Meta Stack Exchange is a question and answer site for meta-discussion of the Stack Exchange family of Q&A websites. It only takes a minute to sign up.
Sign up to join this community
The analytics for the beta site I moderate (Lego Answers) provides unreliable search keywords. The first ten or so results are unique, but every search keyword after has 41 views, and there are hundreds of results with supposedly '41' views. Some of the referring sites also have 41 views.
Something doesn't seem right. I can understand each search keyword having one or two visits each.
Part of the reason is due to a sampling error; the data are taken in "samples" so some keywords are overemphasized and some are de-emphasized. The sampling also assumes a certain degree of uniform distribution; this means that you see a lot of "41 views" because the sampling percentage, when multiplied out to the whole, yielded 41 as the value. Or, at least, that's how it's been explained to me. Assuming the explanation I was given is correct, then the "bug" you're seeing is actually a statistical error revealing itself. However, as some have pointed out, as the site ages (read: the dataset gets larger) these errors will be minimized more efficiently.
Long story short: trust nothing! Remember the adage: There are lies, damned lies, and then there are statistics.
