CopyrightX, an online course run out of Harvard this spring as part of the EdX program, was unusual in a couple of ways. It might not strictly be called a MOOC—a massive open online course—because it wasn’t open. More than four thousand people applied, and enrollment was capped at 500. Half of the selected students were women. There were equal number of students from the United States and from other countries. Students outside the US came from 70 different countries, in total. The youngest student was 13, the oldest 83. Although CopyrightX was a class about copyright law, only 30 of the 500 students were lawyers.
And these students stuck with the course. Most online courses have an appalling rate of attrition. Students start out with the best of intentions. But week after week, lecture after long, academic lecture, commitment flags. Usually around 90 percent of students drop out. For CopyrightX, two-thirds of those students made it through until the end. Fully half of them took the final exam —a typically grueling and mind-bending, four-hour law school take-home test, not much different from the exam Prof. Fisher gave Harvard Law School students.
“We built this course in part to engage people on a sustained basis, and they stuck with it,” Fisher says. The course’s design helped deliver that outcome: a team at Harvard spent months before the course thinking about how it would work. They focused on creating small groups, led by Harvard Law teaching fellows, that would meet regularly online to discuss course materials. The teaching fellows and the project’s coordinator, Nathaniel Levy, met regularly during the spring term to assess how the tools they were using to engage online students were working. But that deliberate design included the choice of materials, too. Fisher teaches real property law, legal history, and the various subfields of intellectual property law, and among those options he chose copyright as the subject of the course for a reason.
“All of the subfields of intellectual property are very important from an economics standpoint now, but copyright is the most culturally significant,” he says. “It influences the most diverse array of cultural practices… It permeates our life. It seemed a subject appropriate for an audience that I expected would be predominantly non-lawyers.”
Many of the students worked in fields where copyright law affected their work and livelihood—film, photography, software design, library science, and music. These are people who copyright law is meant to serve and who, proponents of copyright reform often say, are being badly served by a law that’s difficult for non-lawyers to understand. But it does seem that—for anyone willing to devote hours each week to law lectures and discussion sections—it is possible.
The CopyrightX class ran parallel to Fisher’s traditional Harvard Law class on copyright, and the two courses shared materials. The recorded lectures that CopyrightX students watched had to be at the same level as lectures prepared for some of the smartest law students in the country, because those students were watching them, too, in place of some of their more traditional lecture sections. If anything, Fisher says, the lectures were “more sophisticated than I congenitally produce in class, because I had more time to prepare them.” Some CopyrightX students initially struggled to adapt to the difficulty of reading dry case law, but, like any law student, they got the hang of it quickly enough. By the end of the course, it was clear that “people without any legal training can embed knowledge of copyright law into their own work and play,” says Fisher.
Although the CopyrightX was successful, limiting enrollment posed a problem. Fisher is committed to openness in education, a principle in tension with the decision to limit the class’ size. As a partial solution, Fisher made all the materials for the course available online, under a Creative Commons license, and invited others to lead their own version of the course.

I was a participant in the Harvard CopyrightX course/experiment. A number of the points in this article are incorrect, as follows:
1) Although it was suggested that the final exam could be completed in 4 hours, one could take up to 24 hours to complete it. Of the class members I communicated with afterwards, none spent less than 8 hours, and one took as long as 16. I would agree, though, that the exam was grueling and mind-bending.
2) The attrition rates for the CopyrightX course were likely distorted by the fact that it was not open enrollment. The article cited that over 4,000 students applied and only 500 were selected. One should assume that Harvard knows how to select law school students who have a high probability of seeing the program through, and that similar heuristics were used, including those that boost the number of females and foreign students in the class. Using their 2/3 survival statistic, it would be 333/4000 = 8.325%, so more than 90% of those that applied did not make it to the end. Actually, in my group, only 1/2 remained and were still attending as of mid-semester, so the 2/3 survival rate may also be inflated. Of course we don't really know how many of the other 3500+ people would have survived to the end, but Harvard should not be claiming a better drop-out rate than traditional MOOCs, since they front-end biased the experimental group. Had it been a random selection of applicants, they might have been able to make that claim, but that was not the case.
3) The class was divided into two groups with respect to the readings. Only half of the class focused on "dry case law" while the others were predominantly given summary articles that discussed legal theory and international perspectives. This may have also affected the retention and success rate.
4) Certainly the opportunity to receive a certificate of completion from Harvard Law School provides high incentive for the participants. Were this course stated as emanating from Rutgers, for example, there might have been a different outcome. Unfortunately, we do not have an adequate control group to assess this factor.
5) Harvard has not yet issued the final exam "grades" which will determine another measure of success of the course. They also plan to analyze the statistics gathered. The author should revisit their conclusions again after this information becomes available.
Still, it was a valuable, though tedious, experiment to have participated in and I was glad for the opportunity to have attended.
#1 Posted by RTMercuri, CJR on Thu 23 May 2013 at 08:41 AM
You provide an interesting statistical analysis, thank you. I also participated in the course and note that (in our tutorial group at least) we were encouraged to participate in whatever means suited us best, choosing from options including - contributing to chat forums, commenting on readings (if using the NB software), or participating in live group meetings.
You observe that "Actually, in my group, only 1/2 remained and were still attending as of mid-semester, so the 2/3 survival rate may also be inflated. " - it is possible that the other half of your group were participating in other ways. I'm unclear how participation is tracked in the software but perhaps they have data on how many hours students were logged in for each week, how many posts they made, etc...
#2 Posted by BHS, CJR on Thu 23 May 2013 at 07:51 PM
You provide an interesting statistical analysis, thank you. I also participated in the course and note that (in our tutorial group at least) we were encouraged to participate in whatever means suited us best, choosing from options including - contributing to chat forums, commenting on readings (if using the NB software), or participating in live group meetings.
You observe that "Actually, in my group, only 1/2 remained and were still attending as of mid-semester, so the 2/3 survival rate may also be inflated. " - it is possible that the other half of your group were participating in other ways. I'm unclear how participation is tracked in the software but perhaps they have data on how many hours students were logged in for each week, how many posts they made, etc...
#3 Posted by BHS, CJR on Thu 23 May 2013 at 07:53 PM
Did Prof. Fisher really say "congenitally"?
#4 Posted by Ral, CJR on Sun 9 Jun 2013 at 06:02 PM