Contact: @Mraish
Research question 1: when IP editors create useful new pages, do they do so at a different rate than registered users? How does the size of those pages compare to those of registered users?
Query: on a per-wiki, per-month basis, split it by namespace and identify if the namespace is a "content" namespace: count the number of pages created by IPs and registered users, and the average (median? geometric average?) page length. Filter out pages that are "immediately" deleted (might require identifying what "immediately" means on a per-wiki basis).
Research Question 2: When IP editors make constructive edits (not reverted w/i 48 hours of being made), do they do so at a different rate than registered users?
(similar query as above, but aggregate edits)
Research question 3: Which of the Wikipedias in the list below have a user group equivalent to the English Wikipedia's "autopatroller" group.
Wikipedias we're querying data for: Arabic (ar), Bengali (bn), Czech (cs), English (en), Spanish (es), Persian (fa), French (fr), Japanese (ja), Dutch (nl), Polish (pl), Portuguese (pt), Russian (ru), Turkish (tr)
Time range for data query: 3 years back
Data format: CSV/TSV
User groups to exclude: bots