Talk:Gold standard (test)

somewhat pedantically
I think there is a POV in saying that the "AMA somewhat pedantically prefers the phrase" --Lux 04:41, 8 May 2006 (UTC)

Contradiction
The start of the article says that a gold standard test is 100% accurate. The last paragraph tells of imperfect yet allegedly gold standard tests. Where does the truth lie? 69.157.111.210 23:50, 24 May 2006 (UTC)

To me it seems there is no contradiction. The second paragraph tells of what a gold standard should aspire to be, and states that in clinical medicine there are no such ideal gold standards; the imperfect (but still very good) tests are referred to as gold standards. --81.132.1.136 20:13, 20 July 2006 (UTC)DB

I think it should be in a different section because it is indeed confusing, especially with the "paradox of the gold standard" explained after. If these tests are sometimes used as bona fide gold standards then the text should mention it. I am guessing the tests mentioned are easier and cheaper to perform than the real gold standard test, whichever it may be (autopsy? surgery?). 132.207.28.121 (talk) 18:04, 9 June 2010 (UTC)

Gold Standard in Machine Learning
The notion of a gold standard is also used in the area of machine learning and especially applications in computational linguistics. However it does refer to a quite similar concept there, so it might not be a good idea to create a new article. In machine learning classification problems (seperate a set of data in classes in an automatic manner) the term "gold standard" refers to a manually classified set of data which is then used to validate the performance of classification algorithms. Any ideas how to bring this in?
 * Isn't it called ground truth in machine learning? Qorilla (talk) 16:27, 6 May 2015 (UTC)


 * Yes, it is indeed called ground truth in machine learning. I've added a note to that effect.  It's interesting to note that there are so many different terms for statistical concepts (between data science, machine learning, information retrieval, mathematical statistics, signal detection theory, etc).  These fields have so much in common, yet we're just now beginning to even note the similarities (let alone consolidate terms and concepts).  "Gold standard" is not a term used in ML or IR (in fact, is it used anywhere outside medicine and medicine-related statistics?).  That said, it might be more appropriate; in other fields, the term "ground truth" refers to an authoritative state of information derived from direct observation, whereas the ML/IR definition is essentially that of the gold standard.  I've spent quite a bit of time trying to streamline the ground truth article and I've been trying to avoid touching this article as well.  So much for that! Adam KatzΔ☎  19:45, 8 January 2016 (UTC)

Gold standard in medical research
Edited the first paragraph to include the gold standard for medical research (e.g., survival to hospital discharge for resuscitation research, decrease in 30 day mortality for other types of medical research). Accidentally clicked 'minor' edit. MoodyGroove 21:54, 31 January 2007 (UTC)MoodyGroove

This needs to be cleaned up
I don;t have the expertise, but this entry needs to be brought up to speed with other epidemiology sections.

Should be linked to the sections of test performance, sensitivity, specificity, positive and negative predictive values. —Preceding unsigned comment added by 204.99.163.8 (talk) 21:14, 18 June 2010 (UTC)

Reference 3
I clicked on reference 3 and since it's not in English (Swedish maybe?) I wasn't able to read what it says. There are some cute pictures of hedgehogs in the first few pages though. Googling "paradox of the gold standard" points back to this page and a few that refer to it. Is this an actual term? —Preceding unsigned comment added by Wolfhound668 (talk • contribs) 18:32, 30 August 2010 (UTC)

I have removed this :

" This term was first coined by Dr. Riley Kennedy. "

It is probably vandalism. --Adam majewski (talk) 14:59, 18 January 2011 (UTC)

Move?
Move to criterion standard as that is what medical journals call it? Biosthmors (talk) 21:15, 16 April 2012 (UTC)