Get the complete book Thinking Strategically about MOOCs: The Role of Massive Open Online Courses in the College and University at Amazon in print or kindle version.
The massive nature of MOOCs offers a new laboratory for approaches to developing competency-based assessment. Many have longed for the opportunity to make use of emerging analytical tools in education. We have large amounts of data that is unstructured and lacks context. MOOCs are poised to provide purposeful data sets about student interaction with course materials. We may have a new opportunity to feed the analyses and make predictive decisions about performance and outcomes in those courses for which such an approach makes sense. In “Can Big Data Analytics Boost Graduation Rates,” Ellis Booker noted that “the advent of computer-mediated and online instruction, especially massive open online courses (MOOCs) with their tens of thousands of students per class, are changing what’s possible.” MOOCs, as a result of their scale, “provide so much data about student interactions, not only with the course material but with teachers and even other students. This massive amount of data can be parsed, compared, merged, modeled and analyzed, with the goal of improving educational outcomes.” Progress may well be made in the refinement of the analytics and development of data collection from MOOC environments. The application of these metrics to the data mined from MOOCs has the potential to teach us a great deal about learning in online environments.
Desire2Learn, a company that provides learning management systems to the education market, is investing big in big data. “In the last three or four years, we built a team of five PhDs who’ve built algorithms and models to predict student performance,” says company CEO John Baker. The company created a “risk quadrant,” a visual representation of how each student is likely to do in an online course. The predictions are made dynamically on a week-to-week basis. After launching a beta version of their analytics program, Baker noted that “depending on the availability of historical data associated with a specific course, we are able to achieve accuracy rates approaching ninety-five percent as early as week two or three.” These results were validated with research data sets, including one from the University of Wisconsin. This research echoes some of the data received from surveys of chief academic officers who hope that MOOCs and other iterations of online learning will afford opportunities to learn more about best practices in online learning.