Building a Teaching and Learning Technology lab

Thanks to our always-helpful government grants office, we were able to find money to develop a technology lab for teaching and learning. We brainstormed for a way to augment our existing installations, which already provide a lot of technology for course instructors. Given our new foray into online learning, lecture capture seemed the best bet--it's focused, manageable under our budget, and can give us an immediate payoff. It's also scalable in application from simple video recording to much more sophisticated productions. It will shine a light on our distribution system for electronic media too, and prompt infrastructure improvements that will have multiplier effects.

"Lecture capture" is of course a terrible name for it. Lecturing means literally reading, which is what our academic forbears did before there were enough books to go around. Like wearing those heavy black robes in the May sun, some ideas are ripe for overhaul.

There's a nice introduction by Educause here (pdf), if you want to see what they have to say.

What if we started from scratch?


It's not just the robes and readings that have seeped into the weave of the academic tapestry. Exposing young minds to the conceptual treasures of our disciplines is constrained by the combinatorics of time and space so that that certain optimizations are so standardized we don't think about them. Like having students sit for fifty minutes at a stretch in classrooms. What if instead we could engage them for ten minutes at a time on focused topics all day long? Another optimization is that all students in the class get the same linear flow of narration, and the discipline itself gets strung out into a one-dimensional string, codified into textbooks and syllabi that have the same linear style.

Imagine if the world wide web were like that: a single linked list of information, so that you had to start at the beginning and work your way forward link by link until you got where you wanted to go. It's a lousy way to organize rich information.

Sometimes concepts do have prerequisites. Math is full of it. But even in math there are optional topics, and opportunities to go into more depth even the simplest subject. If you want, you can think of the long multiplication process you learned in third grade as a convolution product, which can be optimized with Fourier transforms. One of my professors in graduate school (David Kammler) took an approach he compared to "Island hopping" as practiced in the Pacific theater in WWII, which consisted of learning certain important topics in depth and assuming that students could then make logical leaps to nearby topics, which could be skipped or treated as exercises.

Imagine instead of a linear trip through a discipline, we captured key ideas and used those as anchors for a network of related facts, concepts, processes, etc. One immediate advantage would be the ability to easily update the network as the discipline changes. Group evolution is back in vogue? Add a node and stick it in there. No need to fell a forest for the new edition of a book. Another advantage is that focused topics can be crowdsourced, as in Wikipedia, with experts in sub-sub-sub-fields filling in blanks ad infinitum. Other thoughts:

• interdisciplinary connections become easier to integrate naturally
• because learning happens in detail, it's easier to assess a small topic than a big one, allowing for faster improvement of the captures and related pedagogy
• custom methods can be applied more easily to particular topics. Think apps. When you get to the bit about singular value decompositions, you can download the stuff you need to see it in action, and then it can go away. It's modular.
• It creates a way for professionals to agree on something. A whole textbook? Never happen. But a five minute tour of Darwinian evolution? Better odds. This standardizes at the right level. You can still customize your curriculum--just match assessments to content.

All of this is probably being done in bits and pieces now, this divide and conquer approach. But it hasn't gelled yet into a common framework supported by real infrastructure and common culture (like say Wikipedia is).

I don't want to go all Utopian here, but let me push this one step further. With a rich network of high quality instruction/assessment/activities modules, the whole idea of a course becomes outdated. I know, I know, that's scary stuff. But if we don't have to have students sitting in desks for fifty minutes at a time, why do we need them for 15 weeks? All that really matters is that they learn and demonstrate mastery of concepts that we think are important. So then our whole bureaucracy can go hard a-port and do something really interesting. All that bricks and mortarboard infrastructure can be focused on motivating students. I've written enough about that already, so I'll stop here.

I should hasten to add that our modest project at my institution has none of these grand ambitions.  I will be selling the idea that a "lecture capture" doesn't need to be 50 minutes long, however, and keeping my eye on the horizon...

Other Reading

Here are some noteworthy articles that speak for themselves:

Strategic enrollment planning:

From Noel-Levitz (papers and reports):

• Six Essentials—and Six Common Mistakes—in Cabinet-Level Strategic Enrollment Planning
• 2009 Student Retention Practices and Strategies at Four-Year and Two-Year Institutions
• 2009 Student Recruitment Practices and Strategies at Four-Year and Two-Year Institutions

On assessment:

• Do college student surveys have any validity? (not as much as you think) (Update: response from NSSE at InsideHigherEd)
  
• Where is the Student Voice in Assessment? (eportfolios use and misuse)

Educational Technology

• The 2009 21st-Century Campus Report:Defining the Vision (article about the report)

Open Courses

A technology article in The Chronicle yesterday describes the Obama administration's plan to develop open courses and give them away. This is a fascinating idea--the sort of thing that I mused about in the post "Unplanned Obsolescence" a few weeks ago. Carnegie Mellon's Open Learning Initiative is given as the model for this plan. The delivery of the course is potentially fully online, but example cited is a hybrid course with content developed by a group of experts and leveraging online capabilities for what I would call the low-complexity part. This is a natural way to optimize the delivery of instruction, and you already see it with course packs delivered from text-book publishers. Rote learning, or learning low-complexity modes of thought can be easily and tirelessly done through computer training. It's potentially more fun too. The Rosetta Stone language software is like this, using instant feedback, images and sound to train vocabulary (my wife used it to learn some Arabic). Where computers fail is at answering open-ended questions. If you've ever called your bank with an odd problem and had to navigate the automated phone tree (press one if you're still breathing...) you know how frustrating it is to try to solve high complexity problems with low complexity tools. So the need for actual instructors still exists, but their time could be applied more usefully to high-complexity tasks. This effect is noted in the article:

Carnegie's materials have already changed how Logan Stark's professor at California Polytechnic State University approaches her widely feared biochemistry-for-nonmajors class. Anya L. Goodman used to work from a prepared lecture, starting with the basics so she didn't lose anyone. Now she puts the burden on students to learn the basics online. She focuses class time on clearing up misconceptions, applying the materials to real life, and working in small groups.

The idea has merit, but there are certainly some big problems to overcome. If the courseware project is run and funded by the government, it may be open, but it surely won't be cheap. How, then do the course materials get updated? This isn't too big of a problem with, say, Euclidean Geometry, but for something like Finance, I imagine the books get re-written all the time. The article supposes that this might continue to be funded by the government, but this doesn't sound like a wonderful idea to me. It would be far better, methinks, if a culture evolved similar to the open source software movement. Rather than a set of "perfect" open courses designed and maintained centrally, a whole ecology of work collocated in some natural place--analogous to sourceforge.net--could grow and evolve, tagged with comments and other metadata. This depends on willing practitioners doing all the work. Faculty members taking the time to update old materials, probably. It seems unlikely on the face of it, but somehow it works for software. It works for Wikipedia.

A really interesting question is how course assessment ties into the courseware. Would it be developed and delivered in parallel, integrated with the materials? Or will assessment remain a second-thought tack-on for another decade? But if it is to be integrated, then the learning objectives need to be clear. There seems to be an opportunity for the assessment profession to get involved with this train before it leaves the station. Polish up your resume.

In a recent report I asked for, a group of twenty-plus institutions like ours had an average total expenditure on instruction-related items of 37% of total budget. We might suppose then that the asymptotic limit for reducing administration (assuming that all academic support is free somehow, libraries and such) is a 63% reduction in the cost of delivering programs. The cost of instruction could be reduced too, if the low-complexity components are off-loaded to technology. Moreover, competition in the fluid digital domain would tend to force prices down. I don't think it's unrealistic to estimate that a bachelor's degree could be delivered for about 25% of what it costs at a traditional college now.

Marshal Smith, senior counselor to the Secretary of Education seems to be the the guiding light behind this open ed plan. You can read his ideas if you have a subscription to Science in his article "Opening Education." You can also browse MIT's version of open courseware here.