Consider the following tags:

  • (4): "the extent to which subsequent studies of the same phenomenon reproduce the results obtained in the original study" (excerpt)

  • (49) - no excerpt

  • (7) - no excerpt

The latter two tags are sometimes used to refer to replication of experiments and sometimes to "replicates" or "replication" in an ANOVA experimental design (when each ANOVA cell contains several values; e.g. see http://www.real-statistics.com/two-way-anova/two-factor-anova-with-replication/).

This is confusing and I thought of tidying up these tags. My suggestion is to move all questions about replication of experiments to [replicability] and get rid of the other two tags altogether. I don't think we necessarily need a tag for "replicates"/"replication" in ANOVA, apart from [anova], [sample-size], etc.


  • 1
    $\begingroup$ With 6 upvotes and no critical comments, I will start implementing it now. $\endgroup$ – amoeba Dec 22 '16 at 21:58
  • 1
    $\begingroup$ While removing [replication] tags I noticed that some of these questions are about "pseudoreplications". I considered creating a [pseudoreplication] tag but in the end decided to use [independence]/[non-independent] tags instead, because it seems that "pseudoreplication" terminology is largely confined to the ecology field, whereas other fields call it simply non-independent / correlated errors, or nested data, or hierarchical data, etc. $\endgroup$ – amoeba Dec 25 '16 at 14:21
  • 2
    $\begingroup$ By the way, if anybody is reading this, why on Earth do we have [independence] and [non-independent] as two separate tags? $\endgroup$ – amoeba Dec 25 '16 at 14:22

Done. I removed and from all threads. (See my comment above about "pseudoreplication"; I mention it in the answer body for future searches.)

The currently has 11 threads.


You must log in to answer this question.

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