Talk:Reliability (statistics)/Archives/2014


Untitled

To me (as a non-specialist)there is an apparent contradiction between the graphics and the text.The target (top right)shows "unreliable but valid", while the text opposite states:"a test that is unreliable cannot possibly be valid" (writing from memory here, as I find I cannot go back to the page without losing what I have written). Correction or clarification needed here! Frankjdavidson (talk) 09:40, 11 March 2013 (UTC)

You may find it helpful while reading or editing articles to look at a bibliography of Intelligence Citations, posted for the use of all Wikipedians who have occasion to edit articles on human intelligence and related issues. I happen to have circulating access to a huge academic research library at a university with an active research program in these issues (and to another library that is one of the ten largest public library systems in the United States) and have been researching these issues since 1989. You are welcome to use these citations for your own research. You can help other Wikipedians by suggesting new sources through comments on that page. It will be extremely helpful for articles on human intelligence to edit them according to the Wikipedia standards for reliable sources for medicine-related articles, as it is important to get these issues as well verified as possible. -- WeijiBaikeBianji (talk, how I edit) 02:09, 2 September 2013 (UTC)

non-coherent statements

the article says: "a test that is not reliable cannot possibly be valid," although the picture to the right shows an "unreliable, but valid" example. Twipley (talk) 16:14, 24 September 2013 (UTC)

Statistics vs. psychometrics

I think the link with psychometrics should be made clearer in the article, perhaps by changing the title to reliability (psychometrics) and beginning the introduction with “In psychometrics, reliability is […]” In fact, the article itself develops ideas, techniques and models originating in psychometrics but the word does not even appear before the categories. Obviously, there are some links between these ideas but it seems to me that statisticians who don't have a background or special interest in psychology are more likely to talk about random error, measurement error, uncertainty or perhaps efficiency than about reliability per se. In fact, I suspect that applied statisticians working in other fields might even find the validity/reliability terminology confusing, so I think it's important to set the context. Also in engineering, reliability apparently refers specifically to consistency over time whereas the notion of reliability developed in this article also applies to consistency across tests or items administered concurrently. What do you think? GL (talk) 08:23, 22 August 2011 (UTC) 

I would suggest to revert this. Reliability (consistency) is also used in other fields, for example in medical and marketing research, with the same meaning. This page is now inconsistent with Validity_(statistics). Reliability in the context of engineering does use statistics but it is not a branch of statistics. It is a branch of engineering . The analogue to this article in different contexts are the Accuracy_and_precision, Measurement_uncertainty pages. SiggyF (talk) 21:16, 29 October 2012 (UTC)

I have compared the use of the word reliability in the context of engineering and statistics. If I search through publications using the words reliability + statistics about 90% relates to reliability in the context of "consistency", mostly medical, health and psychological. If I search through publications using the words reliability + engineering I find the reliability used in the context that Pkgx is referring to, in reliability in the context of "non-failure". Here the term is often used as "reliability engineering" more often than just reliability. If nobody else objects I'll change it back to Reliability (statistics). — Preceding unsigned comment added by SiggyF (talkcontribs) 08:37, 8 August 2014 (UTC)