I strongly agree with everyone. There's a big difference between the two terms, and what we're talking about is precision! It's disappointing that an ISO committee would be confused. I'm sure no more explanation is needed, but I was briefly a high-school math
teacher, and I can't resist adding some.
(1) Attached is a photo I've used to help students understand the concept of significant figures; it shows the label on the package of a hammer I own.. Of course 226.72 grams is much more precise than 8 ounces, but I doubt it's any more acccurate!
(2) One of my favorite examples of the difference between accuracy and precision is: The mathematical constant pi to 10 decimal places is 3.14159 26536, but, as you know, it is often approximated by 3-1/7. Rounded to 10 decimal places, 3-1/7 is 3.14285
71429, which is much more precise than 3.14159, with only 5 decimal places. But it is considerably less accurate, i.e., it’s not nearly as close to the actual value as 3.14159 is.
On Feb 5, 2016, at 4:34 PM, John Hostage <[log in to unmask]
From: Discussion of the Developing Date/Time Standards [[log in to unmask]] on behalf of Nathan Harrenstein [[log in to unmask]]
Sent: Friday, February 05, 2016 16:20
To: [log in to unmask]
Subject: Re: Accuracy vs. Precision
I think precision is the correct term here. According to the Oxford dictionary, precision is the "refinement in a measurement, calculation, or specification, especially as represented by the number of digits given". In this case, we are refining the number
of possible dates as we add components (Y -> YM -> YMD).
In addition if we have an event which occurred on 1965-07-07, we can also say that the event occurred in 1965-07 or 1965 and we would not be inaccurate, only less precise.
On Fri, Feb 5, 2016 at 11:11 AM, Denenberg, Ray <[log in to unmask]> wrote:
There is a technical matter I need to raise, but first, an update.
The work on a Committee Draft of ISO 8601 Part 2 is progressing nicely and it might be available for comment within several weeks (although with ISO, you never know what sort of administrative matter might hold things up). The plan is that 8601 Part 2 will
include all of EDTF levels 1 and 2 functionality (all functionality retained; some changes in syntax) at least for the Committee Draft; it is impossible to predict what will survive the balloting process. The main body will not mention or refer to EDTF.
It will however define the two levels (1 and 2) and indicate that these levels are defined for reference by profiles. There will be an Annex (informative) for “Community Profiles of ISO 8601”. The Annex will be two parts: the first will describe profiles
in general. The second will be the EDTF profile, which will define the three levels (0, 1, and 2), explicitly listing the level 0 features, and calling out the main body to describe levels 1 and 2. Of course EDTF will be the only profile, initially, but there
will be procedures for adding other profiles.
The technical matter: ISO wants the term “precision” replaced by “accuracy”. I have objected to this change and I would like to hear your views.
The request for this change is based on http://www.astro.lu.se/Education/utb/ASTM11/lecture2.pdf which I don’t think justifies it at all.
Following is part of my objection which I submitted:
I'm sorry but I disagree with the change from precision to accuracy.
……. I have researched this a bit further and I conclude that precision is correct. For example see
“Accuracy refers to the closeness of a measured value to a standard or known value. For example, if in lab you obtain a weight measurement of 3.2 kg for a given substance, but the actual or known weight is 10 kg, then your measurement is not accurate. In this
case, your measurement is not close to the known value.
“Precision refers to the closeness of two or more measurements to each other. Using the example above, if you weigh a given substance five times, and get 3.2 kg each time, then your measurement is very precise. Precision is independent of accuracy. You can
be very precise but inaccurate, as described above. You can also be accurate but imprecise.”
So let’s say the “actual” date is 1965-07-07. Then an estimated date of 1965-07-06 is more accurate than 1965-07-05. If we estimate that date with “month precision” we get 1965-07, and we will get 1965-07 whether the actual date is 1965-07-07, 1965-07-06,
1965-07-05 or any other date in that month. So, for any date in that month, 1965-07, in terms of decade precision is precise. “Month” is then a measure of precision, not accuracy.
Anyway I welcome opinions on this, it is quite possible that my reasoning is faulty. If nobody responds I will assume nobody cares, and I will drop the objection.
Woodrow Wilson Indiana Teaching Fellow
Adjunct Associate Professor of Informatics
Visiting Scientist, Research Technologies
Indiana University Bloomington