3. Data Life Cycle
3.4.2 Precision & Accuracy
PRECISION & ACCURACY:
one can be very precise but… also very inaccurate.
Often incorrectly used as synonyms, these two terms have very different meanings as illustrated by this diagram.
ISO standard 5725 1 refers to trueness as a better word for describing the principles and methods used for measuring and assessing the accuracy of a test. Trueness is defined as the degree of agreement between the arithmetic mean of a large number of test results and an accepted reference value; precision refers to the agreement between test results.
Data management errors are generally defined by their source:
- human: errors in reading, typing, transcribing or in judgement;
- environmental: errors due to ambiant conditions having an influence on measurements. E.g.: air temperature;
- equipment: errors due to equipment limitations and/or capacities.
TO LIMIT ERRORS:
- Ensure staff is appropriately trained;
- Perform quality control tests. E.g.: comparing measurements and observations to references.
Errors can also be called "systemic" when they present similarities in occurrence and pattern ("bias" is also used) or, on the contrary, called "random" when no particular trend is detected.
- ISO 5725-1:1994. Accuracy (Trueness and Precision) of Measurement Methods and Results.