I am surprised to see that a covariate tag does not yet exist, as it would be more applicable than any of the other tags I had to settle for when posting this question: Should covariates that are not statistically significant be 'kept in' when creating a model?

Note: covariance exists, but does not seem to be a match at all, as it does not distinguish between IVs that are in or not in a model.

Side note: I am tagging this as a feature request, since I am asking for a change, but it's a tiny feature request...which could be implemented by any one of many, many site users! ;)

  • 1
    $\begingroup$ I feel obliged to point out here that covariance is not remotely the same thing as a covariate. That they are different concepts has nothing to do w/ whether any given IV is or is not in a model. $\endgroup$ Commented Aug 4, 2013 at 0:15
  • 1
    $\begingroup$ @gung I completely agree. That's exactly why I did not want to use "covariance", the closest tag that appeared when I started to type "covariate". $\endgroup$
    – A.M.
    Commented Aug 4, 2013 at 0:34

1 Answer 1


Currently, we have:

(x 23) and (x 298), both with excerpt and wiki filled in.

It was discussed in comments that these tags have different meanings. If anyone sees the opportunity to improve them, please go ahead.

  • 2
    $\begingroup$ They both have tag wiki excerpts that seem perfectly adequate. $\endgroup$ Commented May 29, 2014 at 23:58

You must log in to answer this question.

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