The Pilot is again asking for guest editorials, so I decided to send one on the poor quality of current science and the reporting of it by the press. Unsurprisingly, they weren’t interested.
So, here is is
I have been in love with science since I was seven years old, but I chose to make a career in health care and to follow science out of love, which is what ‘amateur’ means.
Now, sixty-five years later, at least as it is reported to the public, science is breaking my heart. It’s much like returning to your hometown to find that your first great love is now an addict, selling herself on the street for her next fix. Something has gone terribly wrong.
It’s not just an intellectual disappointment. In this increasingly complex world, public policy is often driven by the public’s perception of science, and that has seen a diminishing connection with reality.
Much of it is simply intellectual laziness on the part of those who bring science to the public. Press reports on science commonly confuse correlation with causation. If A and B occur together, it may be that A caused B but it could as easily be that B caused A, or both A and B were caused by some other factor, or it could be simply coincidence. Correlation can be a starting point in investigations, but it should never be presumed to proof of cause, yet most articles on science report it as such.
Another problem is reliance on authority or consensus. Physicist Richard P. Feynman warned that progress in science begins with the belief in the ignorance of experts.
For forty years, experts told us that eating foods containing cholesterol was the cause of high cholesterol in the blood, and there was hardly any disagreement with that consensus. But it was never true, and low fat diets doctors prescribed turned out to be a large contributor to the rise in type two diabetes.
A more insidious problem is the corruption of science by politics. Researchers no longer rely on university salaries for their sole income, they are dependent on research grants, and those who provide the grants, whether government or private, have an agenda. We would like to believe scientists are incorruptible, and most probably are, but the funding goes to those who support the agenda.
It is the duty of amateurs to hold the experts, and those who report on science, to a high standard of intellectual rigor, as too much of the scientific community has been compromised by the grant system.
The basis for that rigor was best explained by Dr Feynman, who described the process by which science advances, We start with a guess at how things work, then we compute the consequences of that guess, and test the results against reality. Feynman asserted that “It doesn’t make any difference how beautiful your guess is, it doesn’t make any difference how smart you are, who made the guess, or what his name is. If it disagrees with experiment, it’s wrong.”
And that’s all there is to it. It doesn’t matter if you work for NASA, or the CDC, or of you have a stack of Nobel prizes, if your theory doesn’t match with experiment, it is wrong. It is not easy to discard a theory that has popular support, but that is the standard to which scientists must be held, by their peers, and by the public.
One of the ways scientists are kept honest is for other investigators to confirm their results independently, but that isn’t possible if researchers withhold their data and methods from other investigators. Any study in which the data and methods are not made public should be suspect.
We can’t rely totally on the scientists themselves. Review of research by peers is supposed to provide a check, but those peers are seeking grants from the same sponsors with the same agenda as the papers they review. Review by researchers from other fields, and by amateurs, is needed.
Editors who present science to the public cannot allow themselves to become cheerleaders for the consensus. They must seek out and include the input of skeptics but ultimately, we must all be better consumers of science reporting.
But when the press asks no questions and the experts look the other way, it is up to the amateurs to cry foul.
18 thoughts on “Poor reporting on science”
Unless the experiment is ill-conceived, and then you’re back where you started.
LikeLiked by 2 people
A couple of thoughts on your essay . . .
Correlation is not causation. True ‘dat. But it can be – and often is – evidence of a causal connection. The correlation of rising sea levels with rising CO2 concentrations MAY be a coincidence or it may be evidence of the power of CO2 to raise temperatures that expand water volumes and melt glaciers.
One thing can vary in correlation with another without a direct causal connection. For example, the concentration of guns in a population is a reliable predictor of the gun homicide rate. The more guns, the more gun deaths. Does that prove that guns “cause” gun deaths when we all know that people kill people? No, it does not. But since the presence of a gun is a necessary condition for a gun death, an increase in guns creates more opportunity for people to kill people. Making increasing guns dangerous even if they do not “cause” gun deaths.
Skepticism of the consensus is better advice for a scientist than for the public. You do not break new ground in science without moving beyond the consensus. But the public, lacking the arcane knowledge to distinguish between
perspicacious outliers and crackpots, is well-advised to trust the consensus. The science of smoking is a good example. So, too, is climate science where, in a close analogue to smoking, there are biased special interests doing their best to discredit the consensus.
Feynman was talking about scientific theories. Not working mathematical models that incorporate many facts and accepted theories in order to understand complex phenomena – such as the climate. If actual data does not support a theory, the theory is wrong. If actual data does not conform to a model, the model needs to be refined – not rejected. Admittedly, it is hard to find a bright line between a “theory” and a “model” but I think it is clear that Feynman had in mind something more fundamental than multi-variant mathematical models based on established theory.
In summary, if the press is guilty of oversimplification in its reporting of science, so too is your critique of that reporting.
LikeLiked by 2 people
RE: “Correlation is not causation. True ‘dat. But it can be – and often is – evidence of a causal connection.”
Nope. Correlation is never evidence of causation.
RE: “Skepticism of the consensus is better advice for a scientist than for the public.”
Only an elitist would think this way. Skepticism of consensus is perfectly good advice for the public.
RE: “I think it is clear that Feynman had in mind something more fundamental than multi-variant mathematical models based on established theory.”
Why would you think such a thing? Feynman’s observation applies quite well to models that don’t work. They may as well be rejected until such time as they are corrected.
Dr. Tabor hasn’t oversimplified anything. He has merely catalogued a few basic ideas about science that any literate person should know.
“Nope. Correlation is never evidence of causation.”
Uh, bullshit. It may not be conclusive evidence of a causal link but it IS evidence. Kind of like “probable cause” in the law. It is enough evidence to warrant further study.
“Why would you think such a thing?”
Because I have read a couple of Feynman’s books and listened to his lectures. I understand the context in which he made that remark.
Skepticism of science by an uninformed public is both unwarranted and dangerous. Just look, for example, at the harm done by anti-vaxxers and their newest incarnation in this pandemic.
“Only an elitist . . .”
I tried to respond to Dr. Tabor’s critique of the press in a respectful and thoughtful way. Leave it to a known ignoramus to interject name-calling. Without much logic to it, I would add.
RE: “[Correlation] may not be conclusive evidence of a causal link but it IS evidence. Kind of like ‘probable cause’ in the law. It is enough evidence to warrant further study.”
That’s a meaningless statement. Just because a correlation may inspire the formation of a hypothesis doesn’t make it evidence of anything. The hard distinction between correlation and causation is fundamental to the process of science.
RE: “Skepticism of science by an uninformed public is both unwarranted and dangerous.”
I disagree. Skepticism is the first step of learning.
RE: “I tried to respond to Dr. Tabor’s critique of the press in a respectful and thoughtful way.”
And you failed. Your comments have all the “logic” of saying, “X is both true and not-true at the same time.”
Keep in mind that the limit on essay length is 650 words and that the intended audience is less sophisticated than this forum, so I don’t think it was really oversimplified for its purpose.
Climate models(though I purposely did not mention them) are exactly what Feynman had in mind, they are literally the computation of the consequences of the underlying theory, If they do not replicate reality 30 years after the base period, they will also not be right 100 years later.
They are simply wrong, as they imply an equilibrium climate sensitivity for CO2 over twice the observed effect.
And Feynman would have rapped your knuckles with a ruler for using the term “established theory.”
It is not clear what underlying theory is you think is disproven when a climate model does not yield exactly the result observed and needs to revised? A climate model is NOT a theory. It is an attempt to apply established theories to very complex data sets to make predictions. By “established theory” I was referring to such things as gravity, quantum interactions, thermodynamics, etc. – the basic stuff of physics. Which Feynman would not wrap me on the knuckles for relying on.
I guess I made it pretty clear that I do not find what you wrote to be compelling. To my mind it is part and parcel of the modern “conservatives” war on science because science so often does not support the preferred agenda. Your amazing statement about “the poor quality of current science” sets that tone. And that tone is continued by your reference to malevolent motives by working scientists.
While you find it easy to see political motives in findings of scientists that you wish were not so, you seem to be unable to see political motives in findings of scientists that you approve of.
LikeLiked by 1 person
A climate model is not a theory, it is the computation of the consequences of a number of theories.
Rather than type all afternoon, take a look at this explanation of climate models.
Click to access Curry-2017.pdf
Note that the models are “tuned” so that they reflect the known values from the mid 20th century. That tuning was in error in that it was presumed that all of the observed warming resulted from CO2.
If any part, however small, was due to natural variability, there is an error in the tuning, and that error will increase at every iteration of the steps of the computation, increasing the size of the error like compound interest.
So, the observed error between the models over the last 30 years falsifies the theory behind the models.
And from my point of view, it is the progressives who are at war with science, clinging to a theory that can be refuted with a single graph.
First you said a climate model is what Feynman had in mind when he used the word “theory.”
I disagreed and said . . .
“A climate model is NOT a theory. It is an attempt to apply established theories to very complex data sets to make predictions.”
Now you say . . .
“A climate model is not a theory, it is the computation of the consequences of a number of theories.”
Generally speaking, when a climate model seems to fail NONE of the theories in its computation is being proven false (a la Feynman) but rather that something is missing from the model or something is there that doesn’t belong or it has been fed bad data.
You are using the word “theory” very broadly. Maybe you mean prediction?
As for the Curry graph proving that progressives – and not you – ignore science, it is not accepted by most climate scientists as explained why here . . .
LikeLiked by 1 person
First, the graph is from Spencer, the explanation of how climate models work is from Curry.
Now, going back to Feynman’s explanation of how science works.
1 Make a guess (theory)
2 Compute the consequences of the theory(that is what the models do)
3 Compare with reality(experiment)
What the Spencer graph shows is that the Tropical mid troposphere heating(the first and greatest measurable consequence) does not occur at the projected rate, if at all.
So, the theory behind the models is disproven.
and the link you provide is totally irrelvant to the issue.
“Make a guess”
So what is the theory that being proposed by climate science – what you call the guess? A “theory” is a proposed cause of natural phenomena. A prediction based on existing theory and available data is NOT a theory.
My theory is that you are really confused. I have plenty of evidence. There are several flaws in the graph that you say proves progressives are at war with science. The article I linked to directly addressed that graph and pointed out those flaws. So, how is that irrelevant? Answer. It is not. That you are really confused is confirmed.
LikeLiked by 1 person
It doesn’t address Spencer’s graph though it does compare a similar graph by Christy.
Spencer’s graph does show the variability of the models and not just the average.
The alternative graphs they present are of surface temperature and sea level, both of which lag tropical mid troposphere by decades.
Tropical mid troposphere is the heat engine of global warming theory. The models vary on how much, but they all agree on a hot spot in that band. And they all project 2 to 3 times the effect of CO2 compared to measurement.
Did you read Curry’s document on how models work?
Consider this, of the millions of cells the atmosphere is divided into, we have actual, concurrent measurements of temperature and humidity for less than 1% The rest are extrapolated from the measured cells, some times hundreds of miles away. So, the models don’t even have a relaible inital state to start from.
Could it be they rejected your essay because of your remarks about Northam’s “political payoffs” with regards to vaccines? Yeah, I saw that and just scratched my head that you, a man of science, presented that with NO PROOF, just conjecture.
LikeLiked by 2 people
The essay was submitted over a week before the comment. so there isn’t even correlation, much less causation.
Fair enough. But it sounded pretty reasonable. That is why it was formed as a question.
Has there ever been a time in history when science hasn’t had to respond to politics? Or power?
The tobacco lobby lied with “science” for decades. The snack food industry? Liquor? Ethanol? Fossil fuels? Heliocentric solar system? COVID?
The point is that when money, and its corollary, power, are in play, science can be attacked, vilified, praised, corroborated. “Doctors” in white lab coats selling the benefits of cigarette brands come to mind.
In my opinion, the vast availability of information and misinformation today complicates the issue. Not to mention that science is evolving at a rapid rate beyond what we are probably capable of dealing with.
Yet that same amount of available sources does provide a tool to ascertain the better “truths”. Only if people are willing and able to make the effort.
Still there is this problem: “Just remember, what you are seeing and what you are reading is not what’s happening…”. Or, as we like to say, “do you believe me or your lyin’ eyes?”
LikeLiked by 3 people
RE: “We must all be better consumers of science reporting.”
One thing people can do is read the papers that news stories are based on. There is usually a link in the news story, and one can learn a lot by following it to the original source.
Most scientific papers adhere to the same basic structure:
Introduction. Describes the purpose of the study (often with historical references), lays out the organization of the material to be presented, may explain the significance of the conclusions.
Methodology. Explains what the authors did, and why.
Results. Presents the research findings. This section may be combined with the methodology section.
Conclusions. States in plain language what the researchers believe they have proved.
The introduction and conclusions are typically accessible to non-scientists.
“Results. Presents the research findings. This section may be combined with the methodology section”
Overall good take. However there is a step missing here and one that Doc Tabor refers to on most things : Peer Review. I know that is not part of the original paper, but it is important to the overall efficacy of the reported results.
Tend to agree with the idea that peer reviewed scientific results go a long way in convincing non-scientists that what they are reading is legit. However, in using the peer review, it is important to be careful of which peers are doing the review and if they have their own pre-conceived notions of how something SHOULD turn out (bias).