From Meta, a Wikimedia project coordination wiki
Jump to navigation
Jump to search
Work log[ edit ]
Archive
Thursday, April 3rd -- Power analysis for experimental period.
Monday, April 7th -- Instrumentation of experimental UX
Wednesday, May 14th -- Checking instrumentation
Monday, May 19th -- Checking instrumentation post deployment.
Tuesday, May 20th -- Checking on small proportion of token'd account creation impressions.
Tuesday, May 27th -- Token stats
Wednesday, May 28th -- Token stats, classes and some preliminary results
Thursday, May 29th -- Fighting with data inconsistencies
Friday, May 30th -- Fighting with data inconsistencies
Thursday, June 5th -- Some preliminary user stats
Tuesday, June 10th -- Productivity analysis (get revisions)
Wednesday, June 11th -- Productivity analysis (comparisons)
Thursday, June 12th -- Productivity analysis (more comparisons)
Friday, June 13th -- Finishing up productivity analysis. (data)
Monday, June 16th -- Productivity analysis plots & revert rates
Wednesday, June 18th -- Funnel data
Friday, June 20th -- Funnel proportions
Monday, June 23rd -- Funnel proportions
Friday, June 27th -- Funnel for all the wikis!
Monday, June 30th -- Block rates
Monday, July 14th -- Checking logging for second study
Thursday, July 17th -- Preliminary data for second experiment
Tuesday, July 22nd -- Productivity results
Registration rates[ edit ]
Registration rates (per eligible anon). The proportion of anonymous users who both clicked edit and registered an account are plotted.
wiki
baseline %
baseline n estimate
v1 %
v1 % $\Delta$
v2 %
v2 % $\Delta$
v1 factor % $\Delta$
v1 n estimate
v2 factor % $\Delta$
v2 n estimate
enwiki
1.367
4754.273
1.871
+0.504
1.59
+0.223
+36.9
6508.355
+16.4
5531.674
Productivity[ edit ]
Article edit rate. The geometric mean number of article revisions saved is plotted by experimental condition.
Productive edit rate. The geometric mean number of productive edits is plotted by experimental condition.
Productive editor proportion. The proportion of productive editors is plotted by experimental conditio
Burden (reverts)[ edit ]
Reverted article revert proportion. The proportion of overall reverted article revisions is plotted by experimental condition. Note that this method for measuring revert proportion over-weights highly active individuals.
Reverted article revert rate. The geometric mean reverted article revisions per user/token is plotted by experimental condition. Note that this method for measuring revert activity does not control for decreased article activity.
Sampled article revert prop. The proportion of reverted sampled article revisions is plotted by experimental condition. Note that this method for measuring revert activity controls for both prolific users and decreased article activity.
Discussion[ edit ]