EDIT: as the results of a few posts I would like to add a few things to clearify. 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.
Niether am I saying that studies are part of a conspiracy to control what people believe. only that the public takes them far too seriously and is having tangable consequences for other people who don't over react to these studies.


So I have been reading alot of results for studies, and they are getting stupider and stupider. they seem to be causing more harm rather than informing the public. example, that fraudelent study were results were fabricated to show that immunation shots cause autism. turns out that was faked, for some kind of monetary gain (don't understand how that worked but thats not the point).

its long, sorry, but reading it all the way through will make the context clear.

some are not fraudelent as much as they are mis leading. heres one, paraphrasing; there was a study while back that said that kids who watch more than an hour of television a day score lower on tests, so the thesis statement of that study was that watching more than an hour of television causes kids to get dumber. thats blatantly misleading. what that study actually proves that kids who watch tv for more than an hour a day are not as intellegent as those who do. not because tv makes you dumber, but because it requires less intellegence to enjoy. the reverse would be saying sudoku makes you smarter, which is misleading because its likely that the people who play sudoku are smarter than those who don,t.

what really got me fired up and to start this thread was hearing the results of a study that said infants that were feed solid food were more likely to have issues with obesity. so what every one takes from that is that giving solid food to babies will some how cause problems with their digestive system which will make it easier to become obese. but that study ingores the fact that the solid food these infants in the test were being given were things from fast food. the problem wasn't the solid food, it was the parents who werent giving the babies healthy solid foods. you can give an infant regular baby food or stuff like carrots and it isn't going to cause them to have problems with obesity.

I am not saying that alll these studies are wrong, they are just lacking enough control subjects. they are taking not of affects and pointing out causes that aren't directly linked. so the for the babies solid food study doesn't mean that ALL solid food will cause obesity, but the fact that the solid food that was being GIVEN to these babies was fatty and high in calories.

so I know some people are going to disagree with me but I really feeled annoyed by the above mentioned and decided to bring it to the forums.

A great man once said - 'There's three types of lies. Lies, damn lies and statistics.

You can bend the numbers and mess with the wording any way you want to make the study/ies work in your favor for whatever reason. Without all the information, and I mean all, I'm reluctant to believe many studies.

Not disagreeing with you. Have you read the book "How to Lie with Statistics" by Darrel Huff released in 1954? There's a short wiki article on the book too.

I've studied information visualization and statistics a little bit, and I've seen how easy it is to present factual looking statistics, in a news article for example, but when you look closer at the graphics, you realize they've cut away a part of the vertical axis or something like that to make proportions all wrong just to emphasize some point.

I definitely agree with basically everything you said.
Like how people think video games make people violent/stupid, it doesn't make you dumb, but playing alot of obscenely violent video games (GTA) does make KIDS sort of weird (just go to my school for a day, you'll know what I'm talking about)

You mean modern scientific studies aren't perfect! You cannot seriously be suggesting that people shouldn't take these studies as complete and udder etched in stone fact! Blasphemy!! .... Now where did I leave that damn sarcasm emoticon.....

Sophestry is a large part of how one argues with the evidence these studies supply thats true. neither am I saying that they are all misinforming. the worst way these seem to be used is by the proponents of holistics and pure natural treatments. specificly how alot of people are refusing to give their kids polio shots because of that fraudlent study that said shots cause autism, also alot of people think polio was cured. infact its been returning in alot of countries because people think they don't have to get those shots any more.



Yeah, the voilent video game aspect too. my mother literally thought that voilent video games caused (not maybe, or could, but did and will) school shootings. and I allways argue angainst those kind of people by pointing out that THEY are the ones who are disconected from reality if they really think a teenager cannot tell the difference from fictional violence and violence in reality.

There's a pretty big difference between a study where someone outright lies (as in, say, the vaccine-autism study) versus a study where something is screened against a huge number of samples, and a type 1 error happens by virtue of statistical chance (like, say, certain drug/chemical studies). One is actively malicious, and the other is someone honestly not knowing any better.

NASA got in trouble with this when they modified their temperature data "by accident" :rolleyes: magically all the years prior to 1970 the temperature was "adjusted down" and after 1970 they were "adjusted" upwards to make it look like there was a huge temperature spread.

Well, guys, looks like all studies everywhere must be lies because of this study right here that looks at a few well publicized studies. So will everyone please for a line behind me as we journey back to the dark ages and once again hit each other with sharp and pointy things. Those of you that have money will be allowed swords and armor. Those that don't will get a kinda pointy stick.



Statistics don't lie. They can be presented in a misleading fashion, the same as anything else. You can't blame the numbers for human error on the part of the reader.

Well, statistics aren't strictly the numbers, but also how they're gathered, processed and displayed for consumer use. Basically, what that quote means is you can use the numbers to prove any point you want regardless of other factors. You can leave parts out, or add parts in to make the numbers appear a specific way.

Not all studies are lies but it depends on who's doing the study and what their agenda is. If somebody doesn't like Smoking they may rig a graph or statistics to show that 2nd hand smoke is worser then Barney the Purple Dinosaur (Now that's a tough choice between the two, one could kill you and the other will make you go insane).

If I see a study or Statistics that's like whoa that can't be true, like for example 80 % of the population in Pittsburg loves the Cleveland Browns. If I see a study like that I'm inclined to research it further to see what the truth really is.

Yes, but in all those cases it leaves a finger print, one that's usually pretty easy to identify. Honestly, I just hate that quote because I've seen it used far too many times as rebuttal for any statistic, rather than actually using some brain cells and looking at the methodology behind a study. Numbers don't lie. The methodology of the numbers can make them say something different than what someone is presenting, but still, numbers don't lie unless they're being totally made up, in which case they're simply a lie.

That's the risk with statistics. You either have all the information available, or you don't. Sure, there are tons of valid statistics that are useful to tons of people. There's also a good chunk that are misleading for whatever reason, be it faulty numbers or a mathematician twisting what you see for their benefit. I agree that each person needs to do the proper research to make an informed decision.

In a recent study I performed, 73% of all studies misrepresent the facts, and I have it on good authority that some are never even conducted.

there's no such thing as faulty numbers, just faulty sampling. If the person performing the study didn't make sure to make their sample random and representative, that is, as Ratwar said, a methodological problem.

The best example of an honest methodological fault was "Dewey Defeats Truman", their sampling was supposed to be random (it wasn't really) and it definitely wasn't representative.

Then of course, there is dishonest methodological fault, where the methodology used to get their "random" and "representative" statistics is purposefully at fault. This isn't something an ethical statistician would do, though, and fairly easy to detect, as Ratwar said.

If you think statistics are innately risky or misleading, you are mistaken. Sample statistics are a valid method of research, and really the only practical one most of the time. You, of course, shouldn't blindly believe all statistics, because there can be both honest and dishonest faults in their methodology, but that is why any good survey will include, if not the entire sample data, at least a portion of it, for you to see if they carried through properly.

Couldn't have said it better myself if I tried.

Sad thing is Statistics are useful piece of information, but they can breed disinformation which is a very useful tool of deception, espionage, and war. You put a % in that study I guarantee you half the people reading that study will believe it. Unfortunately these are the people that are truly blind willing to follow anything and everything because it tells them to do so. If a published statistic came out in a medical journal saying that if you eat 1/2 teaspoon of dirt every day will help you live longer people would freaking do it.

Be careful whenever you hear the word "study". Like when you hear on commercials "Studies show..." The first thing that pops into most people's minds when they hear this word is guys in white coats in a lab doing a scientific examination of the subject matter, even when it doesn't make sense for the particular subject. In reality a study could be anybody, qualified or not, spending five minutes thinking about the subject.

You said udder instead of utter. *giggles*

Are you mad at the studies themselves or peoples misinterpretation of them? The biggest problem I've noticed in peoples interpretation is assuming correlation=causation. The media usually does this first, as many people don't bother to read the "raw" studies themselves. Then the 'telephone effect' happens and people develop very unrealistic beliefs.



I love arbitrarily creating statistics too.

I really have a problem with a few posts in this thread - the first post (you aren't citing any actual studies here), and squeekers1234's post (NASA? What study? What were the consequences of this?)

Plus the undertone of simply dismissing a study out of hand (like many people do on these forums in regards to videogames and aggression) because it doesn't "seem right" is likewise disturbing. Scientists put a lot of work into their research - and to dismiss it without either directly pointing out the error in their methodology (with the research text in plain view) is a mistake.

And don't even get me started on the topic title. It makes it sound like the whole scientific community is trying to dupe people into believing false information, where in many cases it is people simply not understanding the scope and limitations that the studies possess.

You do know just how much information / statistical evidence it takes to get a study published in a peer-reviewed medical journal, right? Ironically, there have been studies that have shown that not playing outside / in the dirt can have adverse effects on a child's immune system (I'll look up the studies later and post them here).

You aint from 'round here, are ya, kid?


Statistics are b.s.
They are meant to influence the opinion of gullible people, and they are a very effective means of achieving that goal.

One group can do a survey on a pool of people, and come up with some figures that say 85% of the population likes vanilla ice cream more than any other flavor.
A second group can then survey the same pool of people with a different set of questions, and come up with some figures that say 65% of the population hates vanilla ice cream.

There is no science involved, it's mathematical smoke and mirrors.
And a six or eight digit government grant to conduct these bogus studies, that is the cherry on top.

Yeah, what a generalization of a topic title. Not to mention the post about statistics being 'lies'. DEFRON's post is spot on.



That 'kid' is 50 years older than you! I mean she's your age.

Anyway, what kind of BS is your post? What on earth are you talking about?

I've studied statistics and claiming there is no science involved is stupid, because it's entirely wrong.

So, your position is that there is no way to gather information from the public on their opinions?

The problem with vaccines is that there was a time when some of them had issues... was being the key word. Now that we've had more experience with them and the technology 'stable' rather than in alpha or beta stage (to borrow terms from the software world), such problems are rare. Unfortunately, lots of people haven't caught up with that second sentence.


That ties into my next point: its quite reasonable for a study aim to prove or disprove a link between two things, and either not try to put forward any explanation, or include a http://wordnetweb.princeton.edu/perl/webwn?s=hypothesis which would require another study. Sadly, media (and politicians, and others) often grab hold of these things and misuse them -- a study which proves that people who commit violent crimes are more likely to play violent video games suddenly becomes a study which proves violent video games cause violent crime, even though the people who did the work had no intention for it to be used that way. [b]That[/i] is not a problem with statistics, it is a problem caused by people grabbing at straws to support their opinions.

Peer-review keeps the studies themselves fairly honest, and looking through the methodology etc. used will generally give even a lay person a decent idea of it's validity.

Hey! I haven't even turned 45 yet! :slap: