Michael Archon Sequoia Nielsen > wrote:
>Blair P. Houghton wrote:
>
>> And it's still ********, because it assumes that all nonfat
>> body tissue weighs the same.
>
>Inaccuracy does not equal "********". All models in natural sciences are
>also just simplified models that approximates nature "good enough". As
>long two samples from the same individual that shows a decline in body
>fat is actually a decline in body fat 95% of the time, it is a good measure.

BMI doesn't tell you that. BMI tells you that their
height or weight changed.

BMI is certainly ********. The correlation between
fat and (height,weight) tuples is very broad. Telling
everyone who is above the BMI calculation that they are
fat is a mistake in a ridiculously large proportion
of cases. But that's what doctors do.

--Blair