View Single Post
  #13   Report Post  
Posted to sci.electronics.repair
Jeff Liebermann Jeff Liebermann is offline
external usenet poster
 
Posts: 4,045
Default Maybe I need another 'scope?

On Mon, 26 Nov 2018 21:16:31 -0500, Phil Hobbs
wrote:

On 11/26/18 5:10 PM, Jeff Liebermann wrote:
https://ndb.nal.usda.gov/ndb/


I'd be super glad to be corrected on those points--I'd far prefer to
have confidence in the data I'm presented with, especially since some of
the conclusions I'm invited to make have serious health repercussions.
Cheers
Phil Hobbs


Well, if I can't convince you that the USDA Food Composition Database
is reasonably accurate by the number of decimal places, voluminous
source data, and the large number of citations, perhaps a "what-if"
sanity check might be helpful. That's what I do when an assertion or
conclusion doesn't quite ring true. I ask myself:
If the USDA database was populated in the 1950's by slave
labor (grad students) and never verified, what might I
also expect to be true or to happen?

1. The data would conflict by food databases from other countries.
For example, Australia:
http://www.foodstandards.gov.au/science/monitoringnutrients/ausnut/ausnutdatafiles/Pages/foodnutrient.aspx
Line number 2653 for
"Banana, cavendish, peeled, raw"
shows 346mg potassium, which is quite close to the USDA nominal figure
of 358mg per 100grams. I haven't checked any other databases, but I
can look around for discrepancies if necessary.

2. Food and supplement producers that rely on potency claims to sell
their products would have an interest in stabilizing the official
figures so that their products would always be higher potency than
typical. A common variation of this need for stability is the
declared weight of the contents of packaged food. The weight can be
greater, but never less than the stated value (unless the listed item
is deemed undesirable). Same with nutritional values.

3. The nutritional values listed have been used in thousands of
health and medical experiments since the database was corrected. If
there were any errors, experiments based on the data would also show
discrepancies or at least large variations in results which would
attract suspicion. I haven't seen any of that in the press.

Enough for now.


--
Jeff Liebermann
150 Felker St #D
http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558