I've been reflecting on the issue of how social science and humanities are distinct from STEM classes, and perhaps the division isn not really one that interposes "Science & Engineering" vs. "the Humanities." Courses in both spheres can provide context and ask students for critical reflection. Another way to frame the issue is to distinguish between online courses that equip students with "work skills" versus those influencing "how we think or live." Classes in "work skills" give students fluency in a specific technical skill set, scientific concept or technology. They empower students with skills in demand by employers, and are an important way to deliver on the promise of online education.

Courses affecting "how we think/live", in contrast, deal with hearts and minds. They influence how we think and believe, how we interact with others. Issues raised in these courses may affect students' identity and sense of self; how they engage with other human beings in their immediate circles; and how they view society and their own role in the world. Ultimately, with these classes, we may be changing norms and values.

This may seem a bit touchy-feely, but this type of education not as groundless or subjective as it might appear at first glance. In every "hearts and minds" class, students first need a foundation of common factual knowledge to build upon. So if we're talking pedagogy, there's certainly a role for quizzes based on facts: "Do you know the basic terminology, key events and information related to this topic?" However, it's also important for instructors to set this foundational learning in a larger context: What are the implications of these facts? What are some lessons learned from these case studies? And perhaps most importantly, how do we apply these lessons to our own lives? Social context is a key difference between "work skills" and "thinking/living", and as a consequence, the model of the classroom will be different.

Given these distinctions, it's important for humanities and social science professors to engage with MOOCs now. Whatever their limitations, MOOCs have already begun to change the conversation about what "education" means. Even though educators in these disciplines may have their reservations about the medium and what it offers, it's important to be part of the discussion so we can shape the future of the field, instead of running to catch up after the CS folks have already defined it.

Humanists and social scientists should take a seat at the table to describe what MOOCs are and also what we hope MOOCs need to become -- to set the agenda for their future development. If universities continue to push online learning (and Stanford seems intent on doing so), we have a stake in shaping that learning experience. We can be proactive in exploring how online tools can be couple with our classrooms, in defending the interests of our students, and in setting priorities for R&D. It would be less than ideal if, in a few years time, we are forced to shoe-horn our classes into a model that's already been dominated by computer scientists.

I have long been a whole-hearted fan of discussion-based learning, but given the power of the online medium to reach many populations around the world and potentially change the nature of education, it's also important to be on the forefront of discussion and take part in shaping the vision, rather than to arrive late to the party.

One issue with MOOCs thus far is that many of them seem to be taking a "one-size fits-all" approach. Many of the earliest MOOCs were computer science courses, and those types of classes have defined the contours of how MOOCs work: video lectures on a topic, followed by quizzes (usually finite-choice or numerical answers) and/or programming assignments. Indeed, MOOCs seem especially fit for computer science education, and since the software developers come from that engineering background, many platform features are geared toward addressing the needs of CS classes. (It's also not to be overlooked that CS instructors and students are readily comfortable with Internet technologies). However, social science and humanities classes are not the same as STEM classes; they have different requirements, and thus ought to be structured differently.

Perhaps I am generalizing a bit here, but (at least in the online education world) many of the science and engineering MOOCs teach students how to understand and use particular equations and solve certain kinds of problems based on that knowledge. Students are tested by completing exams and problem sets. These are discrete skills, for discrete tasks, and the questions usually have one correct answer. As mentioned, MOOC platform tools are largely built around this model.

In contrast, for social science and humanities classes, it is not only about teaching facts or getting students to "the right answer." Teachers challenge students to think critically, sometimes questioning knowledge that is presently held as self-evident or taken for granted. The experience can be pretty jarring.

Therefore, in traditional real-world (i.e. non-MOOC) classrooms, the pedagogical form is often distinct from the one employed in STEM lectures. The favored model for political science or literature classes, for example, is the seminar format, with the emphasis on discussion-based learning. A student does not simply go off on his own, read the textbook to acquire factual knowledge, and complete an exam by solving for the correct answers. Lectures are certainly used in political science, but these are supplemented by discussion section.

It is not the case that every student always participates in the conversation; just ask the next frustrated PhD who is TA-ing a class about how silent the room sometimes seems. But even active listening to others engaging in discussion helps students think through the material, as they may privately raise questions and objections. It is not a lone-wolf model of learning that you can get away with in a math or chemistry class. Being forced to construct arguments based on evidence, defend them, and hear multiple perspectives on the same issue, are all useful means of interrogating new concepts introduced in class. There is, of course, room for reflection and "alone time" and literature and political science students are just as often found with nose in book when first reading the material. But the social nature of classes, modeling the interaction of ideas in an engaged debate, after the reading is completed, is recognized as a key part of the learning process.

If you're asking people to think differently, it's simply less effective to learn in a vacuum. Some STEM classes do fit in this interactive realm as well -- for instance, courses on sustainability often benefit from discussions and asking students to consider how issues inter-relate. Perhaps it's because for many environmental problems, there is also no "one right answer." We are asking students to absorb facts, comprehend the complexity of issues, and reframe those issues while understanding that there may be multiple solutions.

I'm not saying one is better or worse, but simply noting that different approaches are required for different kinds of learning. We make this distinction in real life classrooms, and we need to make the distinction in online learning as well.



P.S. I recognize that in actual science and engineering classrooms, educators do aim to teach students to be critical thinkers. The goal is usually described as having students comprehend the underlying principles so they can apply them in diverse contexts, be these engineering scenarios, biological mysteries, or a mathematical challenge where computational power can be brought to bear. But in MOOCs especially, much of the focus is on discrete, testable skill sets. These skills can be assessed, and students presumably demonstrate gains in scores as they learn the material. (While this issue prevalent in MOOCs, it is hardly limited to them. Sit through a run-of-the-mill freshman math lecture for another example).

I was recently told by a Stanford Lytics Lab researcher that the pre-course survey for Democratic Development had much higher rates of participation than other online classes. "This is by far the highest response I've seen on a MOOC survey," he wrote, wondering how we got "that many responses."

It was a very high priority for us to obtain demographic information about the MOOC participants. We felt it critical to know where students hailed from, in order to better shape the course for them -- structuring forums, noting future subtitle needs -- and to understand the international reach of Democratic Development. (It turns out there were participants from 185 different countries and territories).

How did we attain such high rates of survey responses? To strongly encourage participation in the survey by your students, you can deploy the following strategies:

Message early. Include the link to the survey in the welcome e-mail to students. Invite their participation!

Message often. Include the survey link in subsequent e-mails to students, either in the body or as a post-script. The series of reminders helps because not everyone reads every e-mail.

Integrate the survey into course material. Add in-video questions reminding students to take the survey. At the beginning of some of the early lectures, I added a question where students were asked either to fill out the survey if they hadn't already, or to check the box that they had completed the survey and move on to the video lecture. (Be sure to include the survey URL in the question). It's an extremely effective, but relatively unobtrusive strategy since students will definitely encounter the videos, in contrast to e-mails that they can easily skip over or miss.

Make it a required assignment. I made taking the pre-course survey the first "quiz" of the MOOC. This was in lieu of an actual quiz, as we wanted to give the students some time to adjust to the course in the first week. We announced that the quiz was ungraded, and that there were no "right "or wrong" answers, but posted it in the "Quizzes" section of Coursera to give it the same gravitas as a required assignment. We subsequently moved the pre-course survey to "Surveys" a few weeks into the course.
_
Use inviting language. I framed the survey as a communal endeavor, a way to participate in the Democracy MOOC community. It was also underscored as a way for students to benefit their own learning, because the survey would help the course staff shape a better experience "for you, the students."

Use these methods to invite participation in your survey and learn more about your students! The findings can be eye-opening (and sometimes plain amusing).

When considering the issue of student engagement with a course, the social aspect of a classroom is highly influential. Silicon Valley likes to package everything into a single device that fits in the user's pocket, so the individual can learn on his own. But in terms of a human experience, the people you are surrounded by and interact with also have a significant impact on your education. (Perhaps that sounds obvious, but engineers glued to screens need to be reminded of that fact from time to time).

Local engagement matters, whether it's with a designated teacher or a group of like-minded peers. Students dedicated to the same purpose already give each other a tacit sense of encouragement and validation: "I am part of a collective." They can also actively remind each other to carry out weekly assignments and responsibilities. Take the experience of prayer groups, "faithful people reminding other faithful to pray," which are highly effective in reminding people to follow through with their common cause of worship. I'm not equating MOOCs to a prayer circle or ascribing any religiosity to online courses, but the phenomenon highlights the value of human interaction and friendly reminders.

This can have a direct impact on the question of attrition: why students don't always make it through the MOOC? If we adopted a game theory mindset, we might outline the costs of "shirking" or "defecting" by not doing your work for the MOOC against the benefits in completing it. Let's examine this issue through the lens of the Democratic Development MOOC: unless you are a very committed democratic activist, you are not going to follow through if the costs (time, effort, brain power) rise. Activists have high intrinsic motivation to want to complete the course because they are already deeply interested in the material, but for other students closer to the edge, how can you raise the "costs of defection"?

Since MOOCs are free (only time investment needed), we might frame the issue in terms of commitment. How do you raise commitment to these classes? Here are some potential strategies:

Reward progress and achievement. The most basic way is to make available a statement of achievement based on completion of assignments. This gives students a tangible sense of recognition, and can also make a difference for students by setting out achievable milestones and standards of achievement. Pokémon, anyone? Gotta catch 'em all! (Cue gamification?)

Host a graduation ceremony. This goes beyond the statement of achievement, as it gives students recognition tied to a unique real-world experience.

Build a sense of community. Let students feel they are part of a larger enterprise of learning. It's especially helpful for forming community when the topic is emotive or heartfelt. If the professor is well-known and charismatic, that is another big plus, because students feel they are in the instructor's orbit. Craft e-mails and announcements with this overarching concept in mind.

Invite student participation. Have them contribute to a Wiki, report on conditions of democracy in their country, or submit photos and videos of local elections. Get past the forum, and start creating products that can have a useful life beyond the end of the class, to motivate student participation further.

Organize a support group or reading group. In addition to the mutual reminders and sense of belonging mentioned above, you are now not only accountable to just yourself, but to a group of friends as well. Leaving a class, or even just sloughing off a week, will have social repercussions.

Create unique events and experiences. Inspire a feeling in students that they are part of a special experience. "If you don't take part in this course, here at this moment in time, you are missing out!" Offer opportunities for engagement sprinkled throughout the quarter. Periodic events bring repeated traffic, and it's simply harder to forget that a course exists if time-sensitive opportunities are baked into the schedule. If you can put out enough new interesting experiences, such as an interview with an influential political leader, or a live webcast where the instructor answers student questions, it helps students stay engaged with the course material and in the community. Through this process, whether it's through lecture videos or other interactions, learning happens.

In the end, if we care about exposure to ideas even more than the completion of specific required elements, then incentivizing student engagement in a multitude of forms becomes a rather fun exercise in imagination.