14
$\begingroup$

We already have a thread for proposing candidates for tag synonyms, but there doesn't seem to be a similar thread for proposing tags for deletion. Similarly to the "synonym" thread, let's use this one to post proposals for tags deletions in answers, so to vote and discuss about them below the answers (rather than by opening new threads).

EDIT

Now I am a moderator and I just got informed that moderator can't remove a tag, it can be done only by removing it from all the questions associated with it.

$\endgroup$
14
  • 5
    $\begingroup$ I welcome this thread. The correct SE-term for tag deletion is "burnination". Mods do not have power to burninate a tag (i.e. to automatically remove it from all threads without pushing them to the front page), so the protocol for burnination is that mods should let SE admins know and the admins do it. As far as I know, no tag has ever been burninated this way on CrossValidated. $\endgroup$
    – amoeba
    Commented Mar 25, 2016 at 15:25
  • 4
    $\begingroup$ One more thing: before burninating a tag, one should better make sure that there are no threads where this is the only tag. If there are, then those should be edited to add some more relevant tags. I'd suggest that whoever proposes a tag deletion should take the responsibility of doing that first. After checking that this tag is never alone, please mention it in the corresponding answer. $\endgroup$
    – amoeba
    Commented Mar 25, 2016 at 15:34
  • 1
    $\begingroup$ Related meta meta thread: What should be our protocol for tag burnination? $\endgroup$
    – amoeba
    Commented Aug 2, 2016 at 11:33
  • $\begingroup$ meta.stats.stackexchange.com/questions/2612 $\endgroup$
    – amoeba
    Commented Feb 9, 2017 at 23:39
  • $\begingroup$ Note for myself: write an answer explaining why [methodology] should be burninated. Also, [analysis] in the link above. $\endgroup$
    – amoeba
    Commented Apr 25, 2017 at 22:12
  • 1
    $\begingroup$ I sometimes see new tags that are obvious typos for existing tags, I just nuke them immeadiately. $\endgroup$
    – kjetil b halvorsen Mod
    Commented Jun 14, 2017 at 12:29
  • $\begingroup$ Congratulations with the diamond, @Tim! I've been away during the final stages of the election and saw the results only today. I'm wondering if you have plans to get some tag burnination going. $\endgroup$
    – amoeba
    Commented Sep 2, 2017 at 21:52
  • $\begingroup$ @amoeba thanks! First I'd need to learn how exactly mod can burinate a tag and then I'd do something with it. $\endgroup$
    – Tim Mod
    Commented Sep 3, 2017 at 9:47
  • $\begingroup$ @amoeba FYI: I just got informed that a moderator can't delete a tag, it can be done only by removing it from all the questions associated with it. $\endgroup$
    – Tim Mod
    Commented Sep 3, 2017 at 11:26
  • 2
    $\begingroup$ I am aware that mods cannot delete tags themselves; however, they can request SE admin team to delete a tag (this is called "burnination") and it is routinely done on other SE sites. So your information is not complete: there is a way that a mod (e.g. you) can make a tag be deleted. See meta.stackoverflow.com/tags/burninate-request/info and links there. $\endgroup$
    – amoeba
    Commented Sep 3, 2017 at 14:31
  • $\begingroup$ @amoeba I went through the tags, in most cases it is not so many questions tagged by them, so we could probably deal with them by un-tagging the questions. I'm not sure if it is worth to ask CMs for this. $\endgroup$
    – Tim Mod
    Commented Sep 4, 2017 at 13:32
  • 2
    $\begingroup$ That is fair enough, but e.g. [basic-concepts] has 200+ threads (see stats.meta.stackexchange.com/questions/2854) and manually untagging is not very attractive. $\endgroup$
    – amoeba
    Commented Sep 4, 2017 at 15:14
  • $\begingroup$ @Tim, shall I delete my answer below, as the tag is now gone? (That is how I've handled the related thread on tag synonyms.) $\endgroup$
    – gung - Reinstate Monica Mod
    Commented Sep 10, 2017 at 1:46
  • $\begingroup$ @gung There are eight implemented answers already deleted here so I assumed we are following the same principle as in the synonym thread. $\endgroup$
    – amoeba
    Commented Sep 11, 2017 at 7:49

4 Answers 4

Reset to default
4
$\begingroup$

Turns out that we have a tag. Its wiki excerpt says

For questions about statistical theory. Always include a more specific tag as well.

This means that it is a meta tag by definition. And so it should go. It does not seem to add any value, is used inconsistently, and is only causing confusion.

$\endgroup$
2
  • $\begingroup$ If I recall correctly, there was discussion on Meta that this tag is actually useful (but I can't find it now). See also here for pourpose of creating it: meta.stats.stackexchange.com/questions/441/… $\endgroup$
    – Tim
    Commented Aug 2, 2016 at 11:25
  • 1
    $\begingroup$ Well, that comment is from 2010 and does not seem very relevant to me anymore. Take a moment to look at the threads tagged with [theory] and you will see that this tag can be removed from any of them without any loss; there is always a better more specific tag(s). $\endgroup$
    – amoeba
    Commented Aug 2, 2016 at 11:27
3
$\begingroup$

with 22 Qs and no wiki excerpt should be eliminated, partially by being replaced with (145), partially by simply being removed.

Currently people use it for questions like "How is independent T-Test valid for unequal samples?" or "Is the following inequality correct? How to prove it?" This does not make sense.

$\endgroup$
1
$\begingroup$

We have a tag, with 26 threads, and no excerpt or wiki. Some of the questions are clearly about the intra-class correlation, but others may be about other topics. The current threads could be retagged, and [repeatability] can be deleted automatically after 24 hours.

$\endgroup$
0
0
$\begingroup$

As discussed in this thread, I think (though popular, with 1820 questions) is not very useful. There's no usage guidance and the tag description is quite vague:

A statistical model is a formalization of relationships between variables in the form of mathematical equations. A statistical model describes how one or more random variables are related to one or more random variables. The model is statistical as the variables are not deterministically but stochastically related

$\endgroup$
6
  • $\begingroup$ As said by @amoeba and myself above, there's no way do delete the tag other then manually removing then from all the 1800 questions. Are you willing to manually remove those tags from 10 questions a day, every day for half a year? $\endgroup$
    – Tim Mod
    Commented Jun 26, 2019 at 15:59
  • $\begingroup$ @Tim Not proposing that we do - I was hoping for something like burnination, which amoeba discusses and recommends in his comment here: stats.meta.stackexchange.com/questions/3031/…. No problem if that's not feasible, but that is far from clear from these threads on the topic. $\endgroup$
    – mkt
    Commented Jun 26, 2019 at 16:40
  • $\begingroup$ @Tim It is not correct that there is no other way; I never said so. $\endgroup$
    – amoeba
    Commented Jun 26, 2019 at 18:56
  • $\begingroup$ @amoeba agree, I oversimplified it, sorry. The other way is way is contacting community moderators that would remove the tag. However CMs have many other, more important things to do then removing one of possibly thousands of tags that are not really needed in SE network of sites. There's really no harm in having such tags, so if we can't do this by ourselves, we should just accept tags like this. $\endgroup$
    – Tim Mod
    Commented Jun 26, 2019 at 19:14
  • 1
    $\begingroup$ @Tim That's fine, but I am wondering what is the status of this opinion. Is it a public SE guideline? A recommendation that CMs privately communicated to the CV mods (e.g. you)? A common understanding between the CV mods based on their personal impressions? Your own personal opinion? :-) I would be interested to know. Thanks. $\endgroup$
    – amoeba
    Commented Jun 26, 2019 at 19:27
  • $\begingroup$ @amoeba my own impression that considers (small) number of CMs, (large) number of SE sites, (exponentially large) number of tags, and observations about their usual response time. $\endgroup$
    – Tim Mod
    Commented Jun 26, 2019 at 19:30

You must log in to answer this question.

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