by Rebecca D. Armstrong, DVM, PhD, director of research subject protection, at the University of California, Berkeley and a member of PRIM&R’s Education Committee
In early June, I was invited to attend a multiday convention, The Asilomar Convention for Learning Research in Higher Education, co-hosted by Stanford University and the Massachusetts Institute of Technology to share my IRB expertise. The event, which was largely supported by the National Science Foundation, tackled the issue of massive open online courses (MOOCs), educational research, and, by extension, human subjects research and IRB review. Knowing next to nothing about MOOCs, what I learned was fascinating.
There is now software that can track an online learner’s every move throughout the learning process, and researchers are hungry to analyze the “big data” that is generated by the more than 50,000 course enrollees from around the world, with the goal of improving learning. Educational researchers are getting data from commercial companies that host MOOCs, such as Cousera and edX, or from their own in-house courses that utilize similar platforms. Typically, at the beginning of a course, enrollees accept (i.e., click on) a basic user agreement, but they may not realize that included in that agreement is language that allows for the use of their data in a research field referred to as learning analytics. This lack of awareness is somewhat akin to the recent Facebook study controversy.
MOOCs, and the abundance of data that results from these courses, are likely to have a significant impact on research in the learning sciences. Those of us involved with the IRB can either be a hinder or facilitate research in this new domain. IRB professionals and institutions need to consider what is “normal” educational practice per 45 CFR 46.101 (b)(1). Where do data domains such as log files and text data fit? What are distance/online learners’ expectations of how their data will be used? Is this research exempt? What about when a learning experience is modified based on real-time analysis of learning data? Does this intervention require the learning experience (and research study) to undergo expedited level review? How should the educational databanks that result from MOOCs be managed? And, how do institutions view those participating in MOOCs? Are they seen the same as students who attend on-campus courses? And, if so, what are the implications of that decision with respect to the protections afforded under the Family Educational Rights and Privacy Act? Do the same rules apply to on-campus courses that utilize MOOCs to generate data that can be analyzed (e.g., key strokes, use of resources, sequence of learning activities, etc.) to improve learning?
In much the same way that attendees at the Asilomar Convention did, those of us in human subjects research compliance need to start thinking about the aforementioned questions and many more. Institutions’ IRBs will need to be proactive and promote collegial science as this field moves forward, while still finding ways to respect privacy and the voluntary participation of learners. To begin this discussion at your institution, I encourage you to review the two-page principles document that was developed by the attendees at the Asilomar Convention. Several of the agreed upon principles are informed by The Belmont Report. In closing, stay tuned! It’s going to be an interesting ride in the realm of educational research, learning analytics, and technology whether at a distance or in your local higher education classrooms.