Bumbling Bob wrote:
>>> We all know obesity is a leading cause of premature death. Here is a
>>> journal article.
>>
>> It's an abstract, not an article.
>
> Correct.

Next time call it an abstract rather than an article.

>> Note that the number of self-reported
>> "vegans" was 83, or 0.114966% of the population surveyed. That small
>> of a sampling is statistically irrelevant for the purposes of
>> comparing to the other samples.
>
> He is either wrong about statistics,

I'm not.

> which I doubt, or lying.

I'm not.

> Statistics
> don't depend on huge sample sizes.

Strawman. I noted that self-described "vegans" made up an infinitesimal
percentage of the entire group surveyed in that study.

> Most statistical studies are done on
> groups of 30 or so.

Ipse dixit and false. Some studies are very small, some are quite large.
The study in question was quite large. I pointed out that out of this
very large survey, just over one-tenth of one-percent of those surveyed
described themselves as vegans.

The issue I've raised is the problem with comparing such a tiny sample
against over 99.8% of those surveyed in such a study. Such a tiny sample
is subject to skewing results, particularly when a sample like that is
of people with an eating disorder and who are being studied for
something altogether different (i.e., the study is a mammography cohort).

> There are measurements called T Tests
> to measure significance.

That's a separate issue from the one I've raised.

A caveat is that the confidence interval relates to the
population sampled. If we have a small sample of part of a
population, or a very small sample of the whole population, then
the confidence interval that is generated is not necessarily
that for the whole population.
http://www.jr2.ox.ac.uk/bandolier/bo...ossary/CI.html

> I'm sure US knew this, but was misleading
> those that don't understand statistics.

I'm not misleading anyone. I'm pointing out the greater probability of
skew in interpreting data from 0.114966% of the subjects of a survey and
comparing it to data from the other >99.8%.

> Unlike me, when I say, you're
> right, it was the abstract, not the animal, Usual Suspects simples lies
> and abuses statistics.

I neither lied nor abused anything.

> In fact, too large a sample size has problems statistically.

Not true, numb nuts. Here's a very understandable set of lecture notes
about sample sizes.

http://www.upa.pdx.edu/IOA/newsom/pa551/lecture4.htm

> The sample size met the standards for the clinical journal.

Which isn't very high. Acceptance criteria are originality, validity of
data, clarity of writing, strength of conclusions, and potential
importance. It's a peer-review journal, not the Bible of nutrition where
every study is hallowed.

http://www.ajcn.org/misc/Info_for_authors.pdf

>>> American Journal of Clinical Nutrition, Vol. 81, No. 6, 1267-1274,
>>> June 2005
>>> © 2005 American Society for Clinical Nutrition
>>> ORIGINAL RESEARCH COMMUNICATION
>>> Risk of overweight and obesity among semivegetarian, lactovegetarian,
>>> and vegan women1,2,3,4
>>> PK Newby, Katherine L Tucker and Alicja Wolk
>>>
>>> 1 From the Jean Mayer US Department of Agriculture Human Nutrition
>>> Research Center on Aging at Tufts University, Boston, MA (PKN and
>>> KLT), and the Division of Nutritional Epidemiology, Department of
>>> Environmental Medicine, Karolinska Institute, Stockholm, Sweden (AW)
>>>
>>> Background: Observational studies suggest that a plant-based diet is
>>> inversely related to body mass index (BMI), overweight, and obesity.
>>>
>>> Objective: Our objective was to examine the BMI (kg/m2) and risk of
>>> overweight and obesity of self-defined semivegetarian,
>>> lactovegetarian, and vegan women.
>>>
>>> Design: Data analyzed in this cross-sectional study were from 55459
>>> healthy women participating in the Swedish Mammography Cohort. Women
>>> were asked whether they considered themselves to be omnivores (n =
>>> 54257), semivegetarians (n = 960), lactovegetarians (n = 159), or
>>> vegans (n = 83), and this question was the main exposure variable in
>>> this study. In secondary analyses, we reclassified women as
>>> lactovegetarians on the basis of food intakes reported on the
>>> food-frequency questionnaire.
>>>
>>> Results: The prevalence of overweight or obesity (BMI ‰¥ 25) was 40%
>>> among omnivores, 29% among both semivegetarians and vegans, and 25%
>>> among lactovegetarians. In multivariate, adjusted logistic regression
>>> analyses, self-identified vegans had a significantly lower risk of
>>> overweight or obesity [odds ratio (OR) = 0.35; 95% CI: 0.18, 0.69]
>>> than did omnivores, as did lactovegetarians (OR = 0.54; 95% CI: 0.35,
>>> 0.85) and semivegetarians (OR = 0.52; 95% CI: 0.43, 0.62). Risk of
>>> overweight or obesity remained significantly lower among
>>> lactovegetarians classified on the basis of the food-frequency
>>> questionnaire (OR = 0.48; 95% CI: 0.30, 0.78).
>>>
>>> Conclusions: Even if vegetarians consume some animal products, our
>>> results suggest that self-identified semivegetarian, lactovegetarian,
>>> and vegan women have a lower risk of overweight and obesity than do
>>> omnivorous women. The advice to consume more plant foods and less
>>> animal products may help individuals control their weight.
>>>
>>> Key Words: Overweight €¢ obesity €¢ BMI €¢ vegetarian €¢ lactovegetarian
>>> €¢ vegan
>>>
>>>
>>>
>>>
>>>