What is the term describing how correct a measurement is?

A “correct” measurement means that the accepted value is the “real” value – within known and allowable variations. Precision is a measure of how consistently you get the same value with repeated measurements. If you measure the same thing and get all kinds of different values, then none of those measurements can really be considered “correct’.

Accuracy is how close the measured values are to the “real” value. Even with poor precision, enough “accurate” values can be averaged into a better approximation of the “teal” value. However, any single measurement cannot be considered “correct” because it will vary around the true value every time.

Putting the two together we can devise and improve measurement systems to give us “correct” answers more confidently. It is better to have a “precision” instrument than a purely “accurate” one – because accuracy contains the implied statistical averaging.

A “precise” measurement that is not accurate can be adjusted to also be accurate. An ‘accurate’ measurement that is still NOT precise will always have a large amount of error associated with any single measurement, even if the "precision" is improved.

Note that “percent error” can be applied to both – it may be the “single measure” you are looking for, as it describes the degree to which you observed value may differ from the “true value”. Error cannot be accurately assessed without multiple measurements either.

A common diagram, well done by http://speakingdata.blogspot.com/