I've recently concluded two years as a research fellow at HarvardX. To bring things to a close, last week I held a workshop with course developers looking at the question: What have we learned from the last two years of MOOC research that could help improve the design of courses?
Over the next few days, I'll release a series of short post on seven general themes from MOOC research that could inform the design of large-scale learning environments in the years ahead.
- MOOC students are diverse, but trend towards auto-didacts
- MOOC students value flexibility, but benefit when they engage frequently
- The best predictor of persistence and completion is intention, though every activity predicts every other activity
- MOOC students (tell us they) leave because they get busy with other things, but we may be able to help them stay on track
- Students learn more from doing than watching
- Lots of student learning activities are happening beyond our observation: including note-taking, socializing, and using other references
- Improving student learning outcomes will require measuring learning, experimenting with different approaches, and baking research into courses from the beginning
1. MOOC students are diverse, but trend towards auto-didacts
Imagine that the 1.5 million participants in the first two years of HarvardX and MITx course were crammed into a single classroom for 100 students. 70 would be men, and 30 women. About half would be in their 20s, over 20 students would be in their 30s, and the remaining would be teens and other folks from 40 through their 80s. Almost 70 students would have bachelors degrees, and many of those would have advanced degrees, but some would have only a high school diploma, or no formal education at all. Two-thirds would be from outside the United States. Nearly 40 would have a teaching background, and 8 would be currently teaching the subject right now. Some other students would be total novices. The diversity of this group is staggering. As my colleague Andrew Ho has argued, there is no physical classroom like this anywhere on Earth.
(Image courtesy Andrew Ho)
Generally speaking, we expect these students to be more affluent and better educated than other students in their home countries. In the U.S., where we have the best data, HarvardX and MITx students live in neighborhoods that are ½ a standard deviation more affluent than typically U.S. residents.
We have some evidence, not yet published, that of those students who actually enroll, affluence is not a strong predictor of completion for older students. But for teenagers and young adults, students from more affluent and educated neighborhoods and families are more likely to complete courses.
So our students are incredibly diverse, and the range of students is broad, but the bulk of students are affluent, well-educated students looking for lifelong learning and career development experiences.
What should course developers do with this knowledge?
One strategy would be to attempt to build courses that were more welcoming and supportive of a more diverse group of students.
Partner with Community-Based Support Structures
One approach is to consider how to provide better support services for students. The LaunchCode project in St. Louis is currently the exemplar program of providing community-based supports to students. LaunchCode provides meeting spaces for students taking CS50x in St. Louis, and it provides tutoring for current students and job placement for graduates.
There is good evidence that close mentoring and support can make a huge difference for students. Online learning resources can have the most impact when combined with human support for the learners who need it most.
Design Courses to Be Accessible to Novices; Avoid the Curse of Expertise
Another approach would be to help faculty make their course content more friendly and accessible for novices. Many faculty, especially at places like Harvard, suffer from the "curse of expertise," where experts can no longer remember the learning progressions that helped them get past novice status towards expertise. For these faculty, making material more accessible can feel like "dumbing down the content," but all novices need on-ramps into complexity. Barbara Oakley recently gave a talk at HarvardX describing a number of strategies that she used in her incredibly popular course, Learning How to Learn, about the cognitive science behind learning. Oakley emphasized the importance of metaphors and analogies as bridges to help students develop conceptual understanding. When students encounter complex ideas, it can be helpful to start by relating those complex concepts to ideas students already understand. Simple analogies can be a bridge to more precise and sophisticated understanding.
One day, it may be possible to have students with different backgrounds have different learning experiences in MOOCs. At present, our incredibly diverse audiences all get the same learning experience. As the edX platform get more sophisticated, it may be possible to develop different learning pathways for different students that more precisely support their needs.
Address Pyschological Barriers for Learners from Diverse Backgrounds
The line of research on stereotype threat helps us understand that for many learners, triggering certain facets of their identity can negatively affect student performance. At the same time, research has uncovered a variety of strategies and interventions that can "innoculate" students against these triggers.
For instance, Rene Kizilcec and colleagues at Stanford have identified a participation gap in Stanford courses, where students from North America, Europe and Oceana perform better in courses than students in South America, Asia, and Africa, even controlling for level of education and other demographic factors. One way to tackle these gaps is to address issues of social identity threat and belonging that might be hindering student participation. Rene has adapted a series of "belonging" interventions, for instance, asking students to write a short letter to a potential future student about what it feels like to belong. Early results suggest that this short "innoculation" can narrow these participation gaps. We have replicated this study in two HarvardX courses and will be examining results.
Target MOOCs at Those Who Learn Best from MOOCs
Another strategy to address student diversity would be simply to try to serve the autodidact elite very well. This strategy basically says, "Look, if the students who do best in our courses are people who are already good learners, lets focus on serving them."
One way that strategy can serve the social good is by targeting professionals who work in the public interest. Many are now well aware from Daniel Seaton's work that a substantial number of MOOC students are teachers; and serving their professional development needs might be one way to benefit society broadly (The Atlantic has a recent article on this idea.) The public health field also strikes me as a domain where this strategy can work well. Many parts of the world have well-trained and well-educated medical professionals working in a context with weak systems for on-going professional development. MOOCs might be one way to benefit those communities.
To be clear, while I present this as one option, I'm not advocating for it. While it may be that some courses are best conducted by targeting experts, educating diverse populations is exactly the kind of grand challenge that organizations like HarvardX and MITx should be taking on.
Overall, student diversity is one of the signature features of large-scale online learning environments, and addressing that diversity should be one of the central themes in the design of large-scale online courses.
For regular updates, follow me on Twitter at @bjfr and for my publications, C.V., and online portfolio, visit EdTechResearcher.