Precisely attuned to this topic, permit me to recommend Dr. Malcolm Kendrick’s excellent book Doctoring Data, which gives a very detailed overview of the many deceptive claims in (primarily) the health field. Many of his observations can be extended to any field of “research.”
To steal one of his examples: when a news headline breathlessly warns that drinking three alcoholic drinks a day doubles your risk of mouth cancer, should you be worried? Problem: the need to evaluate such claims. That requires looking up the study (if you have access!), knowing how to untwist the pretzel logic, etc. In this particular case the claim may be true, but the more relevant context is that one is far, far more likely to die of heart disease than oral cancer. And finally, there is some evidence that moderate alcohol consumption may actually reduce cardio disease. But even that claim needs quantification. A cardinal sin is to confuse association with causation. E.g. it’s not the one glass of red wine with dinner nightly that is protective; far more likely it’s that person who drink one, and only one such daily glass also do a bunch of other stuff that correlates with a healthy lifestyle. But the person buying red wine extract at the health food store, who otherwise does everything else wrong, probably won’t benefit.
It’s worth mentioning that he cites most of his sources. Furthermore most of those are easily verified by looking them up. Some academic databases may require a trip to a university’s library or to pay an exorbitant fee to access an article. Having the time (and a university to visit) I've checked some of his work, and other than the occasional typo, it rings true.
Much of the same information, albeit in a disjointed form, can be found at his website
Nietzsche addendum: In his excellent Beyond Good and Evil, he ponders what he terms the prejudices of philosophers. Nietzsche covers a lot of ground and don’t provide many examples. A core argument here is that ALL human behavior is driven by the subconscious, natural instinct. A partial list of those might be: the desire for survival, to reproduce, to seek security, food, acceptance of one’s fellows, to fit into a hierarchy (rule or be ruled), and so forth. Nietzsche argues persuasively that even apparently “scientific” thought, such as the reasoning of a philosopher, is likely driven by unconscious needs. To a large extent, if the above is true, then in large part it explains why researchers set up their investigation to favor a certain preconceived notion.
Finally, it’s worth noting that the above need not be deliberate intent to deceive oneself or others; it may be entirely innocent, and the persons doing it may be no more guilty of misconduct than would be a sleepwalker. Of course, it’s equally possible that deliberate intent to deceive is operant, or anything in between.
Finally, it should be apparent that “psychology” is always in play not just in the individual, but in human groups, be they a married couple or a large organization.
I tend to think in terms of 1. Soundness of the industry prior art and data/sampled domain, 2. Critical path of questions asked, and 3. Mode and form of inference being drawn. If one gets those things correct, then I will listen to the statistical argument. In the case you cited, where alcohol consumption is related against mouth cancer, the question posed is an orphan question. It bears no prior art, is 4 or 5 questions down the line (being asked too soon), and uses mild linear induction (probably with Cox Hazard Ratio metrics) to draw inference.
This is not how actual science is done - statistics or no. That is why I begin with the scientist's ability to prosecute an argument first. That tells me whether or not they truly understand the subject in relation to the scientific method - or are they merely a stats tech - running heuristics. This is what I do when investors have brought me in to get a research lab out of first gear, or complete a line of research. Statistics is a very large cul-de-sac, which can delude one into believing they are doing science 'because their stats are sound' - when nothing is further from the truth.
"It is not the credulous or those prone to magical thinking who are the primary opponents of science, but rather those who insist on disseminating false realities under the guise of its virtue."
It seems to me that science is being used as a weapon, it has been infiltrated by corporations who have the money to manipulate it to their own ends. What has happened over the past 4 years I find horrific in that a special version of "science" has been used to subjugate the populations. No questions to be asked, no discussion, no dissent we are told, we must only follow their "science".
I think it is safe to say if government/bug business/corporate media is involved, it is NOT science. Going forward, real trustable science will be done by private actors who are willing to be completely transparent about their biases and funding.
You have described exactly the procedure that recently drove perceptions that masks can protect against respiratory infections and that hydroxychloroquine can’t effectively treat Covid-19 symptoms, to cite two examples of “follow the lie-ence” -- thank you.
a friend scientist told me years ago that she studied bio-chemistry because the science is settled. I told her from all we learned in school, probably 90 percent was no longer there. That caused a silence. And then I asked how come there were still so many scientists. If the science is settled scientists are unnecessary costs.
Re: 1. Concoct the claim your club wants to be true.
I am struggling to understand the basics of "Club" Biology.
The engineering examples, bridges and planes,
What about social/cultural engineering?
Can the principles and practice of ES be appropriately applied to us?
Precisely attuned to this topic, permit me to recommend Dr. Malcolm Kendrick’s excellent book Doctoring Data, which gives a very detailed overview of the many deceptive claims in (primarily) the health field. Many of his observations can be extended to any field of “research.”
To steal one of his examples: when a news headline breathlessly warns that drinking three alcoholic drinks a day doubles your risk of mouth cancer, should you be worried? Problem: the need to evaluate such claims. That requires looking up the study (if you have access!), knowing how to untwist the pretzel logic, etc. In this particular case the claim may be true, but the more relevant context is that one is far, far more likely to die of heart disease than oral cancer. And finally, there is some evidence that moderate alcohol consumption may actually reduce cardio disease. But even that claim needs quantification. A cardinal sin is to confuse association with causation. E.g. it’s not the one glass of red wine with dinner nightly that is protective; far more likely it’s that person who drink one, and only one such daily glass also do a bunch of other stuff that correlates with a healthy lifestyle. But the person buying red wine extract at the health food store, who otherwise does everything else wrong, probably won’t benefit.
It’s worth mentioning that he cites most of his sources. Furthermore most of those are easily verified by looking them up. Some academic databases may require a trip to a university’s library or to pay an exorbitant fee to access an article. Having the time (and a university to visit) I've checked some of his work, and other than the occasional typo, it rings true.
Much of the same information, albeit in a disjointed form, can be found at his website
https://drmalcolmkendrick.org/
Nietzsche addendum: In his excellent Beyond Good and Evil, he ponders what he terms the prejudices of philosophers. Nietzsche covers a lot of ground and don’t provide many examples. A core argument here is that ALL human behavior is driven by the subconscious, natural instinct. A partial list of those might be: the desire for survival, to reproduce, to seek security, food, acceptance of one’s fellows, to fit into a hierarchy (rule or be ruled), and so forth. Nietzsche argues persuasively that even apparently “scientific” thought, such as the reasoning of a philosopher, is likely driven by unconscious needs. To a large extent, if the above is true, then in large part it explains why researchers set up their investigation to favor a certain preconceived notion.
Finally, it’s worth noting that the above need not be deliberate intent to deceive oneself or others; it may be entirely innocent, and the persons doing it may be no more guilty of misconduct than would be a sleepwalker. Of course, it’s equally possible that deliberate intent to deceive is operant, or anything in between.
Finally, it should be apparent that “psychology” is always in play not just in the individual, but in human groups, be they a married couple or a large organization.
Very good resources, SD
I tend to think in terms of 1. Soundness of the industry prior art and data/sampled domain, 2. Critical path of questions asked, and 3. Mode and form of inference being drawn. If one gets those things correct, then I will listen to the statistical argument. In the case you cited, where alcohol consumption is related against mouth cancer, the question posed is an orphan question. It bears no prior art, is 4 or 5 questions down the line (being asked too soon), and uses mild linear induction (probably with Cox Hazard Ratio metrics) to draw inference.
This is not how actual science is done - statistics or no. That is why I begin with the scientist's ability to prosecute an argument first. That tells me whether or not they truly understand the subject in relation to the scientific method - or are they merely a stats tech - running heuristics. This is what I do when investors have brought me in to get a research lab out of first gear, or complete a line of research. Statistics is a very large cul-de-sac, which can delude one into believing they are doing science 'because their stats are sound' - when nothing is further from the truth.
Brilliant
Ethical Skeptic is a boss
I wonder how Thomas Kuhn would feel about the state of science today....
Every single time I hear the phrase “peer reviewed” I cringe. It means NOTHING to me. One word in the breakdown that stood-out to me was “interns”.
Peer review is little more than ideological gatekeeping.
Magic 8-Ball says, “It is decidedly so.” (Also says, “Outlook not so good.”)
Powerful.
What a great synopsis of how "Science" has been operating for the last 50 years at least.
Hopefully change is a coming.
"It is not the credulous or those prone to magical thinking who are the primary opponents of science, but rather those who insist on disseminating false realities under the guise of its virtue."
Love the quote!
Is that from J.P.Moreland ?
I love your graphic and will use it in my memes!
Brilliant!
Thank you.
It seems to me that science is being used as a weapon, it has been infiltrated by corporations who have the money to manipulate it to their own ends. What has happened over the past 4 years I find horrific in that a special version of "science" has been used to subjugate the populations. No questions to be asked, no discussion, no dissent we are told, we must only follow their "science".
I think it is safe to say if government/bug business/corporate media is involved, it is NOT science. Going forward, real trustable science will be done by private actors who are willing to be completely transparent about their biases and funding.
Beautiful. Couldn’t be more timely. We needed this. Thank you.
You have described exactly the procedure that recently drove perceptions that masks can protect against respiratory infections and that hydroxychloroquine can’t effectively treat Covid-19 symptoms, to cite two examples of “follow the lie-ence” -- thank you.
a friend scientist told me years ago that she studied bio-chemistry because the science is settled. I told her from all we learned in school, probably 90 percent was no longer there. That caused a silence. And then I asked how come there were still so many scientists. If the science is settled scientists are unnecessary costs.
Your friend might be smart, but they also ignorant :)
book smart not streetsmart