12
$\begingroup$

(402) is an R package that has (498), (168), and (10) functions.

(119) is an R package that has (196) as a function.

Both packages fit [generalized] linear (and nonlinear) mixed models. Many questions deal with both packages.

Is it useful to have all these separate tags?

The problem is that it seems that, e.g., [lme4] and [lmer] are often used almost interchangeably. And in general, such a deep tag nesting is a bit annoying; e.g., every Q with [lmer] could/should also be tagged with [lme4]+[mixed-models]+[r], i.e., this automatically eats up 4 out of 5 possible tags.

So one could consider mapping all functions to the respective packages. This would end up with [lme4] with 906 Qs and [nlme] with 294 Qs. Or perhaps even map everything together to [nlme-lme4] (or something like that) with 1138 Qs.

Or is it a bad idea?

$\endgroup$
  • 2
    $\begingroup$ Seems like a great idea to me ! $\endgroup$ – Robert Long Aug 29 '16 at 11:33
  • $\begingroup$ @RobertLong Would you prefer merges into [nlme] and [lme4] or a full merge into [nlme-lme4]? $\endgroup$ – amoeba says Reinstate Monica Aug 29 '16 at 11:40
  • $\begingroup$ I think the important, broader, question here is how we should deal w/ nested tags. Should they all be used when appropriate, or only the deepest level? Etc. $\endgroup$ – gung - Reinstate Monica Aug 29 '16 at 11:56
  • 5
    $\begingroup$ I'd prefer the full merge because all too often questions where the OP used nmle are not actually specific to nlme at all. $\endgroup$ – Robert Long Aug 29 '16 at 13:09
  • $\begingroup$ I wonder whether the first term in the new mega-tag should start with lme as I think that is possibly what more people may start typing. $\endgroup$ – mdewey Aug 29 '16 at 15:02
  • 1
    $\begingroup$ @mdewey So [lme4-nlme]? Or [lme4-etc]? My personal favourite is [lme4-and-all-that]. $\endgroup$ – amoeba says Reinstate Monica Aug 29 '16 at 15:06
  • $\begingroup$ I rather like the and-all-that variant but it might be thought too flippant for serious people like statisticians. No real preference otherwise. $\endgroup$ – mdewey Aug 29 '16 at 15:09
  • $\begingroup$ @mdewey In fact, whether the new tag starts with lme or has lme in the middle should not really matter because tag suggestions appear based on full text search across tag names (try it). That said, lme4 is a more common package, so it can have the priority. $\endgroup$ – amoeba says Reinstate Monica Aug 30 '16 at 0:53
  • 3
    $\begingroup$ I'd vote for [lme4-nlme] since the rest of "all that" is much less used than the two main packages. lme4 is nowadays more popular, but nlme is still popular. Moreover, "all that" can be ambiguous and I imagine that many people could consider very different things as "all that" (e.g. anything related to mixed-effect models). $\endgroup$ – Tim Aug 30 '16 at 12:24
  • $\begingroup$ +1. Some years ago one of these tags intrigued me a bit. meta.stats.stackexchange.com/questions/1689/… $\endgroup$ – Andre Silva Aug 30 '16 at 15:51
  • $\begingroup$ Possible duplicate of Redundant tags: mixed effects and related models $\endgroup$ – smillig Sep 6 '16 at 11:57
  • 1
    $\begingroup$ @smillig This Q is definitely not a duplicate of that one. This one raised a new (albeit related) specific issue, it was discussed, and the solution has already been implemented. $\endgroup$ – amoeba says Reinstate Monica Sep 6 '16 at 12:40
11
$\begingroup$

Yes, mapping the function to the packages is a good idea. Maybe we should make all of them as synonyms to [lme4-nlme] ? This is the simplest choice and anyone trying to use one of the listed tags would not have problems with finding appropriate one since they all will be synonyms to the more general tag.

Moreover, those functions are used in quite similar contexts, packages have similar authors etc.

EDIT: the tag was created and everything seems to work. I just added wiki to it.

$\endgroup$
  • $\begingroup$ +1 for merging everything together, see also @RobertLong's comments (supporting this) above. He has been contributing quite a bit to these tags. $\endgroup$ – amoeba says Reinstate Monica Aug 29 '16 at 13:11
  • $\begingroup$ By the way, it would be good to have @BenBolker's opinion (who is by far the most prolific answerer in [mixed-model]), but it seems he never visits Meta. $\endgroup$ – amoeba says Reinstate Monica Aug 29 '16 at 13:43
  • 2
    $\begingroup$ I agree that this is reasonable. $\endgroup$ – Ben Bolker Aug 30 '16 at 18:02
  • 1
    $\begingroup$ Tim, do you want to edit your answer to change [nlme-lme4] to [lme4-nlme] as per your comment to the OP? $\endgroup$ – amoeba says Reinstate Monica Aug 30 '16 at 19:47
  • $\begingroup$ @amoeba done, it seems I treat them so much as synonyms that I exchanged the names :) $\endgroup$ – Tim Aug 30 '16 at 20:01
  • 2
    $\begingroup$ I've created the six mappings to lme4-nlme. I'll wait a little before merging (changing tags on existing questions) in case I messed up or in case anyone has second thoughts. $\endgroup$ – Scortchi - Reinstate Monica Aug 31 '16 at 13:10
  • $\begingroup$ @Scortchi Thanks a lot! I see the synonyms created, but I see zero threads when I go to stats.stackexchange.com/tags/lme4-nlme. Why is that? I am confused. That's not how un-merged tag synonyms usually work. $\endgroup$ – amoeba says Reinstate Monica Aug 31 '16 at 19:42
  • $\begingroup$ @Scortchi I edited one thread to add [glmer] to it and it was automatically changed into [lme4-nlme]. So the synonym works, but the new tag still appears with only 1 thread now, it's very strange. $\endgroup$ – amoeba says Reinstate Monica Aug 31 '16 at 20:03
  • $\begingroup$ @amoeba now it works, it seems it needed to refresh. $\endgroup$ – Tim Aug 31 '16 at 20:04
  • $\begingroup$ @amoeba: 1397 posts now with the new tag $\endgroup$ – Scortchi - Reinstate Monica Sep 1 '16 at 9:51

You must log in to answer this question.

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