Wikipedia:Education program/Research

September 2013
Results and preliminary analysis are available for a study of plagiarism on English Wikipedia, comparing the rates of plagiarism by student editors in the US and Canada education programs with those of other groups of editors. In short, the data shows student editors plagiarized at a lower rate than other newcomers, and at a higher rate than admins and very prolific editors.

Update, October 2012
The results of the article quality part of the project are now available. Articles that spring 2012 students contributed to improved by an average of 6.5 points on a 26-point scale, with 87.9% of all articles showing some improvement.

Preliminary results about what factors lead to success in classes are now available. Additional research will be done on this front in the coming months.

Background
The Wikimedia Foundation is planning a research project to evaluate each course in the Wikipedia Education Program in the United States and Canada so we can determine what factors make courses more successful. We have worked with more than 150 classes at this point, and we'd like to find out patterns among our most successful courses so we can tailor our future professor selection around courses that will be the most successful.

How we are defining what a "successful course" is:
 * Students add a large amount of content to Wikipedia (based on bytes added)
 * Content students add is of high quality (based on quality ratings from a sample of students done by Wikipedians)
 * Sustainability of professors and Ambassadors (whether they return or intend to return in future terms; satisfaction level with program)

At the same time as Wikipedians are assessing the quality of the random sample of student articles, Wikimedia contractors are reaching out to every professor we've worked with, working to determine a set of datapoints. The assessment of quality, quantity, and sustainability will give us a ranking of which classes are most and least successful. We will run successful and not successful classes against those quantitative datapoints.

We're looking for common markers of a successful course — is it class size? Number of Ambassadors? Wikipedia experience of the professor? Public or private schools? Age level of the students? We have some hypotheses, but we are doing this research project to prove or disprove our ideas. We will use the outcomes to determine which courses we want to work with in the future.

Get involved!
Interested in helping out? Volunteer to help assess student articles. See Ambassadors/Research/Article quality for more details.