Accuracy and precision

Accuracy and precision are measurements within science and engineering of how well results fit with reality and how reliable they are. Accuracy is a concept that describes the degree to which a measurement approximates the actual value, whereas precision measures how reproducible a result is.

The poetry of reality
Science
We must know.
We will know.
A view from the
shoulders of giants.
v - t - e

Low accuracy or precision

An experiment or test with low accuracy or precision is still useful if the other is reasonably high. A test with low precision can be improved with more samples as each data point will still average out as an accurate value (diagram 1). On the other hand, a test with low accuracy can still be useful if it is precise, because we can conclude that there must be a reproducible and systematic error in the measurement causing the lack of accuracy (diagram 2). In science, this is particularly useful for refining theories and experiments.

Visual representation
gollark: I mean, in *my* case, I find random giant companies having access to stuff like my browsing history creepy, which is a good reason to me. Other people might not think this. But there are other reasons.
gollark: Sure? It's a bit loosely defined but I guess so.
gollark: Also, there aren't "objective reason"s to do anything. The most you can say objectively is that "X is good/problematic because it satisfies/goes against Y goal", or maybe "I consider Y goal/X thing important".
gollark: People should probably consider privacy more seriously than most actually *do*, at least. A lot of people say they care a bit but then ignore it.
gollark: <@126590786945941504> Maybe they should.

See also
This science-related article is a stub.
You can help RationalWiki by expanding it.
This article is issued from Rationalwiki. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.