5
$\begingroup$

The tag is meant for reliability of measurements, but today Calculating Failure Rate with Some Failure Cause Data being Vague (and certainly before) somebody used it in the meaning of "reliability theory", as used in engineering and other fields, and covered by books like System Reliability Theory Statistical Applications.

What should be done here, put a different tag, or are there existing tags for posts like the one linked?

 EDIT 

After a day, it seems the question was interpreted very narrowly. I meant not just that specific post, but the tag name invites misinterpretation. We could change the name, maybe measurement-reliability or something similar, and then maybe later, for the technological use, systems-reliability ?

$\endgroup$
3
$\begingroup$

I edited the question to change the tag. The issue is somewhat ambiguous, but the question asks about estimating a proportion, so I put the tag on it. I think this is the first thing to do when a tag is inappropriately applied to a question.

Relative to this specific tag-inappropriate usage, yhe tag excerpt has usage guidance that is clear enough, IMO. I suspect people just don't read the excerpts. If someone adds the tag incorrectly anyway, I would probably just edit. If you want, you can also leave a comment.

$\endgroup$
  • $\begingroup$ OK, but still there is the question how to signal to askers about "reliability theory" that the reliability is the wrong one ... $\endgroup$ – kjetil b halvorsen Jun 24 at 20:34
  • $\begingroup$ The tag excerpt has usage guidance that is clear enough, IMO. I suspect people just don't read the excerpts. When someone has placed the tag incorrectly anyway, I usually just edit. If you want, you can also leave a comment. $\endgroup$ – gung Jun 24 at 20:41
2
$\begingroup$

So I have two responses to the main post and one follow up comment that has not been brought up yet. My opinion is that if we want to distinguish between the different kinds of reliability, then yes, we should have separate tags.

1) Reliability Theory actually has two different definitions (general statistical reliability and educational reliability). The latter reliability is more biological in that it is specific to psychology, teaching, education, and a few biological fields: https://chfasoa.uni.edu/reliabilityandvalidity.htm I am not sure how the tagging process works, but it probably necessary to have more than one tag we can distinguish the different definitions of reliability. After posting this I realized that there is more than one tag.

2) Inter-rater reliability is a type of statistical reliability and a part of reliability theory. I personally think all inter-rater reliability questions are also reliability questions, but hey I don't know everything. So at least for me, reliability indeed leads to some misinterpretation if we are trying to keep questions about reliability and inter-rater reliability separate.

A heads up about reliability:

3) There can be a cross-over between educational reliability and statistical reliability. Here is a very specific recent research example involving Bayesian Statistical reliability: https://arxiv.org/abs/1404.3656 . I know no other specific examples; however, so the cross-over would be very narrow on StackExchange to questions about applying better math to grading in education. I would imagine this means that there should be some kind of reference or link in the question if the two are mixed. This is of a heads up statement for the mods and administrators rather than something actionable, as I can find little examples of this happening in the stack-exchange (but there could be deleted posts I don't see).

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .