13
$\begingroup$

There seem to be quite a few questions that refer to Fisher's or Stouffer's method for combining p-values, but without the corresponding tag. Would it help the site to tag those questions? Or should that tag perhaps be merged/aliased with the tag? (I don't have enough rep to edit directly, someone will have to review and accept the edits.)

Examples (changed):

Examples (not yet changed):

The following search seems to be helpful, too: https://stats.stackexchange.com/search?q=%5Bp-value%5D+combining

$\endgroup$
12
$\begingroup$

Yes it would.

As it happens, the [combining-p-values] tag did not exist until around one week ago, when I created it and single-handedly tagged around 10 questions with it. There was some activity in several questions about Fisher's method and it prompted me to create this tag; I felt that [meta-analysis] is too general. I was not sure what is the best name for the new tag and asked in chat if this name is okay; @gung replied that it sounds fine.

I originally estimated that around 30 threads should be tagged with this tag and was going to re-tag them all, but got distracted (and lazy). So to properly finish what I started, other relevant threads should be found and re-tagged. Your list is helpful here; in fact, you can do it yourself by suggesting edits.


Several things to keep in mind about re-tagging old threads:

  1. Don't re-tag a lot of questions (say, more than 5) in one go. This will fill the front page with the similar old questions and people can get annoyed. It's better to do it in several steps, e.g. 5 questions at a time each day.

  2. If you edit an old question, proof-read everything. Don't just put a tag on it; check the spelling and the grammar; remove "thank you"s; consider if the title can be improved (it often can) by making it more precise; capitalize the first letter of the title and put a question mark in the end if it is formulated as a question; etc. etc.

  3. After you read a question anyway, you will most likely either want to upvote or to downvote it. Do it. (Don't be afraid to downvote old confusing unhelpful questions.) The same goes for its answers.

  4. In the process of reading all the old threads on a certain topic, you might discover some duplicate questions. Vote to close them as such. Also vote to close as unclear questions that are unclear.

It is partially because I felt I didn't have time to deal with points 2-4 that I did not finish the re-tagging. In general, I find that tidying old questions can take a lot of time (but is also a very good thing to do).

$\endgroup$
  • 3
    $\begingroup$ (+1) There's also the consideration that "combining p values" is likely to occur to a lot of people who haven't heard of "meta-analysis". $\endgroup$ – Scortchi - Reinstate Monica Aug 21 '15 at 10:53
  • 1
    $\begingroup$ @gung: No problem with the edit, but my remark was definitely not intended to be a "snark" (if I understand this word correctly and it refers to a critical comment)! I just found it curious that just several days after I created this tag somebody came across it and started wondering; it felt like a coincidence. So I meant "funny" in a good way :) $\endgroup$ – amoeba says Reinstate Monica Aug 21 '15 at 16:41
  • $\begingroup$ @amoeba: No offense taken. I've only learned about combining p-values in the past few weeks... $\endgroup$ – krlmlr Aug 21 '15 at 17:08
  • 1
    $\begingroup$ Started the first bunch of edits. Can't vote today anymore, though, will vote later. $\endgroup$ – krlmlr Aug 21 '15 at 17:22
  • $\begingroup$ My apologies, @amoeba, perhaps I misinterpreted. It came across as testy to me. $\endgroup$ – gung - Reinstate Monica Aug 21 '15 at 17:32

You must log in to answer this question.

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