Jump to content

Talk:Measurement uncertainty

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Page overhaul

[edit]

The new Wikipedia entry on Measurement uncertainty has been prepared to reflect, and to be consistent with, the work of Working Group 1 Measurement uncertainty of the Joint Committee for Guides in Metrology (JCGM). The working group has international responsibility for maintaining and promoting the Guide to the expression of uncertainty in measurement (GUM), which is the primary document regarding measurement uncertainty. The working group has had no formal involvement in the preparation of the entry, which replaces the previous entry with the agreement of the main author of that entry. --Az628 (talk) 16:56, 28 March 2010 (UTC)[reply]

This entry is needlessly long winding and overemphasizes the role of the GUM. It is suggested that JCGM creates its own page and puts a link on this page to that page, instead of explain its current stage, work plan, structure, etc. There are many basic textbooks on statistics, probability theory, error analysis, etc. that are more accurate, detailed and informative, in a more concise manner, to which reference could be made. Some terminology seems to deliberately avoid standard statistics terminology, e.g., propagations of distributions instead of the well-known term variate transformation; expectation instead of mean value. 86.169.8.171 (talk) 01:03, 30 September 2013 (UTC)[reply]

As of November 2014 I feel that the article is quite good. It is concise and clear especially for those who are not experts in uncertainty. If there is an underlying controversy with respect to GUM it should be noted, but for those who have to implement such work in a practical situation, this article is a very good start. Rjo min (talk) 23:10, 29 November 2014 (UTC)[reply]

The example for parenthetic notation is incorrect according to IUPAC

[edit]

The page suggests that "6.67428(67) ... stands for (6.674 28 ± 0.000 67) ..." but the IUPAC guideline (page 83 of the "green book" available as a PDF from http://old.iupac.org/publications/books/gbook/index.html) says that the +- notation should be 2*sigma and the parenthetic notation should be 1*sigma. The page cites the IUPAC as an authority, so I think the example should be changed, but I am reluctant to do that, without discussion. For more on this, please see the example in the green book on page 83, section 4.1, about 1/4 of the way down the page. --Dankelley (talk) 18:20, 16 July 2009 (UTC)[reply]

Missing Kline & McClintock

[edit]

The Kline & McClintock method is a useful tool for uncertainty analysis. Should be covered. Kline & McClintock is the standard for quantifying measurement uncertainty taught in US engineering curricula. Tjcognata (talk) 19:52, 30 March 2008 (UTC)[reply]

Does it differ from the GUM method? The GUM is an ISO standard and as such also applicable in the United States. --DrTorstenHenning (talk) 07:41, 31 March 2008 (UTC)[reply]
Of that I'm not certain, I have not till here been introduced to the GUM method. But, then, as for standards, the unfortunate thing is that there are so many. Tjcognata (talk) 04:49, 5 April 2008 (UTC)[reply]
Yes. From what I gleaned of this article GUM is applied strictly to measurements to evaluate the "true" value and its uncertainty. K&M is a constant odds method for predicting the uncertainty of values which are a function of uncertain measurements. While not strictly a measurement uncertainty, it is sometimes referred to as such since the value is often what an experimenter desires to "measure". Perhaps not appropriate as a part of this article, my apologies. Tjcognata (talk) 05:47, 5 April 2008 (UTC)[reply]
As far as I know, the GUM does not speak about "true" values. I do not see how K&McC differs from the propagation of measurement uncertainties according to the GUM, would be interesting to have some input from someone who knows both the valid standard and what is taught at engineering schools (which, at least in my experience, is all too often different from each other). --DrTorstenHenning (talk) 15:52, 5 April 2008 (UTC)[reply]
Propogation of uncertainty is precisely the function of K&M - If you think it can add to the discussion, I'll set aside time next weekend or that following to bring a discussion of K&M into the article. That should invite further input from anyone familiar with the merits of both GUM & K&M. Tjcognata (talk) 19:39, 6 April 2008 (UTC)[reply]
Found a very useful reference which I'll have to leaf through a bit more thoroughly later. Apparently Kline and McClintock pioneered the modern concept of uncertainty propagation, which is likely why professors refer to the method by their names. ISO's GUM and ANSI's "Test Uncertainty" standards, which apparently are identical save in terminology, have their roots in and may be largely based upon the work of K&M in their consideration of the propagation of uncertainty. Tjcognata (talk) 02:48, 7 April 2008 (UTC)[reply]

Page needs an overhaul

[edit]

Looks like this page needs at least a section (if not a major overhaul) written less academically--more toward the practical. The theoretical text seems content rich, but still could use a bit of tweaking. —EncMstr 07:15, 26 January 2006 (UTC)[reply]

Definitely needs an overhaul, for a start the first sentence is very misleading, worst case uncertainties aren't often quoted in practice. A level confidence of say 95% or 99% would be more normal. JMiall 13:09, 13 July 2006 (UTC)[reply]

I've added a bunch of links and conformed to some of the style conventions. The theoretical text is NOT content-rich at all. It is seriously deficient. It's very vague. That's the major weakness of this article. Michael Hardy 14:57, 13 July 2006 (UTC)[reply]

This page is rubish

[edit]

This page is rubish and needs a complet overhal Mat.roesslein 14:49, 14 June 2007 (UTC)[reply]

I disagree. The issue of random and systematic errors is important and this article highlights the contoversial nature of GUM.Petergans (talk) 10:16, 8 January 2008 (UTC)[reply]

Was this a spelling mistake?

[edit]

I changed respectice to respective. This was the sentence: As is known, weights cause two effects: firstly, they shift the numerical values of the estimators, and, secondly, they reduce the respectice uncertainties. Is respectice some technical, rare, word that is spelled correctly? Saros136 (talk) 10:22, 21 January 2008 (UTC)[reply]

You were right in correcting the spelling. More interesting, however, is the question why you undid my rearrangement of the external links. A certain individual who has a personal grudge against the GUM keeps flooding the external links section with links to his website and specific documents on his website and to his book. I think this should be made transparent, possibly as one of the first steps on the road to to cleanup. Do you disagree? --DrTorstenHenning (talk) 12:22, 21 January 2008 (UTC)[reply]
Thanks for the confirmation. I didn't do anything to the links on purpose. I'm sorry about that.
I don't know what has been going on with the article, I haven't paid attention to it. And I didn't look at any links section. Reading it last night, I did notice sort of an argumentative tone, so your comments don't surprise me. And no, I don't disagree.
On the plus side though, it's good to mention any controversy, and I was glad that it covered random vs.systematic mistakes. It's unrealistic to treat errors as random. Saros136 (talk) 20:13, 21 January 2008 (UTC)[reply]
Every single paper of site liked to "Alternative approach" is the work of Dr. Michael Grabe! Nor do I think that his approach is so very different from GUM – he advocates treating type II uncertainty seperately from type I uncertainty during the measurement and calculation stage (before production of the final result), and you won't find anything different in GUM! Physchim62 (talk) 11:01, 5 October 2008 (UTC)[reply]
[edit]

Hello fellow Wikipedians,

I have just modified 2 external links on Measurement uncertainty. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 21:54, 23 January 2018 (UTC)[reply]

[edit]

I am researching on "errors and mistakes" from a multi-disciplinary perspective. Least count and experimental error are strongly related with the precision or resolution of the instruments and, after computations, with error propagation, in several "Experimental Physics" works. Here, in Wikipedia, experimental error is redirected to observational error. The latter states that it is different from measurement uncertainty. And, this article does not mention least count even once.

Is measurement uncertainty related to "least count"? Or are both, experimental/observational error AND measurement uncertainty related to it? George Rodney Maruri Game (talk) 21:11, 13 April 2023 (UTC)[reply]

The concept you refer to is essentially the resolution of a (counting) instrument. That could be an individual source of uncertainty associated with an instriment reading, in teh same way that scale resolution usually would be. But it is very unlikely that either would be large contributions to MU; if they were, you'd probably be using the wrong instrument for the job. SLR Ellison (talk) 09:44, 27 June 2024 (UTC)[reply]

'Further reading' and 'external links' seem to overlap

[edit]

Several of the 'external links' are to documents or pages that fall into the 'further reading' category. Should all the links to documents and the like be moved to 'further reading', or is there reason to keep them as they are? SLR Ellison (talk) 09:46, 27 June 2024 (UTC)[reply]