Page MenuHomePhabricator

Report New editors per month in 2014 Oct-Dec
Closed, ResolvedPublic


Generate number (average over 3 months) for this report:

  • Historical data available from Wikistats based on legacy definition, complete data for Q2 is missing
  • Vital Signs has counts per project, no historical data and no notion of deduplication
  • Recommendation: use the same extrapolation ErikZ is suggesting for TAE (EZ) T88403#1030929

Event Timeline

kevinator raised the priority of this task from to Needs Triage.
kevinator updated the task description. (Show Details)
kevinator added a project: Analytics.

Erik described the extrapolation method here: T88403#1030929

We don't have Wikimedia wide deduplicated counts for new editors, like we have for TAE.

@ezachte, as Kevin said, this is about the data that already exists , just not yet for December 2014. Could you try to adapt your approximation method from T88403#1034921 to this one, too?

I don't think we can apply this estimation technique used for other wikistats metrics to new editors.
This metric is much more influenced by dump-over-dump adjustments, not in the range of tenths of percents mainly for the last month, but more several percents, also (but less) for earlier recent months.

Which makes sense. Deleted articles impact TAE en edit count much less.
'New editors' zooms in on the very segment of our editor base where edit patrollers do their thing.
That makes the data too volatile for this technique imo.

Per ErikZ's argument, recorded in the email thread over the weekend, I have marked this as "TBD" in the published version of the scorecard, while still linking to the source where people will hopefully be able to find the numbers soon.