and look like they addressing the same topic. The latter seems to be a more specific name and the description accompanying it seems a little better. While we are at it on meta there is [tag] and [tags]. Those probably could be merged too.

  • 4
    $\begingroup$ A synonym has been created and the tags have been merged. $\endgroup$
    – whuber Mod
    Jul 24, 2013 at 2:11
  • $\begingroup$ Thanks, @whuber, would you mind doing the [tag] tag on meta, too? I submitted it as a synonym when I answered this. $\endgroup$ Jul 24, 2013 at 2:16
  • $\begingroup$ @gung Thanks, done. $\endgroup$
    – whuber Mod
    Jul 24, 2013 at 2:18
  • 1
    $\begingroup$ As this suggestion has been implemented long ago, consider accepting gung's answer to make this thread appear "resolved". $\endgroup$
    – amoeba
    Dec 11, 2015 at 11:32

1 Answer 1


+1, I agree. Note however, that there are 79 threads categorized under [kolmogorov-smirnov], but only 9 under [kolmogorov-smirnov-test]. In addition, there is a slight preference for tags with shorter names. These facts suggest that [kolmogorov-smirnov-test] should be the synonym and [kolmogorov-smirnov] should be the master.

  • $\begingroup$ I didn't know there was a preference for shorter. I suspect that 'kolmogorov-smirnov' gets more because that tag comes up first. But note that it is "mann-whitney-u-test". I'm sure there are counter examples. One interesting case is spearman and spearman-rho where the former is described as being about the person and the later about the statistic. $\endgroup$ Jul 24, 2013 at 1:56
  • 1
    $\begingroup$ I'm sure there are counter examples. I also think this is a sufficiently weak preference that it could be over-ridden, if there were some reason. In this case, however, I don't see such a reason (but I'm always open to a persuasive argument). $\endgroup$ Jul 24, 2013 at 2:00
  • $\begingroup$ That spearman one is at least partly my fault; I sought to distinguish spearman from spearman-rho, because Spearman was responsible for more things relevant to CV than just the nonparametric correlation; it seemed to make sense to make the two tags refer to different things. The alternative (which may well be preferable) would be to simply make spearman a synonym of spearman-rho and leave the (small number of) other references untagged. $\endgroup$
    – Glen_b
    Aug 2, 2013 at 7:50
  • $\begingroup$ @Glen_b, I don't follow you. What does spearman have to do with this? $\endgroup$ Aug 2, 2013 at 12:22
  • $\begingroup$ That was in response to rpierce's comment immediately under your answer, which raises spearman vs spearman-rho. $\endgroup$
    – Glen_b
    Aug 3, 2013 at 1:34
  • $\begingroup$ @Glen_b, oh. That seems like a reasonable case, though. It's not like your tag was charles-edward-spearman. The man & the statistic are different & could legitimately have 2 tags. $\endgroup$ Aug 3, 2013 at 1:47

You must log in to answer this question.

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