Accuracy is how close a measured value comes to the true (accepted) value of a quantity, and it is limited by the least count of the measuring instrument — the smallest quantity that instrument can reliably distinguish (e.g. 0.1 mL for a typical burette, giving a reading like 10.2±0.1 mL). Error in a single measurement is expressed as absolute error (observed value − true value) or, more usefully, as relative error (absolute error ÷ true value, as a percentage). Precision, in contrast, is how closely a SET of repeated measurements of the same quantity agree with EACH OTHER, regardless of whether they are close to the true value; it is expressed through absolute deviation (a single reading minus the set's own mean), mean absolute deviation (the average of those deviations across the whole set), and relative deviation (mean absolute deviation ÷ the mean, as a percentage). High precision — tightly clustered repeated readings — is a necessary condition for high accuracy, but does not by itself guarantee it: a set of readings can be very precise yet still be systematically off from the true value, meaning it is precise but not accurate. SI units, the internationally standardised measurement units (e.g. the kilogram for mass), underpin how these measured and true values are reported and compared.