Precision And Accuracy: What Is The Difference?

Precision and accuracy are two words that are commonly used in the context of science. People of science often use these words interchangeably. However, strictly speaking, they mean different things.

Here is the short version of the difference between them:

1. On the one hand, the functionally correct usage of ‘accuracy’ refers to how close a measurement is to the correct/true value.

2. On the other hand, the functionally correct usage of ‘precision’ refers to how close a measurement is to other measurements repeated under the same conditions.

If that is all there is to it, why am I bothering to write an essay about this topic? Well, there is more than what meets the eye here.

To make things clear or to throw a wrench in the works (depends on how you look at it), there also exists an ISO 5725 definition of accuracy. This can get complicated. So, I will treat the ISO version of accuracy later in the essay.

Let us start by covering some fundamental concepts that we will be needing to understand the scientific use of accuracy and precision better. But don’t worry. In this essay, I will be focussing on practical situations rather than boring theory. Let us begin.

This essay is supported by Generatebg

Matt Vs. Cheat

If you have been following my work, you would recognise Matt and Cheat straight away. They are in-house fictional characters that I employ to illustrate concepts and puzzles.

In this essay, these two make their appearance because of a recent competition that they got into with each other. You see, both of them got archery kits recently and have been at it. Each of them has his own strengths and weaknesses.

Based on their performance, your task would be to figure out who is the better archer. This should be straight forward, right? Well, let us find out.

---

What is the Functional Meaning of Precision?

Up first, Matt would like to present to you, his last ten arrow hit marks. His bullseye board looks as follows:

Notice how Matt’s hit marks are not close to the bullseye. We will come back to this point in a bit. But first, let us focus on another aspect of his bullseye board. Matt’s hits are actually very close to each other.

In short, Matt’s hits are fairly precise, because each hit has little deviation from any other hit. Based on his performance, would you rate Matt as a good archer?

As you ponder that question, let us see what Cheat has to offer.

What is the Functional Meaning of Accuracy?

It is now Cheat’s turn to present his results. He too shows you his bullseye board with his last ten hit marks. They look as follows:

Notice how they are not as tightly grouped as Matt’s hits. But at the same time, they are much closer to the bullseye.

In short, Cheat’s hits are fairly accurate, because each hit has little deviation from the bullseye. Based on this performance, would you rate Cheat as a good archer?

And between the two, which one is the better archer? I am sure that you have an answer to that question by now. For now, just hold on to it at the back of your mind.

---

Precision and Accuracy: What is the Difference?

So far, with the help of Matt and Cheat, I have laid out the functional meanings of precision and accuracy.

To decide which is better really depends on the requirement. In other words, there is often no straightforward answer.

The answer typically depends on what the goal (in the scientific setting) is. Let us consider Matt and Cheat once again. While Cheat’s hits are pretty close to the bullseye, Matt’s hits are more tightly clustered.

Purely based on their respective performances, I would say Cheat is the better archer. He is objectively closer to the target after all. But often in scientific settings, results akin to Matt’s performance are preferable.

You see, Matt might be an inferior archer to Cheat now, but he just needs some ‘calibration’. If Matt can calibrate his accuracy better, he already has the precision to hit the bullseye ten out of ten times.

By now, you would think that things are pretty clear as far as precision and accuracy are concerned. However, ISO 5725 has other plans.

Precision and Accuracy: What Does ISO 5725 Have to Say About Them?

So far, we have used to the words precision and accuracy to convey the following meanings:

1. Accuracy: How close is the measurement to the objectively correct/true value?

2. Precision: How close is the measurement to other measurements repeated under the same conditions.

In a technical sense, “accuracy” tracks systematic observational errors. ISO 5725 calls this “trueness” instead of accuracy. So, as per ISO 5725, “trueness” measures how close a measurement is to the correct/true value.

In a technical sense, “precision” tracks random observational errors. ISO 5725 is happy with the definition of precision as it is. But here comes its most important contribution.

ISO 5725 defines accuracy as a description/measure of BOTH trueness and precision. So, in a scientific sense, one cannot simply use accuracy and precision interchangeably.

Precision is a subset of accuracy. Consequently, accuracy is a superset of precision. However, precision and trueness are not directly connected.

For a measurement to be scientifically accurate, not only does it need to be close to the correct/true value, but it also has to be close to other measurements repeated under the same conditions.

---

If you’d like to get notified when interesting content gets published here, consider subscribing.

Further reading that might interest you: 

• Barnum Statement: How To Really Debunk Astrology
• Molyneux’s Problem – Can You Really Solve This Challenge?
• How Imagination Helps You Get Good At Mental Math?

If you would like to support me as an author, consider contributing on Patreon.