30

The dictionary for accuracy says:

  1. The quality or state of being correct or precise.
  2. The ability to perform a task with precision.

And for precision:

  1. The quality, condition, or fact of being exact and accurate.
  2. Marked by or adapted for accuracy and exactness

Are these nouns interchangeable? If not, how to decide whether to use one or another?

3
  • 1
    Actually they aren't verbs... They are nouns. :)
    – Alenanno
    Jun 19, 2011 at 22:13
  • "Well, you wore out your welcome / With random precision..." (Pink Floyd, Shine on, you crazy diamond)
    – MT_Head
    Jun 19, 2011 at 22:18
  • uow! verbs was an awful mistake lol Jun 19, 2011 at 22:41

6 Answers 6

73

The Wikipedia article explains the difference in the best way possible:

In the fields of science, engineering, industry and statistics, the accuracy of a measurement system is the degree of closeness of measurements of a quantity to that quantity's actual (true) value. The precision of a measurement system, also called reproducibility or repeatability, is the degree to which repeated measurements under unchanged conditions show the same results. Although the two words can be synonymous in colloquial use, they are deliberately contrasted in the context of the scientific method.

High accuracy, but low precision:

High accuracy, but low precision

High precision, but low accuracy:

High precision, but low accuracy

2
  • 1
    you might want to add that one can be extremely precise but not accurate or even wrong (the images show that, but some text might help the OP); this was a pet peeve of one of my early science teachers, who drilled the difference into our little heads. In the case of your hypothetical ruler, you might get one with markings down to the micron level, and you might make very precise measurements - but imagine that your elbow kept getting jostled as you measured. Your accuracy would obviously suffer, while your numbers would still have many digits of (useless, deceptive) precision.
    – MT_Head
    Jun 19, 2011 at 22:33
  • 1
    That's why I decided to remove the ruler part, because there was the human part involved and that's very hypothetical, as you proposed. Feel free to edit my answer and add what you propose, I believe this community to be a work of all of us, I'll be happy if you can add your two cents. I believe your reputation allows this, so go on, please!
    – Frantisek
    Jun 19, 2011 at 22:42
22

3.14 is accurate, 3.1111111111111111 is precise, 3.14159265 is both

4
  • 2
    An even better example than the images in @RiMMER's answer.
    – Oben Sonne
    Jun 20, 2011 at 8:51
  • 1
    3.14 is accurate, but 3.11111111 is not precise in the context of this answer. It is less accurate than 3.14. (Although 3.14159265 is both accurate AND precise!) It seems misleading to refer to a measurable quantity as "precise" when it is less accurate than an alternative that has a lesser degree of precision. Jun 20, 2011 at 13:46
  • 4
    @Feral: I think cindi meant by including more digits in 3.1111111111111111 she's been specified a ~15+ digits of precision just it happened to be less accurate than 3.14 (which with only three digits isn't particularly precise; but more accurate).
    – dr jimbob
    Jun 20, 2011 at 13:54
  • 2
    @Feral: 3.1111111111111111 is very precise: the hypothetical author has specified the number very exactly, to (too) many digits of precision. Of course the author happens to be hopelessly wrong: inaccurate. Jun 22, 2011 at 5:56
14

In non-technical usage:

  • 'precision' refers to the exactness of a statement, how sharp the boundary is between a statement being true or false (not whether it is true or false but the lack of vagueness). The antonym of 'precise' is 'vague'.
  • 'accuracy' refers to how correct a statement is, how close it is to being true.

In science and engineering:

  • 'precision' is how close a number of attempts are to each other (RiMMERs answer), or the number of digits used in presenting a numerical result.
  • 'accuracy' is the how close the average of a set of answers is to the correct answer (RiMMER's answer).

In statistics, these both have very particular technical meanings in the context of 2 by 2 contingency tables:

  • 'precision' is the ratio of true positives to all positives TP/(TP+FP), also known as the Positive Predictive Value or PPV.
  • 'accuracy' is the ratio of true positives and true negatives to the total (TP+TN)/(TP+FP+FN+TN).
2
  • Nice additional distinction for statistical context, as a separate or sub-class definition (in addition to overall science and engineering)! Jun 20, 2011 at 13:39
  • +1 for the explanation of how the two words differ in non-technical usage. Aug 16, 2021 at 12:00
11

Having read RiMmer's excellent distinction, I'd say that, outside science, these words are sometimes interchangeable, but not always; where they are contrasted, I believe the distinction often hinges on intention (focus/method: precision) versus result (success/failure: accuracy).


If you are precise, you display attention to detail as much as you can, i.e. you do your best within the limited range of your abilities to be precise. But whether you are successful may also depend on other factors beyond your power.

Embroidering cushions is very precise work.

Work can only be precise: accuracy wouldn't be quite appropriate, because failure is hardly relevant or even possible. A good result is taken for granted, because it is well within one's abilities to succeed.

She calculated precisely how much sulphur she would need according to the formula to make gold.

She does what she can; but will she make gold? This sentence says nothing about success or failure.


If you are accurate, you produce results that are close to your target; whether you tried hard and did your best is irrelevant.

She calculated accurately how much sulphur she would need to make gold.

This implies that she successfully calculated how much sulphur she needed.

Your prediction of 1–2 % economic growth was accurate, imprecise though it was. You were right; but next time please provide a more precise estimate.


This is not to say that they always differ on this point; sometimes they are equivalent; and sometimes precision isn't so much about intention as it is a neutral opposite to blurred or general. But I do believe accuracy is mostly about result.

1
  • +1 for the intention/outcome distinction. It rings true for me, and in matters like this, by definition one's personal distinctions must be valid (even if no-one else appreciates them! :) Feb 8, 2012 at 1:38
4

Is this a fair response? - I've seen people put up Powerpoint Slides suggesting the market for say, tablet computing devices, will be $527,658,749 by the year 2015. While that number looks pretty precise, it certainly won't be accurate. This is an important distinction that exemplifies these two words are not synonyms (even when I see some dictionaries offering them to be)

1
  • 2
    not a full response, but certainly a good example Jun 20, 2011 at 18:25
0

I surprised nobody added picture of all 4 possibilities! Picture from St Olaf College.

enter image description here

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.