ABSTRACT
Data can be described in terms of: • Accuracy: how close the measured estimates are to the exact or true value (1). For example, when measur-
ing the prevalence of cancer, different studies may produce different values for prevalence. The most accurate studies are those in which the measured value is as close as possible to the true prevalence of cancer
• Precision: the reproducibility of a set of measurements (2). A precise sample estimate will have a very small random error of estimation. For example, a machine that measures blood glucose levels in blood samples is precise if it will consistently give the same measurement of blood glucose from the same individual’s blood sample. However, this machine may not be accurate if it is calibrated such that it always underestimates the value of the blood glucose. Thus a precise but inaccurate method of gathering data will produce a systematic bias in the results.
