7
$\begingroup$

The current tag has no wiki or excerpt and its 130 questions cover a wide variety of topics. I propose the following

  1. many of the entries are about random allocation in the context of clinical trials. I propose to create a new tag and re-tag them assuming I have enough reputation to create a tag. I know many people prefer the phrase random assignment but allocation is the term used in the CONSORT guidelines for reporting clinical trials
  2. There are other questions which are about some other aspect of clinical trials like analysis, sample size determination, which have probably been tagged [randomization] since they are about randomized clinical trials. Since we have a tag I propose to re-tag them where they do not already have that tag
  3. Some are about and I propose to re-tag them with that tag
  4. Some are about random generation, properties of random number generators and we have a tag for that so I would re-tag them with that
  5. Some will probably be dealt with as @Amoeba cleans up permutation

Assuming this meets with approval I will do all of this and then re-post with suggestions for the residue of questions.

| |
$\endgroup$
  • 1
    $\begingroup$ Re #5: I was not going to look at the [randomization] tag. Questions about permutation aka randomiztion test should be retagged with [permutation-test]. We can create the [randomization-test] tag and make it a synonym of [permutation-test] for convenience of future visitors, but if you are going through these questions please retag to [permutation-test]. $\endgroup$ – amoeba Feb 13 '17 at 20:07
  • $\begingroup$ Re #1: I am not sure "random-allocation" is the optimal tag name for that. I would guess that some questions can be about randomized experiments that have nothing to do with clinical trials. That's why I thought of "randomized-experiment" as a more general name... $\endgroup$ – amoeba Feb 13 '17 at 20:11
  • $\begingroup$ @amoeba if they are about experiments I will leave them for now but most are about clinical trials. About permutation I meant the ones so tagged will be cleared without further action from me. If they are not so tagged I will report back. $\endgroup$ – mdewey Feb 13 '17 at 21:13
  • 1
    $\begingroup$ Clinical trials are experiments. We could tag [randomized-experiment] + [clinical-trial], and we could additionally make [random-allocation] a synonym of [randomized-experiment]. I am not insisting on this tag name, just discussing. $\endgroup$ – amoeba Feb 13 '17 at 21:35
  • 1
    $\begingroup$ Update: I see that both terms are widely used, en.wikipedia.org/wiki/Random_assignment (also "random allocation") and en.wikipedia.org/wiki/Randomized_experiment. I guess it does not matter what name we pick for the main tag, but it would be handy to have other two as synonyms. $\endgroup$ – amoeba Feb 13 '17 at 21:39
  • $\begingroup$ @amoeba my feeling is that since most of the usage is people with clinical trials I would avoid experimentation. Although logically trials are experiments they never use that word. I think perhaps Josef Mengele en.wikipedia.org/wiki/Josef_Mengele is to responsible.The only other use which seems to occur although I have not added them up is randomized-blocks. $\endgroup$ – mdewey Feb 14 '17 at 9:43
  • $\begingroup$ OK, random-allocation as the master tag is fine with me (I still think it's a good idea to make synonyms as mentioned above). $\endgroup$ – amoeba Feb 14 '17 at 9:46
  • 1
    $\begingroup$ When you are done, could you please also take a look at randomized tag with 10 Qs and no excerpt. $\endgroup$ – amoeba Feb 22 '17 at 9:39
  • 1
    $\begingroup$ I know @amoeba tagged me at some point to look at this, and it looks like I am very late to the party. $\endgroup$ – StasK May 28 '17 at 20:26
2
$\begingroup$

I have now dealt with most of these. Just to report what happened: about half were re-tagged , a handful each were re-tagged , , , , (in alphabetical order). In response to comments I also went through the much shorter list tagged where most were about randomised clinical trials.

Loose ends: there are six which I have not done and one in randomized. It would be enormously helpful if someone looked at them and gave an opinion as I cannot see what to do with them. There are two about randomised black designs which could be dealt with together but the others are distinct.

There is also the suggestion that we have synonyms for some of the tags but I am not sure quite how to do this and I do not have much reputation in .

| |
$\endgroup$
  • $\begingroup$ It looks like [randomization] tag can be simply removed from all the remaining 6 questions. Some are about exp design (but have this tag already), others are about various stuff but we don't have a special tag for it (which is okay). The 1 remaining question in [randomized] I would re-tag to exp-design. $\endgroup$ – amoeba Feb 26 '17 at 20:44
  • $\begingroup$ Great job, by the way! $\endgroup$ – amoeba Feb 26 '17 at 20:46
  • $\begingroup$ Regarding synonyms, after some reflection I don't think that we need to make [random-assignment] one: it would also start with "random-a", so I can't imagine people failing to find the right tag. However, [randomized-experiment]->[random-allocation] can potentially be helpful, what do you think? $\endgroup$ – amoeba Feb 28 '17 at 12:08
  • $\begingroup$ @amoeba yes, that seems a good idea but I am not sure how to do it or whether I can. If you can set it up I think we have to vote for it? $\endgroup$ – mdewey Feb 28 '17 at 17:11
  • $\begingroup$ You can't. But you can post it as a suggestion in meta.stats.stackexchange.com/questions/1200 and if it gets several upvotes we can try to implement it or perhaps one of the mods will do it. $\endgroup$ – amoeba Feb 28 '17 at 20:22
  • $\begingroup$ Other than that, I think you can accept your answer :-) $\endgroup$ – amoeba Mar 1 '17 at 12:21
  • $\begingroup$ By the way, there is a couple of pending big retagging projects in meta.stats.stackexchange.com/questions/tagged/…, would you be interested in any of them? E.g. I've just edited meta.stats.stackexchange.com/questions/2979 to update and to bump. $\endgroup$ – amoeba Mar 2 '17 at 10:13
  • $\begingroup$ @amoeba I am currently doing cooks-distance but will have a look at rank-correlation unless you finish them first. $\endgroup$ – mdewey Mar 3 '17 at 17:21

You must log in to answer this question.

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