My ire is pointed towards the people who are trying to sophisticate these statistics towards their own beliefs as they try to push it on others. go back 40 years there used to be all kinds of studies that "proved" homosixuality was susceptable and that smoking improved health.

its not necessarily true that numbers don't lie, they don't say any thing, they are just there. if there was a study in a town (and 1000) people were surveyed which pet they preferred and the results were 6 out of 10 like dogs (this is all hypothetical, an example to show how numbers can be wrong) does it mean that statistic applies to the whole world. of course not, not many people would say that it does. yet an overwealming number of people now are afraid to give their infants solid foods because they are convinced that it will cause obesity, even if that solid food was something like carrots or apples.

or... just or.... it could mean that the kids who are watching TV for more then an hour a day forget to study and has nothing to do with intelligence

But I'll bet you were born at a very early age. :tongue:

Statistics and polls can be used to a particular positions perceived advantage...no doubt about it. But to say that they are used to spread misinformation is a stretch. imo. They could be I suppose. Depends on who you want to listen to, that's all.

You would need a large enough random sample of people from various cultures. If you took a large enough random sample from people from all over the world and then found that 6/10 people liked dogs you could say that 6/10 people worldwide like dogs.

It's like looking at 1 class in high school to see who does their homework, let's say everyone does. Then saying that everyone in the highschool does their homework, which is obviously wrong. If you take a random sample of all of the highschool and then find that 40% do their homework, you can say that 40% of people in that high school do their homework.

To say that statistics don't mean anything is flawed at best, all statistics mean something (for example you now know 6/10 people in that village like dogs), it's just people not being able to interpret grounds for an accurate study that is the problem. Have you ever taken a psych 101 class? I know mine covered various statistical trickery and what to watch out for, as well as what makes for a good study.






Exactly, correlation =! causation. Is it that video games cause violence or that more violent people are drawn to videogames? (That's just an example founded on no evidence swaying either way)

the point of this post is that some of these studies are arrogant to say. THIS HAPPENS BECAUSE OF THIS, never mind that our study only ranged 10 years and no more than a couple thousand people (which isn't a big number when you are applying something to the rest of humanity). but I main Ire was not towards the studies themself (which allow people to interpret them as fact, being what i have issue with on that side) but the people who are causing a series threat of pandemic by refusing to give their children polio shots because they think their more informed than their doctor because they saw a study (which turned out to be a fraud).

don't know how to make links so I'll just put the url for people who want to see some stories that cite what my OP was about.

http://www.huffingtonpost.com/glenn-d-braunstein-md/andrew-wakefields-vaccina_b_816208.html

http://www.businessweek.com/lifestyle/content/healthday/649632.html

I, uh, totally didn't think you were sixty five :shifty:.

See, without studies, any argument breaks down to people using words with little or no context. How do you know the number is overwhelming?

EDIT: Unless phlooph's age has been the subject of extensive scientific study, you two need to take that conversation to PM please.

Misleading stats. are where someone says something like "100% of the people who participated in our studies showed improvement" when 100% could mean 100% of 1 person. The numbers don't lie, but the truth is sort of half.

Obviously not. I've yet to hear of or see a method of harvesting data from a group of individuals that is anywhere near scientific.
And it is the act of calling a study "scientific" that irks me.

If opinion polls or social studies were scientific, there would be no "margin of error". The results would be concrete, and repeatable. And, there would be no conflicting results between multiple studies conducted on the same subject.

And I stand by my opinion that most government funded studies are cooked from the very beginning, and the results are pre-determined.
It's just a matter of wording the question the right way, and cyphering the numbers the right way, and it will stand up in a court of law.

I am not saying that they are all wrong, but their not all right either. The reason I think (and yes, this is an opinion not a study, but that doesn't disqulify the validity of what I am saying just because I havent conducted a study of my own.) that there are overwealming numbers of people who are affraid because there was a huge scare over the matter when that study came out. it was all over the news (and yes, I am aware that the news is part of the misinterpreting problem, but that IS the point. thousands of people saw this on the news, and people panic when the news tells them too.) but aside from the publicity and media perception of a scare my aunt is a nurse and they have been having problems with younger (natural and hollistic types) refusing to consent to their children getting polio shots and other vaccinations. I have an ex who just had a baby and she is like this, she won't get the baby a vaccination. (wont even take her to a doctor, at all.) because she has seen studies that say; vaccines cause autism, tv degrades your intellegence, giving babies solid food will cause them to be obese.

the context of this topic is that people are being to ready to believe in what these studies imply and go over board, with out considering whethor or not they fill the demographic in which the study took place, or that the study is implying a link that later may be discredited.

Small sample size isn't necessarily a fatal flaw, as long as conclusions are adjusted correctly. A chapter of a book I have covers a study where they ended up with only ten subjects, due to strict eligibility criteria. That doesn't invalidate the conclusions, which were presented as having limited reliability, but being suggestive of ___, and meriting another run with more people. The conclusions were valid, since it wasn't claiming to have proven anything (other than the merit of repeating it with a larger sampling), and they were tentative.

this is exactly what I mean. most studies are not saying that their results are incontravertable, but there needs to be some kind of common sense put into peoples minds when they read these results as they should be conscous that the people in what ever study may not apply to them, most are targeted at specifc demographics but the general public and the media treat them as if the study blanketed every one.. Like the study about the babies and obesety, the results of that study may be linked to region (or what kind of solid foods being given like i originally said) rather than just solid food in general.

So, your venting is due more to media than science? Id est, people taking things out of context and blowing them out of proportion.

Ssenkrad, please don't take this the wrong way, but I'm not sure you really understand the purpose of polls.

You see, to eliminate margin of error, you'd have to actually ask every single person. This is logistically impossible, so we take a large enough sample to get an idea of what people think, and then say, "Well, here's what we got, and based on the number of people we surveyed, we think that this is the range that the actual number lies in." That's what a margin of error is. It isn't there because it's unscientific; it's there because of the logistical needs to not survey every single person.

To make this clearer, in engineering, to calculate the strength of a given material (like Steel or Concrete), samples of the batch of the material is taken out and tested for strength (and a variety of other properties). This gives us a representative strength of the entire batch of material with a margin of error. The results aren't repeatable (especially in the case of concrete, that stuff is like Russian Roulette as far as strength is concerned), but it's enough that we can safely use it for building. For these materials, there's nothing unscientific about how everything is done. The stress is applied via machines that are controlled via computers, yet a margin of error still exists because we can't test an entire batch of the material in question (because the material is tested in such a way that it is either destroyed, or permanently damages).

If you believe that Government funded studies are 'cooked', I suspect that there's little that can be done to change your mind. That being said, I hope you understand that the vast majority of research done in a University setting in the United States is in some way funded through government grants. This includes (I apologize for reusing an Engineer example, that is my background) some of the leading Structural Modeling software and the research that the US highway system is based off of. To be honest with you, I am far more likely to trust a government funded study than a study conducted by a research group that depends on Corporate funding. Mainly because Corporate funds are often tied to a result. If you don't get the right ones, your funding is gone. Government funding is less susceptible to that in my experience.

Or even better: Rich kids can afford to do cool stuff like paragliding and snowboarding.
Poor kids can't.
Poor kids spend more time watching TV.
Poor kids can't afford private tutors either.

This teaches us correlation does not always equal causation.

Sometimes, not all that much. Google for Dr. John Ioannidis -- this guy has made a career of going around and informing the public when people do things like screen a disease against thousands of genes and, by sheer chance, hit a type 1 error in the process that they think is significant. No malicious intent, it's just honestly not knowing any better.

http://www.businessinsider.com/736-of-all-statistics-are-made-up-2010-2

Has he ever had an article features in Nature or another reputable science journal?

When that happens i always get the mental image of a person bursting trough the doors of a scientific convention and saying loudly: "You're doing it wrong. Trust me, i'm from the internet!" :hehe:

I find it odd that people just seem averse to the whole field of medicine. Like my sister doesn't like medicine. She says she feels worse when she takes it because she doesn't trust it. Why? This isn't some mystery story where the corporations are all maligned against the public spoon-feeding us poison and raking in the cash.

It seems people will believe horse[censored] if it has numbers. The worst part is, we use it as an agent to further the cycle. But stats are always something you should check. Hell, if you feel like something pertains to you, you should effing double-check it. The whole Autism is linked to Immunization was, IMO a HUGE scandal. But there are still people who wholeheartedly believe it. I guess I'm saying, numbers fool us, but it is our own friggin dogmatic belief in the factual presentation of anything that will continue to drag us down.

It haunts us in the form of babies who will never get immunized due to fear of autism...

Since the whole Autism to Vaccines thing has been thoroughly debunked with factual presentation, wouldn't it be more true to say that dogmatic belief is in fact what brings us down and not the factual presentation of information?

And then there are those of us who swing the opposite way and begged doctors to give their children Small Pox vaccine, but were denied because we "eradicated" the disease years ago. Well, we didn't, and I wouldn't be surprised if we had some kind of sweeping plague of Small Pox sometime in the future, and I am the last generation to have the vaccine. Feel sorry for those after me who might fall victim to it.

Ehh... lol YES. I'm kinda tired here. So for the record WHAT HE SAID ^^^^^

I don't know, has he? Why don't you check? I see two in Nature subsidiaries and one in Lancet by glancing at the first page of scholar.google.

We gave a dog lover and a cat lover a gun, and the cat lover shot someone! Our study concludes that cat lovers are violent, impulsive murderers!

N.B. our study subjects were chosen from a nursing home and mental institution, respectively.