Bogus Causality Claims in the Media

The people that do these studies have to make a living. Bad news=more funding.

I wonder where these kids are getting all the second hand smoke. Smoking has been banned in all high schools and every other public place. They must really think we're dumb. My generation grew up surrounded by the smoke of our parents.................where are all the deaf people?

I'll take some second hand smoke any day over being trapped in a car with one of these kids that have the stereo so loud it rattles the windows in surrounding buildings. Duh................

I just read an article about environmental groups that are working hard to get all public fire works displays banned in the US, because they spew carbon and contribute to global warming. They are actually making progress. One bill stalled in congress, but now that the EPA has been granted such sweeping power, policy can be set without congress.

Could you repeat that please? OH ME GOD! Me eyes are deef!

http://www.mendhak.com/102-smoking-is-good-for-you.aspx

i wish i had seen this before giving up

Thank's for the laugh.

"Dr. Lalwani speculated that second-hand smoke could lead to hearing loss by reducing the amount of oxygen-rich blood reaching the delicate structures of the inner ear. "Tobacco smoke is known to have a bad effect on blood vessels … and the inner ear is incredibly sensitive to blood flow," he said."

If it has such a bad effect, and the inner is is so incredibly sensitive to blood flow, how come all smokers aren't stone deaf?

what?

It needs to be combined with high moisture content in the air. The tube of Eustachius gets clogged that way.

And yet snorkeling keeps them squeaky clean? Oh the causality conundrums!

ARE SMOKERS BOTHERED BY THOSE LOUD HARLEY PIPES FROM ANOTHER THREAD???

I quit 5 years ago. I'm glad I did, but sometimes..................I still want one.

What I'd like to know is, has your hearing been restored after quitting?

Nope. I'm still as deaf as a rock band member.

However, I still hear well enough to seriously want to choke the life out of these a**holes who turn the base up in their car radios till my rear view mirror vibrates.

You mean those teenage a**holes blasting the music on their way to the drive-through for twenty plates of fries - hell, second hand smoke isn't even required for the cotinine level on that diet.

Makes me wonder whether potatoes, tomatoes, black tea and especially instant tea, also cause hearing loss...

Cars with bass! I hate that too. I'm working on an EMP generator as a solution.....

Yes

Why did you quit Harleys?

"Why did you quit Harleys?"

Every time I tried to smoke one the tailpipe burnt my lips.

Sound Effect - Rimshot.wav

I can't see how a car stereo could possibly hurt eardrums

Thankfully I sustained no hearing loss from the second-hand phone book shredding.

I'm sure that guy's hearing is fine. I didn't see any cigarette smoke drifting into the video.

what pisses me off to no end is that these people try to pass there data off as being factual, when in truth, it's only a theory at best. most of the time they're just blowing smoke up your ass. the sad part is that people accually beliieve this garbage.

if these type of people did'nt make these outrageous claims, there funding would stop. it's all about money and job security. it's sad, but i believe the majority of people are nieve enough to swallow this garbage

The issue isn't really about finding false causality claims. the difficulty is finding articles without such claims. People are taught to use statistical phrasology in their reports. very few use them correctly. They are also taught to be positive. Look at the emotive dfference in "You have a 2.5 % chance of winning up to $111,000,000 on the next power ball drawing" and "You have a 97.5 % probability of losing on the next powerball drawing. Causality is a statistical term, but you seldom see in a report what percentage defines causality versus corelation. Almost all write-ups on the dangers, or non dangers of smoking use statistical terms with no reported basis. Normally, little is actually stated; much is implied ; and more is left to be infered. In the article being discussed and with the data supplied, the conclusion could been that people with poor hearing have a tendency to smoke.

Funny no one on this thread has mentioned anything about . . . GW.

Not while they have our phones tapped

Is it the science or the science journalism that's the problem? Shouldn't there always be something about corroborating studies (or lack thereof) whenever the subject is some kind of scientific study?

In my opinion it's both the science and the science journalism at fault.

The science is, for starters, driven by a social/political agenda, namely the "denormalization" of smoking and the demonization of second hand smoke. This particular social agenda has been seriously tarnished by the manner in which they have repeatedly abused statistics and "science" in order to make their social case. In fact they have lost all credibility in my eyes because of this, and now have the same status as "Reefer Madness" (remember that one?) and similar propaganda.

There is big corporate money behind the "denormalization" campaign. If they want to fund weak science to fuel the campaign, they can do so and get away with it, as long as the science journalists continue to accept and publish it without critical comment. The social agenda - creating irrational fear of smoke, in the public - is achieved.

I don't believe for a second that they would fund a study which is actually designed to corroborate the causality claim in this case. Ha ha! Only a fool would even consider it. They know it is bogus. It's the uneducated public they are terrorizing with these lies. You would think that publicizing the grounds for rational fears would be enough - but apparently that is not the case - and this is where the foul and corrupt intentions of the backers of this agenda is showing through.

It hardly surprises me that there are scientists to be hired by the dozens, to concoct weak scientific studies and draw illegitimate - and even absurd - conclusions, to prop up a social plan that big pharma buys into. Sad, but a fact of life.

It is more disheartening that the science journalists don't have the skills, apparently, to tell the difference between science and utter hogwash.

Really!? You are in favor of second hand smoke? Really? I agree, there is a social agenda (I'm a willing participant) against smoking and second hand smoke, but that's because it's noxious, plus it's a known health hazard. "Denormalization"? Really? Where are corporate profits in reducing the market for a product? You think big pharma is making more money off quit smoking patches than with big cancer and COPD drugs that insurance pays for?

I read a print article today on the same study in my local paper - yes, real paper delivered to my door - that did not come across as positive that there was significant causality as the linked article did. It did say the research "controlled for other factors", whatever that means. Other than that it described the study and said the subjects with higher levels of the chemical had measurably poorer hearing. Pretty bland stuff, just the basic facts. Of course, I neglected to expend a bunch of my personal energy reading more than that into it. Still no mention of other studies, or why it was done in the first place, and no healthy skepticism.

Without venturing to informally survey your hearing and other senses, I notice your use of the word "Really?" in a similar way that a deaf person says "What?", and using it to bracket a completely made up suggestion that I am "in favor of second hand smoke". _{are you hard of reading?}

"denormalization" is exactly the term used. Really.

I have no respect for this campaign. If the reason for it was concern about the risks, why haven't they stuck to the facts. It should be enough, but it isn't enough for them. They have crossed into some very murky territory, where smokers are the new "witches" - irrationally blamed for things they could not possibly cause.

As for big pharma, I don't have the inside track, but I know they are ruthless about suppressing any kind of competition, or anything that will get in the way of their long term plans. How much do you think they stand to make from the boomer generation afflicted with Parkinsons and Alzheimers? I'd say there's a chunk of change there. Nicotine has value as a treament for cognitive impairment of various kinds, and smokers have reduced rates of these demented conditions. I notice the media is forecasting an "epidemic of alzheimers".. there's money motive in there somewhere, I'm pretty sure.

I don't see the link [presentation of information] between smoking & reduced rates of dementia. I probably just missed it

Nicotine & smoking are not the same

there probably aren't as many demented smokers, they die off before then

of course there is going to be more Alzheimers, people are living much longer & we are better at diagnosing all sorts of conditions, which would have been described as getting old

Re: smoking and reduced rates of dementia - I saw this commented but did not find the original research. There must have been a study of some kind, because subsequent studies on nicotine and cotinine based their hypothesis on this prior observation.

There's a reference to the "early reports" of protective effects here. And this interesting study, on the effect of the source of funding on the results of studies of cognitive effects of smoking, found positive results associated with tobacco industry funding, while results without industry funding were "split".

However, all the research being conducted now seems unequivocal in claiming the opposite: that smoking is correlated with, and blamed as the cause of, a substantially higher risk of age-related cognitive decline and alzheimers.

Another explanation for the discrepancy is given in this systemic review study, which found that case-control studies and cohort studies produced opposite results. Evidently, cohort studies are now the biggie and are considered better than case-control. Meanwhile, the leap to causality is seen everywhere, including the conclusions here: "If new results confirm that smoking is associated with increased risk of AD, then smoking prevention and cessation should become public health priorities in the fight against dementia."

I have yet to see any explanation, how or why it is that nicotine and cotinine are confirmed to have cognitive benefits, while their consumption by smoking is now claimed to cause exactly the opposite effect....

This doesn't make sense, therefore raises a red flag, IMHO, about the role of the "agenda" of the day, similar to the "hearing loss" issue.

The whole business has such a bad smell on it.

your cognitive abilities go way down after a stroke

not being able to breath also makes thinking a challenge

you seem to be guilty of the same behavior you decry in your original post

I'm not making any claims of causality here. Just reporting what I read, and sayin I'm not satisfied.

I'm with you. Cigarette smoke is a dead horse. Funding, carrying out, and publishing frivolous studies to find more and more minuscule things that are caused by cigarette smoke is a waste of time at best, bad, scare tactic science at worst.

All while the government continues to subsidize tobacco.

Astonishing! And I thought it had boiled down to the smokers by now, subsidizing the government by paying the addiction taxes.

The government is addicted to taxing and spending. They've got their feet on both sides of so many fences, I don't even think they know how many.

Can you say "Special Interest", Grasshopper? Lobbyist, maybe? Lawyer for sure.

All it takes is enough money to purchase the votes needed to get a subsidy in the first place. Once entrenched, your on easy street.

Farmers make money, lawyers make money, tobacco companies make money, lawyers make money, insurance companies make money, lawyers make money, hospitals make money, and so it goes..........

Don't get me started.

It has surprised me for a long long time that big tobacco hasn't forced the legalization of pot.

It actually goes back to the depression. Farm subsidies were intended to be temporary, until farmers could take care of themselves. Welfare was intended to be temporary, until people could go to work and get on their feet. Once the government puts a policy in place, it never seems to go away.

Now we've got a welfare nation and a government that insists that they have to borrow more money to take care of us.

Garthh,

You might find this article interesting - from the news today - where misreporting, that is, misplaced "causality" claims, are discussed by dissenting experts on the very subject of 'Alzheimers prevention'.

Point made by the experts cited, it is the controlled studies (not the cohort studies) that allow causality to be reasonably well established.

wait wait

here's the best one

Their estimates for the risk factors worldwide differed because some behaviors and conditions are more common than in the United States. So low education accounted for 19 percent, or 6.5 million cases, worldwide, while physical inactivity accounted for 13 percent and obesity 2 percent.

I'll summarize

Uneducated people, forget things

I saw one bit that may actually mean something

The panel found the strongest evidence for only one conclusion: that the herb gingko biloba does not prevent Alzheimer's. There was moderate evidence that neither vitamin E nor drugs called cholinesterase inhibitors, used to treat dementia symptoms, decrease risk of Alzheimer's. And there was moderate evidence that the gene ApoE4 significantly increases Alzheimer's risk, as does menopause therapy with estrogens and progestins.

You got it! Cheers.

butwaitthere'smore.

Here's a fun headline:

Genetically modified tobacco plants to fight HIV?

Clinical trials haven't started yet; just thought I'd add more combustibles to the fire with a little GMO twist to set Kramarat off.

After they extract the drugs, do they smoke the leftovers?

_{This is the only smoking smiley I could find.}

You are a naughty girl!!!

Now all they have to do is make cigarettes out of the modified tobacco plants to ward off HIV, allow smoking in bars again................we can all go back the fun times of the 70s.

I had a physicians assistant tell me the best way to quit smoking cigarettes was chewing tobacco

the little pouch things

I'm still smoking. After taking a drag I never blow the smoke into the wind though...........................it's the second hand stuff that kills.

Never liked chewing tobacco......................baseball either.

Probably the reason I didn't care about building nets to save the fans.

I'd think a better way to quit smoking than chewing those little pouches would just be to chew the cigarettes themselves.

"It is more disheartening that the science journalists don't have the skills, apparently, to tell the difference between science and utter hogwash."

And a complete tragedy when journalists invent utter hogwash links for 'impact'.

Good catch.

Sadly just one of many examples of flawed journalism/science reporting, re-posted on CR4, that I have been moved to comment on.

Of course it go's both ways.

Should be. But the funny thing about statistical analysis is that no study ever studies the same sample twice. And the methodology is geared to reflect the proof the statistician is looking for, rather than just the compilation of raw data.
Add to this the problem of the subjects themselves and the fact that no two are ever exactly alike, then you have a recipe for disaster if you expect any kind of scientific results.
People are not like photons or electrons or atoms. They do not possess an intrinsic similarity that is not also a product of their genetics, location, economic status, and all the other things that go into making each of our lives different. But statistical studies are usually modeled to show a preponderance of similarity among the population under study in order to prove the existence of a similarity, or to disprove the existence of a similarity.
Here is the rub - even though a similarity can often be observed, no single causality can ever be attributed to it decisively because the seeing of the similarity is the function of the observer, the statistician, not the subjects polled or studied! This is precisely why so many statisticians can make a living giving their clients what they want to see!
If you want to see what percentage of the population of the Earth is poor, then you must first define poor, then you can see poor people and tell them apart from everyone else. The statistic will be valid, but only for that definition. It's the observer, the statistician, who decides what definition of poor to use.
If you are a physicist looking for an electron, you must first define what that is so you know it when you see it. This is a normal consequence of science. But unlike physics, the statistician is not limited to definitions that are repeatably verifiable when studying the general populus. He is limited only by his imagination and what he can glean from his "data". Physicists must produce data and definitions that can be verified independently by other physicists. Statisticians only publish to their readers, and do not face generally, any sort of repeatable scrutiny.
Statistics in physics is revealing. Statistics when the subjects are people are manipulative. Statistics involving people are always accompanied by an agenda. Statistics involving pet theories in some sciences can also have an agenda, but thankfully, real scientists in the field have an agenda of their own, truth. If a theory can withstand multiple studies, then it's probably a good model. Studies of people very rarely get that far.
I read a book called "Fruitless Fall" about the colony collapse syndrome affecting bees. The author waded through a great deal of data and studies, only to find that there was truth in all things, but not the whole picture in all things. Each theory had it's merits, but not one could explain the entire phenomenon. But together, they added up to a chain of causality that MIGHT be correct. (It's a great book by the way.)
I think every poll conducted should be conducted by different study groups on the same sample at the same time. THEN, massage the data each independently and publish simultaneously. The differences in outcome would astound even the most adamant statisticians!

I think we may have read the same book, just wish I could recall the title, (Lies, Damn lies, and Statistics maybe?) about statistics. All statistical analysis are influenced by the question they seek to answer because the question also has to contain the definitions. The questions in polling contain the answers the questioners seek. And, someone had to have an agenda to think the question(s) needed answering in the first place!

The general public is not educated well enough in the proper abuse of statistics to have the appropriate amount of skepticism with articles like this one.

I disagree with you about testing the same sample more than once, though. Often tests are repeated with the same group but with different researchers to determine repeatability or to eliminate other factors. There are other problems with that approach with subjects that can learn though. What is wrong with most statistical studies involving humans, or other "cute" creatures, is the sample size is too small to develop a general "true" statement. So how do we know they didn't just find a group that had hearing loss that was also, coincidently, exposed to 2nd hand smoke? Test more subjects and develop a better base line. How do you do that? That's the rub. If you need money from some outside source, then the agendas have to catered to.

I also wonder at the use of the noun "Statisticians" so negatively. Statistics is just a math tool used by lots of different disciplines to analyze data. The only people i know who think of themselves as statisticians are into baseball or fantasy football, etc.

Really good scientific statisticians are fine. What irks me is how the math is used to justify a cause, rather than to show cause, in political and government "statistics". I know a couple of them and they are really nice people with a very jaded approach to political "truth", because they don't get the work if they don't prove the "facts" in the way they are hired to do. Just sayin'. Not a smack on the art and science of statistics. I like statistical analysis. It can be very interesting and even enlightening, provided you don't look for too much correlation and stick to the interpretation of the raw data.

I like Keno, and Poker, and stats "in-your-head-and-on-the-fly" is part of the challenge and the fun for me.

Could the book have been "How to Lie with Statistics" by Darrell Huff, copyright 1954? I am currently readig it on my NOOK.

A google search to refresh my memory turned up several titles but I think it was:

Damned Lies and Statistics: Untangling Numbers from the Media, Politicians, and Activists by Joel Best ©2001

I found it to be pretty entertaining for non-fiction. My wife has it and uses it in her 8th grade language arts classes and it's a lot more current than 1954. (A tool to help teach critical thinking!)

I now have it on my Kindle

I find myself thinking of a phrase often attributed to Mark Twain "There are three kinds of lies: lies, dam*ned lies, and statistics!"

Tom D.

thats 66.66666 % true!

And the other one I love; "43.7%* of all statistics are made up on the spot".

_{* insert your favorite number here if you disagree}

Tom D.

Some people would object to that number 23.8% of the time.
I know a Math prof who is also a statistician for the Ca state assembly. He said to me, "My job is to make my client right. Statistics, done properly, can do that for anyone!"

Define "true"!

As an introduction to correlation, my statistics professor offered up this...

There is a (I don't remember the actual number, but it was WAY high) percent correlation between bananas imported into the US and the number of women pregnant in the US. Obviously importing bananas doesn't cause more pregnancies.(Or does it???)

There is a huge difference between a correlation and causation. I've watched for this ever since I learned the difference. Unfortunately, it's easy to spot all too many "studies" that fall into this trap. They use words like "linked" and "related".. and in far too many "cause". Once a correlation is found, then some theory is proposed that to explain it. It is sad a researcher falls into this, but the thing that makes it worse is that there is no peer criticism to put it in check.

"Obviously importing bananas doesn't cause more pregnancies.(Or does it???)"

Just depends on how she eats it.

Sound Effect - Rimshot.wav

^{(Giggles again and hides in corner)}

You giggle?!

Ω

It goes well beyond the media.

Since you guys started talking about marijuana. One of the most egregious cases of bogus or non science, that was specifically implemented to target a particular group of people, was the controlled substances act.

Nixon's absolute hatred of hippies, combined with their association with marijuana, placed marijuana as a schedule I narcotic, carrying maximum prison time.

Marijuana, according to the schedule, is worse than cocaine. It remains a schedule I drug to this day. Long after Nixon fell in disgrace.

Back in college during the mid 90's one of my dorm buddies had a statistics class with a professor that made his class work rather fun at times.

One of his assignments to my buddy one week was to record all of the supposed deaths by what ever means that are purported on TV commercials over a week end. From that and assuming that no one dies of more than one thing, car crashes, assorted cancers, starvation, etc etc.. Calculate how long the average life span of every person in the world should be according to the numbers given by the media if that many people died of that many things as often as they say they do.

According to the class average numbers they came up with I was told they came up with a mean life span of no more than roughly 4 years per person for every one on the planet. The media lies, Go figure..

here's an example

http://www.newswise.com/articles/view/578948/?sc=swhn

the researchers make some claims, but come clean in the last paragraph...

Nearly 90% accurate to tell true from false reviews, that's pretty good. I hope they do go further with it, but sadly, as soon as it's published, the fake reviewers will change their tunes.

it would be situational

the tells would be different for say electronics

if I go to amazon & look at reviews, I feel like I can get a good idea by reading both ends of the spectrum. you can tell who the fanboys are, so I guess there are commonalities

but as you say some reporter will give it the full sound byte treatment sans qualifiers

Vat? Vat dija say?