Does anyone have any good examples of bad science masquerading as good science that they'd like to point out?
That is to say topics published in reputable sounding journals, yet the scientific basis is lacking. There are a variety of examples of so-called evidence based approaches in the medical sciences that are based on very weak objective evidence for example. Pharmaceutical corporations that cherry pick studies for example.
Oh and I also kindly ask not to discuss evolution in this thread as I don't want it to be derailed.
i would love for this thread to continue. I do not have any examples at the moment, but can you give us an example to start us off.
ok, i think i have one, its a reach:
i was watching "the universe" a series on history channel (i know its not a published paper), and there was an episode on parallel universe. It included string theory, membranes etc. The episode felt more like science fiction than scince fact. I just felt that the scientist were taking too many liberties with their theories that were based on very, and i mean very little data. Even though i liked their theories, it felt more like theories based on theories, it did not feel like science to me.
I will refrain from going into evolution directly, but carbon dating, evolutionary epochs, and the dating of the planets and universe has always made me a little uncomfortable. I know that there is little data to go on, and a lot of inferences have to be made, but how do these inferences differentiate us (modern scientist) from those living in the ptolemaic universe. i guess its all part of the scientific process.
its 3am on my end, and i am half asleep, give me a break. no, bring on the heat. i should not have posted if i was not ready to back it. do answer the original question first though.
I think it is BS (bad science) because nearly all conclusion are based on insufficient evidence, perjury, fabrications and defunct methodology. One should not model a chaotic system using a 20 year calibration period, with no validation period, to predict 100 - 200 years in the future.
Truth be told, I would loose my job if my employers knew I had this stance.
Both string theory and E8 theory are completely mathematical models that have little basis in experimentation or observation. They seek to change reality to fit their model, since there are no experiments that would require them to change the theory.
I agree about anthropogenic global warming. While I think that human activity most certainly does influence the global climate (it would be even more egotistical to think we have no influence), the science based around this truism is shaky - we really can't say for sure what influence humans are having and to what degree. The climate is a non-linear, complex dynamic system for which no single cause at a single point in time can be pointed at as the origin of climate change. I'd be more worried if the climate wasn't changing.
Psychology may not be be bad science, but it's not good science. It depends on a tabula rasa model, where different past events cause people to behave different ways. Using scientific methodology for subjective experience makes several unfounded assumptions: A) everyone experiences everything the same way; B) self reporting is accurate; C) their models of subjective experience correspond to objective reality (while many of them do a good portion of the time, none of them do all the time, therefore they can't be used to make accurate predictions); D) our past experience is the only thing that influences who we are. I think psychology is valid, and that it makes useful models based on observations and empirical testing, but it's a very soft science - or maybe hard philosophy?
As someone mentioned: pharmaceuticals. It's a business more than a science. They aren't interested in making discoveries, unless it helps their bottom line. Pharmaceutical companies are the top grossing businesses in the world, they sell shoddy products, they employ short term testing for drugs with long term side effects, and there is little account of the placebo effect in a lot of research.
As far as climate change goes, there's enough solid evidence that there's *some* change, but the whole matter is so entangled in politics, that there's enough fabrications, propaganda, etc. on both sides. In this situation the presence of bad science for both sides is expected and does not refute their respective points - to refute it you need to refute the actual cases of proper science or prove that none exist.
The results are presented as too definite considering the study does not filter out a lot of other factors. The most obvious problem is that it does not consider the effect of the parents' gender in addition to their sexual preferences.
lol... if you're getting your science from news articles, then all studies are bad studies If you read about it in a non-scientific magazine or saw it on the news, odds are it's bad science in some way or another.
How 'bad' does the science have to be? I mean... ideally, every statistical healthy study would be better if every human on the planet, and clearly a sample size of 5 isn't enough to draw a sound conclusion, because no single person has one property, so you can't statistically balance out their other characteristics enough to draw a conclusion about the thing you're studying... but where do you draw the line between bad science and good, between those two?
The results are presented as too definite considering the study does not filter out a lot of other factors. The most obvious problem is that it does not consider the effect of the parents' gender in addition to their sexual preferences.
There are a variety of potential biases in such a study. First, the selection of the original cohort.
Between 1986 and 1992, prospective lesbian mothers who were inseminating or pregnant through DI were recruited via announcements that were distributed at lesbian events, in women's bookstores, and in lesbian newspapers throughout the metropolitan areas of Boston, Washington, DC, and San Francisco.
Now if the data controlled for factors such as educational, other sociodemographic background, intelligence etc. Then the results would be more solid. Although the genetic influences of the sperm donor is still hard to control for.
Although:
No CBCL differences were found among adolescent offspring who were conceived by known, as-yet-unknown, and permanently unknown donors (Wilks's {lambda} = .70, F14,78 = .80, P = .752) or between offspring whose mothers were still together and offspring whose mothers had separated (Wilks's {lambda} = .69, F14,52 = 1.68, P = .088).
During pregnancy, the prospective mothers took classes and formed support groups to learn about childrearing.21 They were actively involved in the education of their children23–25 and aspired to remain close to them, however unique their interests, orientations, and preferences may be.21–25 To the extent that the NLLFS mothers may have achieved this goal, numerous studies showed that having a satisfying relationship with one's parents is associated with a more favorable adolescent adjustment.*
Secondly, instead of objective evidence, a questionaire was used:
Achenbach's CBCL/6–18 (Child Behaviour Checklist). The controls were the original Achenbach sample.
How 'bad' does the science have to be? I mean... ideally, every statistical healthy study would be better if every human on the planet, and clearly a sample size of 5 isn't enough to draw a sound conclusion, because no single person has one property, so you can't statistically balance out their other characteristics enough to draw a conclusion about the thing you're studying... but where do you draw the line between bad science and good, between those two?
It really depends on what is being measured. A sample size of 10 can be quite suggestive if the difference in outcomes is death. Large sample sizes (hundreds to tens of thousands) are needed however if many of the potential biases are to be eliminated. Questionaire based 'science' for example, should have sample sizes in the thousands.
Secondly, the rarity of the event must also be considered - very large sample sizes are needed to examine the true risk of rare, but serious reactions to vaccinations for example.
I think it is BS (bad science) because nearly all conclusion are based on insufficient evidence, perjury, fabrications and defunct methodology. One should not model a chaotic system using a 20 year calibration period, with no validation period, to predict 100 - 200 years in the future.
Truth be told, I would loose my job if my employers knew I had this stance.
So you accept the hypothesis, but believe the data is inconclusive? What would it take to prove such a thing - I realize our computational output is a tiny fraction of the actual complexity of such a system.
Both string theory and E8 theory are completely mathematical models that have little basis in experimentation or observation. They seek to change reality to fit their model, since there are no experiments that would require them to change the theory.
They do have some basis, since they attempt to predict much of the phenomena already explained by the standard model / general relativity. But given some of the triumphs of theoretical physics in the 20th century, is it not possible that we will make some sort of major breakthrough that allows such a theory to explain physics that the standard model (+general relativity) cannot? Or do you think it is a blind path?
Advocates claim some sort of Occams Razor argument in favour of such theories, but such arguments are never scientific, but instead merely based on aesthetics.
Psychology may not be be bad science, but it's not good science. It depends on a tabula rasa model, where different past events cause people to behave different ways. Using scientific methodology for subjective experience makes several unfounded assumptions: A) everyone experiences everything the same way; B) self reporting is accurate; C) their models of subjective experience correspond to objective reality (while many of them do a good portion of the time, none of them do all the time, therefore they can't be used to make accurate predictions); D) our past experience is the only thing that influences who we are. I think psychology is valid, and that it makes useful models based on observations and empirical testing, but it's a very soft science - or maybe hard philosophy?
The problem with most non-pharmacological research in psychology, is that it is never placebo controlled. The same results that would be ridiculed in a pharma study are somehow accepted in psychology as evidence-based!?! The many claims of cognitive behavioral therapy for example. Results are almost always based on subjective questionnaires and not objective measurements - who are they trying to fool?
As someone mentioned: pharmaceuticals. It's a business more than a science. They aren't interested in making discoveries, unless it helps their bottom line. Pharmaceutical companies are the top grossing businesses in the world, they sell shoddy products, they employ short term testing for drugs with long term side effects, and there is little account of the placebo effect in a lot of research.
There has also been a recent controversy about anthropology:
Anthropologists have been thrown into turmoil about the nature and future of their profession after a decision by the American Anthropological Association at its recent annual meeting to strip the word “science” from a statement of its long-range plan.
So you accept the hypothesis, but believe the data is inconclusive? What would it take to prove such a thing - I realize our computational output is a tiny fraction of the actual complexity of such a system.
I don't accept the hypothesis of anthropogenic global warming as being plausible. Though, it is just relative to the magnitude of temperature change that increased carbon dioxide concentrations in the atmosphere cause. One simply can't state that there has been no warming due to man made emissions of carbon dioxide gas.
Over the last 400 years the mean global temperature has increased by 1.5K. Under the IPCC models using the most liberal values of climate sensitivity (all unknown factors) and feedback factor, the warming attributed to increases in carbon dioxide concentrations is 0.2K. Roughly 13%.
The forcing attributed from carbon dioxide gas concentrations accounts for 30% of the warming calculated in their models. The other 70% is attributed to positive feedback mechanisms and climate sensitivity 'k'. The feedback factor used is an aggregation of all known and quantified feedback mechanisms, while the climate sensitivity factor and carbon dioxide forcing coefficients are calculated using the calibration period as outlined above. These is strange because forcing due to changes in carbon dioxide concentrations can be calculated in a lab using the stefan boltzmann equation.
I have seen k values range from 0.05 to 0.71 depending on algorithms used. Common feedback factors used range from 2 to 4.5. They choose a water vapour feedback factor of 1.71 but most studies conclude that it is in fact a negative feedback instead of a positive. Hence, they use grossly exaggerated feedback factors.
Miskolczi's paper describes that there is in fact a net negative feedback which stabilizes the climate and disproves the so called "tipping point."
Studies have displayed that the warming over the calibration period of 1980 to 1998 is attributed to el-nino portion of the Southern Pacific Oscillation and a abnormal decrease in cloud cover preventing less short wave radiation to reflected back into space. If this is true the magnitude of forcing attribution to carbon dioxide concentrations is somewhat less than predicted under the models. Other researchers state the warming experienced over the period was due to the ozone hole and nothing to do with carbon dioxide emissions.
Empirically determining the forcing rather than trying to use predictive models would be a start. They use the results of their model to prove that increased carbon dioxide concentrations cause increased mean global temperatures but even to do so implies that they used the conclusion to prove the conclusion for the models rely on concluded relationships to operate.
Anecdotally, all one has to do is look through geological climate history to see that there is little correlation between carbon dioxide concentrations and mean global temperature.
All other hypotheses based on how the climate will behave in response increased mean global temperatures are just purely exercises in mental masturbation. The system is chaotic, there is not way in hell they can predict to any degree of certainty long term changes in weather patterns.
---
I thought the hypothesis was sound until I researched it.
There is a lot of bullcrap out there, but it isn't bad science masquerading as good science unless it has the air of science about it. Acupuncture is one of them. It never pretended to be part of the scientific system, not as far as I know, anyway. Maybe acupuncture really does work better than a placebo, but its roots are in Chinese myths, not in the modern scientific framework, so I wouldn't put it in the same category as lobotomies or leeching or homeopathy or chiropractice--all with the air of science, nothing to do with traditions.
If they manage to get published, then it is science, in my opinion. It starts to become good science when it makes good sense. For those of us who don't trust our ability to examine the evidence and make a good decision or just don't want to take the time, it becomes good science when many other professional scientists agree with it.
What about the psychiatry then? Many professional scientists 'agree' with it, yet it has consistently failed to produce coherent models of mental disease and uses subjective measures for both diagnosis and to measure the success of treatments.
What about the psychiatry then? Many professional scientists 'agree' with it, yet it has consistently failed to produce coherent models of mental disease and uses subjective measures for both diagnosis and to measure the success of treatments.
I have little familiarity with psychiatry, but I have low expectations for a scientific field that is less than 100 years old and is all about fixing problems with the most complex thing in the universe. I know that there are at least some methods within psychiatry that actually produce measurable and productive results, such as curing arachnophobia, or other "irrational" fears.
That whole thing was blown way out of proportion by outlets like the New York Times. There was a lot of knee-jerk reaction from both scientifically-inclined anthropologists (myself included) and the pure post-modernist crowd. The dust seems to have settled on that one until someone else decides to kick it up again, which seems to happen every five years or so.
I have little familiarity with psychiatry, but I have low expectations for a scientific field that is less than 100 years old and is all about fixing problems with the most complex thing in the universe. I know that there are at least some methods within psychiatry that actually produce measurable and productive results, such as curing arachnophobia, or other "irrational" fears.
But can psychiatry adequately account for it's definition of 'irrational fear'? In the case of spiders, it can be easily argued that an irrational LACK of fear would be far more potentially dangerous...
Galileo claimed to have proof of the Copernican solar system model when in fact what he had was data of the moons of Jupiter and the phases of Venus. He believed that Copernicus was right (or that some variant of a heliocentric model was right), but he couldn't prove it. Kepler, using Brahe's data, came as close as anyone without modern telescopes could come to proving it, but Galileo hated Kepler and refused to share his own telescopes with him. So, the one person who could probably have helped Galileo was shunned. And Galileo was an even bigger narcissistic ass to the Vatican, which certainly didn't make things any easier on him once the crap started to hit the fan over his publications. People like to blame the Catholics, but, Galileo kept pour gas on himself even as the Church was holding the match. (Galileo believed that the Pope, who had been a good friend of his back before he was the Bishop of Rome, would help Galileo out in a pinch--turns out the Pope's loyalty isn't to his old friends if it comes to that or the Church.)
So, bad science, bad politics, and bad intellectual honesty all in one package.
The rigirous definition of a Scientific theory says that it must be testable, and no aspect of string theory is testable, despite how much I want it to be true.
Great example. Galileo's not one of my favorite historical figures. Apparently he also plagiarized a Jesuit correspondant, which upset the Jesuits to the point that they wouldn't help him out. And if Urban VIII abused his power a bit, well, anyone would have been tempted to do that.
I have seen a whole bunch of bad science in the form of a correlation presented as causation. I don't have any examples handy, but many of them are articles on parenting.
The "study" that showed immunizations causing autism were really bad science that hurt a lot of people. Many in my community don't immunize their kids as a result.
Galileo, by saying that the Vatican was not the point around which the universe spins, was tearing the entire world apart with what he said.
The Pope was the grandson of the instigator of the Renaissance, the thing that led to scientific inquiry. I do not see how, first, Pope Medici could with any degree of good conscience turn his back on his grandfather's legacy; and, two, nearly allow his childhood teacher to be executed.
The Medici begin the Renaissance and two generations later begin the Inquisition to try and put the lid back on the Pandora's box it wrought.
The Copernican theory had been around since 1543, and did not bother the Church because it was very useful. It wasn't until 1633 that Galileo was punished for espousing it. The idea was hardly new.
The Pope was the grandson of the instigator of the Renaissance, the thing that led to scientific inquiry. I do not see how, first, Pope Medici could with any degree of good conscience turn his back on his grandfather's legacy; and, two, nearly allow his childhood teacher to be executed.
The Medici begin the Renaissance and two generations later begin the Inquisition to try and put the lid back on the Pandora's box it wrought.
Galileo wasn't attacking the church, he was attacking the pope, who as a former friend, should have been able to expect better treatment than public humiliation as a "Simplicio." The pope had been trained as an engineer, not exactly an untutored lout.
When Galileo's book was published, Urban demanded the inquisition find him guilty of something. They came back with an accusation of "poor judgment" which apparently wasn't sufficiently actionable. The pope demanded they find him guilty of worse, so they did.
I don't think Galileo was striking a blow for science, he was trying to impress everyone. He insulted his friends, plagiarized his allies, and showed basically zero political savvy. The pope, on the other hand, abused his power by ordering the inquisition to help him take revenge. But as Prof. Richard Olson pointed out, "misuse of power is neither confined to religious figures nor does it seem to most to be an essential feature of religion." His account of the whole Galileo affair is fascinating, and at least that part of his book on Science and Religion can be read online; it is a part of the introduction. The rest of the book is good, too.
I don't accept the hypothesis of anthropogenic global warming as being plausible. Though, it is just relative to the magnitude of temperature change that increased carbon dioxide concentrations in the atmosphere cause. One simply can't state that there has been no warming due to man made emissions of carbon dioxide gas.
Over the last 400 years the mean global temperature has increased by 1.5K. Under the IPCC models using the most liberal values of climate sensitivity (all unknown factors) and feedback factor, the warming attributed to increases in carbon dioxide concentrations is 0.2K. Roughly 13%.
The forcing attributed from carbon dioxide gas concentrations accounts for 30% of the warming calculated in their models. The other 70% is attributed to positive feedback mechanisms and climate sensitivity 'k'. The feedback factor used is an aggregation of all known and quantified feedback mechanisms, while the climate sensitivity factor and carbon dioxide forcing coefficients are calculated using the calibration period as outlined above. These is strange because forcing due to changes in carbon dioxide concentrations can be calculated in a lab using the stefan boltzmann equation.
I have seen k values range from 0.05 to 0.71 depending on algorithms used. Common feedback factors used range from 2 to 4.5. They choose a water vapour feedback factor of 1.71 but most studies conclude that it is in fact a negative feedback instead of a positive. Hence, they use grossly exaggerated feedback factors.
Miskolczi's paper describes that there is in fact a net negative feedback which stabilizes the climate and disproves the so called "tipping point."
Studies have displayed that the warming over the calibration period of 1980 to 1998 is attributed to el-nino portion of the Southern Pacific Oscillation and a abnormal decrease in cloud cover preventing less short wave radiation to reflected back into space. If this is true the magnitude of forcing attribution to carbon dioxide concentrations is somewhat less than predicted under the models. Other researchers state the warming experienced over the period was due to the ozone hole and nothing to do with carbon dioxide emissions.
Empirically determining the forcing rather than trying to use predictive models would be a start. They use the results of their model to prove that increased carbon dioxide concentrations cause increased mean global temperatures but even to do so implies that they used the conclusion to prove the conclusion for the models rely on concluded relationships to operate.
Anecdotally, all one has to do is look through geological climate history to see that there is little correlation between carbon dioxide concentrations and mean global temperature.
All other hypotheses based on how the climate will behave in response increased mean global temperatures are just purely exercises in mental masturbation. The system is chaotic, there is not way in hell they can predict to any degree of certainty long term changes in weather patterns.
I wouldnt be so sure, I used to support the anti-global warming movement quite vigorously, but after dozens of arguments where I could simply not prove that they global warming groups were wrong, It really made me reconsider which side I should take. Should we really be attacking these people? They only want to protect the world, and the byproduct is a much healthier world with a huge reduction in pollutants.
When it comes to the research, I found it astounding that even without any extra water coming from glaciers and the poles, a few degree increase would lead to an expansion of water which appeared very small, but when multiplied by the average depth of the ocean, lead to a twenty foot rise in sea levels. This is just from the expansion of existing sea water. You can find the charts and data yourself and confirm for yourself.
When it comes to the ice, I was astounded to find that Greenland did not have sufficient altitude to protect its ice sheets from the summer heat. The latitude at which all ice will melt can be measured fairly easily, by its average temperature. The ice will recede to the general point where that average temperature exists, which itself would be moved hundreds of miles north.
When it comes to feedback mechanisms, shouldnt we be more safe than sorry? Also, I dont expect you to know this, but if you have ever worked in the petroleum industry, they have insane models of how permafrost works. I was working with one of the engineers who worked on the Alaska pipeline, and he was telling me about how they once accidentally collapsed a mountain because they didnt waterproof the pipe's insulation and so it funneled water to the base of a the mountain and wore it away. He pulled out some charts which showed the different results of different kinds of insulation, and it was astounding. A small change in insulation would cause dozens, if not hundreds of feet of the surrounding permafrost to melt. Not only that, but without insulation, the pipe would melt so much of the permafrost that it could not be measured on the maximum scale of the instruments used.
Perhaps global warming has not been proven, but its quite obvious that even small changes can make huge differences. If a pipe which is 40 degrees hotter than the surrounding can melt everything in sight, its fair to say a few degrees would melt at least a few extra inches of permafrost, which amounts to a crapload of methane.
Really, do you really expect nothing to happen? Even in the last 20 years Ive seen every place I lived in change dramatically. It doesnt rain anymore where it used to rain every day, there is overwhelming amount of brush on the forest floor, where there has not been brush for all of recorded history. There are no mosquitos where they used to be extremely prevalent. The local lakes have lowered so much that the ground rose, and now the shore is hundreds of feet out. You have the ecosystems changing at almost every place on earth, its just undeniable.
BTW, I was a park Ranger, and the differences have been noted in every environment in the entire state I served in. The Rangers in my current state tell me the same, its nothing like what it was when they were young.
It is really undeniable that something is happening, regardless of if its from CO2. How likely is it that dumping all this crap into the atmosphere wont do anything. Really? You wouldnt raise the density of any chemical in any manufacturing process by 12%, it would obviously ruin the product, so why wouldnt the same thing happen to the atmosphere?
Sigh. No. I do not subscribe to the religion of 'everything that man does is evil and hence destroys'. Noting changes in the environment's datum as proof is philosophically incorrect. You're confusing correlation and causality.
This site uses cookies to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies. We have no personalisation nor analytics --- especially no Google.
