Research:SuggestBot - Adding Information (2011)
- Morten Warncke-Wang (primary investigator and caretaker of SuggestBot)
- John Riedl
- Dan Cosley
- Sam Tucker
This project will investigate the effects of adding information to SuggestBot's recommendations of articles to edit. We plan to add information about the readership and estimated quality of articles, as well as specific suggestions on how to improve articles.
The study will alter the recommendation messages sent to users on English Wikipedia who have signed up to receive these messages regularly (see Getting Recommendations Regularly for specifics on this sign-up process). We will also contact a dozen of these users for participation in a short design study, where we will ask questions on how specific suggestions for article improvements can be added. Lastly, we plan on contacting about 20 users for short interviews about the usefulness of the information added to the recommendations.
We plan to write a peer-reviewed research paper based on our results, of which a summary will be published on our research group's website.
Wikimedia Policies, Ethics, and Human Subjects Protection
This research project has been approved by the University of Minnesota's IRB, with reference number 1108E03504. It has also been up for discussion on the English Wikipedia's Bot Approval Group and received approval, see en:Wikipedia:Bots/Requests for approval/SuggestBot 5 for details.
Benefits for the Wikimedia community
This study aims to improve the quality of SuggestBot's recommendations, which is a service currently available to all registered users on English, Norwegian, and Swedish Wikipedia.
- Sept 2011: Commence part 1 of the study, adding general statistics to article recommendations
- Sept 2011: Contact users for participation in the design study.
- Oct 2011: Commence part 2 of the study, adding specific suggestions for article improvements.
- Nov 2011 onwards: Data analysis of results and research paper writing.
This study is funded by NSF grants IIS 08-08692 and IIS 09-68483.