Stats for article counts, editors and edits 2001-2018:
https://stats.wikimedia.org/EN/ChartsWikipediaZZ.htm#2

Stats for article here: https://xtools.wmflabs.org/articleinfo/en.wikipedia.org/2021_storming_of_the_United_States_Capitol

Thanks again @nshahquinn-wmf. The monthly average pageview for 2020-4 through 2020-11 is 16.2B, I updated the spreadsheet with the corrected pageviews using Neil's updated data set from Turnilo.

I believe Group A sees Home and Group B sees Explore, @Dbrant can verify if this is correct.

Results from 89 day data set show very little difference between Group A, which was presented with reward interstitials, and Group B which was not.

Thanks @nshahquinn-wmf, @ASvitak I made a spreadsheet for 2020 Monthly Wikipedia Pageviews: https://docs.google.com/spreadsheets/d/1g_oDXJbD4vpfEHWVBfG9PL9V0XjeN4fpW4AsvRKXicM/edit?usp=sharing

Looking at avg. session count and avg. session length between both groups shows no appreciable difference. Data for 30 day period:

Hi @ASvitak Editor count data spreadsheet is updated through 2020-11
https://docs.google.com/spreadsheets/d/1o0TTvYsLL4Ef_DDAHrr-jkLXdTpd7sX_ZzFqfBFnfUA/edit#gid=1264761950

Draft Schema doc: https://docs.google.com/presentation/d/1b6Q1a4ahs2kM8I53drUXBudQoT4rerZUe-LkJrWoZS4/edit?usp=sharing
Draft Schema: https://meta.wikimedia.org/wiki/Schema:MobileWikiAppWatchlist

The COVID-19 related pages list has been updated as of 2020-12-08 to include new pages and pages related to COVID-19 vaccines.

Will resume QA when new iOS schemas are added to MEP.

Resolving this ticket - see https://phabricator.wikimedia.org/T266056 for trend/new change data analysis.

Completed a second round of analysis for test group users using data from 2020-10-26 to 2020-12-03.

Dashboard on Superset

Data from 2020-09-03 - 2020-11-03

Thank you Cooltey. I'll add version numbers with errors I find going forward.

There are some errors showing up in Logstash/Kibana snapshot here that I suspect to be part of the drop in Create Account events, these events won't make it to the schema, there are ~1000+ of these event per day. Kibana is a very useful tool (it is linked by default on every Schema Discussion page).

Spreadsheet with requested data is 2020 Election Related Pages - T266719.

This required a change for the integer value being sent and a change on the schema to make the value for type just a string and not an enum value. Schema is updated to Revision 20636844.
Confirmed that data is populating tables as expected with type and position for events that utilize those values.

@Tsevener I am in the process of post deployment QA, noticing that I am not seeing any recent_change or thank events in the data on event.mobilewikiappioslivingdoc, so not yet seeing data for event.type or event.position - I need to look at event logs (I'm just looking in the event.mobilewikiappioslivingdoc table ) but are you seeing any 'type' or 'position' data going out on your end?

Noting here since the question of allowable differences in data between Legacy and MEP tables has been raised. @mpopov and I discussed allowing a 2.5% variance in counts as passable for data between sets of tables. We arrived at 2.5 as reasonable because some variance is to be expected but anything above 2.5 may show where data collection is possibly due to bugs, based on experience and expectations.

Update:
Instrumented in next update:

• dismiss rates for notifications
• We can see open rates, (CTR if there is a link). Notification itself is a link or can have actions at the bottom that can go to different destinations
• Send/deliver rates, do we have a way to see those metrics? Sum of clicks + dismisses = all impressions