OT-Precison vs acccuracy
Close
Login to Your Account
Page 1 of 3 123 LastLast
Results 1 to 20 of 56
  1. #1
    Join Date
    Aug 2004
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    3,013
    Post Thanks / Like
    Likes (Given)
    259
    Likes (Received)
    1417

    Default OT-Precision vs acccuracy

    My wife and I went to buy a new thermometer for the porch yesterday and out of a box of 20 or so thermometers, none of them displayed the same temperature. She said that they could not be accurate because they all read differently. A very astute observation on her part. I told her that they could not be accurate because there was no precision. This started a lively discussion about accuracy versus precision. It ended with me trying to explain how calibration achieves accuracy only after precision is exhibited. Anybody else ever had this issue (not trying to explain something to my wife)?
    Last edited by crossthread; 01-15-2022 at 08:10 AM.

  2. #2
    Join Date
    Feb 2007
    Location
    Ontario, Canada
    Posts
    169
    Post Thanks / Like
    Likes (Given)
    357
    Likes (Received)
    32

    Default

    The best comparison I heard was this: The statement "Christmas falls during the last week of December" is accurate; vs. "Christmas falls on October 23 at 9:43 PM" is precise, but not accurate.

  3. #3
    Join Date
    Dec 2008
    Location
    Vt USA
    Posts
    9,940
    Post Thanks / Like
    Likes (Given)
    1826
    Likes (Received)
    3734

    Default

    well, it could be it's just a repeatability issue.


    ;-)

    How large is the group
    How close to the bull is the group center
    is the group in the same spot off the bull every time?

  4. #4
    Join Date
    Jun 2012
    Location
    Michigan
    Posts
    6,393
    Post Thanks / Like
    Likes (Given)
    5966
    Likes (Received)
    4089

    Default

    I would try to explain it this way to someone who isn't in our line of work:
    Precision - to make a part exactly to size one time.
    Accuracy - the ability to repeat that size on a part over and over and over.

  5. Likes crossthread, kustomizer liked this post
  6. #5
    Join Date
    Dec 2008
    Location
    Vt USA
    Posts
    9,940
    Post Thanks / Like
    Likes (Given)
    1826
    Likes (Received)
    3734

    Default

    Quote Originally Posted by Mtndew View Post
    I would try to explain it this way to someone who isn't in our line of work:
    Precision - to make a part exactly to size one time.
    Accuracy - the ability to repeat that size on a part over and over and over.

    how would you describe repeatability then?

    The stool needs three legs

  7. #6
    Join Date
    Mar 2005
    Country
    UNITED STATES
    State/Province
    Indiana
    Posts
    6,652
    Post Thanks / Like
    Likes (Given)
    2392
    Likes (Received)
    2641

    Default

    The target series of images is a pretty good visual aid for this one.

    Wikipedia has a decent description of this:

    In a set of measurements, accuracy is closeness of the measurements to a specific value, while precision is the closeness of the measurements to each other.

  8. #7
    Join Date
    Jun 2012
    Location
    Michigan
    Posts
    6,393
    Post Thanks / Like
    Likes (Given)
    5966
    Likes (Received)
    4089

    Default

    Quote Originally Posted by CalG View Post
    how would you describe repeatability then?

    The stool needs three legs
    That's on you, I didn't mention repeatability.

  9. #8
    Join Date
    Feb 2004
    Location
    peekskill, NY
    Posts
    27,825
    Post Thanks / Like
    Likes (Given)
    1
    Likes (Received)
    6609

    Default

    Old cryogenic engineer's maxim: "Man with one thermometer knows the temperature. Man with three thermometers has no idea what the temperature is."

  10. #9
    Join Date
    Oct 2007
    Country
    SPAIN
    Posts
    6,423
    Post Thanks / Like
    Likes (Given)
    2358
    Likes (Received)
    1264

    Default

    Quote Originally Posted by eKretz View Post
    The target series of images is a pretty good visual aid for this one.

    Wikipedia has a decent description of this:


    That's interesting....In the UK there's a lot of company names with the suffix of "Precision"....ie "Fred Bloggs Precision" or "Bloggs Precision Engineers".
    I always thought, they must have too(?), that Precision was hand in hand with Accuracy.
    Afterall, there's no point being highly precisely wrong?

  11. #10
    Join Date
    Dec 2008
    Location
    Vt USA
    Posts
    9,940
    Post Thanks / Like
    Likes (Given)
    1826
    Likes (Received)
    3734

    Default

    Quote Originally Posted by Mtndew View Post
    That's on you, I didn't mention repeatability.

    Well, you must expect pushback when you conflate the two qualities.

    I didn't do that.

  12. #11
    Join Date
    Jun 2012
    Location
    Michigan
    Posts
    6,393
    Post Thanks / Like
    Likes (Given)
    5966
    Likes (Received)
    4089

    Default

    Quote Originally Posted by CalG View Post
    Well, you must expect pushback when you conflate the two qualities.

    I didn't do that.
    No, but you're trying to make something out of nothing.

  13. #12
    Join Date
    Mar 2005
    Country
    UNITED STATES
    State/Province
    Indiana
    Posts
    6,652
    Post Thanks / Like
    Likes (Given)
    2392
    Likes (Received)
    2641

    Default

    Quote Originally Posted by barbter View Post
    That's interesting....In the UK there's a lot of company names with the suffix of "Precision"....ie "Fred Bloggs Precision" or "Bloggs Precision Engineers".
    I always thought, they must have too(?), that Precision was hand in hand with Accuracy.
    Afterall, there's no point being highly precisely wrong?
    Generally you need good accuracy and ​precision if you want to do good work.

  14. Likes mhajicek liked this post
  15. #13
    Join Date
    Oct 2007
    Country
    SPAIN
    Posts
    6,423
    Post Thanks / Like
    Likes (Given)
    2358
    Likes (Received)
    1264

    Default

    Quote Originally Posted by eKretz View Post
    Generally you need good accuracy and ​precision if you want to do good work.
    Yes - but I was under the understanding, that if you had accuracy, you automatically had precision.
    I guess I never knew the true meaning of the word!

  16. #14
    Join Date
    Sep 2004
    Location
    south SF Bay area, California
    Posts
    2,240
    Post Thanks / Like
    Likes (Given)
    245
    Likes (Received)
    809

    Default

    I've never been comfortable with the "conventional wisdom" thesis that high accuracy can co-exist with low precision in the absence of "dumb luck". Neither have I been comfortable with the idea that high precision can co-exist with low accuracy. That said, I do fully agree that the average of a series of measurements that are individually both accurate and precise is, very probably, a more accurate and precise quantification of the true value than are the individual measurements of the series.

    Having thusly trivialized and dismissed my iconoclastic mindset, I'll argue that precision and "repeatability" -- which is necessarily quantified as NON-repeatability -- are very nearly synonyms. At the same time, I'll point out that excellent "repeatability" of a measurement can whisper of coarse measuring-instrument graduations.

  17. #15
    Join Date
    Aug 2004
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    3,013
    Post Thanks / Like
    Likes (Given)
    259
    Likes (Received)
    1417

    Default

    I believe you are correct barbter. If you have accuracy then you must have precision and calibration is what makes accuracy out of precision.

  18. #16
    Join Date
    Oct 2007
    Country
    SPAIN
    Posts
    6,423
    Post Thanks / Like
    Likes (Given)
    2358
    Likes (Received)
    1264

    Default

    Quote Originally Posted by John Garner View Post

    Having thusly trivialized and dismissed my iconoclastic mindset, I'll argue that precision and "repeatability" -- which is necessarily quantified as NON-repeatability -- are very nearly synonyms. At the same time, I'll point out that excellent "repeatability" of a measurement can whisper of coarse measuring-instrument graduations.
    That's my thinking TBH....The "Low Accuracy High Precision" is effectively only really showing repeatability (and inaccurate repeatability).

    Quote Originally Posted by crossthread View Post
    I believe you are correct barbter. If you have accuracy then you must have precision and calibration is what makes accuracy out of precision.
    Very well put.

  19. #17
    Join Date
    Oct 2015
    Country
    UNITED STATES
    State/Province
    California
    Posts
    20
    Post Thanks / Like
    Likes (Given)
    2
    Likes (Received)
    9

    Default

    I dislike the word "precision" because it is used so loosely. I was raised using 3 words: repeatability, resolution and accuracy. I made a chart similar to eKretz's in an attempt to teach high school students. The context was a discussion of the Phalanx anti-missile gun on the USS Stark, and its failure against the Iraqi exocet missile, circa 1987 (so many things to learn: Saddam was our friend, France makes impressive weapons, stuff doesn't work if it isn't turned on). Here are some of my slides. Summary is: precision = resolution.
    screen-shot-2022-01-14-11.13.42-am.jpg

  20. #18
    Join Date
    Jan 2007
    Location
    wales.uk
    Posts
    2,014
    Post Thanks / Like
    Likes (Given)
    467
    Likes (Received)
    518

    Default

    So your saying everything I make is precisely wrong and accurately correct ( the inspector said that not me)
    Mark

  21. Likes barbter liked this post
  22. #19
    Join Date
    Aug 2004
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    3,013
    Post Thanks / Like
    Likes (Given)
    259
    Likes (Received)
    1417

    Default

    When I think of "resolution" in this context, it means not enough data points to determine precision.

  23. Likes barbter liked this post
  24. #20
    Join Date
    Oct 2007
    Country
    SPAIN
    Posts
    6,423
    Post Thanks / Like
    Likes (Given)
    2358
    Likes (Received)
    1264

    Default

    Quote Originally Posted by crossthread View Post
    When I think of "resolution" in this context, it means not enough data points to determine precision.
    Meetoo.
    Because that's what half stepping was then invented for. And then microstepping.
    Wasn't it?


Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •