Harvard and MIT have made a suite of open source interactive visualization tools for the data of moocs. This is a follow up of their release of a series of working papers:
The technologies, dubbed Insights, draw upon near real-time, de-identified data of course registrants, dynamically updating at frequent intervals.
Developed collaboratively by Sergiy Nesterko, a Research Fellow at HarvardX, and Daniel Seaton, a Postdoctoral Research Fellow at MIT’s Office of Digital Learning, the tools showcase the potential promise of “big data” generated from massively open online courses (MOOCs) to help advance learning science.
Harnessing information on 1,055,562 total registrants, the tools provide an intuitive way to review individual courses across a number of metrics, including global enrollment, certificate attainment, gender and age composition, and education levels, all viewable country-by-country on a world map. The tools also feature HarvardX-wide and MITx-wide statistics.
“These tools provide our faculty and course developers, as well as the general public, usable, visually interpretable data from course offerings in near-real time,” says Nesterko. “This can help to guide instruction while courses are running and deepen our understanding of the impact of courses after they are complete.”
For example, Insights shows how HarvardX and MITx course registrations from China, highlighting how technical barriers such as the blocking of YouTube and firewalls influence participation. In addition, the tools can shed light on common claims like MOOCs serving primarily male, highly educated students. It turns out to be less black and white, as there is dramatic variability in gender and education across different countries and courses.
“Our hope is that these tools will be useful to everyone, including researchers, journalists, course developers, and the public, for the purpose of understanding how these courses work and who our students are,” says Seaton.
Moreover, the code is open source and the aggregated datasets are downloadable, allowing interested users to modify and build upon existing functionality. In the future, the pair plans to expand the Insights toolkit to include distinctions between “auditors” and “enrollees.”