by Judith Curry
This list will help non-scientists to interrogate advisers and to grasp the limitations of evidence – William J. Sutherland, David Spiegelhalter and Mark A. Burgman.
Nature has published a very interesting comment, titled Twenty tips for interpreting scientific evidence. Excerpts:
Perhaps we could teach science to politicians? It is an attractive idea, but which busy politician has sufficient time? The research relevant to the topic of the day is interpreted for them by advisers or external advocates.
In this context, we suggest that the immediate priority is to improve policy-makers’ understanding of the imperfect nature of science. The essential skills are to be able to intelligently interrogate experts and advisers, and to understand the quality, limitations and biases of evidence.
To this end, we suggest 20 concepts that should be part of the education of civil servants, politicians, policy advisers and journalists — and anyone else who may have to interact with science or scientists. Politicians with a healthy scepticism of scientific advocates might simply prefer to arm themselves with this critical set of knowledge.
Differences and chance cause variation. The real world varies unpredictably. Science is mostly about discovering what causes the patterns we see. Why is it hotter this decade than last? There are many explanations for such trends, so the main challenge of research is teasing apart the importance of the process of interest from the innumerable other sources of variation.
No measurement is exact. Practically all measurements have some error. If the measurement process were repeated, one might record a different result. In some cases, the measurement error might be large compared with real differences. Results should be presented with a precision that is appropriate for the associated error, to avoid implying an unjustified degree of accuracy.
Bias is rife. Experimental design or measuring devices may produce atypical results in a given direction. Confirmation bias arises when scientists find evidence for a favoured theory and then become insufficiently critical of their own results, or cease searching for contrary evidence.
Bigger is usually better for sample size. The average taken from a large number of observations will usually be more informative than the average taken from a smaller number of observations. That is, as we accumulate evidence, our knowledge improves. This is especially important when studies are clouded by substantial amounts of natural variation and measurement error.
Correlation does not imply causation. It is tempting to assume that one pattern causes another. However, the correlation might be coincidental, or it might be a result of both patterns being caused by a third factor — a ‘confounding’ or ‘lurking’ variable.
Regression to the mean can mislead. Extreme patterns in data are likely to be, at least in part, anomalies attributable to chance or error.
Extrapolating beyond the data is risky. Patterns found within a given range do not necessarily apply outside that range.
Scientists are human. Scientists have a vested interest in promoting their work, often for status and further research funding, although sometimes for direct financial gain. This can lead to selective reporting of results and occasionally, exaggeration. Peer review is not infallible: journal editors might favour positive findings and newsworthiness. Multiple, independent sources of evidence and replication are much more convincing.
Feelings influence risk perception. Broadly, risk can be thought of as the likelihood of an event occurring in some time frame, multiplied by the consequences should the event occur. People’s risk perception is influenced disproportionately by many things, including the rarity of the event, how much control they believe they have, the adverseness of the outcomes, and whether the risk is voluntarily or not.
Data can be dredged or cherry picked. Evidence can be arranged to support one point of view. The question to ask is: ‘What am I not being told?’
JC comments: I really like the idea behind this article:
What we offer is a simple list of ideas that could help decision-makers to parse how evidence can contribute to a decision, and potentially to avoid undue influence by those with vested interests.
I suspect this article will not be appreciated by scientists who are playing power politics with their expertise, or by advocates promoting scientism with cherry-picked evidence.
I picked 10 of the 20 tips that I thought were of greatest relevance to the climate change debate. So, what do you think of this list? What would you add?
If we apply Popper’s falsifiability to the assumptions?
”why is hotter this decade than last?”
that’s a loaded question, this decade is as hot as the last one, no more / no less. Nobody monitors on the whole planet – nobody knows the precise temp to save his life
Well, you’re right, it’s not very precise but you don’t wanna go above 107 deg F or below 90 deg F for very long.
=============
kim, I did record 108F once, while in hospital with multiple fractures and multiple internal haemorrhages, but the nurses decided that that was impossible and threw away the thermometer. The fact that they also placed me near an open window in sub-freezing temperatures perhaps helped my own temperature to fall.
Kim, and will not! if one takes the Warmist & skeptics by time traveling machine to 2100 and they see that is same temp as today – when they come back – they will be still spieling the same crap
“Considering that the climate models are the only support for the AGW premise, and the AGW premise is the only support for the climate models, exposing this simple fabrication is all that needs to be done to put an end to this circular argument that forms the basis for the entire climate change lunacy.” (Norm Kalmanovitch)
Just wrong, plain wrong.
Do some homework for a change.
Anyone that is incapable of even taking the statement with ‘a grain of salt’ without resorting to ad hom attacks cannot be expected to capable of maintaining the degree of skepticism that science requires.
I just disagree with the statement that climate models are the only support for the AGW premise as the instrumental record for one is evidence for AGW, secondly I made no ad hominem attack towards either Wagathon or Norm Kalmanovitch but since you asked.
Wagathon constantly rants against progressive principles therefore his opinions mean nothing.
there, satisfied?
Or should I take everything you say with a grain of salt?
The MWP and the LIA existed. The IPCC’s ‘progressive’ principles were on display for all to see when they showcased Mann’s ‘hockey stick’ to support its case that the productive were guilty of causing global warming. It’s not my opinion that the ‘hockey stick’ is scientific fraud. It’s a fact. But the Left doesn’t care about truth
“the instrumental record for one is evidence for AGW”
No, at best it may be evidence for GW. All the instrument tells you it that the readings were higher or lower. It is silent on the causes. For that you need a model (in the statistical sense) of some sort.
And as for models, Mike Flynn, as Mandy Rice-Davies famously said, “Well, they would say that, wouldn’t they?”
Mike Flynn,
You are right that the instrumental record is not the whole case but it is evidence for AGW, along with the rise in CO2 and the sources of that rise in CO2 and the fact that CO2 emits infared radiation make a pretty good case for AGW.
And its all models, all the way down.
And did you notice I said emits rather than absorbs?
The MWP and the LIA existed!!!!!!!
This instrumented Warming is well inside the Bounds of those last two warmings and there are many proxy and historical records to support this as fact.
AGW is not supported by data that stays inside historic bounds.
That is not reasonable!!!
Wagaton, the only reason the Warmist are succeeding is because: the skeptics don’t want to acknowledge that: LIA, MWP were only on the northern hemisphere – the whole planet overall cannot get warmer or colder; because oxygen & nitrogen by shrinking when colder / expanding instantly when gets warmer than normal, are regulating overall temp to be always the same = Skeptic’s fault!
Abstract. Increasing paleoclimatic evidence suggests that the Little Ice Age (LIA) was a global climate change event. Understanding the forcings and associated climate system feedbacks of the LIA is made difficult by the scarcity of Southern Hemisphere paleoclimate records. We use a new glaciochemical record of a coastal ice core from Mt. Erebus Saddle, Antarctica, to reconstruct atmospheric and oceanic conditions in the Ross Sea sector of Antarctica over the past five centuries…
Wagathon | November 21, 2013 at 10:29 pm said: ” Understanding the forcings and associated climate system feedbacks of the LIA is made difficult by the scarcity of Southern Hemisphere paleoclimate records. We use a new glaciochemical record of a coastal ice core from Mt. Erebus Saddle, Antarctica, to reconstruct atmospheric and oceanic conditions in the Ross Sea sector of Antarctica over the past five centuries…;;
Wagaton, all ”proxy” climate records are: they find what they want to find, because the truth is irrelevant for proxy – proxy is Pagan belief same as warmer in 2100, religion, not facts
If LIA was global – you have to recognize that: when oxygen & nitrogen get colder -> troposphere instantly shrink -> release less heat and equalizes in a jiffy . Only the ”Skeptic’s ”pagan beliefs are keeping the Warmist to flourish. The truth will eventually win.
wiki:
Although it only provides anecdotal evidence, in 1675, the Spanish explorer Antonio de Vea entered San Rafael Lagoon through Río Témpanos (Spanish for “Ice Floe River”) without mentioning any ice floe and stated that the San Rafael Glacier did not reach far into the lagoon. In 1766, another expedition noticed that the glacier did reach the lagoon and calved into large icebergs. Hans Steffen visited the area in 1898, noticing that the glacier penetrated far into the lagoon. Such historical records indicate a general cooling in the area between 1675 and 1898, and “The recognition of the LIA in northern Patagonia, through the use of documentary sources, provides important, independent evidence for the occurrence of this phenomenon in the region.”[62] As of 2001, the border of the glacier has significantly retreated compared to the borders of 1675.[62]
Wagathon | November 21, 2013 at 11:40 pm said: ”Although it only provides anecdotal evidence, in 1675”
Wagaton, because of some ”anecdotal evidences” ,the Warmist are ripping off billions of dollars – because the ”Skeptics” can’t admit that anecdotal crap is all crap. Glaciers are getting bigger when is more moisture available; same as: when is more moisture in the air = more rain. Glaciers are freeze-drying the ”available” moisture from the air.
2] two years ago river Danube was frozen for 2 weeks, that doesn’t mean that was ice age on the WHOLE planet. Wagaton, only the proxy crap is making the Warmist successful, shame skeptics, shame…
“Several studies, such as Lamb 1965, Grove 1996, 2001, and Ogilvie et al. 2001 have suggested that the mediaeval warm period and the little ice age were climatic anomalies on a global scale and not merely regional phenomena.” ~MoB
Such a finding should not come as a big surprise. It’s the Sun, stupid.
@bob droegen
“…. as the instrumental record for one is evidence for AGW.”
As Mike Flynn pointed out it is, at best, evidence of GW; evidence for the ‘A’ part is pretty thin on the ground.
My question, and it is an actual question, is: given all the adjustments that have been made to the instrumental record, and there have been multiple passes, does the instrumental record, meaning the actual, original data output of the actual instruments in place at the time, still exist? Or is the ‘corrected’ data all we have left?
bob droege | November 21, 2013 at 8:44 am |
“the fact that CO2 emits infared radiation make a pretty good case for AGW”
Can you show me experimental evidence that a liquid body of water free to evaporate can be heated from above by 12 micrometer radiation? If not then over 70% of the greenhouse effect is unproven speculation.
Good luck, Bob. I’ve yet to find such an experiment and I’ve been googling and asking around in the climate change blogosphere for a couple of years now.
Wagathon | November 22, 2013 at 10:41 am said:’“Several studies, such as Lamb 1965, Grove 1996, 2001”
Wagaton, Hubert Lam was one of the first conmen – in the 60-70’s he was promoting: ice age by year 2000, because of CO2 dimming effect. I hope you have defrosted from his ”ice age”
Wagaton, con job goes with the climatology profession; they can’t produce something to package and sell it in the shop, or for export – instead they have very fertile imagination – inventing crap. Global warming for year 2100 is same as Hubert’s ice age for year 2000, only it costs much more,.
only I’m correct: warming, or cooling of the WHOLE planet is not possible for more than a day – because oxygen &nitrogen expand extra, INSTANTLY -> troposphere release more heat and equalize in a jiffy!!! http://globalwarmingdenier.wordpress.com/climate/
You would make a good point there but for the fact that we live on a water world.
Herman Alexander Pope | November 21, 2013 at 10:20 am said:
The MWP and the LIA existed!!!!!!!
Herman MWP and LIA existed only on the northern hemisphere – when was warmer there – on the southern hemisphere was colder, to balance
Cannot get colder on both hemispheres simultaneously for more than a day – because oxygen & nitrogen (troposphere) shrinks and equalizes in a jiffy: http://globalwarmingdenier.wordpress.com/2012/08/25/skeptics-stinky-skeletons-from-their-closet/
Wagathon | November 23, 2013 at 11:35 pm said: ”You would make a good point there but for the fact that we live on a water world”
Wagaton, water makes ”milder climate” not warmer, or colder, but as a shock-absorber = cooler days / warmer nights.
Absence of water makes hotter days, but colder nights those two factors cancel each other; if the D/H were recording the temperature for every minute in 24h, would have seen that overall is same temp in the desert and the jungle.
Wagaton, you should be interested what water vapor really does: http://globalwarmingdenier.wordpress.com/water-vapor-h2o/
The climate alarmists’ job is really tough but what makes it easier for them is that us common folks have rightly lived in a state of fear our entire lives that global temperatures would actually hit a plateau. And, guess what? They did!
It would be nice to know what advice the authors have concerning a claimed consensus?
How about:
Identification of a problem is not the same as identification of a solution.
One other thing. There’s a difference between the truth, the whole truth, and nothing but the truth. Keep that in mind when somebody tries to claim that xxx is ‘basic physics’.
Everyone talking to a politician has an agenda. We are so used to worrying about the agendas of politicians that we forget this.
That’s really good, Tom. we do need reminding.
I would add to that list a lack of ethics in some scientists behabior. Scientist are smart enough to notice if their claims are appropriate or not. In my doc:
https://docs.google.com/file/d/0B4r_7eooq1u2VHpYemRBV3FQRjA
it is shown that some “scientific” claims of IPCC’s climate change are not appropriate. In this case we just have to contact the responsibles (the leading authors of the WGI AR5 key chapters) and challenge them.
This challenging thing should be done by an independent scientific council (or by a judge) (of course not by an individual) in an open-to-the-public trial.
Yes, what they are doing is a crime and it should be tried in a Texas Court.
If a Texas Corporation understand that IPCC’s manipulation of science acts against their corporative interests, I guess that they have the right to sue those scientists responsible of WGI AR5.
But I was thinking in trials like that one in the UK (in October 2007) where the High Court ruled on the teaching of Al Gore’s documentary: http://news.bbc.co.uk/2/hi/7037671.stm
I do not intend to imprison IPCC’s responsibles of science manipulation. I will be happy enough if an independent scientific council (or a judge) challenge them in an open-to-the-public trial.
I thought this one was also very relevant: Seek replication, not pseudoreplication
Especially in light of the refusal by some to share data and methods.
This is the beginning of the article I read and one of the best points imho anyway.
“One of the big problems in science journalism is the tendency to hype scientific research. You’re familiar with the routine: A new study comes out on, say, how coffee might lead to a slight increase in a particular disease. Then, plastered all over the front pages of websites and newspapers are headlines like, “Too Much Coffee Will Kill You!” Of course, the following week, a different study will report that coffee might protect you from another disease, and the media hysteria plays out all over again, just in the opposite direction.
This is bad. Poor science journalism misleads the public and policymakers. Is there a way to prevent such hype?”
The older we get the more this rings true.
“Poor science journalism misleads the public and policymakers. Is there a way to prevent such hype?”
Yeah, make all news free.
This is one of the effects of data dredging. All sorts of spurious correlations get reported, while paying scant regard to any sort of causal mechanism – or lack of.
I think it’s a case of too many people chasing after too few research grants.
I think it is a case of scientific illeteracy – both among journalist and the public.
“Science Journalism” can be manipulated, and has been. Here is an early marketing game plan from Futerra:
Futerra. “New Rules: New Game.” futerra, October 12, 2006. http://www.futerra.co.uk/downloads/NewRules_NewGame.pdf
“These short rules are communications techniques which pull together the most effective strategies for changing people’s behaviour. They are based on a huge body of international psychological, sociological and marketing studies, gathered and analysed by Futerra. We’ve taken great concepts with terrible titles like ‘psychological reactance’ and ‘symbolic self-completion’ and translated them into simple-to-use communications tools to motivate behaviour change.”
An updated version of “The Rules of the Game” can be found in the ClimateGate FOIA2009 zip file.
I think this is a good checklist. I would add, watch out for the influence of “groupthink,” as mentioned in the MAD posting.
1. Best available knowledge is not the same as adequate knowledge.
2. There is no reason why you should know something. There may be reasons why you should want to know.
3. If human nature has always prompted scholars and intellectuals to cleave to inadequate theories for the sake of having something to cleave to, it is likely that scholars and intellectuals are doing the same things right now. (This dictum depends on the assumption that human nature has not changed.)
4. The assumption that human nature has not changed is the most reliable of all assumptions. All other assumptions are less reliable than that one.
mosomoso – In the context as described by JC, you are on the right track. A non-scientific politician has no chance of being able to work anything out from the science. Even a scientific one is likely to be out of their depth very quickly. Some issues may be testable against plain common sense, but in general what the politicians most need to rely on (apart from their advisers) is their knowledge of human nature. The first rule is Do No Harm. The first question is Cui Bono (who benefits). The next question is of the people putting a science-based case – are they behaving like dependable people, eg, are they prepared to discuss the issue openly with opponents and are they open to their theories being tested, or are they using bullying tactics, trying to silence opponents, using ad homs, etc? It isn’t easy, and it would be a mistake to portray it as being easy.
moso and Mike, nice to see applied commonsense.
moso, look for me at the Gabba tomorow, below the big screen, in black and white stripes, laughing as England power ahead.
Faustino, I suspected you’d have on a black and white striped apron – and be pushing a food trolley to cater to the diva demands of the English cricketers. Does KP like his crust on or off? Don’t get it wrong!
Faustino
Black and white stripes??!!
Its about time we had a cricket thread instead of wasting our time on all this science stuff
tonyb
Good list and a good place to discuss. Its seems that data interpretation issues makes up the majority of the list and the others would be generally applicable to any form of communication but especially advertising in the general population domain.
“Bigger is usually better for sample size. The average taken from a large number of observations will usually be more informative than the average taken from a smaller number of observations. That is, as we accumulate evidence, our knowledge improves.”
Correct as far as this statement goes. HOWEVER our knowledge can never improve beyond the basic knowledge/accuracy of the measuring instrument. A million readings of any thing with 1% accuracy is still 1% accurate.
Yes, engineers do routinely increase the “signal to noise ratio” SNR in a system by averaging multiple measurements of a single signal with a single measuring device (a voltage meter for example). This is subject to strict limits, the noise to be reduced must be stochastic (fixed in nature WRT frequency/amplitude) and the sampling frequency and averaging duration must be carefully selected (generally at a lower frequency) to reduce the noise.
If you do not observe these limitations you can actually make the SNR worse.
Also, this averaging of large numbers of samples does not increase the accuracy, if you average a million readings from a 1% accurate voltage meter you will NOT reach the accuracy provided by a 0.1% voltage meter. There will be less noise in your averaged data set, but it is still anywhere from -1% to +1% away from “truth”. A single reading from a meter with 0.1% accuracy will be closer to “truth” than a million readings from a 1% accurate meter.
Cheers, Kevin
http://rankexploits.com/musings/2009/rounding-of-individual-measurements-in-an-average/
Steven, in engineering and machining “accuracy” and “precision” have very specific meanings that are distinct and in no way equal. Lot’s and lot’s of digits is NOT the same thing as accuracy (or knowledge of truth).
This is engineering 101 and you should be glad that engineers apply it rigorously so that plane you get on can actually take off and land safely more than 99.999% of the time.
Rounding is a totally different topic than accuracy and is also covered in engineering 101.
Billions of temperature records taken with thermometers with a basic accuracy of 1 degree are only accurate to 1 degree regardless of any “massaging” of the data.
Cheers, Kevin.
Mosh and Kevin
It appears to me that it is a common “logic trap” to believe that thousands of readings of thermometers (or satellite altimetry or anything else) with an accuracy of +/-X will be inherently more accurate than +/-X just because there are more readings.
Max
And, going further still, accuracy is not the same thing as correctness. It’s possible to be wrong to 100 decimal places.
Steven Mosher,
There is a difference between averaging and instrumentation error. You are assuming that instrumentation error is uniform over the measurement range and/or the error is equally distributed. Life’s not that simple.
Take the change from LIQ to digital readings. The LIQ has a different mean error than digital. The closer you get to a lower or upper limit the more the bias changes. Digital instruments have some power source like a battery that preforms best at its rated temperature so the error is based on that temperature. In the Arctic/Antarctic in winter the normal digital thermometer would have a warm bias, they only go so low. Remember Anthony Watts noting that Alaska should have had a new low temperature record but the instrument bottomed out.
UAH is at its limits near the poles also, it will have a bias. GISS adds more high latitude stations and interpolates, it will have a bias. The trick is determining the real bias/uncertainty, not playing with a spreadsheet.
Steve, imagine that of the 1000 thermometers you have X come from Acme, Y come from GE, Z come from Edison and DN come from miscellaneous sources.
Now;
the Acme thermometers have a normal distribution around the mean
the GE thermometers have a Poisson distribution that biases them to read low.
the Edison thermometers have a very narrow distribution at manufacture, but there is a drift of 0.1 p/a as the sliver slowly oxidizes.
The DN thermometers have no documentation.
You have 1,000 readings, but you have no idea of the ratio of X,Y,Z and DN, nor do you know how often the units are changed.
“Steven, in engineering and machining “accuracy” and “precision” have very specific meanings that are distinct and in no way equal. Lot’s and lot’s of digits is NOT the same thing as accuracy (or knowledge of truth)”
who ever said they were.
I think you’re missing my point in posting the link.
let’s see if any einstein here can get it. I doubt it. But give it a go.
Hint. dont view my link as
1. a counter argument
2. an endorsement.
you’ll have to think. this is a test
Kevin,
There is a basic difference between resolution and accuracy. If a thermometer can only resolve to 1 degree, or even if a single thermometer is only accurate to 1 degree, then if a large number of thermometers read both high and low for their bias (i.e., the bias is not one sided), a large number of reading averaged from different thermometers will give greater accuracy (by the square root of N thermometers). The key is presence or absence of average bias, not resolution or individual accuracy.
Leonard,
“a large number of reading averaged from different thermometers will give greater accuracy”
NO, NO, NO;
Corrected text follows;
“a large number of reading averaged from different thermometers MIGHT (if all assumptions about statistical distributions agree with real world measurements, and errors are distributed as assumed a-priori) give greater accuracy”
MIGHT, if, and only if, all the statistical assumptions are correct.
When I fly on a plane I don’t give a darn about all the assumptions being correct. I want to know for sure that all the measurements used to build said plane are known to match “truth” (as defined by rigorous standards bodies).
I do not care that “by averaging” only 1 of 1000 planes is “likely to bee too weak to actually fly”. I want to know that every plane certified to fly is strong enough to do so.
This is the fundamental difference between engineering and climate “science”. And engineer looks for all the reasons some real object may fail, a climate “scientist” looks for all the reasons a theory “may” be correct.
I prefer planes that really can fly, instead of planes that “might possibly fly”, if all the statistical assumptions are just right, and the models “predict” it can fly.
Cheers, Kevin.
Kevin,
My favorite analogy for explaining the difference is that of the Marine Forward Observer.
He has at his disposal weapons systems that can hit from 5 to 0.5 meters of the designated coordinates. That is precision.
Making sure those coordinates are correct is accuracy. Often an impercise 25 meter circular impact error on the target is preferrable to 0.5 meters. Accuracy trumps precision.
This does not mean that one should not pursue both. The ultimate objective is tying precision to accuracy.
A relevant example is determining changes in temperature. Assuming we have 10000 equally equipped measuring sites read with the same procedure, the accuracy of the determination of the temperature change is better by almost a factor of 100 for the average of those sites than it is for any single site.
The only reason that it’s not better by exactly the factor of 100 is that thermometers have a small error also in the coefficient that relates the change in reading to the change in temperature. This error is, however, in most cases much smaller than the error in the value of the temperature itself (i.e. the rise in the temperature needed to raise the reading from just reaching 20C to just reaching 21C is usually much closer to 1C than the full values 20C and 21C are to the real temperatures).
“Assuming we have 10000 equally equipped measuring sites”
Which we do not. This is where I have a philosophical problem with Cowtan and Way; what is being measured when you measure land temperature, SST and the microwave emission of atmospheric oxygen?
Cowtan and Way is a different case. It adds to HadCRUT4 some information from other empirical sources plus a model. That kind of additions surely reduce the accuracy meaning that the HadCRUT4 is more accurate for what it covers than C&W is for the full coverage.
Yep, this point could have been better expressed. Selection of an appropriate sample is more important than gross numbers, at least in the field that I know something about, which is human population surveys.
Two of the 20 points of the original article are related to your concern (Controls are important and Randomization avoids bias). They were in the other half left out by Judith as less applicable to climate science.
Not for human population surveys, Pekka. When you poll the public you can’t have a control group in any meaningful sense; and getting a representative sample population is the opposite of randomisation.
I have no idea how large and effect this has, but I have seen it, particularly in iterative floating point calculations.
Nova, Joanne. “WARNING: Using a Different Computer Could Change the Climate Catastrophe.” Scientific. JoNova: Science, Carbon, Climate and Tax, July 28, 2013. http://joannenova.com.au/2013/07/oops-same-climate-models-produce-different-results-on-different-computers/
“When the same model code with the same data is run in a different computing environment (hardware, operating system, compiler, libraries, optimizer), the results can differ significantly. So even if reviewers or critics obtained a climate model, they could not replicate the results without knowing exactly what computing environment the model was originally run in.
“This raises that telling question: What kind of planet do we live on? Do we have a Intel Earth or an IBM one? It matters. They get different weather, apparently.
“There is a chaotic element (or two) involved, and the famous random butterfly effect on the planet’s surface is also mirrored in the way the code is handled. There is a binary butterfly effect. But don’t for a moment think that this “mirroring” is useful: these are different butterflies, and two random events don’t produce order, they produce chaos squared.”
I think of it as a “significant figure” barrier. Your results can only be as accurate as your least significant factor. No exceptions.
“JC comments: I really like the idea behind this article:
What we offer is a simple list of ideas that could help decision-makers to parse how evidence can contribute to a decision, and potentially to avoid undue influence by those with vested interests.
I suspect this article will not be appreciated by scientists who are playing power politics with their expertise, or by advocates promoting scientism with cherry-picked evidence.”
Touché. The ‘electorate’ is the real ‘battle ground’. However, the education of the ‘policy makers’ in ‘science etiquette’ is also of primary importance. Would they understand the implication behind “decision-makers to parse how evidence can contribute to a decision”? I doubt it.
Best regards, Ray Dart.
(data not shown)
Experiment not done
“a typical image is shown in figure 2”
The best image I have ever taken is shown in figure 2.
“n=7”
We did it 12 times, but 5 of the runs gave us the wrong answer.
“It can be clearly seen”
Squint, in low light, and look out the corner of your eye
“Samples were taken every hour for analysis”
Ignore the fact that our methods section clearly show we need 90 minutes to run each sample before starting the next.
(Personal correspondence with lead author)
The old fraud was drunk at the San Francisco meeting and admitted his work was bollocks
“All authors contributed equally”
The Ph.D. student did all the experiments and the post-Doc wrote it up. The stats were done by the boss who managed to get a p<0.05 by rounding and throwing away two outliners.
"In light of preliminary experiments we"
We tried everything we could think of to get this result, who would have guessed that microwaving the buffers* would work!
*Have actually done this.
Thanks, Doc, I’m sure that reading that has been curative.
Doc
You missed ‘it is widely accepted’
we couldn’t locate the reference
“We show for the first time”
We have never read a paper that isn’t online and all the work that was done up to 1985 might as well have never existed.
No offense, but anyone who needs to be told any of those 20 tips shouldn’t be making policy anyway.
Edit: “whether the risk is [taken] voluntarily or not”
#11 – Causes precede (and cannot follow) effects.
(Jim2)And we need to keep in mind pathological science.
From the link:
Irving Langmuir1 has identified several recurring patterns in cases of pathological science:
The maximum effect that is observed is produced by a causative agent of barely detectable intensity, and the magnitude of the effect is substantially independent of the intensity of the cause.
The effect is of a magnitude that remains close to the limit of detectability, or many measurements are necessary because of the very low statistical significance of the results.
Theories outside the field’s paradigm are suggested.
Criticisms are met by ad hoc excuses thought up on the spur of the moment.
The ratio of supporters to critics rises and then falls gradually to oblivion.
To these we may add the following:
The remarkable result is specific for a “special” system.
Some special technique or equipment is involved.
The result requires a stunning departure from the paradigms that fully determine results in all other comparable systems, including those studied by the authors. — N.J.T.
http://www.columbia.edu/cu/21stC/issue-3.4/turro.sb1.html
And that was included in:
http://www.columbia.edu/cu/21stC/issue-3.4/turro.html
More from that article:
Advice for the working revolutionist
Clearly, scientific progress would be impossible if researchers always played it safe within a dominant paradigm, discarding disturbing results or shying away from daring hypotheses. Some of today’s most robust discoveries and most promising research subjects–manned space flight, wave-particle duality, C60 (buckminsterfullerene or “buckyball”) molecules, high-temperature superconductivity, ad infinitum–once struck mainstream scientific opinion as completely implausible. Working researchers have practical steps they can take to lower the chances that today’s “eureka!” will be tomorrow’s Ig Nobel:
Always generate and test several plausible hypotheses to explain a result.
Use imaginative experimental design to increase objectivities and decrease the chances that the initial observation contains artifacts.
Let the best available paradigm be your guide, until you’re certain that your results require revision of the paradigm.
Be conservative about the concepts of statistical significance and margin of error, especially when analyzing phenomena on the threshold between signal and noise.
Reproduce, reproduce, reproduce.
Discuss surprising findings openly with peers (through both formal and informal channels, inside and outside one’s own specialty), and make constructive use of the critiques that arise.
When discussing research with non-scientists–especially those holding microphones, cameras, notebooks, or checkbooks–avoid the temptations to overinterpret results, oversimplify your explanations, or promise the moon in practical applications.
If further studies falsify your hypothesis, acknowledge it with grace and learn from the experience. Blind leads are nothing to be ashamed of; they are inseparable from the progress of science. Any number of pathological investigations give way eventually to one like quantum mechanics–which necessitated a few adjustments to the law of conservation of mass but ultimately withstood criticism, explained results that Newtonian theory couldn’t explain, and revolutionized physics. The same communal corrective processes that falsified one theory verified the other; that’s how science operates and why it almost always works.
Do the unthinkable: Try your very best to find faults in your experiment or to falsify your interpretation. If this is done fairly, objectively, and passionately, even if you turn out to be wrong, you will be true to your science, and you will be admired by the community for your intellectual courage and dedication to the scientific ethos.
Jim(2), pretty much everything you have listed there could be applied to the field of xenohormone research.
We have to get rid of all sorts of xenoestrogens/androgens to observe effects; so most plastic pippet tips are out. The effects you see are subtle and typically you need to remove all the natural hormones in the system to see a reproducible effect.
Thing is, hormone mimitics are real, we just don’t know how much harm they might be doing.
It kind of sounds like the effect of CO2 in a sea of water vapor, Doc. I mean, if you have to eliminate all of the natural hormone, how big can the effect of the mimics be?
I like that, CO2 vs Water Vapor – WV wins I’m thinking.
Doc;
Sounds cruel: “throwing away two outliners”. Because they lie out beyond the margins? :p
Brian H
Throwing away two outliers from today’s paradigm could be tossing out the Black Swan, which will become tomorrow’s paradigm.
Max.
:)
I killed a Golden Goose because it laid Carbon Eggs.
================
Um, read carefully. “outliners”
Measurements and estimates are completely different things.
Jim Cripwell
You know the difference between “estimates” and “measurements”, as a physicist. I know it, as a chemical engineer.
But there are folks out there that have lived in the virtual world of computer models so long that this distinction has become foggy. They equate computer simulations with experiments (or with scientific evidence) and estimates with measurements.
Often they use arguments, such as: “well, every measurement also involves some estimation and every estimate some measurement, so there is really no difference”.
But the difference is fundamental.
Max
Max, you write “But the difference is fundamental. ”
I know this, You know this. All physicists know this. Steven Mosher insists that there is no categorical difference between the two, and he will never admit he is wrong. However, you might try and convince John Carpenter. I have tried to no avail.
Jim Cripwell
Here is the problem, as I see it.
You and I live in a real world. We observe and measure what is going on around us and draw conclusions from the physical observations we make. Through our different educational and career paths we have learned that the best thought out hypothesis is worthless if it cannot be corroborated by real-time physical observations or reproducible experimentation, confirmed by actual measurements.
There are others who live in a virtual world – the world of computer models fed by theoretical deliberations. These individuals are so immersed in this virtual world that they begin believing that computer model outputs based on theoretical “physics” are the same as actual physical observations. Ergo, to them one computer simulation can validate another and a computer-derived “estimate” is the same as a physical “measurement”.
The two worlds have a hard time communicating with each other – since they speak two entirely different languages.
Max
manacker: “well, every measurement also involves some estimation and every estimate some measurement, so there is really no difference”.
So where exactly is the difference? Is your reported body temperature, from a sublingually placed thermometer, a “measurement” or an “estimate”. How about the blood concentration of something (e.g. BUN) in a medical document?
We have asked many times, but no one has supplied any examples of “measurement” that don’t depend on approximate mathematical relationships between estimands, and on estimated parameters. What has been provided have been examples of estimation procedures with a diversity of accuracies and precisions.
http://en.wikipedia.org/wiki/Measurement#Difficulties
“Information theory recognises that all data are inexact and statistical in nature. Thus the definition of measurement is: “A set of observations that reduce uncertainty where the result is expressed as a quantity.”[11] This definition is implied in what scientists actually do when they measure something and report both the mean and statistics of the measurements. In practical terms, one begins with an initial guess as to the value of a quantity, and then, using various methods and instruments, reduces the uncertainty in the value. Note that in this view, unlike the positivist representational theory, all measurements are uncertain, so instead of assigning one value, a range of values is assigned to a measurement. This also implies that there is not a clear or neat distinction between estimation and measurement. Ascertaining the degree measurement error is also a basic facet of metrology, and sources of errors are divided into systematic and non-systematic.”
So, there is not a clear or neat distinction between estimation and measurement.
MattStat, “So where exactly is the difference? Is your reported body temperature, from a sublingually placed thermometer, a “measurement” or an “estimate”. How about the blood concentration of something (e.g. BUN) in a medical document?”
I like that, you are measuring a proxy for some other function in order to estimate the impact on the desired function. There needs to be adjustments made to the “measure” that can impact the “estimate”, so you need to consider your frame of reference, the measurement. with respect to the object.
That is Thermo 101 rule three. KISS, ASSUME and FOR.
Since the powers at be selected 255K (240Wm-2) as the object, the measure, surface temperature is amplified by 288( +/-?)/255(+/-?)=1.29 +/- ? %. Using anomaly you can get your accuracy/precision down to almost unbelievably low uncertainty and at the end you still have that +/-?% uncertainty because of the selected reference. Since energy is involved you also have 390/240=1.625 +/-?% uncertainty relative to that “measure”. The end result is you have ~+/-0.25C reference slop versus ~0.4C measured impact.
You can reduce the slop by comparing results versus different frames of reference. So if SST indicates one rate of change you can’t just highlight land surface temperature because they are related by an amplification factor due to T^4
So climate science is using sub lingual when they actually have rectal readings and are anal about sticking with a failing frame of reference :)
:)
Jim
Several months ago I posted a comment from Mosh -which I directed to you- where he confirmed estimates and measurements are two separate things.
You obviously never saw it
tonyb
Thanks, Tony. I did not see it. I hope Steven is reading this, and he will confirm that estimates and measurements are two different things. Maybe John Carpenter will as well.
climatereason: where he confirmed estimates and measurements are two separate things.
How about your answer to this question: is blood pressure measured or estimated? How about the density or half-life of an isotope of Uranium, the gravitational constant, or the oscillatory frequency of a laser beam?
Beware of press releases as they often misrepresent science with sensationalist spin.
Good article. Does this mean that the mainstream sceince media is preparing for a change of previous pro-AGW reporting?
I like the 20 tips, and JC’s pick of the ten relevant. So maybe we need more perhaps 25? Some sugestions:
We don’t know everything (Science is never settled) – that’s why we keep investing in scientific research. We will know more in another hundred years, by which time a lot of what we think we know now will be proven to be nonsense.
Science funding supports a process not an outcome – well done science doesn’t pretend to know the outcome of research, diversity of research perspectives is a good thing.
Proper experiments produce the most robust scientific evidence – good experimental design can eliminate many of the problems mentioned in earlier tips. Observational studies and computer models are not a patch on proper experiments.
The article states: The average taken from a large number of observations will usually be more informative than the average taken from a smaller number of observations.
There are a number of assumptions required for this to be true — imagine, for example, averaging the gross domestic product of the US for ted years and then trying to “be more informative” with a fifty year average.
The first lesson for everyone should be how scientists lie and, in particular, how many of them pretend to understand subjects like statistics that they don’t.
“Correlation does not imply causation”
Of course the above is true, but what of the converse. If two functions are not correlated then there is good reason to believe they are independent. Apply this to historical climate data and the correlation of global average temperature with CO2 concentration is:
(a) 1910 to 1940 good correlation
(b) 1940 to 1970 zero or negative correlation
(c) 1970 to 1997 good correlation
(d) 1997 to 2013 zero correlation
What does the above imply? The relation is either very odd or discontinuous, or could be called an on/off relationship. That is, sometimes there is anthropogenic climate change and sometimes there isn’t.
This is exactly what I have been saying for years. Would anyone from either camp like to comment on this?
Assuming your (a)-(d) is correct, there is that lurking variable(s).
I’d call (a) zero to poor correlation, leaving ‘the last quarter of the last century’ as the best period of correlation since The Little Ice Age. The biggest lurking variable is the oceanic oscillations; attribution of temperature rise to CO2 seems a Post Hoc, Ergo Propter Hoc logical fallacy.
The likes of Muller, who might ascribe all of the temperature recovery from the Little Ice Age to AnthroGHGs, seem to disagree.
Were those such people correct, few have yet examined the ramifications of just how cold it would now be without the effect of AnthroGHGs.
We’d be far better off if it turns out that natural variability dominates, and that AnthroCO2 has a trace(cross that out and substitute ‘minor’) effect.
===========================
kim –
Here you go: Detrended 3.0 ECS.
If 3.0 ECS is true, then the Little Ice-age deepens w/o anthro CO2. And with Canadian and Soviet wheat production going to zero in the 1960’s, there’s 100’s of millions facing a Paul Ehrlich famine scenario.
(Note: the detrending would be better if power and/or exponential detrending were a WFT option.)
This one is better. Detrended 3.0 ECS II
Alexander Biggs
You raise a good point, but kim is right in saying that
(a) 1910 to 1940 poor correlation
So we are left with 1970 to (let’s stretch it to) 2000 good correlation.
And the IPCC logic goes as follows
1. Our models cannot explain the early 20thC warming cycle
2. We know that CO2 caused the statistically indistinguishable late 20thC warming cycle
3. How do we know this?
4. Because our models cannot explain it any other way.
[You can add in that the models can also not explain the mid-century cooling cycle or the current “pause” in warming despite unabated human GHG emissions and concentrations reaching record levels.]
Three strikes – yer out!
Max
Kim and Manacker: Thank you for your replies. On the contrary, the correlation 1910 to 1940 is very good. Since the beginning of the 20th century CO2 has always increased, remember Henry Ford made 15 million Model T’s between 1908 and 1928. the low octane fuel then available made these cars not very efficient producers of power, but quite efficient producers of CO2. Second, stripped of random noise by the BOM’s 11 year central moving average filter,the mean value of global average temperature shows a quite uniform steady rise of nearly 0.5C in 30 years. The IPCC missed this because they only started their analysis after 1961,following their charter from the politicians of the UNFCCC. See my theoretical climate model’s first figure underlined above. You will also notice that the 1970 to 1997 rise is almost an exact copy in both slope and extent of the 1910 to 1940 rise, which shows the oceans to be very good copiers. albeit with a 30 year delay (my hypothesis).
If my analysis is correct I can think of no single differential equation style model that could satisfy the stop/start data. Where is the fallacy? Of course, CO2 is a single compound gas. but for IR spectrum, there are thousands of variants due to its isotopic nature. These variants could explain different vibration IR absorption frequencies and explain the on/off nature of climate change. If you can think of a better explanation, please advise, but don’t expect to get anywhere without quantum mechanics.
Alexander Biggs
According to IPCC, “most” of the late 20thC warming cycle can be “explained” by increasing GHG concentrations (primarily CO2).
IPCC does not make this claim for the statistically indistinguishable early 20thC warming cycle; in fact, it states that the models cannot explain this warming period (AR4, Ch.9, p.691):
The problem with the early 20thC warming cycle is that there was not enough increase in human GHGs to have caused a significant part of the warming, so it had to be caused by something else.
GH forcing and warming are logarithmically proportional to the GHG concentration at the end of the period divided by that at the beginning of the period.
CO2 concentration was 299 ppmv in 1910 and 310 ppmv in 1940 (Siegenthaler et al. ice core data).
It was 324 ppmv in 1970 and 369 ppmv in 2000 (Mauna Loa).
ln(310/299) = 0.036
ln(369/324) = 1.300
1.300/0.036 = 3.6 times the GHG forcing in 1970-2000 as in 1910-1940.
So the argument for substantial GH warming in 1910-1940 is much weaker than for the period 1970-2000.
That was the point kim and I were making.
Max
http://popesclimatetheory.com/page38.html
CO2 stopped following temperature about 5000 years ago.
I received this data from:
Thomas C. Peterson, Ph.D.
Chief Scientist
NOAA’s National Climatic Data Center
the graph says temperature is very stable but points at a very unstable plot of temperature
CO2 went up by just 20ppm (less than 10%) over several millennia while the Milankovitch forcing went down. CO2 has now gone up 43% more in 150 years — a period over which the change in Milankovitch forcing is negligible.
I luv NOAA. They ran into enemy lead with my father’s USMC regiment to plant the flag on top of Mt. Suribachi, Iwo Jima. Gotta luv geeks who will run into hot lead with leathernecks. They NOAA geeks planted some scientific device next to it.
But somebody please send him CO2 data after 280 ppm.
I was going to suggest that P-N.
“the graph says temperature is very stable but points at a very unstable plot of temperature”
Most of the temperature variation is within 2 degrees.
The 10 tips are pretty good.
But, while “scientists are human” and “data can be dredged or cherry picked” go in the direction of “data can be fudged to support a preconceived agenda”, they still does not completely cover it. And this is a real issue IMO.
The forced consensus process not only “fudges” the data, “cherry picks” data supporting the agenda and ignores or rejects data that conflict, it does so with the explicit purpose of supporting a preconceived political agenda.
When scientific claims are used to support a political agenda, we no longer have objective science at play, we have “agenda driven science”.
So the tip is:
Look for a political agenda behind the scientific claims. If such an agenda can be identified, be extremely cautious in accepting as objectively correct any scientific claims, which support this political agenda.
Max
Yep.
You bet Max, always look for the hidden agenda! This applies to everything, including communicatons from the grandkids! ;)
Crucial point, Max.
anyone who thinks all scientists are human has never had to read referees comments. Some scientists are definitely inhuman.
Max, that’s the exact rule that I apply to Wikipedia. If the subject is dry and esoteric, I tend to assume until proven otherwise that it’s reliable. OTOH, if it’s a contentious (particularly when political) subject, I consider Wiki useless until demonstrated otherwise.
Whatever applies to Wikipedia applies to ‘experts’ in general. If there’s a political/ideological connection, consider the information to be suspect.
I think, Dr. Curry, you should add to the list that you’ve done everything the list warns against. You picked the ten items “of greatest importance” to extol?
No, as the tenth item warned against, you cherry-picked.
This would be an example of that rife bias point.
While no measurement is exact, no general tip is exact, either, yet you grab onto the ten that tell your narrative and milk them for all they’re worth.
Bigger really is better; the list doesn’t appear to be meant to be used by picking and choosing only the parts the reader likes best.
The first four tips you choose to skip clearly are areas you frequently violate, and encourage in your denizens: failing to check fallacy, failure to respect control groups, failure to randomize, pseudoreplication. Small wonder you don’t find them important enough to mention. Which would be a failing of “the main challenge of research is teasing apart the importance of the process of interest” — which in this case is getting the point of the article, to give non-scientists the tools to arrive at decisions based on scientific evidence, not to block anyone from coming to any decision ever.
Looking at the rest of the points you apparently thought could be skipped over by DECISION MAKERS, one finds exactly the tips that most facilitate evidence-based decision-making.
Significance is significant. Expressed as P, statistical significance is a measure of how likely a result is to occur by chance. Thus P = 0.01 means there is a 1-in-100 probability that what looks like an effect of the treatment could have occurred randomly, and in truth there was no effect at all. Typically, scientists report results as significant when the P-value of the test is less than 0.05 (1 in 20).
Separate no effect from non-significance. The lack of a statistically significant result (say a P-value > 0.05) does not mean that there was no underlying effect: it means that no effect was detected. A small study may not have the power to detect a real difference. For example, tests of cotton and potato crops that were genetically modified to produce a toxin to protect them from damaging insects suggested that there were no adverse effects on beneficial insects such as pollinators. Yet none of the experiments had large enough sample sizes to detect impacts on beneficial species had there been any5.
Effect size matters. Small responses are less likely to be detected. A study with many replicates might result in a statistically significant result but have a small effect size (and so, perhaps, be unimportant). The importance of an effect size is a biological, physical or social question, and not a statistical one. In the 1990s, the editor of the US journal Epidemiology asked authors to stop using statistical significance in submitted manuscripts because authors were routinely misinterpreting the meaning of significance tests, resulting in ineffective or misguided recommendations for public-health policy6.
Study relevance limits generalizations. The relevance of a study depends on how much the conditions under which it is done resemble the conditions of the issue under consideration. For example, there are limits to the generalizations that one can make from animal or laboratory experiments to humans.
Why is that?
Did your feelings influence your perceptions of what is important in conveying the meaning of evidence to policy people?
I don’t mean to extrapolate beyond the data, but on the face, one could conclude you’re trying to pull a fast one, Dr. Curry.
Bart R
Your logic is flawed.
To “cherry pick” scientific data in order to get a desired message across, while ignoring or rejecting scientific data which do not support the message, is a violation of the scientific process and method, which dictates that any scientific hypothesis should be challenged.
To select a series of premises out of a longer list for simplification or in order to emphasize a point is not a violation of the scientific process.
Max
Sometimes, it’s just like a giant spring goes ‘Boing!’
=========
I am uncomfortable excerpting more than 50% of someone else’s article; there are copyright issues also. My strategy is to motivate people to go to the original link to read the entire article.
And then God created paraphrases.
Stealth activism
Never believe someone’s perfectly reasonable explanation when you can instead ascribe it to some ulterior motive, huh guys?
More seriously, what would I add to the list?
21. As a group, scientists tend to be more skeptical.
22. Scientists’ conclusions tend to err on the side of least drama.
23. Simplicity, parsimony, universality equals truth. The foundation of scientific philosophy was laid 300 years ago by Newton, Hook, and Halley with the principle: The explanation with fewest unexplained assumptions (simplicity), fewest exceptions (parsimony), and that applies most universally is held to be accurate or very nearly true until new evidence requires it to be amended.
24. Obligation. What science tells us is true, we are obliged to act on as true.
25. Precaution. Where evidence is insufficient to accept an explanation, it is a precaution to treat the worst possible outcome as likely while pursuing better understanding, as precautionary actions tend to be the most expensive.
Oy!
Someone’s taking my name in vain! Or are they?
Oh well, I suppose imitation is the sincerest form of flattery.
My work here is done. I’m vanishing for a while.
Live well,and prosper,
Mike Flynn.
Flynn said:
Oh, that’s what he calls it.
To the politicians, bureaucrats and the media ;
Scientists are just another bunch of quite ordinary people just like all of those other people from every walk of life that your profession deals with on a daily basis.
Regard them and treat them and their claims and beliefs as you would all others.
And applicable to all; it is wise to remember that no pedestal is ever too high to not have it kicked out from under some self important little tin gods.
Science is the father of knowledge, but opinion breeds ignorance.
Hippocrates
Science is wonderfully equipped to answer the question ‘How?’ but it gets terribly confused when you ask the question ‘Why?’
Erwin Chargaff
i like check lists – except Schneider people-lists, who’s in
and who’s out – and we know where yer live- kinda’ lists.
This looks like a good sorta’ list ter Professor Curry (and
ter some serfs too.) Can’t do no harm offerin’ it, might do
some good. One ter Four ‘n -Six ter Ten, yep. Number
Five, ‘correlations don’t mean causation,’ yep, but
don’t ferget ter point out, (5b) that non – correlation
sure knocks a hole in yer theory …. like rising temps
goin’ tergether with rising CO2 like a horse and
carriage when they don’t. Addend: ‘ Say, is yer test
replicated?’
Too bad that the deniers can’t test out any of those 20 tips on their own stuff. They still lack an alternate theory for GW that can even pass rudimentary acceptability.
CSALT passes with flying colors.
http://contextearth.com/2013/11/21/variational-principles-in-thermodynamics/
Agree, the deniers (or anthropowarmists) can’t test any of those 20 tips. Luckily, the nature is conducting the necessary experiments.
Edim
Yeah. And the tests aren’t coming out the way Webby thought they should.
Max
Plus one Edim.
Nature rules ‘n don’t you fergeit
it. beth the serf.
“Plus one”
Was that the number of tips that EDim’s theory passed?
1 out of 20 isn’t so hot.
Webster, “The CSALT model of the global temperature anomaly has no right to work as well as it does.”
Don’t short your self. if you invert SOI, scale it a touch and tweak a lag you can remove ENSO since SOI is a proxy for ENSO. If what causes ENSO/SOI has a longer term trend, then that trend will remain because indexing removes the trend. As long as you assume a spherical cow has five udders, you get a perfect fit. ;)
Using atmospheric pressure removes the trend because long trends in pressure cannot be sustained.
CO2 forcing can be sustained.
Cappy do you understand the 97% ?
Hi WHT
“They still lack an alternate theory for GW that can even pass rudimentary acceptability.”
Well, I’ll give it a shot, but since I am not a scientist maybe this should be filed under ‘opinion’ rather than ‘theory’:
According to a variety of sources, including proxies, written records, and instrumental records ‘climate’ has changed continuously on all time scales examined. Current variations in climate, however defined, are well within the range of historical climate variations that occurred when the output of the Human Carbon Volcano (HCV) was essentially zero. Therefore, whatever the actual influence of the HCV on the climate, its practical influence is negligible. We should continue to supply our energy from whatever sources make economic sense and ignore their ‘carbon signatures’ completely.
A corollary to the above theorem is: When, following some unpleasant weather event like a hurricane, typhoon, tornado, flood, drouth, whatever, someone grabs the nearest microphone or calls a press conference and announces that said unpleasant weather event was almost certainly caused by the HCV and that we can no longer afford to delay the implementation of the taxes and regulations that will prevent similar incidents in the future, that someone has been self-identified as a person whose advice on any subject whatsoever should be ignored. With the possible exception of ‘Psst! Zip your fly!’
“Scientific claims to guide policymakers?”
But what came first, the chicken or the egg? (What is the cause and what is the effect?)
Could it be that policymakers (i.e. politicians) saw the opportunity to implement a universal tax on carbon emissions, to take control of global energy and gain billions in new revenues, which could be shuffled around for pet projects or to reward supporters, for which they needed a scientific justification, so they doled out large sums of taxpayer money to fund the scientific studies to support the agenda?
“Policy makers to encourage and fund scientists to provide scientific claims to support the preconceived agenda of policy makers?”
Could we have a case of “causation” here?
[And puh-leez don’t come with the “conspiracy” canard.]
Max
John Brignell has a piece listing indicators of when you might want to question whether you are being provided the best available information.
http://www.numberwatch.co.uk/lying.htm
Thanks for flagging that excellent piece, Bob.
John Brignell knows of what he speaks.
How about “extraordinary claims require extraordinary evidence?” That would be my five-minute university on epistemology. Period. (Uh, maybe that’s no longer an effective form of emphasis.)
Period is usually some form of cherry pick I’m afraid NW. BTW I appreciate your econ stats common sense approach to what has been put here for discussion. That plus your civility. The ones who indulge in verbals will never persuade or convince anyone to agree with their POV, IMO.
tips for interpreting scientific claims
1. do a science course
2. do a second science course.
3 do a philosophy course
4 go back and do a third science course.
If you do not understand the claims do some research
Scientists are very good at finding what the funders want to find.
What would you add?
Science is based on proof, not consensus.
More practical than all those in my experience is the following rule of thumb: most of the time the impression a “study” makes on you is wrong. All along the chain, from basic methodology to the wording of the press release, there are opportunities for the truth to be mangled and, believe it or not, completely reversed.
When someone tells you that “studies show” something, grab you wallet (and your skepticism). Most likely, they don’t show that at all.
Having spent 40 years working with politicians and policymakers on science intensive issues I find this list simpleminded to the point of uselessness. It ignores the fact that elected officials and their staffs are experts at what they do. As such they are seldom fooled, while this list suggests they are.
Well the real audience for this article is scientists (they are main audience for Nature); perhaps some useful hints for scientists in terms of how they should engage with policy makers
even there is conveys the misconception that science is more important than it is. Politicians and policymakers are experts in social controversy, something scientists have little understanding of. Politicians and policymakers are trying to solve social problems, which are about as far away from scientific problems as one can get. They are doing so using complex democratic machinery that scientists have no understanding of. They are professional advocates in a world of advocacy because that is how democracy works, or any large scale multi-group governing system for that matter.
Sorry I got called away and posted the above comment without editing it. It should begin “Even there it…”
Ironically the authors mention then ignore the actual working mechanism, which is that a politician heavily into a science intensive issue will hear from the various advocacy scientists in the case, who will criticize one another along the lines of this list. If the science is controversial that is easily learned and most of what needs to be known by the politician as far as the science is concerned. The scientific specifics of the controversy are not important to the politician. If the science is not controversial then it is probably irrelevant.
Many scientists do certainly see science misrepresented by the media. They (or I) feel that early results of science are presented as final – and then contradicted by the next science news.
I was recently at a presentation given by a science journalist to graduate students on how to tell about their research in public. The dilemma between making it simple and interesting enough on the one hand and accurate on the other came out clearly – with no simple solution.
One thing that I have learned slowly over years is that I should not worry too much about inaccurate reporting on issues close to myself. They are mostly not as important to others. Others are not so disturbed or influenced by the errors that I feel to be catastrophic. They have also learned that newspapers simplify things and try to make news more dramatic than it really is.
While I can almost accept the content of the above paragraph, I do still worry about the overall impression news media give on science, and also about the help many university press releases give the journalists in that misrepresentation.
Not sure about this, but might add as a point that records (much loved by media) are not as unlikely as you might think. There is a small, recondite, probability literature on this, but the principle can be made by the ‘famous’ arcsine law for the maximum of Brownian motion/random walk. The most probable times for a maximum (ie record) are ‘now’ or back at the beginning of the records. The probability distribution for the maximum is U-shaped, peaking at the start and end of the interval. Even if temperatures, storm strengths or whatever are just a pure random walk, and I am not saying they are please note, the fact of records or near-records is not really evidence of great wonders and apocalypse.
Please correct – that should be probability density, not distribution, is U-shaped
“Perhaps we could teach science to politicians? ”
The first premise of the article is totally wrong.
What we need is teach science to scientists.
Too many climate scientists make unsupported claims, or do dubious science, while most of their peers applaud.
It is biased science that is the main problem, not ignorant politicians.
This always gets loud guffaws, but I will repeat it again for the umteenth time:
If the scientific claim cannot be explained such that an 8th-grade college-prep student can understand, then the scientist does not really know what they are talking about.
David Wojick is semi-correct. Policy makers are expert in taking the position that ensures they remain policy makers. They are very good at cherry-picking the science to reach these goals. Their nimble minds that understand the science are what allows them to run between the raindrops with the politically expedient position.
Howard, you write ” Policy makers are expert in taking the position that ensures they remain policy makers. They are very good at cherry-picking the science to reach these goals.”
It was a Canadian Prime Minister, John Diefenbaker, who said “It is a long road that has no ashcans”. What you describe only goes on until reality bites. This now seems to be happening in the UK. The Climate Change Act is hobbling the UK economy, and George Osborne and David Cameron know it. They seem to be desperately trying to find a way out of their troubles. I wish them the best f luck.
15 year old Kristen Byrnes pretty well demolished Al Gore and his crockumentary years ago. A society that ignores truths that is obvious to its children has big problems and finding a way out may not be possible. The best thing than can happen to America is to watch dead and dying Europe actually go belly-up.
i would add something about limitations of computer simulations… and their very nature.
Not all scientific claims are equally reliable.
Some scientific claims are backed by a mountain of evidence and established theory. When astronomers say that an asteroid is about to pass a million kilometers from earth, they usually have it dead on. Other claims are tentative at best, and have a high likelihood of being overturned or significantly modified. When someone claims that broccoli reduces colon cancer, for example, take that broccoli with a large grain of salt.
The scientific claims I enjoy most is when over-the-top alarmists say–e.g., it’s already too late: even if we act now catastrophe is certain. Climate alarmists have already pulled that one out of the hat. On its face it seems like inept marketing but actually it is a desperate Hail Mary attempt to shame us all for not doing what we surely should have long ago and then counting on our innate desires to be positive about something such as, maybe it’s not be too late if we act quickly. It is all a part of the the fallacy of believing we must act now because the Earth can’t wait any longer.
Moderate amounts of butter and multiple but moderate amounts of grains of salt are good for humans. Large amounts of BS are bad for humans, especially poor in the third world. Good luck on getting massive reparations on an imaginary problem out of the overtaxed populace in the US. May be the Brits will go for it but not the Irish. I don’t know about the Scotts, I was saddened at the deal to release the Pan Am 103 bomber for an oil deal with Libya. Those are not the Scott’s of the Border Clans. They are tight fisted but too proud to sell honor for a little oil. Should use their own fracking oil.
Scott
Moderate amounts of butter and multiple but moderate amounts of grains of salt are good for humans…
You’re almost describing bacon. Maybe we can bring back the US economy if we wrap Chevy Volts in bacon and sell them to Brits before they catch on that the global warming scare was a big load.
Wag,
sorry, broccoli better than bacon, but bacon in little amounts still ok.
Need to restore sanity to the debate. Chevy volt is nice and uses big plants generating electricity by natural gas in CA. hydrogen economy on the way. Just around the corner. Fusion is also around the corner. Wait till the wind energy kills the golden eagles in CA and then the BS hits the fan. What happened to natural gas vehicles? Hope the BS hits fan of more discourse first but that is another thread. So much BS and so many golden eagles to kill. I do wish we could figure out how to make solar cheaper. Lots of sunny parking lots and roofs to cover.
Scott
Now you’re talking about the fallacy of the Left: if you wrap BS in enough bacon you sell it to enough low-information voters to elect the devil hisself.
In its purest form, the precautionary-principle is a recipe for nihilism and mental paralysis.
Borrowing two quotes from Richard Dawkins:
-“The essence of life is statistical improbability on a colossal scale.” -“However many ways there may be of being alive, it is certain that there are vastly more ways of being dead, or rather not alive.”
It is not sensible to spend time trying to consider all possible things that might go wrong with a given course of action. Life, and the universe in general, is just too short. It is always possible to find more and more potential disasters lurking round every corner, waiting to make a disaster even worse than we thought.
Of course, it is not sensible to abandon caution altogether either. But choosing to explore what might be wrong at the expense of what is already in place (and appears to be working… otherwise we would be dead) is a losing strategy in the long term.
Practical up-shot for scientists (policymakers)?
Try to steer clear of promoting research that appears solely predicated on the proposition “We are doing X but this might cause the undesirable result Y in the long term”, where “Maybe we should stop doing X” is the stated, or unstated corollary.
Of course proponents of stopping doing X may also say “We already think Y is happening”. And they might not be wrong. No easy answer.
But there are many more ways of being wrong than being right. [As well as being too easy] And when you frequently hear the same refrain from the same, or similar sources, then it’s time to ask the question more forcefully-”Is this another case of being repeatedly lead astray by the excessive use of the precautionary principle?
——————————-
“Don’t center on your anxieties, Obi-Wan. Keep your concentration here and now, where it belongs.”
“But Master Yoda says I should be mindful of the future.”
“But not at the expense of the moment. Be mindful of the living Force, young Padawan.”
———————————
If only it was that easy.
True, true the Left’s view of the world is really pretty simple: a free man cannot be trusted.
This piece begins … “Perhaps we could teach science to politicians?”
I ask – for what reason?
I would submit that most politicians have no real interest in learning science.
Politicians are, in aggregate, a “win at any cost” bunch. Their position on a subject is, at best, based on the polling of their constituents. They care little for the truth, or fairness, only wanting the appearance of being truthful or fair.
Politicians want power and control and it doesn’t really matter to them if they have the truth on their side as long as they win.
Perhaps we could teach politics to scientists, but I fear this would corrupt the good scientists and the rest, well, the rest have learned the art of politics on their own.
Here’s a radical thought which could be worth playing with. Maybe there is no climate science to “communicate”. Maybe we have confused the aspiration with the reality. (And maybe Publish-or-Perish hasn’t helped.)
Just as scholars once believed in the existence of medical science without practising anatomy – and “communicated” their opinions on the human body very vigorously – we now have scholars who pronounce on the behaviour of the atmosphere while knowing a bit about it, the biosphere and the outer lithosphere, far less about the hydrosphere, and less again about the great bulk of the planet – which happens to be mostly hot and mushy.
That’s without mentioning orbits, sun etc.
No climate science to “communicate”. Not a criticism. Just a thought to play with, right? Can’t hurt to consider the possibility.
Pingback: Social cost of carbon | Climate Etc.
There is no doubt that a running average destroys data and distorts the shape if the temperature curve. In particular, the height of El Nino peaks is reduced according to the width of the peak and this may give a wrong impression of where the curve is going. To avoid this I use a semi-transparent magic marker (or its computer equivalent) as explained in my book “What Warming?” You should have sufficient resolution to see the raw shape of the curve which means enough to show the fuzz surrounding the curve. The fuzz is not noise but is caused by global cloudiness variations and has a maximum amplitude. Real noise is comprised of large spikes sticking out of that fuzz. This was the clue I used that allowed me to identify some non-random noise in GISTEMP data. It turned out that theses noise spikes, all of which occurred during the first month of a year, were also present in HadCRUT and NCDC databases, all at the exact same locations as in GISTEMP. There is no doubt that they were put there by an out-of-control computer program that was used to process these data for reasons we were never told. About a dozen or more of them are easily identifiable in the interval between 1979 and 2010. I did not check the data before 1979 when the satellite era begins but someone should check to see how far back this data manipulation extends. Also, data guardians, whose product we now know is contaminated, owe us an explanation of how and why the data were processed and why they lost control of their software that created the spikes. Some of them are large enough to locally twist the curve out of shape.
There is no doubt that a running average destroys data and distorts the shape of the temperature curve. In particular, the height of El Nino peaks is reduced according to the width of the peak and this may give a wrong impression of where the curve is going. To avoid this I use a semi-transparent magic marker (or its computer equivalent) as explained in my book “What Warming?” You should have sufficient resolution to see the raw shape of the curve which means enough to show the fuzz surrounding the curve. The fuzz is not noise but is caused by global cloudiness variations and has a maximum amplitude. Real noise is comprised of large spikes sticking out of that fuzz. This was the clue I used that allowed me to identify some non-random noise in GISTEMP data. It turned out that theses noise spikes, all of which occurred during the first month of a year, were also present in HadCRUT and NCDC databases, all at the exact same locations as in GISTEMP. There is no doubt that they were put there by an out-of-control computer program that was used to process these data for reasons we were never told. About a dozen or more of them are easily identifiable in the interval between 1979 and 2010. I did not check the data before 1979 when the satellite era begins but someone should check to see how far back this data manipulation extends. Also, data guardians, whose product we now know is contaminated, owe us an explanation of how and why the data were processed and why they lost control of their software that created the spikes. Some of them are large enough to locally twist the curve out of shape.
Pingback: Weekly Climate and Energy News Roundup | Watts Up With That?
Pingback: Policy cooking | Climatemonitor
Pingback: benefits of CO2 outweigh the deficits … | pindanpost