2
$\begingroup$

Precision is usually understood as the number of significant figures in some experiment. Accuracy is the difference between the best measurement and the real value.

How are precision and accuracy determined from a given experiment? Or equivalently, how are systematical errors and statistical errors calculated. Statistical errors are usually treated approaching "the normal distribution". But I am not sure how systematics are determined...Perhaps making some "pattern" measurements?

$\endgroup$
0
$\begingroup$

the most precise instruments available to us must be calibrated before they can be accurate. ie, used on values we assume are already known.

the difference between the mean of measured values and that of the actual value is the systematic error, also known as the offset in calibration.

$\endgroup$
2
  • $\begingroup$ How is the "actual value" determined?What if the "actual value" is NOT known? Does it mean there is no systematical error? $\endgroup$
    – riemannium
    Sep 10 '13 at 12:37
  • $\begingroup$ if you go down to the fundamentals: what is a kilogram, second or meter...these are all DEFINED not measured values. In practice systematic error can be completely removed via calibration. $\endgroup$
    – gregsan
    Sep 10 '13 at 14:13
0
$\begingroup$

Since the real value is not known, accuracy cannot be measured - it is a concept used for mathematical analysis: constructing estimators, evaluating their bias, convergence, etc.

I suggest reading this chapter for learning basic statistical concepts.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.