Precision VS accuracy

In English, there’s a distinction between precision and accuracy:

We learn these words on level 20 and 28:

Since there’s a difference in English, I’m wondering if there’s also a difference in Japanese, even though WaniKani teaches these words as synonyms?

15 Likes

精度 is a word that truly encompasses both English words, as can be confirmed checking a monolingual definition. The usage of both is not a result of blurry English meanings.

正確 only means “accurate” in Japanese, and has “precise” listed only in the sense that English speakers use these words interchangeably in colloquial situations. You can make this “accuracy” by adding さ for 正確さ.

The Japanese word that only means “precise” in the strict sense is 精密. However, you might see “accurate” listed due to the above mentioned colloquial usage of the English. You can make this “precision” by adding さ for 精密さ.

16 Likes

you learn something new every day here :smiley:
In my head, these two have always been synonyms…

12 Likes

I think that in everyday situations, as @Leebo suggests, many native English speakers use these words interchangeably.

As a physics teacher, I have to explicitly teach that they are not the same. (I also use the same diagram!) Most students will not already know this.

20 Likes

i remember getting upset at this lesson too. i’m a statistician!

but it’s true a lot of people actually use the words for one or the other interchangeably.

such is life.

7 Likes

I used to do target rifle shooting, so the difference was important to me too. Essentially, accuracy is in getting the sights wound in properly, but precision is all down to the shooter. The more precise you can be, the easier it is to make the rifle accurate.

8 Likes

Me too! But not since I left university. As a cadet, marksman badges were always judged on precision.

So… if you hit the same distant corner in five consecutive shots, you got a top score? :thinking:

2 Likes

I don’t see why not. You’re clearly a great shooter, but the calibration was just off.

3 Likes

I mean… sure, why not. I’m not going to argue against a Finn when it comes to sharp-shooting. :grin:

2 Likes

I do wonder at this - people definitely don’t understand the difference, but I feel like they often intuitively pick the right one in normal conversation (not that I can justify that quantitively or anything). I think it just feels off to use the wrong one in a lot of common sentences because of colocations or something.

2 Likes

Yes - You were scored by far apart your two furthest shots landed.

As @Belthazar points out, the rest is just down to adjusting the sights.
That takes time and cadet’s range time was very limited

Starting quizzes of my sixth form students would suggest otherwise :confused:

6 Likes

I’ve found the exact opposite. lol

2 Likes

Phrases like “Be precise in your answer.” or “Can you be more precise?” are where I hear the term used most often in non-technical conversation. It tends to take on the same meaning as “accurate”. I suspect this is likely due to accuracy generally mattering more than precision in daily life.

However, I could come up with a folk explanation of how the two might have been conflated. When explaining an idea and trying convey the most accurate meaning, we often fumble around additional utterances to help us. So long as these utterances are within the scope of what we mean, we are being precise. But, the real hope in such a case is to be accurate. Thus, a link between the two.

Obviously, that’s a folk explanation, not a linguistic explanation. Nonetheless, it was a fun exercise in creativity. Personally, I’d like to know whether the history of their technical usage is derived from their meanings, or if the meanings were created so as to allow additional accuracy. :wink:

1 Like

I disagree - when someon asks you to be precise, they’re normally asking you to give more detail (precision) rather than be more correct (accuracy). E.g. if someone asks for a more precise time, they’re generally asking you to narrow the range of times, or if they ask precisely what happened, they’re generally asking you to narrow the range of possible events.

I think you might be mixing the two up in your usage here :thinking:

2 Likes

perhaps it makes more sense if we take the precise/accurate discussion away from target-shooting:

if i tell my gps to take me to a specific location, i have to be precise. there is no “right” answer for where i want to go. if i ask the gps to take me to main street, the gps might take me to any number of main streets, and do so accurately. but because i wasn’t precise enough and didn’t tell it which town’s main street i wanted to go to, i’ve not arrived where i want to be.

dunno, just thoughts passing through my mind ^^

1 Like

Not at all. But, that’s the whole point I was making, isn’t it.
To show how easily the meanings could be conflated.

Accuracy - The question of how close something is to the actual answer.
Precision - The question of how close a result (whether accurate or not) can be repeated within error.

(Perhaps I could have been more accurate in what I actually intended to convey.)


In this sense, sure, it fits the actual technical usage. But, that’s the only one I can think of where that’s a guaranteed fit.

This one could be either. Are they asking for an accurate account, or looking to assure that the detail remains the same over multiple accounts? That completely depends on context.


The conversation I was imagining was, e.g., Alice, trying to describe a new concept she has generated. The listener, e.g., Bob, has come up with his own conceptualization of what Alice means to convey. When Bob requests that Alice be more precise, he really means to assure that his central idea is near to Alice’s. (He’s looking to shift from Target 4 to Target 3.) That’s an attempt at Bob trying to gain better accuracy, not necessarily better precision. Of course, it does turn out that this comes from Alice being more precise. (She’s shifting from Target 3 to Target 1.)

  • Thus, from Bob’s point of view, he is asking for more accuracy. (Where’s the target, Alice?)
  • From Alice’s point of view, Bob is asking for more precision. (Here’s different ways to see it, Bob.)

It completely depends on point of view. Ultimately, from outside the conversation, it turns out that Bob is asking for both more accuracy and more precision. Though, he might never see it that way.

1 Like

if you only have one word to describe both the concepts they’re accurate by definition (they mean both of the concepts) but they’re imprecise (you can’t tell which one they mean).

adding the new word means you can be more precise, but if people can’t use them properly, then they become less accurate in the meaning they convey, right?

I don’t think precision in the statistical sense makes sense to apply to a normal conversation, so I tend to think of it more as how strictly/exactly you can describe something…

you’re mixing the question that’s being asked with the intent behind it (I feel like you kind of acknowlege this at the end when you basically say bob needs alice to be more precise).

bob can’t tell how far away his conceptualisation is from alices without increasing the precision, so he requests precision until he’s able to tell whether his idea is accurate or not. Whether his end goal is more accuracy of his internal idea is besides the point - his request is for more information. I think this is pretty analogous to repeating a measurement to reduce the size of your confidence intervals.

if what he wanted was actually more accuracy from alice, that would be implying that he thought alice was describing the wrong thing.

2 Likes

Pretty much, yeah.

Fair enough.

I will not deny that I mixed them, but that is also much of the point. I argue that this mixing could very well be what is happening in either Alice’s or Bob’s mind, or perhaps even an external observer’s. And, as such, might very well lead to the definitional conflation whose existence, or lack thereof, we’ve been investigating. Thus, Bob’s goal is integral to this, albeit, preposterous folk theory.


Given you’ve mentioned confidence intervals, I’m guessing you’re a statistician, either by trade or by hobby. Recognizing that, I understand your point of view, here, in that it’s about narrowing scope.

In regard to definitions, I tend to be a linguist by hobby. (I’m a computer scientist and logician by trade.) Thus, my analysis thereof usually involves viewing the dualistic purposes of language and utterance. More specifically, what is said versus what is meant or intended.

Two very different points of view, which can, as we have seen, lead to quite different evaluations. Both of which are perfectly valid within their context. It’s a fun little exercise to see them. Thanks! :wink:

2 Likes

Wow, I didn’t expect such an interest regarding my question :stuck_out_tongue: Thanks everyone for your comments!

After discussing this with a Japanese friend and checking Wikipedia, this is my conclusion (in increasing order of scientific-y):

  1. For everyday conversation, 精度 seems to be the most natural way to describe the concept of “precision” (according to my Japanese friend).
  2. If you want to take it one step further, you can distinguish “accuracy” and “precision” like the image I posted in the beginning. I.e. such that “accuracy” means “close to the target” and “precision” means “several times the same value”. In this case: “accuracy” = 正確度 and “precision” = 精密度・精度 (synonyms).
  3. If you want to be as precise as possible, you include a third parameter: “trueness”, such that “accuracy = precision + trueness”. I.e. now “trueness” takes on the role of “being close to the real value” and “accuracy” is rather used as an overall term. In this case, “accuracy” = 正確度, “precision” = 精密度・精度 (again synonyms) and “trueness” = 真度.
3 Likes