Editor engagement experiments/Results

This page hosts an overview of the results of past E3 experiments in chronological order. Click on the title of the experiment to read the full report.

What we tested	Hypotheses/Research questions	Results
Post-edit feedback
Added a simple confirmation or gratitude to newbies upon completion of an edit. (R:PEF1)	Does receiving feedback increase the number of edits in new contributors? Does feedback lead to larger contributions?	New contributors receiving feedback produced a marginally significant increase in number of edits and made significantly larger contributions than the control group. No significant difference was found in the size of content removed across conditions. Read more...
	Does feedback affect the rate at which newcomers are blocked? Does feedback affect the success rate of newcomers?	Receiving feedback doesn't seem to affect the quality of new contributors. We found no significant difference across the three conditions when comparing the rate of blocked users or the success rate (number of reverted edits over total contributions) for editors in each group. Read more...
Timestamp position modification
Added a prominent link at the top of a Wikipedia article displaying the timestamp of the last revision and linking to the history page.	The addition of the modified timestamp produces a significant increase in impressions of an article history	Adding the timestamp link as a supplementary entry point nearly doubled (+96.2%) the overall number of impressions of revision histories. Anonymous editors and readers landed on the history page more than twice as often (+120.6%) compared to the control.
	The addition of the modified timestamp decreases clicks on the history tab	We found no evidence of "cannibalization" on the history tab for the experimental group. On the contrary, there was evidence of increased activity on the history tab (+46.5%) compared to the control.
Necromancy
A/B tested emails to different groups of lapsed editors to encourage them to return to Wikipedia.
	Emailing former contributors and reminding them about Wikipedia will encourage them to return and edit.	Contributors who had lapsed for a long period (one year) and a short period (one month) were not significantly affected by emailing. A small effect was observed for contributors who had lapsed for three months; however, the numbers were not significantly higher than the natural variance in return rates.
Editor milestones
Measured the effect of giving barnstar for users' 1,000th edit
	Giving a barnstar to editors who make 1,000 edits to articles will increase their short- and/or long-term editing activity.	No significant effect was observed in short- or long-term editing activity.
Template A/B testing
A/B tested different user warnings and notices.
	Changing the content (tone and length of prose) in a user talk message increases the likelihood of retaining the good-faith new users who receive it.	Good-faith new users who received templates that were shorter and more personal went on to make more edits on average than those who received the default longer, impersonal messages.