Unwarranted precision

[pne]

I was taking a multi-vitamin tablet this morning and looked at the packaging.

What I saw made me sad due to the unwarranted precision. Clearly, they measured the amounts present in a single tablet (of nominally 45 g) and then converted that to quantities in 100 g (a measure that's usually present in such tables of nutritional information, probably so that you can compare quantities of nutrients in different items regardless of the typical serving size of each), always representing them as numbers with one decimal place.

This means, for example, that the tablets supposedly contain "4'444.4 µg" of folic acid per 100g.

Really? Is that really accurate to the nearest 100 nanograms?

I'll note that the quantity per tablet is "200 µg", which is presumably "2×10² µg" and not 2.00×10² µg, let alone 2.0000×10² µg with five significant figures, as implied by the quantity per 100g.

I also note that the tablets contain "2.0 mg" of vitamin B6 but "1 µg" of vitamin B12, implying that they were aware of precision there—yet those two measurements turned into 44.4 mg and 22.2 µg, respectively, both implying three significant figures. (And a precision accurate to the nearest 100 nanograms in the second figure, again.)

Honestly, people. Learn about the difference between precision and accuracy, and don't claim more precision than is warranted by the accuracy.

Comments

MICRO-grams

[jixel.livejournal.com]

the mu is micro or 1,000,000th

milligram is 1,000th
nano is one billionth

http://en.wikipedia.org/wiki/Nano-

Re: MICRO-grams

[pne.livejournal.com]

I'm not sure what your point is?

(Are you doubting that some nutrients are present in the tablets only in microgram quantities?)

[jixel.livejournal.com]

sorry i guess i didn't do the math. i thought you were confusing the mu symbol with nanograms.

truly they need to provide the statistical precision of their measurements - you may actually have a legitimate case to claim to them of mislabeling

and i need to read more thoroughly before commenting!

i recall once a friend had me realize how meaningless all the digits of pi really are when you consider the scale as you have done here, in most cases only a few digits are really meaningful at all (and perhaps explains how fractional approximations were acceptable enough in antiquity)

[pne.livejournal.com]

i recall once a friend had me realize how meaningless all the digits of pi really are when you consider the scale as you have done here, in most cases only a few digits are really meaningful at all

Ah, right.

I read somewhere how many digits of pi you need to calculate the circumference of the known universe given its diameter to an error of less than one atomic nucleus -- I forgot the exact number, but it was around 80, I think. Certainly a lot lower than the number of digits currently known :)

[jixel.livejournal.com]

is that really the chinese (or other) character for "bitchy" lol

[pne.livejournal.com]

:) No, unfortunately not.

I couldn't think of a good character to use for "bitchy" so the mood just inherits the mood picture for the "angry" mood (see the tree of moods, and which inherits from which, here (http://www.livejournal.com/moodlist.bml?moodtheme=17068&ownerid=725716&mode=tree)).

The character does mean "(be) angry" in Japanese.

(The Chinese meaning is "rage, fury" according to zhongwen.com (http://zhongwen.com/d/171/d227.htm).)