I often use these two words when I try to describe my experimental results. I wonder whether they can be used in the same situation. I searched these two words with WikiDiff, but still cannot get a clear explanation.
If you imagine the target to be made up of many squares, I've understood precision to be the size of each square, and accuracy to be how close you got to the desired square. This is based on the practice of precision being specified to a certain number of significant figures in scientific measurements.
However, Wikipedia notes that ISO 5725 provides the following definition:
According to ISO 5725-1, the general term "accuracy" is used to describe the closeness of a measurement to the true value. When the term is applied to sets of measurements of the same measurand, it involves a component of random error and a component of systematic error. In this case trueness is the closeness of the mean of a set of measurement results to the actual (true) value and precision is the closeness of agreement among a set of results.
Precise is "my watch says it is 10:35:06". But is this accurate? Maybe, maybe not. It depends on how carefully your watch was last set. You can get a precise experimental result, which is not at all accurate because someone bumped into the equipment and misaligned something or loosened a connection
Following is a recent, famous example of a (fairly) precise result of the speed of neutrinos which turned out to be wildly inaccurate. From Science:
Enough already. Five different teams of physicists have now independently verified that elusive subatomic particles called neutrinos do not travel faster than light. New results, announced today in Japan, contradict those announced last September by a 170-member crew working with the OPERA particle detector in Italy's subterranean Gran Sasso National Laboratory. The OPERA team made headlines after they suggested neutrinos traveled 0.002% faster than light, thus violating Einstein's theory of special relativity. The OPERA results were debunked months ago, however. So instead of the nail in the coffin of faster-than-light neutrinos, the new suite of results is more like the sod planted atop their grave.
The OPERA team had timed neutrinos fired through Earth from the European particle physics laboratory, CERN, near Geneva, Switzerland, and found that they made the 730-kilometer trip to Gran Sasso 60 nanosecond faster than they would traveling at light speed. But in February, the OPERA team also discovered that a loose fiber optic cable had introduced a delay in their timing system that explained the effect. (Emphasis added.)
This result is famous, but things like this happen all the time in labs. That is one reason why results need to be independently confirmed.
In 1923 Theophilus Shickel Painter, 1889 - 1969, an eminent cytologist, reported that there were precisely 48 human chromosomes. This was published in a respected scientific journal and was accepted in the scientific community for a period of 33 years.
In 1956, Tijo and Levan published their work and were puzzled to discover they could only find precisely 46 chromosomes.
Not until Ford and Harrison also published their findings of 46 chromosomes, again in 1956, did it dawn on everyone that Painter's original data was inaccurate.
The difference between precision and accuracy is that the first is a matter of what is expressed, and the second is a matter of what is executed.
OED-3. . . (subscription required)
Precise : Characterized by definiteness or exactness of expression
OED-3 . . (subscription required)
Accurate : Executed with care; careful.