Food Studies & Statistics

greatwestescape

A lot of people are citing, this study or this study based on what they have heard. I've added a few things below.

Keep in mind this is OVERLY simplistic

General Education

1) Studies do not PROVE anything. They show the strength (positive or negative) between 2 or more variables.
---> Abstracts & media will say "the most recent research 'indicates' blah blah blah". This does not tell you the strength of the relationship or any other vital information.

2) Conditions more times then not do not reflect real life
---> Research conditions more times then not use HIGH AMOUNTS of the substance over short periods which do not reflect day to day diets

---> Let's use Omega 3 studies. You hear on the news, OMG OMEGA 3 IS THE BOMB! What do you do? You run out to your health food store and buy a bottle of OMEGA 3 capsules. For argument sakes, let's say 2-4 capsules a day gets you 600mg of DHA and 250mg of EPA. You're now getting all the benefits, that the study said, right??? NO! Most of the studies use HIGH doses to check for significance. For Omega 3 studies with the highest amount of significance, this is around 3-5g of OMEGA 3 per day.

3) Meta-Studies, a lot of medical & fitness researchers will use meta-analysis for fitness & nutrition. If you see "meta-analysis" in the title, take it with a HUGE GRAIN OF SALT.
---> Meta-analysis is extremely difficult & most statisticians/researchers should prob. not partake in meta-analysis. You need very solid methods & criterion along with a very narrow scope.

4) Nutritional studies more times then not have questionable research methods.
---> Without getting complicated, to assesses something you need a scale, inventory or some type of measure. A lot of nutritional studies re-purpose scales without testing their validity & reliability. The researchers use the measures under the guise that, "hey other studies have used this". This isn't done intentionally, the researchers just don't know any better.

5) Samples
-- Always ask, where did they get there participants from? Do they represent the population or only a specific portion?
-- If it involves the word "multiple case studies" be careful
-- n < 30, or the sample size is below 30 (e.g. 8, 10, 12) you have to be careful.
-- A good sample size is always about 34, a lot of places will say 30, but 34 is the magic number

6) Use common sense
---> If it doesn't sound right, it prob isn't.
---> Most things if they are not a naturally toxic product (e.g. arsenic), will not negatively effect your health if taken in moderation.
---> We are all different, studies try to generalize to a population. Studies try to generalize to 68% of the population based on a bell curve. If you do not fit into the 68% in terms of body fat % (i refuse to use BMI), height, weekly exercise (in hours), lifestyle, diet etc the study will not apply to you as much unless the study is not stratified (e.g. it is targeted towards a specific population).

I've kept this overly simplistic. I have generalized & exaggerated in some respects to get the point across. I'm sure i missed a few things as I wrote this off the top of my head.

ONE BIG THING! Well at least to me,

It's just about what you eat, but how you feel. If you're grumpy every day because of your regime, i'm guessing more people then not will fail at their goal long term. Who cares if you shed 20lb in 45 days if you put back on 30lb by years end. You need to feel good about yourself. You need that morale, you need that support whether it comes externally or internally. Mental healthiness is just as important as your physical health! Make a lifestyle change, not a, on again, off again diet!