Don’t Always Believe the Science

LiviuX says on July 14th, 2008 at 9:10 am

What do you want? To belive religion?

Blog with news : http://liviux.byethost15.com/

James Massey says on July 14th, 2008 at 9:35 am

Science and engineering are very different things. Inventing lightbulbs, planes and wheels has very very little to do with science. Nothing is proposed/tested. People were inventing things long long before the scientific method was itself invented.

But here here on these nonsense measures like BMI. It is enfuriating to see TV doctors measure people’s BMI, totally ignoring that it is meant for group and population comparison, nothing more.

Anthony Judd says on July 14th, 2008 at 10:01 am

So in a country of 21,000,000 people they tested 0.06 percent of the population which means that they didn’t test 99.94 percent of us! I have two questions:

1. How do they know that the 0.06 percent is representative of the 99.94 that they didn’t test?

I hope that there was a more subtle question there about sampling methodology or selection bias which I missed, because otherwise I am appalled at this question from a science graduate.

You have a BSc right? So you know how sampling works don’t you?
Say for example when a newspaper publishes an opinion poll – you realise that they don’t ask every voter who they would vote for don’t you? The results are usually still indicative of the final election result because a random sample of sufficient size should show the same trends as the larger population, so we can infer facts about the larger population from our sample.

A sample size of 14000 individuals (if selected randomly) should give a beautiful picture of trends in the greater population.

Leo says on July 14th, 2008 at 10:04 am

“1. How do they know that the 0.06 percent is representative of the 99.94 that they didn’t test?”?

Significance levels depend on the absolute size of your sample, not to its relation with the size of the population. If you want to estimate the weight of the population of Luxemburg or China the size of sample is the same.
I know that this is not intuitive. But this is basic statistics.

Leo says on July 14th, 2008 at 10:06 am

Ops, Anthony has noticed the same mistake 3 minutes before me….

Julio says on July 14th, 2008 at 10:14 am

Don’t always believe in History!

It is highly disputable that either Edison or the Write Brothers are the actual inventors of the light-bulb and airplane, respectively.

Thomas Edison is said to be the inventor of the first “practical light bulb”. Yeah, right… but what does it mean? Does it mean that someone elso invented a light-bulb first but it was not considered “practical”? Is a 14-hour only Light-bulb, like Edison’s, really anything that practical? Whoe did invent the actuial Thungsten filament bulb we still use today?

The Wright Brothers story is all but right. They only came out with pictures claiming their “invention” after Santos Dumont flight in 1906. While the last flew in front of thousands of people, the firsts were never able to show what they did. “It crashed”, they said, “we can’t rebuild it”.

Prasanna says on July 14th, 2008 at 10:52 am

I agree.
I get reminded about this conversation between two charecters(docs) in comedy series ‘Scrubs’:
——
Man1:
Ah, come on, statistics show
that kids whose parents stay together…
“Statistics show”

Man2:
who… who cares,
what statistic’s showing?
Look at medicine. 80% of people,with pancreatic cancer, die within
5 years, 95% of appendectomies,occur with zero complications. But…
we both know pancreatic cancer patient that lived,
and appendix patients
that unfortunately… passed.

Statistics mean nothing to the individual.

You’re either gonna be a good
parent to that kid, or you’re not.
—-
I think you r saying the same idea conveyed in this conversation.

Jari says on July 14th, 2008 at 11:07 am

The BMI (item no 2) wasn’t designed for the purpose you mention. It was originally used to measure oxygen consumption and, as far as I know, never intended as a scale from underweight to obese.

dmat says on July 14th, 2008 at 11:11 am

haha aussies are fat

James says on July 14th, 2008 at 11:41 am

The main conclusion I would draw is that the examples given are not really science. They are ideas that are being ’supported’ by misleading terminology or graphs and other non-rigorous arguments. Just because it sounds plausible doesn’t make it real – and that’s what the crooks feed off. It’s the same as bad statistics being used to support political arguments either way or the selective quoting of religious texts out of context to support an argument. Science is not to blame, but rather the public’s reluctance to think critically about the information delivered to them, and those who maliciously (or ignorantly) take advantage of this.

Less harshly, there are some who purport an opinion and show the evidence to support it, but unless they state clearly that it is an opinion and at least acknowledge any evidence against their point of view, then they too are guilty of being misleading.

Art Gonzalez says on July 14th, 2008 at 12:11 pm

Great info. I would emphasize though the importance of proper hydration for health iprovement. A technique I was taught about a year ago is to drink six glasses of water (it’s difficult when you are starting!) first thing in the morning. Then wait about 45 minutes or an hour (while you are getting ready) to eat or drink anything else. Then continue drinking clean pure water the rest of the day as normal. This technique stimulates all the internal organs and brings them rapidly to prime condition. You will feel the difference in a few days. Also another tip is to take ice cold showers. It will feel you with super energy and also make you sleep better at night.

Art Gonzalez
Check my Squidoo Lens at: Quantum Knights

djc0 says on July 14th, 2008 at 12:28 pm

Lifehacker, you should make it clear that is an article about *medical* science, and a somewhat poorly written one at that.

In addition to the criticism about survey ignorance (this guys a university lecturer?!?!?), the author misses the point that most health metrics, such as BMI and calorie intake, are simply general guides that the population as a whole can use (and often good for many, e.g. those who are not rugby players). Your doctor always holds the final word as to your state-of-health – doesn’t everyone know that?

I can tell this was written by a “motivational speaker”. Step 1: everyone is lying to you. Step 2: trust me! (since i’ve been honest with you so far , e.g. revealing step 1).

Tim says on July 14th, 2008 at 12:37 pm

Anthony beat me to it. Somebody’s never heard of statistics. Perhaps you actually have quite the aversion to it as well, because statistics is at the core of all these things you call junk science. All those charts you’re talking about are based on correlation, not causation. Having a certain height/weight ratio is statistically correlated with good health. It does not indicate that if you lose weight to get into the target range you will become more healthy, nor that if you gain enough weight to go outside the ratio you will become less healthy. There could be some completely independent third factor involved (e.g. an eating disorder or weight-lifting & muscle-building). But as your article points out, one does have to be aware of these measures being pushed people as the solution to health.

Jason says on July 14th, 2008 at 1:03 pm

As a scientist (physiologist), I disagree with your logic in this post.

You consistently lay out your article as though science is flawed and you give reasons why. While I won’t argue that there are flaws in the scientific method, none of the issues on your list are flawed science. They are all mischaracterizations of the science or misinterpretations of what the science means. With so many misunderstandings among of the general public of how science works and what it means, it only makes things worse when you play fast and loose with the word “science” for the sake of a catchy post.

PS – The words “belief” and “science” in your title do not belong anywhere near each other, which relates to the same problem mentioned above.

Steve says on July 14th, 2008 at 1:18 pm

While I would agree with your assessment of using these guidelines as “correct”, I have to agree with Jason about the misuse of the word science here. We’ve witnessed the decline of critical thinking and the rise of pseudoscience and outright quackery with CAM and Homeopathy. Please be more responsible before calling some of these “general guidelines” science. Science is evidence based – always look at the evidence. We really don’t want any more reasons for people to distrust evidence based medicine or sound nutritional guidelines.

Mickey says on July 14th, 2008 at 1:52 pm

“Science is an incredibly valuable and necessary part of our existence, survival and development here on the big blue ball and I’m passionate about it. I’m also passionate about not being mislead or misinformed. We can learn and benefit so much from so many clever people in the world of science but like anything that involves humans, it’s flawed.”

Make the bit about not being mislead or misinformed the main point of this post; it is not nearly emphasized enough, which is why you’ve gotten the flak above that I was planning to give.

Esgo says on July 14th, 2008 at 2:16 pm

This article is trying to compare measurements for a regular person (that rarely exercises) with an extreme one, in this case a rugby player. Do the measurements work for regular people? Yes. Do these measurements work for athletes who regularly work out and stay fit for their sport? No. Because everyone’s situation is different, these measurements are only guidelines. If you are confident in the fact that Australia is indeed made up of mostly rugby-type players that have an average body fat lower than the rest of the world, but by these measurements are considered fat, why do you care what other people think?

What this article should be really trying to convey is that everything should be tailored to your current lifestyle and situation, not that science is flawed. This was a dismally written article – even I can come to the obvious conclusion that athletes are healthy.

Andrew says on July 14th, 2008 at 2:59 pm

Good blog!

Win says on July 14th, 2008 at 3:51 pm

Well the heading is clear enough: Don’t ALWAYS believe in Science. I think one is dump if he/she DOES always believe in Science or anything. The best thing is to question everything and anything. As reliable as Science is, there are always 2 sides of 1 problem and Science can’t always (if not a lot of times) prove itself or disprove the other side of the argument. On that alone, Science is somewhat opinionated.
Not to mention, this is a BLOG! It’s ideas and opinions people. You are reading it because it’s personal! Agree or don’t agree and take away what you want.
At the end of the day, really, you only take to heart what you want, not what Scientists say! Look at yourself and you’ll find examples of what I meant.

Kyle Maxwell says on July 14th, 2008 at 6:09 pm

I came here to post on the issue of sampling, but Anthony covered it quite thoroughly. I have a degree in Statistics and was shocked to see this fundamental lack of understanding of how science and mathematics work. It throws the entire rest of the article into doubt. Since you’re just looking at the “man on the street” version of the study, you clearly haven’t looked at the confidence intervals and error margins that would be listed in the actual scientific study.

Most of the rest of your “counter-examples” are really just pointing out that various health guidelines can’t be taken as hard-and-fast rules. This doesn’t invalidate the science behind those guidelines. Rather, those guidelines of course have statistical outliers and must be evaluated intelligently and critically.

I think you fundamentally misunderstand how science works, or have listened to “experts” who also misunderstand. One single indicator is just one piece of evidence. To correctly evaluate the health of an individual (or any other complex system), multiple indicators must point in the same direction and within the appropriate confidence intervals.

But hey, if you were to dig deeply into all this and actually critically evaluate the tools in question, then you wouldn’t be able to write a blog post based on confirmation bias with lots of ad hominem attacks.

brendan says on July 14th, 2008 at 8:25 pm

I have to agree with the commenters above: this article is very scary in it’s obvious lack of understanding of basic statistics, and in it’s misrepresentation of the difference between scientific research and science reporting, and in it’s misunderstanding of the value of indicators like height/weight ratios, BMI etc.

I wont talk about the statistics since that has been covered well already: suffice to say there is a well established way of saying exactly how well a sample represents the population and it is shocking that someone who is presenting themselves as a sports scientist doesn’t know that.

As for confusing scientific research with science reporting, that is also quite unforgivable for someone claiming to be a scientist. Many of the examples given in this article are not about research, they are popular media examples: and popular media is about entertainment, not facts or science.

Finally, the author appears to be deliberately misrepresenting the use of indicators like BMI. These are designed as diagnosis tools that are easier, quicker and cheaper to calculate than doing a full medical investigation into the state of a person’s health. They idea is that anyone can calculate their BMI (just to use an example) and figure out approximately if they are overweight or not – and perhaps do something to lose or add weight if required. This is only meant to be an easy way to get an approximation of a person’s state of health: if they want a more precise assessment they need to get a thorough checkup by a doctor, sports instructor, physio etc.

I hope there will be better quality writing on this in future.

le says on July 15th, 2008 at 6:23 am

circa 1050 people is (statistically) enough to measure statisticall stuff in every (or almost every – china/india etc may be to big :) ) country. For 20 milions people in australia, about 1000 polled people is enough.

Ruth says on July 15th, 2008 at 7:25 pm

I’ve never been a proponent of the “one size fits all” approach to weight and health, and that, to a certain extent, is what all these measurements are. Sure, it’s nice to have guidelines but I’d much prefer a more personal approach to my care, and an integrative approach to eating and weight loss such as that put forward by Pamela McDonald’s APO E Gene diet. For me, having my treatment based on my genetic blueprint, instead of on a sample of what other people think is pretty much a no-brainer.

Marcin says on July 18th, 2008 at 8:09 pm

One sentence, two mistakes.

Following different sources (wiki included) Thomas didn’t invent light bulb but “only” patented it on Jan 27, 1880 (patent filled on Nov 4,1879). Sir Joseph Wilson Swan did it before. He received a British patent for his light bulb in 1878.

The same about the Write Brothers. What about Richard William Pearse and his true flight on 31 March 1903 or his hop (91m) on March 31, 1902 comparing with Write’s jumps on Dec 17, 1903( 36,6 m – first one and 270 metres – the best one)?

There in nothing more to say about sampling methodology… (all explanation in comments).

Samuel says on July 20th, 2008 at 10:04 am

@James Massey
“Nothing is proposed/tested.”
Oh… For sure… So your car was never tested.

@Win
Yes, this is a blog. My complaint (and many of those previous commenters, I guess) is not about freedom of speech, but of it´s responsibility.

Everyone has an oppinion, but posting, on a idea-agreggator like this website, such an ill-formed one is not appropriate.
It is ill-formed one because it seems like the first thing that came to the author’s mind. I doubt he ever searched for anything that proved him right or wrong.
A simple, shallow research would have made him rewrite his entire article.

The main idea is also bad. One should NEVER belive in science. That´s not it´s purpose. If you disagree, go ahead and test it. There is no absolute truth in science (_not_ever_). If you disagree, please do your statistics homework once more.

It really scares me that he’s a “qualified exercise scientist” and university lecturer. We all hope that he does much better there than he did here.

@Craig
I’m sorry if we (commenters) are being too harsh. If you´re good at writing, then please take in account that we´re educated and deserve a greater share of your skills and effort.

Michael Gorsline says on July 28th, 2008 at 2:05 am

I highly recommend Defending Science Within Reason: Between Scientism and Cynicism. This is a tough topic, and it demands careful thinking. Understanding exactly what science and what can be expected from it is essential. I’ve never seen clearer thinking about this than by Susan Haack. I was pretty well read in science prior to reading her, and I’ve never learned as much as I did about the whole enterprise and what distinguishes science from other kinds of inquiry like journalism or history. I think you’ll appreciate her overarching metaphor, which helps to account for some of your examples, why they went awry and how that fits into the whole picture.

Michael Gorsline says on July 28th, 2008 at 2:10 am

@James Massey, you are right that there are differences, but you are dead wrong about them not testing anything. That, as a coincidence, is Haack’s central point is that science is continuous with other kinds of inquiry. Inquiry being any investigation into the nature of how things really are in the world. Science tends is more thorough than other kinds of inquiry, but it is a simply not true that no other kinds of inquiry such as engineering don’t use testing (i.e. falsification).

Jeff says on July 28th, 2008 at 11:09 am

The only solution to bad science is good science. When science fails to find truth, it was likely still the best option available at the time.

Darren says on July 30th, 2008 at 11:52 am

You’re combating ignorance with more ignorance. First, understand that science is a process. We come up with an idea or model, we test it, and the more the evidence backs it up, the more we trust that idea or model. Science doesn’t ever say “this is a fact”, it says “based on the evidence, this seems to be the way x works” — if something becomes a “theory”, it means that *lots* of evidence supports it, and attempts to prove it wrong (falsify) all fail.

That doesn’t mean someone won’t find a flaw in it some day.

Second, and more importantly, the examples you cite are not problems with the scientific theories, they are problems with how people (trainers, doctors, insurance companies, etc.) *interpret* the science. For example, the science says “if you look at height vs. weight for a population (BMI), you get a pretty good idea of the overall health of the population.” So far, that’s held up very well. Unfortunately, a lot of people use it as an absolute measure of individual health, and that doesn’t work so well.

I think that better than “don’t trust the science” is “don’t blindly accept advice” — one should always check up on advice (even from professionals) before following it. Even something that applies perfectly well to most people might not be a good idea for you.