Research:Characterizing Wikipedia Reader Behaviour/Robustness across languages/Communications

From Meta, a Wikimedia project coordination wiki

Template for communicating the results of Part 1[edit]

Dear XX Wikipedia language community,

Between 2017-06-22 and 2017-06-29 a sample of the readers of your Wikipedia language participated in a survey [LINK TO VILLAGE PUMP ANNOUNCEMENT BY POINT OF CONTACT]. The results of this survey will help you learn about the information need, familiarity (prior knowledge), and motivations of the readers of Wikipedia in your language. In this report, we share some information about the first part of the analysis with you.

Survey questions The readers were asked 3 questions. These questions are the 3 questions developed as part of an earlier study where we build a taxonomy for Wikipedia readers. You can see the exact questions asked at LINK TO LOCAL LANGUAGE SURVEY SCREENSHOT.

The order at which the survey questions were shown as well as the order at which the responses were presented were shuffled for each participant.

Data For the duration of the survey, we sampled the reader requests to your Wikipedia language at a rate of INSERT THE RATE FROM Sampling rate FROM Research:Characterizing Wikipedia Reader Behaviour/Robustness across languages#Design and the response count was INSERT RESPONSE COUNT FROM THE SAME TABLE.

Privacy statement A privacy statement was designed for this surey. You can read more about it at CHOSE THE LANGUAGE SPECIFIC PRIVACY STATEMENT FROM https://wikimediafoundation.org/wiki/Survey_Privacy_Statement_for_Schema_Revision_15266417.

Results You can see the results of the survey for your language at LINK TO Research:Characterizing Wikipedia Reader Behaviour/Robustness across languages#All languages SPECIFIC TO YOUR LANGUAGE and the results for all languages for comparison at LINK TO Research:Characterizing Wikipedia Reader Behaviour/Robustness across languages#All languages.

In the first set of plots (specific to your language), you want to focus on the green bars. (The blue bars indicate the survey responses before the results were debiased according to the strategy described in Appendix of https://arxiv.org/abs/1702.05379 . For example "intrinsic learning" will tell you what percentage of the Wikipedia readers in your language come to Wikipedia because they are motivated to learn. Or, "in-depth" in information need plot tells you what percentage of your Wikipedia language readers come to your language for in-depth reading of the articles.

What's next? We will continue this research in the coming months and will share more insights with you as we learn more about the readers of you Wikipedia language. In the mean time, if you want to be in contact with us, please talk with us at Research talk:Characterizing Wikipedia Reader Behaviour/Robustness across languages or through your point of contact IF THE POINT OF CONTACT AGREES, LINK TO Research:Characterizing Wikipedia Reader Behaviour/Robustness across languages#Participating languages