Accuracy is an agreement of a measured value with an expected value. While a precise measurement may speak highly of an instrument's quality, an accurate reading will not reflect on the quality. In this case, an experiment could yield inaccurate results yet be highly precise. The danger when evaluating an experiment is that some errors are not random.
Therefore an experiment with a low degree of precision can provide accurate values where appropriate statics are applied.Ĭonversely, precision cannot be improved by taking repeated measurements but it is impossible to quantify precision without experimental repeats.
(This assumes that errors are randomly above and below the true value to the same degree). Number of MeasurementsĪccuracy can be improved by taking repeat measurements and taking an average. (c) is precise because the shots are clustered together, but not accurate because they are not close to the target. (a) is neither accurate nor precise because the shots are not close to the bull's-eye but all over the place. However, it is important to note that it is not possible to reliably achieve accuracy without precision.Īnother picture showing the difference between accuracy and precision using the target-shooting metaphor. This is how results can indicate precision but not necessarily accuracy. When all or most arrows are grouped tightly together, the shots fired can be considered precise since they all landed near the same spot, if not necessarily near the bull's-eye. How precise the shots are depends on how often the arrows land near each other on the target. The closer an arrow is to the bull's eye, the more accurate the shot. Accuracy describes how close the arrows are to the bull's-eye. For this example, consider results from a round of target practice.Īrrows are fired at a target, and measurements are taken in relation to the bull's eye at the center of the target. The device was reasonably precise but not very accurate.Īccuracy is the degree of correctness, while precision is how strict that correctness is (or isn't) - how reproducible results are. It had 37 gear wheels and could follow the movements of the moon, including the irregular orbit of the moon where the moon’s velocity is higher in its perigee than in its apogee. This distribution shows no impressive tendency toward a particular value (lack of precision) but each value does come close to the actual temperature (high accuracy).Īnother example is the Antikythera mechanism, an ancient Greek analog computer that was used to predict astronomical positions and eclipses. A temperature sensor is tested 10 times in the refrigerator. A clock is precise when it marks the seconds exactly and unvaryingly but may still be inaccurate if it shows the wrong time.Ī measurement may be accurate but not precise, precise but not accurate, or neither, or both.Īn example of bad precision with good accuracy might be a lab refrigerator that holds a constant temperature of 38.0F. But the shots are only accurate if they hit the bull’s eye. When a sharpshooter fires at a target, if the bullets strike close together-clustered, rather than spread out-that is precise shooting. precision involves an ideal of meticulousness and consistency, while accuracy implies real-world truth. In his review of Simon Winchester's The Perfectionists: How Precision Engineers Created the Modern World, James Gleick writes for the New York Review of Books