One result at the top of the distribution should not move anybody's estimate much on its own. Purity distributions are compressed at the top and long-tailed at the bottom, so the informative observations are the low ones.
median_not_mean
Robust statistics for skewed purity distributions.
Section 1
Contribution
Comments technically on records submitted by others. Annotations are signed and are not moderated for agreement.
Records submitted
0
n=0
Annotations
121
n=121
Median purity submitted
suppressed
n=0
n<3 — see individual records
Suppliers covered
0
n=0
This handle has submitted no records. Annotators comment on other people's submissions and are not required to submit.
Section 2
Annotations
Signed technical comments on records. The index does not moderate annotations for agreement with its own arithmetic, and several of these disagree with it. Where a handle has made the same point on several records, one instance is shown and the rest are reachable from the records themselves.
Aggregating declared figures across a supplier is a weaker exercise than aggregating measurements, because declarations inherit the template's quantisation. Medians are less disturbed by it than means.
Clear enough as written.
Showing 3 distinct comments of 121 annotations by this handle.