Everyday Systems

Good article, well written and very true.
http://bradpilon.com/2009/08/what-is-sc ... earch.html

I enjoyed that; thanks for sharing. The next time I read about medical/diet/exercise research findings, I will be able to put it into better perspective.

Interesting article

I love the quote from the English teacher. I'll have to use that one. I always encourage my students to look at the specifics of a study to determine who conducted it (which is a search for possible bias), the conditions of the study, and how many participants were in the study. Simply throwing around statistics is not enough. As Mark Twain once said, "There are three types of lies: lies, damn lies, and statistics."

This article also made me think of my younger years when I would often let common sense guide my attempts to lose weight or other goals. In those days, we didn't have a computer or access to the Internet, and my home town library was abyssal as far as research material. So, I used my common sense. And you know what? I was generally more successful than I am now. It seems like now I will think of a plan, and then I'll do some research on it, and I'll read something that discourages me. It's almost as if I would have been better off not knowing! (Nah--but it's better to take everything I read with a grain of salt).

Part of the problem is that researchers get tunnel vision and look at one particular thing -- ignoring everything else. But nothing exists in a vacuum. Another part is that those who interpret the research look at it from their own paradigm and what they're trying to "prove." I don't pay much attention to research anymore. I look at what has worked for people in the past (or currently) and determine how I can adapt to that. One thing's pretty certain in terms of diet -- neither low-carb or low-fat work for the long haul.

wosnes wrote:Part of the problem is that researchers get tunnel vision and look at one particular thing -- ignoring everything else.

I must disagree with you here; that's how scientific research is (or should be) done. To study one variable, it must be viewed under controlled conditions to ensure that any output changes observed are as a result of the input variable changed and not some other, uncontrolled, variation. Combinations and interactions of variables can only effectively be studied when every one of them is understood on its own. Whether this is valid in terms of nutritional research is another question altogether.

The problem I see is not with the research, which is generally reported soberly and with all intent to produce a meaningful result, but with the media reporting of it, which sensationalises everything to try to make it relevant or exciting and distorts the research conclusions in the process. Qualifications and allowances made during research are lost in the conversion to soundbite status, though the most important thing lost is generally the purpose of the research. Scientific attempts to better understand the functioning of a thing are often reported as though it was an attempt to cure everything in the world all at once. People then get annoyed when they try to apply results found in a vacuum to the real world, which, as well we know, doesn't adhere to lab conditions.

ThomsonsPier wrote:

wosnes wrote:Part of the problem is that researchers get tunnel vision and look at one particular thing -- ignoring everything else.

I must disagree with you here; that's how scientific research is (or should be) done. To study one variable, it must be viewed under controlled conditions to ensure that any output changes observed are as a result of the input variable changed and not some other, uncontrolled, variation. Combinations and interactions of variables can only effectively be studied when every one of them is understood on its own. Whether this is valid in terms of nutritional research is another question altogether.

The problem I see is not with the research, which is generally reported soberly and with all intent to produce a meaningful result, but with the media reporting of it, which sensationalises everything to try to make it relevant or exciting and distorts the research conclusions in the process. Qualifications and allowances made during research are lost in the conversion to soundbite status, though the most important thing lost is generally the purpose of the research. Scientific attempts to better understand the functioning of a thing are often reported as though it was an attempt to cure everything in the world all at once. People then get annoyed when they try to apply results found in a vacuum to the real world, which, as well we know, doesn't adhere to lab conditions.

One of the comments to that article mentioned Ancel Keys research and how it was "fundamentally flawed." The problem is that Keys studied the effects of saturated fat and cholesterol intake on cardiac health and found that the more one consumed, the more likely one was to have cardiovascular disease. The problem is that he ignored the rest of the diet.

When you limit one thing in the diet (say meat), something else increases. And that something else may be equally as important as what is limited. Even at the time Keys did his studies people in the U.S. were eating less plant foods and more processed foods than people in the Mediterranean or Japan (the two healthiest areas). So, was it the absence of meat or the abundance of plant foods that was contributing to their greater health? Or maybe it was the absence of processed foods.

Thanks to Keys the lipid hypothesis has guided nutrition advice for the last 50 years. We should have gotten healthier, but we haven't. We've gotten sicker.

ThomsonsPier wrote:

wosnes wrote:Part of the problem is that researchers get tunnel vision and look at one particular thing -- ignoring everything else.

Combinations and interactions of variables can only effectively be studied when every one of them is understood on its own. Whether this is valid in terms of nutritional research is another question altogether.

I think this is an important question. More and more - in issues of health and social science - we are finding that the interactions and combinations are diffferent from what we might expect from the individual variables. (And in social science, that level of control is often difficult or even unethical to achieve.)

The scientific method is valuable, but it does have limitations.