Dis-Integrating the LMS

Key Takeaways

I know what you’re thinking after reading the title for this column. Take a page from Jack Bauer: Put a little C-4 in your data center, right next to the server farm for your LMS. Add an ignition switch, connect to the explosive, get to a safe distance, then push the button. Kablooey! Admit it; you’re smiling. How much stress is involved in supporting the LMS? Getting rid of that in an instant sounds good. Enjoy the moment, but know we have to move on. That’s not what this column is about.

Many CIOs regard the LMS as the academic counterpart of the ERP. However, there is a basic difference between the two. The activities the ERP supports are fundamentally transactional: Does the student’s course grade get reported? Does the employee get reimbursed for campus-supported travel? The ERP affects the ease with which transactions occur and how the data so generated are used subsequently, determining the ROI from the ERP. The ERP does not affect the nature of the transactions themselves.

It is different with the LMS. Teaching and learning are not fundamentally transactional. An instructor’s assignment can be highly illuminating; alternatively, it might be busy work. Either way, the work must be turned in for the student to receive credit. Clearly, the transaction itself does not discriminate by the learning the assignment engenders. The LMS does have potential to influence the nature of the work itself, for example through how it affects student participation and instructor response, partly determining ROI for the LMS. Dis-integration is a way to increase ROI by strengthening this effect.

Think back to teaching with the web before the LMS. Ten years ago in a large section of Intermediate Microeconomics I used three different web-based applications. For each lecture there were online quizzes to be done in Mallard, a sophisticated quiz engine. I kept my grade book in Excel, but Mallard had a way to display a CSV file so that students could see their information. The next software was WebBoard. I had the students divided into teams; each got their own board for submitting the longer homework and getting feedback from the TAs. A text chat feature supported online office hours. The final application was FrontPage, where I published a class website with syllabus, lecture notes, old exams, and other course information. Dial-up was still the rule, so most of the content was pretty Spartan. It was quite functional, nonetheless, and the various pieces fit together nicely to form an organic whole.

Soon thereafter my campus formed a Center for Educational Technologies. In addition to Mallard and WebBoard, the CET supported CourseInfo and WebCT (now part of Blackboard), RealAudio streaming, and customized web pages. At about the same time my campus created a CIO position, with the main academic computing organization and CET reporting to the CIO. With that I began my education of IT more broadly. We had not yet moved to an ERP. There was a lot of discussion about best of breed versus integrated solutions. Having essentially no experience with administrative computing applications, I had no feel for this argument as it pertained there. But for teaching and learning apps, I was definitely a best-of-breed kind of guy; with multiple tools it’s easier to match the tool to the task, and good tools make for a better fit.

Over the next several years we moved toward an integrated approach with the LMS. The reasons were many and varied. Mostly it was not by design, though when it became time to move to an enterprise LMS, there was considerable planning. Scaling drove much of this move. Then, too, there was a need to make the supported software more secure. And students wanted one URL for all their courses. All these reasons are valid, and moving to an integrated approach helped us address them.

In case it is not obvious, however, none of these reasons deal directly with learning and how technology might facilitate that. Further, the late majority faculty who were fueling the growth we witnessed were themselves not inclined to use the LMS intensively. Much of the use was content push—distributing the syllabus and PowerPoint presentations to students. (This is documented in Glenda Morgan’s ECAR study of LMS use and reaffirmed in the various ECAR studies of undergraduate student technology use.) I will posit an underlying explanation for why this happened.

For long-time instructors there can be considerable lock-in to a teaching approach that features periodic revision of the syllabus, say when a new textbook is adopted, at which time topic coverage may change and there may also be revision of lectures, assignments, and exams. The lecture mode persists, however, and there is no desire on the part of the instructor to make other changes in the interim between syllabus revisions. Viewed this way, the LMS serves as an affirming technology of traditional teaching. The instructor doesn’t challenge the LMS very much, and, in turn, the LMS doesn’t challenge the instructor. The student gets the convenience benefit from electronic distribution of documents (and grades) but little more.

Consider a more experimental alternative, for example the approach posed by Nancy Chism, where she grafts Donald Schon’s Reflective Practitioner onto a semester (or quarter) system. Each offering of a course is an experiment of something new to try, while the implementation provides fodder for reflection that fuels the next experiment. This cycle of experiment, experience, reflection, and further experiment is something we should encourage, because it will lead to better learning for students and instructors alike. The direct evidence for this assertion may be slight (how do people outside the course know whether deep learning occurs?), but Atul Gawande’s book Better provides a convincing case for the analogous argument when applied to medicine, where objective comparisons of performance are available. If the argument works there, why shouldn’t it also work for teaching and learning?

Armed with that insight, an initial impulse might be to return to the best-of-breed approach of the 1990s as a way to foster experimentation, but You Can’t Go Home Again, certainly not with a built-out LMS that has tools with parallels in many other applications. Here I’m arguing for dis-integration through the behavior of instructors and students who opt to use freely available alternatives for some of their needs. For example, instructors might put their PowerPoint presentations into SlideShare and embed them in a discussion forum in the LMS so that the presentation itself can be viewed online, thereby encouraging students to comment on it. Or, instructors might ask students to submit their assignments via a link to an online document in Buzzword, where there is no need to download or re-upload to make comments on the paper, hence the electronic communication should encourage revision according to the precepts of Writing Across the Curriculum. In other words, dis-integration via instructor and student behavior is a natural path to the experiments that Chism advocates.

CIOs might feel uncomfortable advocating for behavioral approaches. In the past the CIO role has been to identify and support well those IT systems the campus embraces. Behavior has been a matter left to the users. While the reluctance is understandable, it’s time to get into the game. EDUCAUSE, for one, now embraces the slogan, “It’s not about information. Or technology. It’s what we do with IT that counts.”

Still feel ill at ease? Perhaps this will help: Think of instruction as highly stylized collaboration. It may be unnerving to tell instructors how they should teach, yet all of us in higher ed collaborate. Shouldn’t CIOs be in the business of learning about the most effective ways to collaborate online and then advocating for those approaches? Indeed, shouldn’t CIOs be doing their own versions of ongoing experiments to foster such learning for themselves and their organizations? Once you walk the walk, it’s not hard to talk the talk and encourage others to do likewise.

Remember, the game is to get the Tragic Tories to accept change by spending some of your own reputational capital. This is how it’s done. If it works with many courses embracing the dis-integrated approach, perhaps the LMS developers will be content to focus on improving those tools in the LMS that continue to get substantial use. Wouldn’t that be nice?