Sunday, May 13, 2007

Why do great teaching ideas spread so slowly? How can we speed that up?

Steve Gilbert, Flora McMartin (a TLT Group senior consultant) and I recently finished a $200K external evaluation for MIT and Microsoft. We studied five well-funded faculty-led development projects from the iCampus program.
  • web services enabling students to use the Internet to control distant laboratory equipment in order to do experiments (iLabs).
  • an interactive, visual approach to learning introductory physics (TEAL),
  • a way to legally use video clips when discussing films online or doing student papers (XMAS),
  • software for teaching computer science that involved sophisticated online homework checking (xTutor), and
  • a new approach to managing the large-scale assessment of student writing (iMOAT). MIT allocated about 10% of its budget ($2.5 million) for dissemination. We were asked to study factors affecting adoption of those promising innovations.
There are about 10,000 universities and colleges in the world. iCampus has been disseminating these projects for a couple years. A naive observer might predict that, by now, that tens of thousands of faculty would be beating down MIT's doors to use each of wonderful freebies. In fact, the number of adoptions of these MIT projects, while growing, is much nearer to ten than to ten thousand. Why do even the best teaching innovations spread so slowly (when they spread at all)?

The biggest barrier: typical faculty members get little preparation, little help and little reward for continually updating and improving all their courses. The best way to improve learning is not to invent your own ideas from scratch. It's to scan the world for successful ideas already tried out by like-minded colleagues teaching comparable courses. And if faculty were better prepared, supported, and rewarded they could do that.

I'll make this claim: academic programs could do much better (in all senses of 'better') if they helped their faculty become the best at a) finding and adapting best practices from peers who teach similar courses, and b) sharing their own best practices with the world.

This kind of support would produce the biggest dividends in disciplines where the field itself is changing due to advances in research, changes in the job market, etc.. In such disciplines, rapid improvement in the academic program is more likely to lead to gains in visibility, enrollment, and respect. However, because teaching options are increasing in all fields, there is no department for which this recommendation is irrelevant.

What do you think? Is continual improvement in teaching important for your program? What kinds of support can the institution and department give faculty to help them discover, screen and adapt the best new practices from peers at other institutions? Please post your comments.


  1. We have invested significant money and resources into developing an infrastructure to support research projects at our institutions. This makes sense as it is an investment to support researchers who generate significant revenue to their host universities. From a strictly business perspective the justification is clear.

    The problem is that the other missions of the university, teaching and public service, are handicapped as these activities are cost centers, not revenue generators. Yet they need a similar support infrastructure for their work, too.

    How do you convincingly make this case?

  2. Steve, I have not yet read your evaluative materials, but perhaps a piece of an answer to your oveall questions is contained in this point:

    I very much need--desperately need!--science amd technology simulations and laboratory equivalents such as ilabs for my work in Africa--and I have just learned about this MIT work because I'm on mailing list to which you've sent information on this blog.

    Perhaps, then, you need some reformulations, some variations of your overall question.

    One variant: why do some innovations find their audiences, their markets, their clienteles quickly, while others of equal quality don't?

    Your emphasis on institutions and how they might encourage teachers to search for new instructional practice makes sense, but there are other ways you might consider. And other questions.

    For example, here's a question: why am I just learning about s set of practices and a group of resources I badly need now? When did MIT start informing the world about these resources? How did they go about the "dissemination"?

    Perhaps a master question is this:What kind of institutions located where need such materials?

    If there is interest, I'd be happy to talk about who I think needs to know about these materials, and how to reach them.

    Meanwhile: how do I get to know all about them, get permission to use them?

    Steve Eskow

  3. In the UK we have found that dissemination through subject based communities works. Our Higher Education Academy Subject Network aligns faculty members in different institutions by their common interest in teaching the subject discipline - ours is the Biosciences. We can also get their pedagogic research some credit by providing a route to publish through our own e-journal. Google 'bioscience education' and we (Centre for Bioscience, the Higher Education Academy) should be the top 2 links.

  4. Great teaching ideas spread so slowly because they are often relatively unimportant for programs in my experience. In K-12, the custodial and socializing functions of schooling take precedence over learning. In higher education, certification is a higher priority for the students; beyond what Philip has noted, career-oriented activities (research, publications) often take precedence for the faculty not (just) because teaching is under-resourced as the result of being a (perceived) cost center but because teaching is often secondary or tertiary to faculty's most important concerns. In the private sector, I've encountered multiple examples in which desire to reduce costs has driven major changes in instructional delivery, whereas desire to improve quality has made little headway.

    One answer to your question is the somewhat painstaking task of locating islands within academia where interest in improving learning is paramount. This is not a very satisfying answer in the long term, however.

    Steve Eskow's post provides another answer: look outside the universe of higher education institutions (HEIs)

    However, I assume that your client is focused on the issue of disseminating within the universe of (US) HEIs. In which case, it appears that you're bumping against a larger cultural question: how to get HEIs more interested in improving the quality of learning? What incentives do they have to do so? In some ways, the issue is similar to what online learning practitioners have been encountering for years.

    I wish that I had better answers, but the previous postings suggest several good directions: provide more resources to support infrastructure and faculty professional development, target faculty on a discipline/ subject-specific basis, and identify other target audiences outside of US HEIs. Beyond that, there is a deeper cultural issue to be grappled with IMO...

  5. Steve E. questioned how it is that he's just now finding about iLabs? Getting the word out about activities within one's own university is a significant challenge. Getting it out across the planet, not just at universities but non-profits, and other communities interested in applying technology to education is even more daunting.

    We did the 'usual' routine - present at educational conferences (Educause, ELI, Sloan-C, etc.) at disciplinary conferences (IEEE, AAEE, etc.), put up a website (, got some collateral communication help from companies (e.g., Microsoft,where we're on their websites especially on MSR's site), did speaking engagements, within the US and internationally, etc. This communications work has been largely concentrated in the last 2.5 years but it's signal to noise ratio, my metaphor for getting 'heard above the din' is clearly marginal.

    Terry noted that discipline communities are important. That's absolutely the case and I admire and envy the collaboration that exists across institutions in the UK around discipline groups. Were that these were more common and visible in the US.

    The iLabs Africa work is sponsored by the Carnegie Foundation and will in the next year be building outreach efforts for both east and west Africa, with hosts from OAU (Nigeria) and University of Dar es Salaam. I'm happy talk off line with anyone interested in pursuing this.

  6. Anonymous2:15 PM

    I hate to be the one criticizing the emperor's fashion sense, but perhaps Steve (Ehrmann) is over- generalizing here. I limit my comments here to the TEAL physics material b/c that is all I have looked at. It seems to me that their failure to disseminate is because pedagogically they are not very good.

    * they are essentially passive: set one or two parameters, sit back, wait and watch what happens.

    * they fail as constructivist learning b/c of the previous point. The first thing I learned in developing simulations is that they are an unparalleled learning experience - but only for their creator. To be effective for teaching they need to take the form of a kit of parts, or a game with a clear goal.

    * they also fail as exposition b/c they don't make clear what interesting stuff is happening.

    * they are a usability nightmare. The link at the bottom of the TEAL Webpage ( is broken. I managed to find them nevertheless by clicking on Visualizations, but the applet itself was three clicks away. Once there, they are heavy downloads - 9 MB for one of the Java applets. When I developed applets for the Harvey Project our limit was a 15 sec download over a 64k b/s modem: about 100 kB.

    Going to, I did not see a lot of evidence that the TEAL material has been widely embraced, even by the other courses in the 8.02x series, except for the 8.02T (for TEAL) demo course. I suspect this has something to do with the point I am trying to make.

    Going back to Steve Ehrmann's thesis, I would argue that successful dissemination of teaching innovations requires:

    1- that they be demonstrably better than traditional alternatives (which TEAL isn't).

    2- they need to be championed by passionate faculty, not pushed from outside or from above. Faculty tend to resist such external pressures, but are likely to listen to their peers.

    3- they need to be open, so that faculty can, if they want, adapt them to their needs and their students.

    Money, in the form of grant support, often seems to have a negative effect on dissemination. It creates a division between haves and have-nots, and when the money runs out, which it always does in spite of lip service to "sustainability," the project dies a quiet death.

    These are the transformative principles I have been trying to promote at, and in this recent article:

    Sorry to be long-winded, but this issue is dear to my heart.

  7. I read with great interest the postings above. In particular, Steve Eskow's comments resonated to our need for content: "Perhaps a master question is this: What kind of institutions located where need such materials? If there is interest, I'd be happy to talk about who I think needs to know about these materials, and how to reach them."

    One way to increase utilization of the resources such as TEAL is to notify the more than 60 digital repositories when new content is available. However, the simplest means of dissemination may be to place them in a repository with appropriate metadata that allows harvesting.

    The Orange Grove, Florida's K20 Digital Repository, ( is operational. We have just begun to share resources with other state repositories in the South and are currently able to do Open Archives Initiative (OAI) harvesting. We are looking for additional repositories to harvest with. From within our repository, faculty can customize personal searches and receive notification when content is added that is within their area of interest. We welcome access to high quality digital resources that Florida faculty can incorporate into their face to face teaching, hybrid courses, and online courses.

    Our repository resources are reviewed for quality and metatagged (described) to ensure discoverability based on the IEEE Learning Object Metadata schema. International metadata standards have been adopted to ensure our interoperability with other repositories for harvesting.

    It is critical for faculty adoption and useage of the resources that they can access them from one location and/or use them from within their learning management system. Basicly, it must be S I M P L E and E A S Y to discover and use resources. It is also critical that our institutions promote the use of digital content and help faculty learn how to use it in their courses for adoption.

    As part of our FIPSE grant to create a blueprint for the creation of statewide digital repositories ( , we would be happy to discuss ways in which we could facilitate the knowledge of and dissemination of high quality content that is critical to the success of any repository. Discipline communities and their development are also of great interest.


What do you think?