ladamic's blog research on information networks and non-researchy random musings


Coursera vs. the classroom

Filed under: CourseWare — ladamic @ 18:53

There are far more education-educated minds than mine contemplating this question, and doing so scientifically. While we’re waiting for their results, I’m just jotting down a few thoughts here based on my experiences as a student, as a professor, and as a Coursera instructor.

The only thing that is more exciting than teaching on Coursera, is taking classes on Coursera. It’s been over a decade since I took my last class. I did vaguely plan to sit in on classes at U. Michigan, with world-class faculty teaching so many subjects of interest. But I never found the time. With Coursera, at least there is the possibility that I could take such classes at any time. I think this is going to be for sure one of the main target audiences for Coursera: folks like me who have gone through lots of schooling, but there are a lot of new developments in their profession that are worth learning, or they found new subject areas of interest that they didn’t think to study back when they were at school.

For example, as an undergrad I wrote a dorky, inexcusable letter to the administration at Caltech to protest the proposed reduction in required physics coursework in favor of introducing a biology requirement. I cringe when I think I may have written something to the effect of ‘Everything can be derived from physics, so why bother with other subjects’. Though I did eventually dive into many subjects beyond physics with gusto, I never did take a biology class, not since 9th grade. What Coursera course am I signed up for now? Genetics. How wonderful to be able to go back and catch up.

I have been thinking about how my Coursera course stacks up against its offline version. I didn’t skimp on much material in the Coursera version, even though it is several weeks shorter. With no one to have a discussion with but the camera, the pace is quicker. If anything, I developed more interactive online demos and data sets to be able to better convey concepts to my invisible audience. These from my perspective worked especially well with the in-video quizzes. The lecture would pause, a quiz question would pop up, and students could play with a demo to derive the answer (technical browser glitches aside). Assignments presented challenges to me. It was difficult to transform them into a multiple choice or programmatically graded format. I couldn’t ask students to stick to the rule: “Don’t just put a number down, interpret it”. Online, a number had to be sufficient. Or the interpretation had to be one of a set of pre-prepared choices. But I did it. It may have worked. I’m not sure.

Then there was the course project. In the offline version this is a multi-step affair afforded by the longer semester format. The students propose the project, I give them feedback. They turn in a mid-project report, I give them feedback. They turn in the final report, I give them feedback and for some I give advice on how to turn it into a publication. On Coursera, the students just did the project, and then graded each others’ work. The grading rubric I constructed for the peer-evaluation was very detailed and left little room for elaborate feedback. At the same time I encouraged students to post their work to the forum. The feedback there was very, very useful. Sure, my feedback is more informed by the literature, and years of my own experience and experience seeing students do such projects. But the students’ feedback had many useful tips and rather creative suggestions. It wasn’t that much worse than what I could have provided. This peer-evaluation thing just might work!

There were other interesting advantages. Since all I needed for class was my laptop, I could take it with me to visit with various researchers in industry and record sessions about the use of course material in practice. Sure, you can have invited speakers in your regular classroom, and that beats a streamed video, but… this was quite practical. The Coursera course also very kindly received technical support and instructional guidance from the creators of the software packages I was using — something that likely would have happened for my small offline course, but where I think the size of the class didn’t hurt either.

The biggest difference between the Coursera and offline versions is in how tailored the course is to the participants. In the offline version I’ll introduce material or discuss in-class applications that the students ask about. And they ask a lot. We go through the same material much more slowly, and that’s because it’s not just me talking. It’s a conversation, about the readings, concepts, knowns and unknowns. There was a lot of interesting conversation in the Coursera forum, but the lectures were pre-recorded, inflexible, and besides, I felt that if I had gone off on a tangent that one student was asking about, it wouldn’t be fair to the other several thousand. In a class of 20-30 I wouldn’t think twice about going on a tangent. One could probably interpolate between the two with Google Hangouts that could accommodate a few select detours.

As I was teaching the Coursera course I regularly received email from students about specific projects they were working on, asking for my advice. I don’t think I answered a single one. I was overwhelmed by work. But I also felt that any one of those emails would take 15-20+ minutes of my time, while I looked up the relevant literature, reviewed it, and composed my thoughts.  And I hate to say it, but my thinking was: this is free course, the students are not entitled to this much of my time individually. In contrast, at UofM students are always welcome to come talk to me during office hours and I would usually answer by email as well. It’s my job, a part of my job I usually enjoy.

Another rather big difference is in the attrition rate. I may have lost up to 1/3 of the registered students in the first week of my offline class, but it was stable thereafter. On Coursera, only 40% watched the first video, I lost many more from there on out. If I had “attended” Coursera instead of Caltech, would I have stuck with it, made it through, or would I have signed up for courses full of good intentions only to abandon them?

In many ways my college-self would have been a suitable student for Coursera. I never spoke up in class (to my students: do as I say, not as I did!). I got a pair of glasses just so that I could have better than 20/20 vision and sit in the very back of the lecture hall. I didn’t understand much of the lectures either (which were usually derivations written on the blackboard). I couldn’t write notes and listen carefully enough at the same time. I learned outside of class, by re-reading the notes and reading the textbook. I especially learned by doing the assignments. I took advantage of office hours and research opportunities. But in short, if the lectures had been recorded rather than live, it wouldn’t have made much difference to me.

But I don’t think I could have graduated without a campus. I’m a procrastinator. I took 6 classes every quarter (except the last one, when I dropped 3D photography in favor of working on my ditch-day stack). Because I knew that I wouldn’t start assignments early, I made sure I had 1 assignment to work on per night (often late into the night). It kept me busy and it kept me happy, and if I got totally stuck, I’d walk down the hall to compare notes with others working on the same assignment. Any flexibility, however, and I would unhappily procrastinate. The summer after freshman year I decided to work on the hundreds (?) of math problems needed to pass out of Math 2B (which would allow me to take Applied Math 95 sophomore year, and ultimately Kip Thorne’s general relativity senior year, which in the end I didn’t do, but anyway…). That was a miserable summer. When I wasn’t working on Math 2B I was miserable because I felt guilty, when I was working on Math 2B I was miserable and lonely. By comparison Ama95 was entertaining and I enjoyed the camaraderie of going through it with fellow students. Perhaps I would have found the same camaraderie if I had taken Math 2B through Coursera and participated on the forum, or joined a Facebook group. I don’t know. In any case, I loved my time at Caltech. Some people left it embittered. Not me though, I wouldn’t trade it for anything.

But back to Coursera. It’s really promising. It needs some work (my SNA course most of all!), but it’s mind-boggling to be able to learn about anything! Anything! What a wonderful time to be human. Now I better get back to my Duolingo Spanish lessons :).



Thanksgiving ingredient network leftovers

Filed under: data, statistics, visualization,papers — Tags: , , — ladamic @ 03:29

Michaeleen Doucleff just wrote a very fun article on our recipe network paper for NPR’s the Salt.

It made me realize that Edwin Teng, Yuru Lin and I have some leftover plots that may be Thanksgiving appropriate. If you don’t have quite the right ingredients handy while cooking Thanksgiving dinner, here is a network of common substitutions as found in reviewers’ comments on a large recipe site (click to see a larger view):

The favorite Thanksgiving ingredients are often recommended as substitutes. e.g. cranberries end up substituting for other kinds of fruits and even somehow for chocolate. In the fat category, olive oil and butter seem to be recommended as substitutes for things such as margarine. Yams are often recommended as a substitute for sweet potatoes (more so than the other way around), etc.

Recently, from my Coursera class, I created region specific networks using data shared by YY Ahn & co. in their flavor network paper. This isn’t a complete set of all regions, but see if you can guess which region is visualized in each of these (mouse over for the answer, your choices are Northern European, Southern European, North American, Latin American, Middle Eastern, South Asian, African, Southeast Asian, East Asian):


Lastly, and most deliciously, here is the network of complementary ingredients for Thanksgiving, created by my collaborator Edwin Teng:

Bon Apetit!

Powered by WordPress