My thanks to Dan and other members of The Faculty Lounge for the opportunity to return. As he mentioned, since my last visit, I have taken on a deanship, and that has led to many wonderful conversations with faculty, staff, students, alumni, and colleagues at other law schools about legal education. Dan has been kind enough to allow me to extend those conversations to this forum.
A topic that has occupied me of late is how online education might work in legal education. David Thomson's wonderful short book Law School 2.0 transformed my thinking on this front, and it propelled me to teach both hybrid and fully online courses. One of the key insights, which is at the heart of the flipped classroom movement, is that by using the online environment to deliver certain basic material and information, the instructor can spend class time on more valuable activities. In other words, faculty members can maximize the value of the time they spend with students. Plus, if students receive release time for material covered online, a faculty member can cover more credit hours with fewer contact hours, which could be seen as a gain in productivity. The faculty member could then devote this additional teaching activities, performing additional service, or conducting research.
With all of this on my mind, I was quite interested to read a recent article in the New England Journal of Medecine that discussed how the so-called IT productivity paradox might apply to greater use of technology in providing health care services. The article has this great summary of the paradox:
During the 1970s and 1980s, the computing capacity of the U.S. economy increased more than a hundredfold while the rate of productivity growth fell dramatically to less than half the rate of the preceding 25 years. The relationship between the rapid increase in IT use and the simultaneous slowdown in productivity became widely known as the “IT productivity paradox,” and economists debated whether investing billions of dollars in IT was worthwhile. The Nobel laureate economist Robert Solow observed in 1987 that “you can see the computer age everywhere but in the productivity statistics.”
The article notes three main reasons that reduced productivity was observed in the aftermath of dramatically increased IT: (1) mismeasurement of productivity, (2) poor useability of IT, and (3) poor implementation of IT. I'll take the first of these in this post, leaving the other two to subsequent posts.
Mismeasurement error can occur when there is an improvement in production that is not captured in the then-existing metric of productivity. For example, when electricity was first used for outdoor lighting and public transit, it allowed increases in quality of service, such as brighter lighting and shorter trip times. While these improvements were of value, they were not captured by the productivity measures of the day. Paul A. David, The Dynamo and the Computer: An Historical Perspective on the Modern Productivity Paradox, 80 Am. Econ. Rev. 355 (1990). This mismeasure could lead to underestimating the value of the technological improvement.
I wonder whether we may face the same issue as technology makes e-learning more widespread. Higher education has a very difficult time judging productivity as it is, largely because it is hard to agree on what output we should be measuring. McPherson, Schapiro, and Winston, Paying the Piper: Productivity, Incentives, and Financing in U.S. Higher Education. For example, we ought to be squeamish about defining the output purely as the number of degrees conferred because that measure takes no account of the quality or learning outcomes of the education.
If higher education has such productivity issues generally, is there hope of measuring the return on investments in the technology of e-learning? Yes and no. No, because the productivity added by e-learning would seem to face the same obstacles as measuring productivity in higher ed generally. Yes, because there have been some attempts to compare the learning outcomes in e-learning courses to in-person courses. See here and here. So while absolute productivity claims may prove difficult, relative measures may be possible.