re being about AGW science – because a lot of what I see being done at WUWT is absolutely the worst sort of analysis, weak and biased and obviously manipulated to produce desired goals. This whole Lewandowsky controversy has been a convenient way for the WUWTs of the world to distract attention from the arctic death spiral, but it hasn’t added to the sum of human knowledge.

—
fn1: record that for posterity!
fn2: bastards!
mpaul Says:

September 27, 2012 at 12:54 pm
faustusnotes Says:

September 26, 2012 at 11:14 pm

DGH, I’m happy to discuss the merits of Lewandowsky’s paper but I refuse to ascribe motives to him or to his research subjects.

But isn’t that the whole point of Lew’s paper — to ascribe motives to the people who question aspects of CAGW? The central question of Lew’s paper is “why do skeptics question the science”. Answer (according to Lew) because they are the kind of people who get drawn into conspiracy theories.

Its ad hominem by survey.

Let me illustrate this. Imagine I got out onto a few blogs that are skeptical and began to discuss the idea that proponents of CAGW are the kind of people who are sexually attracted to farm animals (or tractors). After a few weeks of conversation, I’d create on-line survey with two questions: question 1, do you believe in CAGW; question 2, are you sexuality attracted to farm animals? I then send this survey to 8 sites; 4 supporting CAGW and 4 skeptical of CAGW. All of the pro-CAGW sites would likely refuse to run my survey — they would find it objectionable. All of the skeptic’s sites would see exactly where I was going and see it as an opportunity to demonstrate the flaws in these kinds of surveys.

In this case, I’m quite certain that my methodology would impact the results. The people who visit the skeptic sites would immediately get my signals and would line up to provide support for my stated objective. Obviously my example is over the top, but I think it helps to illustrate the methodology problem with Lew’s paper.

If you doubt this is the case, I’d be happy to run the experiment.
faustusnotes Says:

September 27, 2012 at 1:06 pm
I have never knowingly had sex with a tractor.

Lewandowsky’s paper aims to investigate cognitive models, not motives – subtly different thing, though from the outside looking in the whole conspiracy theory thing looks like an ad hominem. I think it’s a legitimate question though if a few of the leading lights of the skeptic movement are claiming an AGW conspiracy, to ask if conspiracy theorizing underlies the reasons why many in the movement reject the science. That’s not ad hominem, it’s investigation.

Another way you could look at it (he says, Gowdinning his own thread): is it ad hominem to ask “do Nazis become nazis because they are racist?” I would think not.
Lady in Red Says:

September 27, 2012 at 1:23 pm
It is *not* “AGW conspiracy.” It is AGW sloppy science, stupidity, and “follow the funding…” If the science were good, the data archived — as mandated by law — if the peer review were not “pal review,” there would be no “skeptics.”

(And, I think that there is no “Arctic death spiral” in our future.)
….Lady in Red
faustusnotes Says:

September 27, 2012 at 3:49 pm
I really can’t say anything to this. The science is robust and they aren’t stupid, and you should be spending more of your skepticism on the sites that are telling you otherwise.
mrsean2k Says:

September 27, 2012 at 9:16 pm
@fn

The data I can understand – to some extent – but that’s a different issue to the code.

I agree with you that the main and most significant reason for “wrongness” isn’t usually a simple coding error; a failure to express requirements and a failure to capture them correctly are more costly and fundamental from a software development perspective.

But just because there are more serious potential sources of error, this doesn’t excuse rigour in other areas.

And “goodwill” be buggered; for anything other than a trivial procedure that can in any case be expressed in, for instance, a blog post, you should be required to pony up your code (“one should be required” I mean).
faustusnotes Says:

September 27, 2012 at 9:58 pm
mrsean2k, if the methods can be reproduced without getting the code, I really don’t think that it gains anything except a loss of goodwill to be required to provide code. Remember, stats software isn’t a programming language per se – it’s a set of procedures that do xyz. If I say “I ran a loglinear regression with population as an offset” you know that I wrote (in R) glm(y~x,family=poisson, link=log, offset=…). Without the data, the code is useless; with the data, the code is superfluous. There are rare cases where this isn’t true (some survey commands, details of methods in advanced heirarchical models), but in most cases what we do is very straightforward. Furthermore, there’s no guarantee that your reviewer uses your package: I use stata usually, which I’m guessing would be incomprehensible beyond a basic level to Steve Mcintyre. The reshape commands in SPSS are gibberish to someone who hasn’t used them. Peer reviewers don’t have time to learn your whole software package just to understand the peer review process – if they can’t see that your results and methods match, or they can’t be clear about what you did from the methods, they say so and get you to rewrite it.

I’m open to the suggestion that scientists should be more forthcoming with this stuff but I’m not open to the idea of doing an extra couple of days work as a peer reviewer, having to prepare code that is accessible in multiple stats packages (there are 4 that are commonly used) and somehow placate an ethics committee that is watching like a hawk for any sign of data loss when I prepare a study. The data protection requirements on any kind of scientific work are already onerous and often prevent one from doing the studies one wants – they would be literally impossible if every study had to be made available to anyone who asked for the data.

This is what professional standing and goodwill is all about – science is built on trust and inevitably that sometimes gets abused but mostly we have managed. It’s only the work of 20 years of skeptics, whipped into a furious lather by the likes of the Catos and Heartlands of the world, that this system of trust is being even slightly doubted.
RomanM Says:

September 28, 2012 at 4:30 am
I have run up against the problem of combining ordinal responses in surveys before and on occasion I have used normal scores (not the same as z-scores) as a solution.

In the case of simple categorical data such as in the Lewandowsky survey questions, one assumes that the responses represent Normal(0,1) censored data. The calculated frequencies of the responses identify the intermediate quantile cut-off points for the four intervals. Finally, the assigned numeric value for a response is the (conditional) expected value of the normal variable given that it is in that interval. The resulting numeric sample has mean zero, however, the standard deviation is typically less than one. The sample can be rescaled if so desired.

The result is to increase the magnitude of the value given to the “extreme” response categories with fewer observations as well as providing reasonable separation of interior categories. This is consistent with the principle that if fewer individuals are choosing an extreme response, the difference between it and an adjacent less extreme response should be greater.

Here is an R function which I cobbled together should you wish to try it.

zscore = function(xdats, rev = F, rescale = F) {
#xdats is a data frame or a matrix of ordinal responses
#rev reverses the scoring
#rescale makes all SDs equal to 1
xdats = as.matrix(xdats)
dsize = dim(xdats)
if (length(rev)==1) rev = rep(rev,dsize[2])
zsc = matrix(NA,dsize[1],dsize[2])
colnames(zsc) = colnames(xdats)
dtabs = vector(“list”,dsize[2])
for (ind in 1:dsize[2]) {
cdata = c(xdats[,ind])
if (!is.factor(cdata)) cdata = as.factor(cdata)
ctab = table(cdata)
cumtab = c(0,cumsum(ctab)/sum(ctab))
tol = 1/(2*sum(ctab))
zvals = (1 – 2*rev[ind]) * diff(dnorm(qnorm(cumtab)))/(diff(cumtab)+(diff(cumtab)<tol))
dats = zvals[cdata]
if (rescale) dats = c(scale(dats))
zsc[,ind] = dats
dtabs[[ind]] = ctab}
list(zscores = zsc, tables = dtabs)}
A.Scott Says:

September 28, 2012 at 4:52 am
A. Scott:

When you only have 10 affirmative responses (6 Strongly Agree and 4 Agree) as with Moon Landing is a Hoax, then even a couple fake responses, are not noises, and clearly have a huge effect.

You’re right, that’s not noise. The sample contains 10 moon landing conspiracy theorists, but they don’t change anything and outliers (and how to manage them) are a common problem in stats, usually easily handled. The noise arises in the bulk of the data that drives the correlations.

I think we are saying largely the same thing … I don’t necessarily disagree removing the responses on CYMoon does not have an appreciable effect on the larger conclusions.

They DO have however a huge effect on the whole Moon Landing hoax and skeptics meme promoted by the title of the paper. There is no legit, valid, statistically sound basis for the title whatsoever.
Carrick Says:

September 28, 2012 at 9:22 am
faustus:

Carrick, a Supplemental Information file doesn’t necessarily contain code, but is for extra data and the like.

It doesn’t necessarily but can and often does. Anyway my comment was in reference to this comment of yours

I think many journals don’t even provide an avenue for publishing code.

Only journals that don’t allow for online published SIs would fall into the category of journals that don’t provide an avenue for publishing code. I’ve looked into this, and the SI is maintained by the journal, not the author, in case somebody was wondering about that.

Regarding this…

This is what professional standing and goodwill is all about – science is built on trust and inevitably that sometimes gets abused but mostly we have managed. It’s only the work of 20 years of skeptics, whipped into a furious lather by the likes of the Catos and Heartlands of the world, that this system of trust is being even slightly doubted.

Of course I have to disagree to some extent.

Science is built on trust also on the ability and willingness to verify the claims that science makes. When science fails to do the later … then trust in that institution is eroded. You want to see trust get rebuilt… push for more not less transparency. Putting up walls in the AR5 is not smart IMO. And yes, to me (as a scientist), transparency does mean vetting and publishing the code, and the data used (not just post-screened data).

As to the second comment on Cato/HL… that claim isn’t very well supported by empirical evidence (compare funding levels to e.g. Greenpeace). Geist’s little maneuver demonstrated just how poorly funded HL was in particular. Almost running on fumes until after their private documents became available, and they were able to use PGs little act as a foil to draw in more funding.

If you want my opinion (and I know everybody does that’s why I provide it so freely :-P), I think the real problem is that the solutions for remediation being offered aren’t particularly palatable to many people, and they saw them as being foisted off on them without any proper public debate on the topic.

You can get people to swallow bitter pills, but it’s a lot harder than to get them to eat candy (or to start smoking).
faustusnotes Says:

September 28, 2012 at 10:36 am
Carrick, by “an avenue” I didn’t mean they don’t let you, I meant that lots of journals don’t provide formatting or publication advice – which suggests that they’re not interested. In the epi/medical field this is largely because code is irrelevant without data (the code, too, could be faked, if you don’t have data to check it on).

While I agree that transparency is a good thing in science and all fields should be pushing for more of it, the approach taken by the skeptic crowd is not the way to get this. Calling people frauds and incompetents, hacking their computers for emails and then publishing irrelevant personal stuff just to smear them, making loud claims of incompetence as soon as you can’t reproduce someone else’s work without thinking about the possibility that you’ve missed something, continuous freedom of information requests for the smoking gun of fraud and collusion, sneering phrases like “pal review,” these don’t get you more transparency, they get you a reputation as a problem. Couple this with a complete absence of skeptical scientific research work, and an ongoing (3 year now?) campaign to claim that the entire temperature record is wrong and all the adjustments false – that has produced one paper and this paper supportive of the status quo – and you aren’t going to get anywhere. I mean right now we have some idiot in the USA (look it up at tamino’s place) presenting sea surface temperature maps as ice extent maps, and pretty much every post Watts has done on the matter has been rashly questioning the data quality, only to be repeatedly shown to be wrong. Do you think this kind of behavior might be just slightly exhausting for the people actually doing the work? People who actually go into the field and install buoys, take ice cores, live in temporary camps collecting snow data that 30 years later are falsely presented as evidence of contamination of the temperature record? If one wants honesty in science, one needs to be honest oneself; if one wants one’s pleas for greater transparency to be respected, not accusing people of lying would be a really good start.

A. Scott, actually I think if you analyze the original data with those moon-landing responses in they are slightly more likely in skeptics than non-skeptics. So the title is okay, and it’s cute (I like cute titles, and I think they’re common in psychology). Of course, if I were presenting my results obtained here I couldn’t possibly write that title. It would have to be something else, like They can take my carbon from my (warm) dead hands: free market ideology and climate change skepticism. (I know, I know, free market ideology and gun nuttery don’t necessarily go together, but free market ideology slogans would be dry and boring).

RomanM, that looks like a method similar to using tetrachoric or polychoric correlations, which are implemented in most stats packages used in the social sciences (I’ve used them in Stata). Perhaps you should try reapplying my code to the data so modified and see what you get?

I think there is a big underlying problem here with PCA, factor analysis and all the other unsupervised learning methods, though: we may understand the maths but somehow something gets lost in translation between the maths and the questions we actually want to ask of our data. As an example: the purpose of FA is to find groups of variables with similar behavior. If there are groups of variables that are showing extreme behavior in the same way, is it necessarily wrong to lump them together even if the numbers aren’t correct? It seems intuitively like it shouldn’t be, but the underlying maths and stats theory suggests we should find another way of doing it. I think the lack of settled opinion on how to handle things like Liekert scales is at least partly representative of the lack of coherency between the theory and the research task – a problem that I don’t think is so noticable in regression, for example, where the research task very closely maps to the underlying maths.

But now we come to an ironic turn, which I would really like you to address because it makes me (slightly) mad. Here you are on my blog, posting up code for and discussing experimental methods for analyzing liekert scales in what is, ultimately, a fairly irrelevant psychology paper. In the last few days over at WUWT, Bob Tisdale has posted up an absolutely woeful “analysis” of the relationship between AMSO and arctic ice (or some such). It contains no error bars or statistical analysis of any kind, and he compares the trends using correlations rather than cross-correlations: an absolute cut-and-dried example of a complete and utter no-no in stats. There are many appreciative comments, including some woman saying “this is how real science is done.” Another time I saw Tisdale delete 40 years from the centre of his data series because it disagreed with him, and this was followed by 100 appreciative comments. Here at this blog we have a few people expressing extreme distrust for mainstream climate science and/or great trust in what goes on at WUWT. Why are you over here discussing experimental methods for improving a paper by a warmist, instead of over there pointing out the huge flaws in the analysis being posted at the world’s most popular site on global warming? Why are you posting R code for something that is not settled, here, when you could be posting R code there for Tisdale to do cross-correlations[1]? When Tony Watts put up a 30 year old picture of a temporary station on the wrong side of the antarctic as proof of UHI somewhere else, it was left to the people who worked there 30 years ago to correct him, and I don’t recall a single skeptic saying “fair play, Tony, best change this post …” I can’t post on WUWT – I was banned for demanding that Watts get the dates of events in his surface station record correct (to within the same decade, mind you, not unreasonable). But you can. Yet you are here, not there. Can you explain to me why you refuse to correct the science on your own side, even when it is done as woefully as Tisdale’s, but you will write 5 or 10 blog posts on some irrelevant paper on skeptics’ mental models?

—
fn1: putting aside the possibility that R’s time series package is broken, surely it’s been fixed by now …
Leo G Says:

September 28, 2012 at 1:10 pm
Well here’s my polling results from 3 Joe Sixpacks and 2 Jane Sixpacks. All believe the science in AGW.

1 Joe and 0 Janes Believe that innoculations cause autism.

2 Joes and 2 Janes Believe that GMO’s are frankenfoods.

All believe that the moon landing was real

And all really , really like beer (including me)

So after running the numbers through my Commodore 64 my results say that people who believe in AGW are drunken, organic eating, conspirators!

🙂
Leo G Says:

September 28, 2012 at 1:13 pm
PS, couldn’t agree more about the lack of sceptism at some sceptic sights. But in Tony’s defence, he does reference a lot of papers that would never see the day of light without his site.
Lady in Red Says:

September 28, 2012 at 1:41 pm
I am frustrated, Mr. Faust. Many very very smart people have given you (and the psych man) gracious time.

I am smart. I don’t suffer fools. I don’t follow authority, per se.

What does that make me? A non-moon landing nut?

If Lewandowsky were doing an honest survey he would have looked to explore possibilities. Instead, he set out to identify crazies, regardless.

You suggest I look at AGW sites. Real Climate? Think Progress/Climate? Where? You imply AGW is “real.” Where, when…?

Here is what I have learned:
AGW types don’t talk, do not engage. They invoke consensus and authority.
The Phil Jones’ computer programmer (Harry Read.Me) is probably dead of exhaustion, trying, for years, to make sense of nonsense. …and no AGW one cared that it was nonsense.
Phil Jones’ comment to Warwick Hughes, in 1996, about not giving him data because Hughes just wanted to challenge it, cut to my heart.
Peter Gleick is an embarrassment writ large. He spoke at Oxford, several months ago, post humiliation.
Jeff Condon and a circle of friends took out Eric Steig’s Nature cover story. The IPCC will ignore that.
RealClimate, when I asked (silly me), after myriad pomposity (gad! do I know pompous!) sent me to a Wm. Connolley piece about how the 1970’s were really about warming, if you were a “real” climate scientist. I worked on the Connelley article for hours before accepting its lunacy: no citations, documentation whatsoever. Lunacy/
NOAA fucks with the surface temperature stations. Watts called ’em on it. Do they repair? Nah: eliminate, more…
Michael Mann? Do you like, respect, him, Faustus? Is there not a piece of your brain reserved for admitting jerks are jerks. Also, Gavin Schmidt. (Gad! Do these men have sex? Do women…? Never mind…)
Joe Romm and ThinkProgress/Climate is hysteria. Anything else?

Faust, if there were a credibility to the AGW story, the climate models would have been proven true. Better, scientists could defend their models, the extent to which this is not just “climate,” but “CLIMATE.” Climate moves, evolves, changes, warms, cools, but this is CLIMATE! Nah, Faustus.

…from the folk who can’t tell me if it will rain tomorrow, can’t predict the seasons… you give me: the models for climate a hundred years from now…?

I smile.

We are in a cycle, probably heading, as predicted, to an ice age. (Fortunately I have a lot of firewood… and will probably be long dead.)

There is no credibility with the AGW folk. None that I have seen. Direct me, if you can.

Once, a very very smart man thought about this, global warming… greenhouse glastnost. “Nah,” Walter Orr Roberts told me. “It’s not happening.”
…Lady in Red
Lady in Red Says:

September 28, 2012 at 2:43 pm
I just visited Lewandowsky’s site. Sad.

Worst: Lewandowsky really is an idiot. Really. In spades.

The man is [snip]!!!!!

….Lady in Red
Carrick Says:

September 28, 2012 at 3:54 pm
faustus:

Calling people frauds and incompetents, hacking their computers for emails and then publishing irrelevant personal stuff just to smear them, making loud claims of incompetence as soon as you can’t reproduce someone else’s work without thinking about the possibility that you’ve missed something, continuous freedom of information requests for the smoking gun of fraud and collusion, sneering phrases like “pal review,” these don’t get you more transparency, they get you a reputation as a problem.

I’m not entirely unsympathetic to this view, but I think there is a bit of back and forth that you are ignoring here. PG is a good example. I don’t think either SkS “hack” nor the climategate “hack” are demonstrable as hacks (I know the bobbies think so, but my opinion of bobbies as IT professionals is about one grade above my opinion about their ability to monitor meters).

Also, regardless of how the SkS forum and climategate emails came out, they were legitimate emails, nobody made these up (unlike PGs infamous faked a$$-covering memo which I think is pretty clear was made up whole cloth, in a bat cave, by an intern, or something like that). The CG emails in particularly legitimately undermined faith in the process when you see people gaming the system as was done on a number of occasions. (Obviously having seen much worse in my own research field, I wasn’t that surprised.)

I do think that McIntyre, Watts etc could cool the rhetoric some, but so could Lewandowsky, Mann and others. I actually think that Lewandowsky and Mann cause more collateral damage to science in general with their bizarre notions of “post-modern science” than any of the skeptics have by poking them in the eye with their rhetoric, poor behavior, and sometimes really crap-assed scientific practices. I can’t say I know or dislike Lewandowsky, but I’ve noticed plenty of (to me) well-deserved critiques of his paper from both sides of the aisle. I happen to know him from some of his previous fruits and that’s biased me a bit to (see him as generally into practicing confirmatory statistics 101).

Mann, well too much attention has been paid to him and his antics, but there has been entirely too much circle-the-wagon mentality on this one, after a what-must-have-been-obvious-as-hell was a convenient-but-erroneous result in his original MBH98 hockey stick paper. Now I know there will be some who will defend this paper, but there really isn’t any question that modern reconstructions including Mann 08 EIV don’t in any respect confirm his paper, rather they show it to be substantively in error. This really raises an issue about all of those earlier papers that came out and the ethics being practiced, partly at the urging of Mann, “demonstrating the validity of his work.”

And by the way …too much attention means just that. There are more important confirmatory tests of AGW, like radiative physics measurements at the surface at top of atmosphere. It’s unfortunate that a blatantly wrong result was used as the poster child so many later publications.

I do think as you apparently don’t that there is a positive role that skeptics can play. Somebody needs to keep the researchers honest if they aren’t willing to police themselves. Can you count how many BS reports there have been in the last two years on extreme weather??? As a statistician you should be aware of the weakness of this approach in looking for confirmatory evidence in the tail of a distribution. .

People who actually go into the field and install buoys, take ice cores, live in temporary camps collecting snow data that 30 years later are falsely presented as evidence of contamination of the temperature record?

Of course that is complete and utter crap (that people fake this data). There was a recent post on WUWT about GISS faking their data again, apparently by somebody who didn’t spend the 15 minutes needed to understand how they generate anomalies, and how when you shift data values as more data comes in, anomaly values can change relative to a fixed baseline.

Anyway, I think there is a lot of room for improvement on both sides. Anthony Watts gave (I thought) a fairly rational interview to PBS, even if I disagree with him about the importance of the UHI corrections (I think they amount to a paltry 3% difference at most in global mean trend).

Maybe if the other side didn’t engage in such extreme rhetoric towards Watts and other critics, you just might see a reduction in the amount of heat coming the other way too.

By the way, “pal review” is a term that has been around for years. It is a real issue. People collude to see friends stay funded and papers that shouldn’t be published, published. Only way to fix this really is more transparency. Whistle blowing mostly gets the whistleblower looking for a new job, and I like my job.

Something you may or may not have heard of is the initiative for training in responsible conduct of research in the US. It has been driven mostly by the antics of people in the medical research community (like Anil Potti who was faking results from a chemotherapy treatment protocol…wtf… and gets to retire with pension and retain his right to practice medicine wtf x 2).

Things never were great in science, we’re human, and the internet is making that more and more clear over time. We need to clean up our own house. If we do so, there’d be less for guests to pick on us about.
faustusnotes Says:

September 28, 2012 at 6:07 pm
Lady in Red and Carrick, I’m seeing a lot of comments about the personalities of people involved in this “debate” but not much about the science. I’m not sure who PG is Carrick, but if you mean Gleick I’m not sure how you can be referring to “back and forth.” The climategate emails were clearly stolen long before Gleick stole anything, so this is hardly a “back and forth.” There was clearly an egg before a chicken, and not everything in the stolen Heartland mails was fake. Note the double-standard here too: it’s fine to steal scientists emails but not lobbyists. This seems like a very strange moral world to me.

Claiming that the CG emails undermined faith in the process is also a little bit like shutting the gate after the horse has bolted. From the tobacco papers we know that Big Tobacco was paying astroturf organizations to undermine scientists a long time before the climategate scandal, and everyone knows Cato, NRO, Heartland, etc. have been attacking climate scientists for years before then. So again, where is the back and forth there? Did Inhofe refer to AGW as a “hoax” before or after climategate? When Mann first published his hockey stick, did he say that anyone who disagreed with him was a fraud and a paedophile? It seems the aggro started in specific quarters…

There’s a strange obsession amongst skeptics with Michael Mann, even though the science of AGW stands up without his contribution. The kind of vituperation reserved for him (didn’t Steyn compare him to a paedophile recently?) is really well, well past the limits of “reasoned” discourse, and hypocritical when compared to the exploits of some of his detractors. Where is the similar disdain from skeptics for, for example, Wegman? Wegman wrote a plagiarized report criticizing the climate science peer review system as “pal review,” that he only got published through a pal, and when the plagiarism was discovered he blamed it on a student who had received no credit in the paper. So many layers of malfeasance, and yet that Wegman report is still linked to at Mcintyre’s place. Why is discredited “science” acceptable from one side but not the other? In comparison, Mann’s original hockeystick has been superceded, in some cases by his own and in some cases by other people’s work – the march of science proceeds at a cautious pace, strangely free of any of these hyperbolic accusations.

Maybe I’m being naive about the rigor of other fields (epidemiology has had its accidents and cleaned itself up about 10 years ago), but it seems like the skeptosphere is not contributing a great deal with this level of vituperation. I mean, really, comparing published scientists to paedophiles? I’ll take their contribution seriously when they start doing serious science – until then, they remain in my view firmly in the box of “political activist.”

Leo G, cute survey. Were they drinking organic beer? Because otherwise I suspect your data is fake.
Lady in Red Says:

September 28, 2012 at 8:45 pm
Mr. Faust…. The Climategate emails were not, I think, “stolen.” They were “released,” by someone upset at the machinations to which Steve McIntyre had been subjected by the “community” for many years.

One of the reasons McIntye will be enshrined in history is his steady patience, never getting mad, going away, for the better part of a decade — McIntyre fought for science. Without his ability to endure the abuses of the mainstream “climate” “science” plotting, evil, cackling crowd, we wouldn’t be where we are today. “They” would still be playing their evil games.

(BTW, there is another Climategate dump on the internet, now, encrypted, but…. I look forward to the release of the key.)

What is germane to this debate is the corruption of “climate” “science.”

Where is the information to convince a lay person that the core of the “science” itself is not corrupted? Where are the “scientists” who can debate the skeptics, convince me of the integrity, consistency, of their work?

I don’t think they exist.

Lewandowsky set up a survey only to prove skeptics are nuts. *I* think they are smart, independent thinkers. One hundred percent of them. But, he didn’t look for that, did he?

Seriously, if there is *anywhere* I might go to find engaging and challenging mainstream AGW folk, tell me.. (As much as the failure of their “science,” I am offended by their slinking, skulking non-engagement and appeals to authority…. all of us are soooo right! Trust us. It is Gollum-like, creepy. That is the appeal of the stupid, failed mind, intellect.)

Where do you find thinking, challenging mainstream AGW, folk willing to stand on their work, engage others, furthering understanding….? Who is willing to engage, in sunlight, beyond the huddling masses appeal to authority?

I am a skeptic and I exhibit characteristics which should make Lewandowsky blow his stack in admiration, respect. But, he didn’t look for positive qualities, did he? A leeetel bit scummy, in my estimation. ….Lady in Red
Carrick Says:

September 28, 2012 at 10:41 pm
PG is Peter Gleick. As to science involving personalities… of course it does. That’s hardly news.

As to ” The climategate emails were clearly stolen long …” you don’t even know if they were “stolen”. We still don’t know how their release occurred, you like to use the word “stolen” (I think) because it absolves you of having to deal with the contents of the climategate emails.

From the tobacco papers we know that Big Tobacco was paying astroturf organizations to undermine scientists a long time before the climategate scandal

First, examine for yourself how effective that was. People still smoke, and the only thing that really impacted smoking rates, is laws regulating smoking in public places. The “information was out there”, people “chose to ignore it”. They didn’t chose to smoke because they “thought it was safe”.

In the case of the evil fossil fuel companies, have you an even bigger problem for your groups pet conspiracy theory…for every dollar spent by fossil fuels on skeptics at least 10 dollars get spent on “warmer” organizations. Google Stanford and Exxon as an example. $100,000,000 + to one organization, and Exxon funds other people to.

One of the things Peter Gleick undermined, that you haven’t picked up on, is the theory that skeptic organizations are well-funded. They aren’t.

Fossil fuel companies properly are energy companies. They sell energy. People are going to need energy even if it’s not from fossil fuel. Energy companies understand this.

I bring up Michael Mann as an example of somebody who is highly visible and damaging to the climate and science community in general. I have no obsession with him, I’m just realistic of the impact people like he have on the public debate.

Mann’s original hockeystick has been superceded, in some cases by his own and in some cases by other people’s work – the march of science proceeds at a cautious pace, strangely free of any of these hyperbolic accusations

You haven’t been paying attention then, if you don’t think there’s been any hyperbola involved in the march of this science. Look up the exchanges between Mann, von Storch and Moberg. LOL.

My point is the original work was badly flawed, and that fact was not only fought, there were numerous peer reviewed papers claiming to replicate it (who like Mann swept known problems under the rug), and the fact it was flawed didn’t stop it from getting used and reused by the political wing of the warmer movement. These are all facts. Interesting that this happened, don’t you think? It may not affect your opinion, but I think your mind is made up and not likely to get confused by the facts. My point is it is damaging to science in general when episodes like this happen, and for people who self identify with a political movement, as you obviously do, that may not matter or you might say “it’s for the common good” or some other bunk to pass it off, but for those of us not politically committed to a movement, it does matter.

As to bringing up Wegman? You’ve got to be kidding me. After launching off on why we should ignore Michael Mann, who people heard of, we’re supposed to focus instead on Wegman? I didn’t read his crappy report when it was written, it was obviously politically motivated, he’s not even a climate scientist so I don’t see any reason for even bringing it up here.

But if you want to discuss it, it’s not even properly peer reviewed literature and written with no external editorial oversight, it was written as a source paper for a US congressman, the main findings of which even though not directly plagiarized were heavily lifted from existing source materials. Very little in the way of original scholarship (the network theory stuff maybe). If McIntyre wants to link it…well it’s his blog. Nobody else voted as to whether it should be linked there or not. Bring that up with him, not me. I don’t even frequent his blog…I’m in the “rare commenter” status there.

As to Penn State and Sandusky…not a relationship I would have said publicly (as being in bad taste and easily misconstrued), but it’s one I think most people have made the link independent to McIntyre’s post.

The problem that it brings up (to me) is how universities are run in general, not that climate scientists are practicing pedophiles (truthfully I think you and others use that latter as an “out” to avoid having to deal with the real issue, just as you like to argue over whether the climategate emails were stolen so you don’t have to discuss what their implications). Anyway, universities hire in professors to manage other professors, so the administration, being made mostly of professors (or ex-professors) is too vested in the interests of its own faculty, and that creates a confusion of the role of the administrator of who he is supposed to be protecting. Maybe this happens less at other countries (it would improve US university administrations if department heads actually had a requirement to know something about administration for example, wild and crazy thought).

I’ll take their contribution seriously when they start doing serious science – until then, they remain in my view firmly in the box of “political activist.”

IMO, one of the most important consequences of skeptical pressure was release of code and data for the surface temperature record, and that has helped substantially the scientific process. It’s led to a half dozen or more new codes being written, and I think improvements to the original code (Clear Climate Code is an example, Nick Stokes code, Mosher’s code,etc.) People are giving talks at conferences (Mosher and Zeke at AGU for example). Are these skeptics? Zeke isn’t, and neither really is Mosher, but that’s my point. When you practice science, you should cease the political activism. Your buddy Lewandowksy could learn something from that model.

Well for them to be a “skeptic” they are an activist, just the same as somebody who is advocating particular political courses for remediation is an activist. As soon as they start practicing science, you shouldn’t be able to find daylight between what they are doing and other scientists.

Otherwise they are still being political activists and not scientists (see Roy Spencer as an unfortunate example of that).
Lady in Red Says:

September 28, 2012 at 10:58 pm
BTW, Heartland et al are chump change. The biggie bucks are government and they are only available for studies about mitigation, or confirmation of climate change… Ring through your nose, and follow the funding…

This is why the present crop of “scientists” is so poor, mere “useful idiots,.” believers in “noble cause.” Corruption?

Michael Mann, poor Gavin Schmidt, and the loon Lewendowsky…? There are not men whose counsel you should seek. Useful idiots.

I am not in a position to defend Bob Tisdale and all of WUWT. I am prepared to defend their commitment to open science.

Where, on the other side…? …Lady in Red
Lady in Red Says:

September 28, 2012 at 11:12 pm
I have to say: I like you, FaustusNotes.

I suspect that, with care, you are redeemable, back to the world of the thinking. Pls breathe. And think. Parse. Really think. …Lady in Red
RomanM Says:

September 29, 2012 at 12:21 am
I came here expecting that you might be interested in discussing statistical analysis under the impression that both of us might exchange statistical ideas and learn something new. Sadly, I was mistaken. You seem to be more interested in the AGW “wars” and beating down those who disagree with your world view than in looking at the statistical science and expanding your horizons.

You ask the specious question of why I seem to concentrate on analyzing the published work of climate science and not the blog postings of others who may be correctly or incorrectly criticizing the work of climate scientists. There are many answers to that question.

The initial reason I came to the entire question of CAGW was that there seemed to be an inordinate number of claims regarding possible catastrophic effects of the current usage of fossil fuels. Some of these claims involved probability assessments in IPCC reports which I knew to be unscientific in character. The science was settled and anyone who disagreed with this was demonized. I came looking for the real evidence and what I have found in the last seven or eight years is very sparse. There was a heavy involvement by non-scientific NGO environmental groups which advocate “solutions” to a “problem” whose seriousness was not substantiated – solutions whose effect would be to produce a world without the resources to cope with any “problem” should it come to pass. Harassing scientists who dared to publish results contrary to those advocated by the CAGW activists has not been uncommon and this increased the level of distrust of whatever results they might produce.

In partial reply to your question, I am not a “web sheriff” nor do I spend a lot of time looking at subject matter in which I am not interested. Climate science publication results are pre-released in exaggerated form to a media very willing to beat the CAGW drum. Furthermore, such results continue their effects when they are referenced in further publications. The effect of those publications is considerable. However, when errors are discovered, scientists are reluctant to admit to them, and rarely do these errors also get corrected in the media where the original message has been spread. Unfortunately, given what I have seen, my distrust in the statistical abilities of many climate scientists has become somewhat stronger and I see no reason why I should not use my professional skills to attempt to identify inappropriate analyses giving unreliable results. It seems unfortunate that this should make you “(slightly) mad”.

In the case of Lewandowsky, I agree that it is “a fairly irrelevant psychology paper”, and might not otherwise pay it much attention, however, its message is already out there in the media. That message affects me personally although I am not prone to believe any conspiracies nor do I consider my political views particularly extreme. So yes, if necessary, I will continue hammering at the paper and the activism of its arrogant lead author.
faustusnotes Says:

September 29, 2012 at 1:04 pm
RomanM, don’t be snippy. We are clearly discussing statistics. I don’t think anyone here is beating anyone’s world view down and yes, the purpose of debate on the internet is to convince other people of your ideas.

Your answer is what I was expecting. You care about how scientists portray skeptics, but not how skeptics portray scientists. You care about the occasional misuse of statistics in a mostly inaccessible, vast scientific literature, but don’t care about the common misuse of statistics in a highly accessible and quite small online community that claims to have something to say about that literature. You claim that scientists “demonize” those who disagree with them (when did Mann call his opponents frauds and paedophiles?) but you don’t care that skeptics demonize scientists. You are worried about non-scientific NGO groups corrupting science in favour of AGW but don’t care about non-scientific NGO groups (including the tobacco industry) corrupting science if it opposes AGW. So you, too, have a worldview, and it appears that you are not interested in being critical of those – on blogs like WUWT – who support it.

Indeed, the blog you most directly contribute to links to the Wegman report, one of the most thoroughly discredited pieces of scientific malfeasance ever produced in connection with AGW. Mcintyre comments at WUWT, but never critically, and never even with the gentlest of corrections. There are people on WUWT who are expecting global cooling to start (Lady in Red has said as much here), and WUWT regularly hosts posts with completely erroneous analyses claiming to show this. Don’t you think you would be doing those people a favour by popping over there and pointing out to them that they are being misled by extremely poor statistics? Or do you care more about how that community is misrepresented on the basis of this paper, than how they deceive themselves through poor science? Do you care at all that the skeptic community is comfortable using stolen, private communications to smear scientists?

For Carrick and you both, I should point out that the IPCC reports are not scientific documents: they are policy documents. They present a summary of the state of the science for policy-makers, and on a scale that has never before been attempted. It’s inevitable that they will be flawed documents, the process by which they are made will necessarily be political, and they will be unsatisfactory to all involved. I doubt that you have got a better method for this summary in mind, or that any method you proposed wouldn’t be corrupted by the nature of the task. This is how policy-making happens, sadly. But AGW is a global, not a local issue, and it can’t be addressed by any single country, so these cross-country policy summaries and conventions are going to be necessary. None of this changes the fact that the arctic is melting at an unprecedented rate, and that this will have effects on the world’s environment, or that we are entering a period of unprecedented food insecurity and environmental change. It only affects what, if anything, we do about it. Lewandowsky’s paper is the same: it is irrelevant to what is happening in the arctic, and yet the entire skeptosphere is focused on this paper, and ignoring the arctic death spiral.

I guess the ultimate point here is simple: you are nit-picking while Rome burns. How do you think that is going to help?
Lady in Red Says:

September 29, 2012 at 1:35 pm
Mr. FaustusNotes…. Hmmmm. You really believe that Goldman Sachs carbon chits will save the planet? Really? How quickly will it be saved?
….Lady in Red
Lady in Red Says:

September 29, 2012 at 1:52 pm
Mr. FaustusNotes…. I have read (and tried to read) Deep Climate on the Wegman Report, which said, I think, Mann is a jerk and couldn’t pass Stats 101 and his friends gave him a “free pass.”

Is that a reasonable summary of Wegman?

DeepClimate said that Wegman had a student “cut and paste” some stuff into the report. Wegman was asked to release documents pertaining to his study of Mann. He did.

Where is the controversy, Faust? ….Lady in Red
Lady in Red Says:

September 29, 2012 at 2:11 pm
M. FaustusNotes…. Let’s do pedofiles…. simple for me. Here’s the original Mark Steyn post, which I thought was witty…. Mann… “a tree ring circus…”

http://www.nationalreview.com/corner/309442/football-and-hockey-mark-steyn

Here’s where we are:

http://www.steynonline.com/5085/i-dont-bluff

Let us go back. Where, oh where, can I find mainstream science which confirms the IPCC reports? Where can I get a sense that (US) science funding for “climate” “science” is more objective than confirmation and mitigation.

I be trying, M. FaustNotes! …smile…. ….Lady in Red
mrsean2k Says:

September 29, 2012 at 7:44 pm
@fn

For the absence of doubt, I’m not suggesting that you need to submit your source in the most common packages / languages used – and I’m not certain how you’ve arrived at that impression.

What I am asserting is that the actual code that the paper has relied upon to produce it’s results should be open to public scrutiny, no more or less.

It scarcely matters how trivial and free from error you may consider the code to be, there is many a slip, etc. etc.
MrPete Says:

September 29, 2012 at 11:35 pm
fn, you believe the science is robust, perhaps summarized in “None of this changes the fact that the arctic is melting at an unprecedented rate.”

Yet:
a) there’s really only funding to ask one side of the question
b) the science is hugely uncertain (go check the LOSU results from IPCC… then see McKittrick’s summary of how that got politically-eliminated from the final report)
c) The uncertainty of “unprecedented rate” should correlate with the solid evidence that (for example) arctic treelines were much further north than they are today, within the last 2k years. Finns have been looking at such things for a while. But that, and much more, is ignored.

Bottom line: so much certainty for the currently-popular meme, yet few if any ask the questions that need to be asked.

You want better auditing of skeptic papers. That too happens, etc check out the gauntlet Loehle et al went through at CA. CA is the serious science site, not WUWT. WUWT is about marketing, IMHO.

But where is the good same-team auditing of warmist work? For the amount of investment, the quality is too often pathetic. And I’m a guy with warmist science relatives…. I am NOT saying everybody’s work is bad. Just there’s way too much of it, and the bad stuff is what gets promoted. Even to the cover of Nature.
mpaul Says:

September 30, 2012 at 5:06 am
faustusnotes Says:

September 27, 2012 at 1:06 pm

Lewandowsky’s paper aims to investigate cognitive models, not motives –

The title of Lewandowsky’s article is “NASA faked the moon landing, Therefore (Climate) Science is a Hoax: An Anatomy of the *Motivated* Rejection of Science”. The root of the word ‘motivated” is motive. This paper is all about trying to assign motives to skeptics. I think the reason that I am so offended by the paper is that is that it attempts to establish that correlation = causation, and then does a really bad job methodologically in establishing the correlation.

For what its worth, I am an engineer by training. I spent my early career in semiconductor manufacturing. I worked with Benoit Mandelbrot at IBM on a project aimed at developing stochastic models to predict defect formation in semi-conductors. While I never became a preeminent expert like Mendelbrot (and, in fact, I’m no longer current in the field), I had a reasonably strong background in the concepts and modeling of chaotic systems.

I am generally skeptical of academics claiming that they can predict the behavior of chaotic systems using deterministic models.

But, skepticism between and among engineers and academics is nothing new. I find that the people at CA, Lucia’s Blackboard, and other skeptic sites tend to be like me — they are engineers skeptical of academics. The characterization by the CAGW community of skeptics as being right-wing, bible thumping conspiracy theorists bought and paid for by some shadowy group of fossil fuel interests is offensive to me. And, the more the CAGW community tries to promote this idea, the more people like me get motivated to expose the defects in the academic’s position. That’s my motivation.

Just a perspective.
diogenes Says:

September 30, 2012 at 8:23 am
for what it is worth, I applaud this blog post. It looks at the data objectively. Unfortunately, I cannot say the same about the scare-quoting hysterics who promote CAGW…some of whom claim to be scientists, eg Michael Mann and Peter Gleick.. They sre the problem and until concerned people like you deal with their bizarre and unscientific utterances, you will not get traction.
faustusnotes Says:

September 30, 2012 at 10:44 am
Sorry for the delayed replies, folks, it’s the weekend here and there is a time difference as well. If you’re still reading…

Lady in Red, I don’t think that Goldman Sachs should be allowed anywhere near the things that the grown ups do. I’m pretty sanguine about whatever methods are used to solve the AGW problem because I think the developed nations are very rich, and the AGW problem will probably not even require the degree of effort we put into WW2. I’m surprised that you think otherwise, and I would like to note here that for all your concerns about AGW “alarmists,” as soon as someone proposes even a moderate intervention – cap and trade, or a carbon tax – skeptics start making the most outrageous claims of “energy poverty” and economic collapse. You can’t claim about alarmists and then oppose them for being “scare-quoting hysterics” in the same breath as you complain about economic collapse. It makes you look – hypocritical as well as alarmist.

Lady in Red, comparing someone to a paedophile is not “funny,” it’s mean. If you can’t apply the same standards of civility to both sides of a debate, why should I listen to any complaints about Lewandowsky smearing your fellow travellers?

mrsean2k, I’m not opposed to presenting code in principle, and everything you say is fine. The problem arises when people see failure to present as a sign that you have something to hide. This attack was very strong against Lewandowsky in respect of his refusal to release the names of the blogs he contacted until he had permission from his ethics committee. As someone who has dealt with information security and privacy issues in a variety of academic and research contexts, what he was saying made absolute 100% sense to me but the blogs that some of you are applauding for their “objectivity” were treating this as suspicious. That tells me a lot about their objectivity, and about how unwilling these bloggers are to engage in honest debate (or how little they know about the field they are “auditing”). The same goes to this issue of code – you may not realize this but some organizations make conditions on how you use code (even which program you can use to analyze data) when they provide you data, and you can’t just go blogging it out there. You should not infer duplicity when you don’t know the whole story.

MrPete, science funding doesn’t work the way you are implying, and you’re misrepresenting the question. Do you think someone should make a funding application to find out why arctic ice is increasing? Because it’s not, so such an application won’t get funded. No, if you want to study arctic ice (and there are experts in this field) you need to apply for funding to study what is happening, which is that it is declining more rapidly than ever seen before. There are people studying every aspect of this, and if you check out Neven’s ice blog or Tamino’s posts, you’ll see extensive discussion of the causes by knowledgable scientists. What you won’t find is people applying for grant funds to study things that are obviously not happening, i.e. a stable ice pack. If it returns to stability a la the wet dreams of the WUWT crowd, then you will see a rush from those same scientists to get grant funds to explain why. But right now we have a set of models that predict gradual ice pack decline against a backdrop of stable ice, and the ice is dropping faster than any model predicts or any historical data suggests is possible. But you think the arctic ice specialists should be applying for funding to study something else?

I don’t know why you think your questions are better posed than those of the ice scientists. Are you better acquainted with the literature on the history of ice pack degradation than Stroeve et al? If not, why are you so unwilling to trust their judgment?

mrpaul, I don’t think you understand what motivated means in the sentence of that paper – I think it’s domain-specific use of a word (like use of the word “censored” in that famous folder that skeptics erroneously take as evidence of censorship). This is the big risk when you go from a field you’re an expert in to one where you’re not. Instead of braying that you’ve been misrepresented and smeared, you should pay some attention to the possibility that you don’t know the field. Take a deep breath and ask some polite questions first.

The characterization of scientists working their day jobs, doing difficult research, as frauds, liars, and paedophiles is offensive to me. The accusations over at WUWT that they are part of a global conspiracy to undermine sovereign governments, and the claims that they destroy the careers of those they don’t agree with, are also offensive to me. Mcintyre’s claims that Lewandowsky was lying just because Mcintyre couldn’t reproduce a trivially obvious result are offensive to me. Today in Australia a major right-wing radio host suggested that our Prime Minister caused her own father’s death through lying about AGW, and his political allies auctioned off a chaff bag (a joke about drowning her like an animal). That’s offensive to me. So if the skeptics want me to see them in a better light, they could try to abstain from applauding skidmarks like Steyn when he compares Mann to a paedophile. They could stop calling serious scientists stupid, and publishing their trivial private emails to shame them. They could stop treating them as “scare-quoting hysterics” when they report on epochal events, and they could try showing them some respect. Until they do that, skeptics should count themselves lucky that they aren’t being compared to paedophiles, accused of killing their own parents from shame, subject to implicit threats of murder by major media figures, having their professional correspondence published, and called frauds and idiots on a daily basis. Lewandowsky is trying to understand the underlying models by which some people reject what he takes as settled science. Instead of getting aggro about that, the skeptic blogs could have cooperated with him. They didn’t, and chose to make premature accusations of fraud. It doesn’t help your case that they are “serious” when they do these things.

diogenes, I think my reply to you is contained in the replies to the others.
MrPete Says:

September 30, 2012 at 9:14 pm
fn, I have no problem with the idea that ice is melting rapidly. However, “never seen before” is not an easy nut to crack. Like mpaul I’m an engineer with enough decades of experience to be wary of academic hubris (not that all academics suffer from this more than anyone else…but for a few in visible positions, there’s no question.)

Humans tend to see patterns everywhere they look, turn those patterns into formulae, and believe they’ve learned something. When historical facts don’t match up, they tend to assume someone just didn’t observe very well in the past.

There’s plenty of funding to study today’s warming. However…

Where’s the huge funding to try hard to disprove the popular assumed causes? In Feynman’s terminology, to ensure we aren’t fooling ourselves.

Where’s the huge funding to dig in on past warming? Arctic treelines are just the tip of the iceberg, so to speak 🙂 …

Go look at “Warm Words,” a funded PR study by MPPR. It’s obviously taken root: LOSU tells us the science is not actually settled at all. But we just act as if it is and people fall in line. That’s the reality today: the actual uncertainty is still huge, but the PR war and the funding war are completely loaded in one direction. There shouldn’t be such “wars” in the first place.

I regularly re-read Feynman’s 1974 Cal Tech speech on scientific integrity. We could greatly benefit from a worldwide shift toward funding that values efforts grounded in that kind of humility: first, ensure we aren’t fooling ourselves, then that we’re not fooling other scientists, and finally that we aren’t fooling the public.

The real message of ClimateGate is completely lost on most people. What we see there is a story revealing an astounding lack of scientific integrity among “gatekeepers” who will do whatever it takes to win a PR war. That’s not science.

The L survey is just another sad example. You and I both know that the study was horribly conceived and executed, from the perspective of scientific integrity. Those of us who are real skeptics would be hardly likely to respond to such a ridiculous survey. I can barely bring myself to answer the surveys required to get my medical expenses paid 😀 …

Beyond the shift toward valuing scientific integrity, there’s one other value shift I (and others) promote: recognizing the need for an incredibly diverse multidisciplinary approach. The serious skeptic community has that in spades over the alarmists. One browse through Jeff Id’s Reader Background thread will tell you the skeptics who post at the “serious” skeptic sites have some pretty reasonable intellectual and educational backgrounds. (Myself: engineer who helped invent a lot of tech you use. Personally acquainted with folks on all sides of AGW — you’d be surprised. Married to a field biologist who laughs at the silliness of some of the climate “experts”… those who mangle data in their models while being apparently clueless and/or uncaring about field reality. Related to glaciology family that have great integrity in their work and have great difficulty even thinking about the hubris evident in today’s “wars.”)

One might think all that is off-topic. But the point is: don’t get lost in statistical analysis of such obviously garbage data. This wasn’t just GIGO. It was GOGIGO. Garbage survey Out gives us Garbage data In. As a result, even the most pristine analysis will still produce Garbage analysis Out… unless the goal is to demonstrate just how bad the GOGI was.
Lady in Red Says:

September 30, 2012 at 11:25 pm
Mr. Faust…. I am so glad we are on the same side, vis a vis Goldman Sachs. ….smile, but:

“Lady in Red, I don’t think that Goldman Sachs should be allowed anywhere near the things that the grown ups do. I’m pretty sanguine about whatever methods are used to solve the AGW problem because I think the developed nations are very rich, and the AGW problem will probably not even require the degree of effort we put into WW2. I’m surprised that you think otherwise, and I would like to note here that for all your concerns about AGW “alarmists,” as soon as someone proposes even a moderate intervention – cap and trade, or a carbon tax – skeptics start making the most outrageous claims of “energy poverty” and economic collapse.”

It is going to take me a bit to, succinctly, explain the import of “cap and trade” or a “carbon tax”, but… Give me a day to think.

Have you ever heard of John McPhee’s book, “The Control of Nature?” It is overwhelming in its arrogance and humility. ….Lady in Red
Lady in Red Says:

September 30, 2012 at 11:49 pm
I like Annie Leonard’s Story of Stuff… Her piece on Cap and Trade, however, is better:

http://www.storyofstuff.org/movies-all/story-of-cap-trade/

….Lady in Red

PS: Again, pls. Breathe. Exhale. Think. Wonder. … before you respond.
Paul Says:

October 1, 2012 at 10:48 am
”Of course, if I were presenting my results obtained here I couldn’t possibly write that title. It would have to be something else, like They can take my carbon from my (warm) dead hands: free market ideology and climate change skepticism. (I know, I know, free market ideology and gun nuttery don’t necessarily go together, but free market ideology slogans would be dry and boring).”
A much better one to link scepticism and the free market would be something like I’m in favour of buying things, but I don’t buy that.

”didn’t Steyn compare him to a paedophile recently?”
Steyn quoted someone else who (almost?) did, then said something to the effect that these comparisons was irrelevant, what mattered was whether the college was habitually shielding its high flying people from proper scrutiny. So the link to the paedophile coach isn’t entirely irrelevant, even though it’s a little tasteless [1].

”I mean, really, comparing published scientists to paedophiles?”
This seems to involve taking a vague memory of an action, then deciding it must have happened the way that suits your argument best and then using it. It’s poor form for a thread on confirmation bias.

As for stealing emails used in “Climategate” [2] I think reasonable people can agree that it’s a grey area. The WikiLeaks release of the US diplomatic communications was handled poorly (i.e. failure to anonymise informants details), and the information was taken without authorisation (and can accurately be described as an act of treason) but the ability to see inside the US governments working is highly valuable. It’s exposed a number of highly dubious (or just flat out immoral) actions on behalf of the US. Do we say that it being stolen means we shouldn’t act on the information in them? Do we say that the information in them excuses the way it was obtained? The same questions with a radically different gut answer apply to the UK phone hacking scandal. [3] “Climategate” falls somewhere between the two.

[1] I’d say very tasteless, but that college appears to have really covered up paedophilia on its campus to win a ball game. In paragraphs describing that sort of action everything else sort of moves down the offensiveness scale.

[2] I just want to note (again) that I hate attaching –gate as a suffix to indicate scandal. If we want to look at lazy, sloppy thinking let’s examine the mindset that does this.

[3] Unless your answer is “Stealing information is always wrong” in which case you’re at least consistent, but you’re also opposed to whistleblowers, which is generally regarded a bad thing.
Peter T Says:

October 1, 2012 at 12:28 pm
Faustus

Favourite coffee mug: Never Try to Teach a Pig to Sing. It Wastes Your Time and Annoys the Pig.
faustusnotes Says:

October 1, 2012 at 11:00 pm
Lady in Red, I don’t think you need to waste time explaining cap and trade here.

MrPete, it’s nice that you know some intelligent and educated skeptics. I know some skeptics whose conspiracy theories, personal behavior and political views would probably make your skin crawl. It doesn’t change anything about the broad scientific consensus. Also, arctic tree lines say nothing about the rate at which ice is declining (or snow-related albedo in the Northern hemisphere), and the possible damage that this could cause. And it’s not hubris to say that the rate is unprecedented if all available evidence suggests it is unprecedented.

Paul, I like your title. It needs a colon, though.

What Steyn actually said is

Not sure I’d have extended that metaphor all the way into the locker-room showers with quite the zeal Mr Simberg does, but he has a point.

which is basically agreement. If your argument is “one skeptic called Mann a paedophile, and another skeptic agreed with him” you haven’t advanced a very robust defense of my point. It’s obviously a metaphor, but it’s an offensive metaphor. Chaff bags are also a metaphor, and apparently the person who used that metaphor didn’t intend to say anyone should kill the prime minister, and the silly old coot may even genuinely believe that, but it doesn’t change the fact that the metaphor is repulsive and people shouldn’t speak this way about others in public life.

I approve of the use of “gate” in the recent “gategate” scandal, because a rich politician calling police plebs cause they won’t open a gate for him is exactly shallow enough to be assigned the “-gate” suffix, and it involved a gate. Otherwise I agree with you.

I think whether stealing data is justified in any case needs to be determined on a case-by-case basis, and obviously we only end up deciding that after the fact of the theft – e.g. if wikileaks had endangered the lives of US soldiers but turned up zero evidence of human rights abuses it would probably be viewed as a completely reckless act. There have now been multiple inquiries that have exonerated the scientists, and the WUWT crowd are now basically just throwing up private emails between scientists about their personal lives and sneering at them (there’s one post in WUWT where Mann mails a mate to point out that his hockey stick is the backdrop of a news conference. How is this advancing debate?) It’s nasty, and it has proven nothing. The “trick” phrase is common scientific parlance, the “decline” has been explained easily, and there is no evidence of fraud at all. So in hindsight, the theft was unjustified and has gained nothing except to further alienate the two “sides” from each other.

This kind of behavior from the skeptosphere is based on a predetermination that all the science is wrong and that anyone presenting it must be a fraud. They had zero evidence to justify that before the data was stolen, and they’ve not had their belief vindicated by the data. But they won’t give up. I don’t see how they can be convinced that fraud is not involved, and that these scientists are doing serious work. And remember, through all of this the skeptosphere has contributed precisely 0 to the scientific literature. It’s politicized grandstanding at its very worst.
faustusnotes Says:

October 1, 2012 at 11:01 pm
Good point, Peter T…
Carrick Says:

October 1, 2012 at 11:57 pm
faustus,

And remember, through all of this the skeptosphere has contributed precisely 0 to the scientific literature

What is your take on people who repeat obviously false statements after it’s been pointed out to them, what they said was false?

And a) I’m not referring to McIntyre, b) look in a mirror if you need a further clue, c) if you want proof about obviously statement is false try “google scholar,” d) if that doesn’t work, read some of the reactions in the climate gate emails to skeptics papers, e) and with that I’m out of here.

Since your’s is the superior intellect you should have no example finding counter examples to your claim.

Bye
Paul Says:

October 2, 2012 at 8:24 am
”which is basically agreement.”
Actually his point was: “If an institution is prepared to cover up systemic statutory rape of minors, what won’t it cover up?” which is an attack on the institution by claiming they’ll support paedophiles therefore they’ll support basically anything. The question of whether they’ll support anything is open (and I don’t agree it’s true), but the support for paedophiles is not open.

It’s sort of like hanging out with a really objectionable bunch of skeptics then claiming that you shouldn’t be judged by the company you keep. That principle can be applied to anyone at Penn State just as easily as anyone at WUWT.

“Chaff bags are also a metaphor, and apparently the person who used that metaphor didn’t intend to say anyone should kill the prime minister, and the silly old coot may even genuinely believe that, but it doesn’t change the fact that the metaphor is repulsive and people shouldn’t speak this way about others in public life.”
Yeah, threats to people’s lives or gloating about their death is really poor taste.
http://blogs.news.com.au/heraldsun/andrewbolt/index.php/heraldsun/comments/if_it_was_a_liberal_tweet_it_would_be_hate_speech/

I’ll accept Gategate as valid. If it turned out that that quote was manufactured I’d also accept Gategategate. But I’d probably draw the line there.

I partially disagree that data theft should be assessed only after the fact. It’s true that we can form a view of whether the theft was right or wrong based on the positive effects it has, but when the data is first stolen the thief has to assess whether to release it. That judgment can only be made based on whether you think it will help or not [1]. So for the climate emails, the best action is release them, query them, be proven wrong, apologise and say you’d still do it all again [4]. So still dwelling on them is the point that’s wrong, not the taking them in the first place.

Remember that even Monbiot’s initial reaction was that the emails were a disgrace. He only retracted that upon further reading and investigation.

[1] Barring the ability to see the perfectly future. In which case such a person should stop stuffing around with political scandals and instead amass great wealth through perfect betting, then use their nigh-infinite money to bring about whatever world they feel like. [2]

[2] I’ll assume that your failure to be rolling in hookers and blow [3] suggests that you don’t have perfect future predicting powers, despite your apparent belief in such a thing.

[3] I’m aware that some people may choose other things to buy. I’m unaware of any reason why they’d do so. 😉

[4] If you did something you believed was right, why the hell wouldn’t you do it again? (Assuming no additional data when making the decision)
faustusnotes Says:

October 3, 2012 at 4:43 pm
Paul, it’s a deliberate and inflammatory way of making that point, and it’s unfair on Mann, who doesn’t get to choose who investigates him or what other issues they have investigated. It’s classic poisoning-the-well argument. The point could have been made without saying “Mann is climate science’s Sandusky.” It’s rhetoric of the lowest kind.

It’s sort of like hanging out with a really objectionable bunch of skeptics then claiming that you shouldn’t be judged by the company you keep. That principle can be applied to anyone at Penn State just as easily as anyone at WUWT.

That really depends on whether the people in the science faculty at Penn State had any idea what was going on in the sports grounds, could have been reasonably expected to be aware of that when they started working there, and/or had any idea that disciplinary and investigative procedures would be corrupt. After all, most people commence work at their chosen institution assuming that it will be capable of reacting appropriately to and properly investigating major ongoing child abuse in any part of the organization. I think it’s reasonable to suppose that when someone applies for a job as a physicist at a uni, they are entitled to assume that a) child abuse isn’t happening extensively in the sports system and b) if it is, proper governance frameworks are in place to ensure that it is dealt with quickly and fairly. I don’t think we should all be expected to investigate these kinds of things before we accept a job, or on a regular (annual?) basis before deciding to remain in the job. In this respect, the comparison with Sandusky is completely ridiculous. We could equally well say that this means I can compare Jeremy Clarkson to Jimmy Savile. I mean, he works in the same organization, right, so he should be aware that it doesn’t properly protect children, and he should consider what would happen if an unsubstantiated claim of child abuse were ever made against him – obviously no investigation by the organization could exonerate him, so he’s better off leaving the BBC and working somewhere else. Like Mann should have from Penn State?

Is that really how people like Steyn think?

As for Andrew Bolt’s cute little post: “If it was a liberal tweet, it would be hate speech.” One would think that by now, having been found guilty of so many misrepresentations of other people on the basis of race, Bolt would be aware of the definition of “hate speech.” Wishing someone were dead is not “hate speech,” so it would not be “hate speech” no matter who said it. Of course, Bolt doesn’t understand basic issues of free speech – this is a man who thinks being forced to apologize for lying about people’s race and saying deliberately slanderous things about their behavior and motivations, without evidence, is “censorship.” But I have to wonder how people buy into his conservative persecution complex with such zeal. Bolt is one of the crew – along with Jones and Abbott – who have been getting away with a constant stream of bile for the past couple of years, including regular death threats, incitements to violence, and all manner of vicious and misogynist insults. Jones is Australia’s most influential radio broadcaster, Abbott is the leader of the Liberal party, and Bolt is a very popular columnist with his own television show. Where, oh where, is the censorship that so afflicts these conservative pundits? How is it, if this censorship is so one way that only Liberals get in trouble for hate speech, that Jones has been able to repeatedly make his chaff bag comments, call for the return of the guillotine, while Abbott has said Gillard “has a target on her head” and called for a “people’s revolt?” How is it if these people are so carefully monitored and repressed by a biased … something (what, exactly, or who, would define that tweet as “hate speech”?) Jones was able to refer to Muslims “infesting our shores” and call them vermin, and Bolt has been able to spend long years calling people idiots, liars, frauds, and talking up all manner of trouble with Aborigines? Bolt has never been censored for hate speech in any way, shape or form – neither have Abbott and Jones. So where does this “it would be hate speech” bullshit come from? That’s a rhetorical question, obviously, since Bolt is a liar and a scoundrel. What mystifies me is how people can nod along to broadcasts from Australia’s most popular radio personality, or tut tut in agreement to blog posts by a very popular columnist, about how they are being censored. Does the contradiction between statement and fact not occur to these dimwits?

I also wonder why this viciousness is a problem only on the conservative side of politics. There is a whole website devoted to cataloguing the nasty things Bolt says, its purpose described here, and even a brief jaunt through it will show that he is full of hate – and not just verbal hate either, the guy who started the website was outed by Bolt’s flying monkeys, then had to deal with a stalker and some credible threats to kill his cat. Bolt, Jones, Laws and Abbott are symptomatic of a serious problem in right-wing politics – at least in Australia – in which high-profile public figures think it’s okay to make death threats, jokes about murder and insurrection, and to regularly say demeaning, crude, insulting, vicious or threatening things about their opponents. Where is the left-wing equivalent of Alan Jones? When has anyone in high office or very visible in public life ever said that Howard should be dumped at sea in a Chaff bag? How does anything David Marr[1] does compare with Bolt’s misrepresentation of facts, attacks on scientists, and open lies about the motivations and characters of people he doesn’t like? And how can they have this ridiculous persecution complex when the right clearly dominates opinion-writing and media in Australia? (Even if you don’t accept “dominates” it is not possible to say that the right is unrepresented or poorly numerically represented when it has (just off the top of my head) Henderson, Devine, Sheehan, Bolt, Blair, Jones, Laws, Hadley, Akerman, Stemanic, and Kelly all writing or broadcasting from mainstream organizations). So how can they combine this persecution complex with this viciousness so effectively, and why do people believe this crap?

—
fn1: And wtf is it with the right’s obssession with David Marr? His most famous essay to date is his vicious attack on Kevin Rudd, who is a lefty. His work on Howard’s asylum policy was completely free of the kind of character assassination Bolt and Jones love. What is their problem with him?
willard (@nevaudit) Says:

October 3, 2012 at 10:39 pm
faustusnotes,

Thank you for your peaceful gesture.

I’m sure you noted that this:

> When dealing with psychological data it’s extremely unwise to infer fake data from outrageous responses.

has yet to be answered and that this lack of answer is quite a tell about the kind of pea and thimble game that is being played in the auditing sciences.

I believe this just reinforces Jonathan Haidt’s point:

> Anyone who values truth should stop worshiping reason. We all need to take a cold, hard look at the evidence and see reasoning for what it is.

http://opinionator.blogs.nytimes.com/2012/09/30/hope-for-reason/

While I still hope for reason (based on evidence like #5), climate blogland is becoming quite depressing, really.
willard (@nevaudit) Says:

October 3, 2012 at 11:35 pm
And why not indulge:

> I have no problem with the idea that ice is melting rapidly. However, “never seen before” is not an easy nut to crack.

Here’s a simple recipe:

> The Black Death in the middle ages is estimated to have killed more of Europe’s population than World War 2. This means that deaths during World War 2 were not unusual, and hence must be due to natural causes, not man-made.

https://www.tumblr.com/neverendingaudit/32753644033

In other words, this is an elementary fallacy.

***

Worse, see how this fallacy is being invoked to focus on your hyperbole “unprecedented”, which might not strictly being a scientific word in our context. See all the content if bypasses:

> You care about how scientists portray skeptics, but not how skeptics portray scientists. […]

> You claim that scientists “demonize” those who disagree with them […] but you don’t care that skeptics demonize scientists.

> You are worried about non-scientific NGO groups corrupting science in favour of AGW but don’t care about non-scientific NGO groups (including the tobacco industry) corrupting science if it opposes AGW. […]

> [T]he Wegman report, one of the most thoroughly discredited pieces of scientific malfeasance ever produced in connection with AGW.

> Mcintyre comments at WUWT, but never critically, and never even with the gentlest of corrections.

> [T]he IPCC reports are not scientific documents: they are policy documents. They present a summary of the state of the science for policy-makers, and on a scale that has never before been attempted. It’s inevitable that they will be flawed documents, the process by which they are made will necessarily be political, and they will be unsatisfactory to all involved.

This “unprecedented” trick is not unprecedented. Not only we’ve seen it with Mann, but we see it all the time. For instance, this is the same trick as Carrick did with what follows his admission that he was not “unsympathetic” with your point regarding what you call the “skeptosphere”, but immediately focusing on the most uncharitable interpretation of one your sentence and try to falsify it.

The fact that people who participate in the skeptosphere do publish papers does not counter the claim that the skeptosphere is quite good at doing what it does which has nothing to do with publishing papers, and that this modus operandi is quite “repugnant”, to borrow an adjective Steve deems proper to use in a scientific debate.

The whole MO is quite repugnant if only because it is based on double standard, the cardinal sin of any rational stance.

And even worse than that, a cursory look at CA’s recent posts should make any reader realize how Auditor reacts to what he perceives as double standards.
A.Scott Says:

October 4, 2012 at 3:45 am
Neither Steyn or the author he was writing about in any way associated Mann with pedophilia.

They both compared the actions by Penn State (or lack thereof) in investigating in the two cases.

There is a huge difference.

It is intellectually dishonest to try to claim either author attempted to link Mann with pedophilia.

You may well disagree whether there was sufficient investigation of Mann – many think there was not – and an honest review shows support, if not proof, there was plenty of reason for folks to believe it was largely a whitewash – that there was no real investigation.
willard (@nevaudit) Says:

October 4, 2012 at 8:58 am
> Neither Steyn or the author he was writing about in any way associated Mann with pedophilia.

I suppose it depends upon what “associate” means.

Here’s the first sentence of that article:

> In the wake of Louis Freeh’s report on Penn State’s complicity in serial rape, Rand Simberg writes of Unhappy Valley’s other scandal: […]

Here’s the second one:

> Not sure I’d have extended that metaphor all the way into the locker-room showers with quite the zeal Mr Simberg does, but he has a point.

The fourth and the fifth:

> And, when the East Anglia emails came out, Penn State felt obliged to “investigate” Professor Mann. Graham Spanier, the Penn State president forced to resign over Sandusky, was the same cove who investigated Mann. And, as with Sandusky and Paterno, the college declined to find one of its star names guilty of any wrongdoing.

The sixth:

> If an institution is prepared to cover up systemic statutory rape of minors, what won’t it cover up?

The seventh:

> Whether or not he’s “the Jerry Sandusky of climate change”, he remains the Michael Mann of climate change, in part because his “investigation” by a deeply corrupt administration was a joke.

This article has seven sentences.

http://www.nationalreview.com/corner/309442/football-and-hockey-mark-steyn

Speaking of association, here’s the wiki characterization of the expression “smear campaign”:

> A smear campaign, smear tactic or simply smear is a metaphor for activity that can harm an individual or group’s reputation by conflation with a stigmatized group.

Let the reader determine how accurate is the expression “link Mann with pedophilia” to judge this Steyn’s op-ed. [fn1]

That there is no “association in any way” might be going a bridge too far.

Maybe it’s just a vocabulary thing.

[fn1] Cf. this other op-ed:

https://www.tumblr.com/neverendingaudit/15511915533
Paul Says:

October 4, 2012 at 9:10 am
”That really depends on whether the people in the science faculty at Penn State had any idea what was going on in the sports grounds, could have been reasonably expected to be aware of that when they started working there, and/or had any idea that disciplinary and investigative procedures would be corrupt.”
At the individual’s insider level this is true. They worked at a place and assessed it based on the available evidence which gave them no reason to think they were hanging around with anyone with issues. Your point that the individual can assume their organisation isn’t hiding child molesters is a reasonable [1].

But from the outsider perspective, an individual can judge an institution based on the overall culture. And saying that there is a culture of cover-up in an institution can be a valid statement. This is a statement about cover-ups, not child molestation.

Take for example the Guardian report on the end of the British Empire in Kenya (was it Kenya?) that showed that the administrators there had deliberately and repeatedly committed atrocities and then attempted to hide them. Based on that, you alleged (and I agreed) that we could assume it was present in other countries where the British were withdrawing, but that the other locations had done a better job hiding them. Furthermore, we assumed that this was not restricted to the withdrawal phase, but was instead a problem that was present root and branch throughout the Empire’s duration. I’m unaware of proof of such systemic abuse in other locations and times, but the burden of proof to demonstrate this was not an institutional problem rests with the Foreign Office (given the evidence, especially the cover-up, suggests it’s an institutional problem).

So to raise a similar objection that Penn State hides problems due to institutional bias is valid and does need to reference the incident that demonstrates it[2]. Personally, I don’t think the claim would be borne out, but I accept it’s not unreasonable to make it. And I accept that because you and I have used the same principle on other situations before and a parallel exists.

Doesn’t mean I like the tone. But I can accept it as a valid complaint/question. The solution is to demonstrate that it’s unfounded. If you think that solution is too much work, then the institution should probably not cover up child abuse in the future [3].

”Where, oh where, is the censorship that so afflicts these conservative pundits?”
Well at a guess it rests in the fact that when one of them was incorrect on the facts of a case his article was read and its tone decided to be sarcastic. Based on that judges sarcasm detector the columnist was found guilty not of slander or libel but of a crime that so far had only applied to Christian pastors who quote the Quran. Or maybe it’s from the threat represented by the government inquiry that has heard ideas like “There should be a fit and proper persons test before you can own a media outlet” and the idea that your blog, when holding climate debates should be a regulated media outlet. An inquiry, furthermore, that is caused by an offense that occurred in another country. [4]

BTW I agree that Bolt’s definition of hate speech is at best weird (i.e. accepting ideologies as valid hate targets) and more probably just dead wrong. For reference, Jones’s comment on the PM was also not hate speech, it was just disgraceful. But to focus on the definition of hate speech rather than the actual content of disgraceful messages is just pointing to a strawman to avoid the real issue. [5]

[1] Though they probably want to keep an careful eye out if they’re working for a church.

[2] It’s possible to hold up a cover-up relating to child abuse without suggesting that everything they do is child abuse. It just demonstrates that the institution does cover ups.

[3] Churches, I’m looking at you. How much trust have you lost and what will it take to get it back?

[4] So it’s roughly like saying Sardusky is a child molester therefore ALF coaches should be investigated.

[5] I’m not going to respond any further to comments on Australian politics in this thread.
faustusnotes Says:

October 4, 2012 at 9:19 am
Carrick, why don’t you try this blog post by Tamino for starters as an example of the skeptosphere’s contribution to the scientific literature: a peer-reviewed journal set up by a skeptic approves a bullshit 7 page long article as a “brief communication” in five days, conveniently ensuring that an article denying sea level rise is published in time for the AR5 cut-off. The article is not only statistically extremely dubious, but it misrepresents the article it purports to reply to (including misrepresenting the purpose, conclusion and methods of that article) and it cites blog posts in its references.

This might also explain your point d) about “reactions in the climategate emails to skeptics papers.”
faustusnotes Says:

October 4, 2012 at 9:21 am
A Scott, the original article explicitly states:

Mann could be said to be the Jerry Sandusky of climate science

and Steyn says it “has a point.” You have to have some pretty serious skills with the English language to convert this into “neither Steyn or the author in any way associated mann with pedophilia.”

They may have a point about the investigative powers of the university, but that doesn’t mean they need to compare Mann directly to Sandusky. That is a rhetorical device, and it’s nasty.
Lady in Red Says:

October 4, 2012 at 9:39 am
Didn’t seem we got to a “good place, ever, did it, Faust?

I cited Ann Leonard, famed for “The Story of Stuff, on cap and trade because she is such a left-wing, eco…. ? ….sometimes to the point of irrationality.

I read “Mann could be said to be the Jerry Sandusky of climate science…” as Steyn writing that Michael Mann fucked science, fucked data, fucked integrity….

And, I thought it was witty. Period. ….Lady in Red
faustusnotes Says:

October 4, 2012 at 9:40 am
Paul, if Steyn had been interested in making the point that Penn U has poor investigative systems, he could have said so. It’s a good point to make, for the reasons you give. But he didn’t make the point in a neutral or polite way: he compared someone who is not a paedophile with a paedophile, and though he has no evidence of any fraud he compared Mann’s scientific work with paedophilia (“molested and tortured data.”) This is exactly what I mean about the tone of conservative debate, and how unhelpful it is. These articles are also wrong for another reason: Penn State U never held a board of investigation inquiry into Sandusky, whose actions were investigated by police. Senior figures in the university were alerted to police investigations but failed to remove Sandusky from roles involving children. But there was no equivalent action to the investigation of Mann. It’s comparing a bungled senior management response to a police investigation and some child protection failures within the university to a board of inquiry into an academic issue. Now, the issues may overlap, but there’s no reason to think they necessarily should. Conflating criminal conspiracies with allegations of academic misconduct or fraud is again inflammatory and deliberately rude.

Incidentally, Steyn is the genius who recently quoted Flashman approvingly as a model for how to handle the situation in Afghanistan. I think he seriously does not understand that Flashman is critical of foreign interventions, doesn’t agree with meddling in other countries, and his method for “Handling” the situation in Afghanistan was to run away, and the devil take the hindmost. Steyn finishes his column by suggesting Flashman as required reading for schools, and suggests “you can be politically correct or a great power – but not both.” He doesn’t seem to have understood that if Flashman were taken by its readers as a model of foreign policy, Britain would never have become a great power because Flashman is opposed to all forms of intervention, and is a rancid coward.

Steyn is not the sharpest knife in the cutlery drawer.

the columnist was found guilty not of slander or libel but of a crime that so far had only applied to Christian pastors who quote the Quran

and was not censored as a result, and in his judgment the judge clearly stated the need to balance freedom of speech issues with the need to be factually correct. Bolt’s column was not taken down, the judge did not order it to be taken down. He ordered Bolt to write a new post apologizing to the victims and correcting the facts. Is that really censorship[1]? The “fit and proper person” test is entirely legitimate, although in the UK at least it’s a paper tiger – Murdoch hasn’t been declared unfit, even though his newspaper spied illegally on the British head of state and interfered with criminal investigations into serious crimes. I wonder what one has to do to be judged unfit? And Bolt and Jones are still warbling their message of hatred on a daily basis, i.e. they have not been censored.

Agreed, let’s not introduce more Aussie politics into a thread that is rapidly warming out of control, but feel free to continue replying on the specific points (to avoid last-wordism).

—
fn1: I’m not sure I’m on board with enforced apologies, but I don’t think they can be called censorship.
Baker Says:

September 20, 2014 at 2:12 pm
Hi Faustus — Do you have any idea why Lewandowsky reported *identical* loadings for every item on his conspiracy variable? See his Table 1, where they report the same loading of .742 for all ten items on the main conspiracy factor.

They even say in the notes “The same loading is shown for all items that entered into a given composite variable.” Why? Is this a weight? How would a weight work in this context, and why not just report the actual loadings as they did for the other variables?
faustusnotes Says:

September 20, 2014 at 11:00 pm
Hi Baker, thanks for commenting. I don’t have any clue why that would have been done, but as you can see from my recent random kingdom-makings, I’m really busy and won’t have time to check out this point for a few days. Give me a few days and I’ll try and get back to it.
faustusnotes Says:

October 5, 2014 at 1:27 am
Hi Baker, I’m not sure if you’re going to see this response – it’s been a long time since you asked and I’ve been too busy and stupid to reply – but for the record, I had a look back at the paper to see if I could figure out your point. Unfortunately I discovered I don’t have a copy here, and the paper online appears to have changed (the link has a note at the top saying a newer version came out on 5-10-2013, I don’t know if that’s American or English dating). I don’t remember seeing equal loadings on all of one dimension, but maybe I just didn’t read the original paper well enough (this was a long time ago!) Anyway, I grabbed the copy of the paper on the UWA website, which I guess is the latest one, and I see what you mean. This does seem very strange. But I found this in the text of the results:

Items loading on each factor [<-this refers to the conspiracy factor and space aliens factors] were summed to form two composite manifest variables. The two composites thus estimate a conspiracist construct without any conceptual relation to the scientific issues under investigation.

This isn’t very clear but seems to be suggesting that they added together all the variables in that conspiracy group (all the variables in my Factor 2, basically) to form a single variable – I guess it would range in value from 0 to 4*the number of variables – and entered that as a single variable into the model. I really don’t think that they did that in the 2012 version of this paper – in that version I (vaguely, now!) recall different loadings for these variables, which I think they entered as a pre-specified single factor in the model. I’m confused as to how the loading in Table 1 came about, since they seem to imply in the results that the conspiracy items had their own loadings and then they summed those items together. But I think what they did was summed the variables together and put the resulting summed variables into a factor analysis (?) and are reporting the loading for the summed variables.

(It’s common practice in Psychology after you have identified a common factor to just add together the variables that are associated with that factor, not weighting them for their loadings, to just get a single composite score that is clinically easy to administer. In this case – if, for example, after the Climate Revolution, I were to use the Faustusnotes Scale to identify people who need to be reduced to Soylent Green, I would administer the Lewandowsky questionnaire and then add together a ~~victim’s~~ person’s answers to the questions on factor 2, without weighting for the loadings, and if it exceeded some arbitrary threshold, I would send them for ~~elimination~~ reeducation. It appears that Lewandowsky has done this for the entire set of conspiracy theory variables before putting them into a factor analysis with the space aliens questions.

This seems particularly terrible. I’m not sure why he would do this. Or why he would not add them together and report them in Table 1 as a single “conspiracy ideation score” with a single loading (there is nothing wrong, in principle, with this, though it’s not a great plan). My impression is that the second version of this paper was rewritten to focus on the Structural Equation Model, while the version I was re-analyzing in 2012 was focused on the factor analysis. So maybe in the new version they don’t really care about getting the exploratory factor analysis right…? Anyway, in my view the exploratory factor analysis was not actually exploratory in the first draft of the paper (it was confirmatory) and is not exploratory in this one either. That’s the best guess I can give. I also can’t make sense of some of the other results in this revised paper. For example, they say

For free-market items, a single factor comprising 5 items (all but FMNotEnvQual) accounted for 56.5% of the variance; the remaining item loaded on a second factor (17.7% variance) by itself and was therefore eliminated.

But what happened to the remaining 25% or so of variance? There are no items left to account for it. It’s really hard to understand this revised paper, but I remember the original one being a classic factor analysis with a single twist. I think the original paper was better …

At the risk of sounding arrogant, I would suggest you privilege my analysis over Lewandowsky’s. Mine shows a very intuitive and reasonable finding: that people who don’t accept global warming tend to be free marketeers, and that there is not a particularly strong conspiracy streak otherwise. A few days reading WUWT will show you the truth of this – they have communists, christians, atheists and libertarians ranting over there. But one thing that is noticeable is that while not a great many denialists are libertarian, almost every libertarian is a denialist. Go figure.

One other thing I’ll say on the conspiracy ideation angle is that when this paper came out there was a huge hullabaloo about how misleading the title was. About a year later WUWT put up a post on the anniversary of the moon landing and at least TWO moon-landing denialist AGW denialists started a huge stoush there. (there was also a geocentrist on that thread!!) TWO may not seem like a big number but moon-landing denialists are very very rare. So maybe – though the evidence from the data didn’t support it – the inflammatory title of the paper wasn’t so far off the mark after all …

And finally, I would like to reiterate that I really hate the way this paper is laid out, with most of the methods in the results section, a really obscurantist turn of phrase, and a lot of information not clearly stated. If you’re used to using Factor Analysis for scale development it’s easy enough to infer what they did (at least in the original paper, where I was able to fairly easily replicate their original results) but it would be really good if this type of research were more clearly laid out and explained. I guess that’s why it’s not in a high impact factor journal …