Saturday, March 23, 2013

Why Online Courses are not just for students

The MOOC spectre

There has been a lot of hand wringing and excitement about the push toward online learning, and especially MOOCs - Massive Open Online Courses. With apologies to Dr. Martin Luther King, Jr., I want to say that I've been to the MOOC, and I don't mind.

MOOCs offer free college courses online. There are a number of organizations offering these MOOCs, all with variations on a basic approach (e.g. Coursera, edX, Udacity). In general, the courses are conducted entirely via the Internet, the instruction happens asynchronously (i.e. not in "real time"), and literally tens of thousands of "students" can be enrolled in any one course. Oh, and they're free.

There are lots of reasons to be excited and scared (depending on your position) about the possibilities and implications of MOOCs. Imagine the possibility of packaging the best instruction by the best professors from the world's best colleges and universities and giving it away for free to anyone with time and an internet connection. A lot of people see MOOCs, and online learning in general, as the coming "disruptive technology" for higher education. Like mp3s and Napster to the music industry, or Orbitz and Travelocity to the travel industry.

While MOOCs are indeed new, online teaching is not. The reality for most colleges and universities is that online instruction is already here. It's not just available through Phoenix or Coursera; it has penetrated the bricks and mortar of nearly every institution in the country. My institution has been offering online and hybrid courses for a number of years now. This coming fall, over 200 of our courses will be offered either fully online or as hybrid courses (mostly online but meeting in person periodically). I'll be teaching my first online courses this fall as well. Online instruction at my institution is entirely voluntary for faculty (at the moment), but there is real professional and social pressure to engage with online tools and to figure out how to adapt them (or adapt to them) for teaching.

MOOCs as Faculty Opportunity

There are lots of questions about how online instruction compares to face-to-face instruction, and what it means for students, but for the moment, I want to draw your attention to the opportunities that MOOCs offer to other instructors. In a nutshell, MOOCs give college instructors an opportunity to see how other college instructors teach online (and maybe to learn something too). This is invaluable. It's an open (and dirty) secret that college faculty are not really taught how to teach. We've been trained to do research (especially those with PhDs), and then we're hired as faculty and someone says, "By the way, you've got to teach these classes too." The pressures on faculty to do online learning have only compounded the depth of this pedagogical ignorance. What an opportunity to be able to see how another professor does the deed.

While on sabbatical, I took the opportunity to try out some MOOCs, initially out of sheer curiosity, but also in a sort of blind assumption that these online courses would offer me an easy and convenient way to brush up on my technical skills for both research and teaching. Between October 2012 and March 2013, I took three courses through Coursera:

  • Computing for Data Analysis, taught by Dr. Roger Peng from the Johns Hopkins Bloomberg School of Public Health. This course was focused on learning how to use R for data analysis. R is a statistical and graphical programming language and software package, and it is probably the hottest thing for researchers around the world who do statistical analysis (and it's free!).
  • An Introduction to Interactive Programming in Python, co-taught by Drs. Joe Warren, Scott Rixner, John Greiner, and Stephen Wong from Rice University. This course was focused on learning how to use Python to build simple, interactive applications. Python is a "high level" programming language (some would say a 'scripting' language) that is used by a wide variety of organizations - from Google and Yahoo to NASA - for an incredible variety of computer applications, especially online. Python is also the primary programming language for interacting with ArcGIS, an industry-standard Geographic Information System (GIS) that I teach and use for research.
  • Data Analysis, taught by Dr. Jeff Leek from the Johns Hopkins Bloomberg School of Public Health. This course was focused on learning how to apply statistics for data analysis using R. When I enrolled, I did not realize that Professors Peng and Leek were from the same institution, but as it turned out, they structured their courses to be complementary. In fact, it was recommended that students enrolling in Dr. Leek's course either have experience with R or else take Dr. Peng's class first.

Summarizing the Coursera experience

The courses all followed a common organizational scheme, which apparently reflects Coursera's pedagogical philosophy. Each course lasted from 4 to 8 weeks.

Lecture

Each week, the instructors posted roughly two hours worth of video lectures, broken up in to 15 to 20 minute segments. These videos were viewable through the course website, on YouTube, or could be downloaded to a personal computer and played offline. Videos were closed captioned, and transcripts of these videos, along with PowerPoint slides and other learning aids, were made available for download.

Testing

Each week, there was a quiz. Quizzes were multiple choice and graded automatically with immediate feedback on the score achieved, as well as what questions I got right and wrong. In most cases, I could retake the quiz up to three times in order improve my score. Maybe "quiz" is the wrong word. Although they were generally 5 to 10 questions at most, and multiple choice, these were not the kinds of quizzes you could just bang through. Each quiz, regardless of the class, took me a minimum of two hours to complete. Seriously.

Assignments and evaluation

There were open-ended exercises or projects too. These ranged from writing small programs that performed some specific task, to doing statistical analyses and writing up reports. These non-quiz assignments required evaluation by a living person. This is where the technology and technique of MOOCs really becomes apparent. Each of these courses had 30,000 to 100,000 people enrolled. I kid you not. Open-ended evaluation of assignments is done through peer evaluation. After submitting a given assignment (before the deadline had passed), my next task was to evaluate from 4 to 5 assignments submitted by my classmates using a rubric form. I was given the chance to look each one over first, and then I went through one by one to rate how well they met each of the rubric items, and in some cases, to provide written feedback on why I thought they had done something imperfectly or really well. After reviewing the assignments submitted by my peers, I then evaluated my own submission. In order to receive a grade for an assignment, I had to evaluate my peers - before the evaluation deadline (usually a week after the submission deadline). Apparently, research on peer assessment and crowd sourcing have shown that these methods can provide accurate feedback and assessment, as well as being a meaningful learning experience. I don't know if the instructors (or their TAs) checked to see if the peer assessments were accurate, but I'm hoping so.

Classroom interaction

Personal interaction and discussion throughout the class were conducted through forums or message boards where participants posted questions and observations and responded to those posts. The forums are a core part of these online courses and crucial to a successful class experience. Forums are the places where you meet your classmates, (potentially) have an opportunity to have a personal interaction with the instructor, and where you raise your hand (metaphorically speaking) to ask a question or offer an opinion or answer a question. Because it is unlikely, if not impossible, for the instructor to answer individual questions, it really falls to the participants in these fora to support each other. I can say with complete honesty that I would not have passed a couple of these courses without the help of my fellow classmates through these fora.

Results may vary

While all the courses followed a common format, the quality of the experiences was not the same. This shouldn't be surprising; online courses can vary in quality just as much as face-to-face courses. I recently read an ironic (if not sadly funny) story about an online course to teach participants how to run online courses that went seriously awry. As the author of that story pointed out, it was only one instance and did not prove anything. However, it does seem to undermine the idea that online courses will simply steamroll over traditional education by way of simple standardization or a universal solution to teaching. Online courses can suck just as much as in-class courses. So there. Of course, I have never heard of an in-class course at my university being canceled in midstream in order to make adjustments. Usually someone has to be physically incapacitated or die (usually the instructor). But I digress.

Lessons learned

Here are some of the lessons I learned from my experience with Coursera and about online teaching:

  • Online courses are no easier than in-class courses. Sure, you can 'attend' class while in your pajamas at home, but that's about the only easier part.
  • Online classes require real time self-discipline from students (and instructors). Because online courses are asynchronous, you don't need to sit in a classroom at a specified time during the week, but you do need to figure out how to set aside adequate time for reviewing videos, reading literature, taking quizzes or tests, completing homework, accomplishing final projects, etc. Without time discipline, you will suffer and perform poorly and you might not learn much. Everyone must understand this before beginning an online course. I spent on average 15 - 20 hours a week on each course - maybe I'm just slow.
  • "Mastery learning" works. Give students immediate and meaningful feedback immediately AND give them the opportunity to re-attempt homework or quizzes. This is most easily accomplished with the automatically graded quizzes, but it still requires forethought on the part of the instructors to provide useful feedback for incorrect answers as well as devising multiple opportunities to try again. Kudos to the crew from Rice University and Dr. Peng on this one.
  • Peer assessment can be a real time saver for grading and it can be a meaningful experience IF the evaluation rubric is set up well. The idea of a grading rubric is simple enough, but it has to be written clearly enough so that anyone (not just the person who wrote it) can understand how to apply it. It helps to provide examples of what constitutes a "superb" versus a "good" analysis (although even those descriptors are so vague as to be useless). A better approach is to clearly identify what qualifies as a minimum level of quality for a given rating (e.g. "The methodology contains these three elements ..."). Also, peers must be able to submit written comments in addition to standardized ratings in order to explain their ratings. It's extremely frustrating to be downgraded for something and not know why or how to improve it. Again, kudos to Rice University and Dr. Peng.
  • Narrated PowerPoint presentation lectures are still just PowerPoint presentations - only less exciting. Most of the video lectures in all three courses were essentially narrated PowerPoint presentations. However, the crew at Rice University did things which relieved some of the lecture tedium and, in some cases, actually made lectures fun. They varied the video lectures between still shots of PowerPoint slides, action shots of their computer screen as they worked through problems, action video of them talking and interacting with each other, and even regular gag skits. The variety in content and approach, and the regular movement on the screen really made a difference in helping to hold my attention. I especially liked the way they were able to integrate these little inset videos showing them talking while the larger image displayed either a slide or the speaker's screen. Good use of the technology and excellent co-teaching. Online instructors, are you listening?
  • Interaction between students is crucial. I found out the hard way that you MUST read through the fora regularly to see what other people are talking about. Lectures and reading material cannot (and should not) make all of the intellectual connections or provide all of the answers. Participating in interactive fora is like participating in critical group activity - you end up feeling smarter and accomplishing more than you would on your lonesome. I was also really impressed by the degree of support that students were willing to give one another (without giving away the answers!). And it is amazing how helpful it can be when students simply share an "aha!" moment. However, the value of these fora, and how to properly engage in them, may not be apparent to everyone. Students must be told how to effectively get help and help themselves in this brave, new online world. Dr. Peng did an excellent job of explaining how to get help and ask questions in an online environment and it's definitely worth watching this lecture.

Although it ate up six months of my life, I am really glad that I took the time to experience how MOOCs work. I have brushed up on some technical skills. These experiences have also helped me, as a college professor, to think about my approach to online teaching and to teaching in general. Whatever else they may prove to be, MOOCs are an opportunity for faculty to learn from other faculty.