Saturday, April 17, 2010

Outsourcing Lectures

Interesting stuff in the NYT:
Some imagine a situation in which the bulk of introductory course materials are online, as videos or interactive environments; students engage with the material when convenient and show up only for smaller seminars. “In an on-demand environment, they’re thinking, ‘Do we really need to show up face to face at 8 a.m. with 500 other students to take Psychology 101?’ ” Mr. Schonfeld says. [...]

Mr. Schonfeld sees still more potential in “unbundling” the four elements of educating: design of a course, delivery of that course, delivery of credit and delivery of a degree. “Traditionally, they’ve all lived in the same institutional setting.” Must all four continue to live together, or can one or more be outsourced?

It's a good question. Large lecture classes are sufficiently impersonal and non-interactive that it'd make little or no difference whether you're there in person or just watching it on video. But of course some professors are much better at lecturing than others. These two points together suggest that the vast majority of large intro classes could actually be vastly improved by replacing their local lectures with links to videos of world-class lectures. The role of the local college would then merely be to provide the Teaching Assistants to (i) run small precept sections where students can discuss the lecture material, ask questions, etc., and (ii) grade assignments.

One possible obstacle to such reform, at least at high-prestige schools/departments, is that students may want to affiliate with famous professors (or "high status folks", as Robin would say), even if they aren't the best teachers. But that still leaves plenty of large lecture courses currently being led by people who are neither famous professors nor brilliant teachers. It seems that these, at least, would benefit from outsourcing the lectures.

Some hope to implement a 'proof of concept' by building alternative educational institutions rather than reforming the existing mainstream:
Edupunks — the term for high-tech do-it-yourself educators who skirt traditional structures — are piloting wiki-type U’s that stitch together open course material from many institutions and combine it with student-to-student interaction. [...] “Having a degree is a signal,” she says. “It’s a signal to employers that you’ve passed a certain bar.” Here’s the radical part: Ms. Paharia doesn’t think degrees are necessary. P2PU is working to come up with alternative signals that indicate to potential employers that an individual is a good thinker and has the skills he or she claims to have — maybe a written report or an online portfolio.

I'd like for this to work, but I suspect it won't, at least if Bryan Caplan is right that employers care about college degrees as a signal of conformity and not just intelligence. Home college makes one look 'weird', even if you can produce high-quality academic work. So I suspect that this is another one of those cases (like open access publishing) where reforming established institutions is more likely to lead to lasting success than is attempting to create new ones.


  1. Here in the UK we have the world's most successful 'distance learning' institution, the Open University. This page is a good introduction; The OU is now in its 41st year and has more undergraduate students than any other single university in the country (Oxford and Cambridge included). Because students study part-time away from the sausage-machine environment of most conventional campuses their efforts in obtaining a degree are strongly welcomed by prospective employers and the academic world alike. A typical bachelor's honours degree will take six years by part-time study, but those six years can be extended over an almost indefinite period to suit the work / study needs of prospective students. I have graduated this year (and I'm off to Cardiff to attend the ceremony next Saturday) after starting with the OU in 1987. How's that for an undergraduate programme?!

  2. I say we release high-quality digital teaching modules under an open license, and pay for top talent to make them. This teaching software would have video lectures inside an interactive "textbook" with manipulable simulation applets to illustrate whatever concept is being discussed, like the flow of electrons through the p/n junction of a transistor. The applets would simulate, in a video-game-like setting, the lab component of a university course. In many ways, this would be not as good as an actual lab, but it could also be more fun. I would love to let students operate simulated equipment like x-ray lasers, particle accelerators and space probe thrusters.

    The content would be designed by a group of the field's experts, and performed by one or several of these. Lectures would allow students many opportunities to interrupt and seek more detail on the concept discussed, either in text, audio or even a secondary "tutor" lecture. There would be a wiki and forum for each "chapter" of each course, where students get help from each other and kind experts. The latter would not be paid, but certainly the programmers, designers, authors and lecturers would be. Where would the money come from?

    We academics to do lots of stuff for free: Publish, review, present, etc.. It's an honor to just contribute, and we’re compensated indirectly through tenure and promotion. I can imagine that being selected as one of the top experts to design a module on, say, Spinoza would be considered a substantial career achievement.

    Still, many aspects of interactive digital courseware would cost real money. Since it's essential that this expensive content would be distributed freely and widely, the investors who fronted the money would be giving away their crown jewels. At first, I thought that grants from governments, universities and philanthropists might cover the costs, and they would still play a role, but the real business model would be based around testing. The idea is actually simple. Many students who work their way through a course might be interested in receiving a certification, to "signal" that they indeed know the material. This would be done much like the GRE, in a supervised classroom somewhere near them, with ID verification and other cheating-prevention measures. The GRE costs about $100; this would cost more, but far less than a similar course in a junior college. Why pay to prove you know physical anthropology and inorganic chemistry? For transferable college credit. US colleges allow students to transfer course credits they earn elsewhere. If rigorous studies could show that the quality of subject mastery of a student who passed one of these tests in no worse than someone with a C from junior college, universities could not possibly justify accepting transfer credits from one but not the other.

    Also, many talented high school students could easily handle college-level work, and if they built up a portfolio of three or four completed courses, their UCLA application would gain heft. Other students would benefit from setting their own pace, others live outside the US but want US credits...

    Every student would have a publicly-visible page on the testing agency's website which proves to everyone which subjects they have mastered, and there would be a link to this on every resume. So the tests would be in demand and could pay for most of the content, which is not to minimize the role of the other potential sources of funding mentioned earlier. This is a business plan that starts with giving away something expensive and yet may end with profit.

    It would give anyone what I believe would be the equivalent of a junior college education (eventually more), irrespective of their location, wealth and schedule. Sure, many of these people would get even further ahead if they get certified, but the knowledge would be there either way. This would be an amazing gift to the world. In the age of AI, wikis, Google and WoW, online teaching should be much richer and cheaper.


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)

Note: only a member of this blog may post a comment.