/sci/ - Science & Math

>>8630224
>well within margin of error
>oh look, it's the denier's latest favorite buzzword
did you actually find a published value on the margin of error for such measurements? or did you just decide "it's kinda small, I think it's within the margin of error"?

but here's a tip (and a dong behind it): compounding multiple measurements reduces total error.
suppose a thermometer is accurate to within a tenth of a degree. now suppose you measure the temperature of the same thing ten times with ten of those thermometers.
see, the margin of error of the compounded average will be significantly less than a tenth of a degree. this is because the likelihood that all the errors line up in the same direction (adding together) rather than have random sign (canceling out) is very low, on the order of one in a thousand. now, if there's some kind of instrumental bias (which would result in all the errors pointing the same way by the same amount) that's a different story. but that's not what margins of error represent.