Sunday, April 21, 2013
Not a Slow News Week
Meanwhile, the AACC conference is going on in San Francisco. I had hoped to make it, but it just wasn’t in the cards this year. Wise and worldly readers who are there, I’m relying on you...
I'm reading your book now and have not yet reached the in-depth discussion about the credit hour. I do wonder, though, if there are some analogies in graduate education. No rule states that a PhD must take N years, for example, and productive students can finish more quickly than the average student. Undergraduates are certainly at a different stage in terms of independent learning and mentoring. Perhaps some relationships can be drawn?
HS lab partner: My own experience in Computer Science graduate school was that increased productivity didn't seem to result in finishing sooner. Again, we seem wedded to time-based criteria, and so if you were in theory, you usually finished in about 4 years; if you were in systems, it was more like 8 years. What changed was the number of publications you had at the end of graduate school.
Competency based college courses go back over 60 years. There is even a company that makes evaluating competency in specific courses a major part of their business.
Back then, it was known as the Keller Plan, named for Fred S. Keller, who, along with collaborators, came up with the basic idea. The Keller plan was in turn derived from the behaviorist theories of learning originated by B. F. Skinner, a professor of psychology at Harvard University.
Basically, the Keller plan eliminates formal classroom lectures and breaks the course down into several separate units (known as modules). Each module is a separate, meaningful unit, and each unit has a set of specific learning objectives written for it. The theory is that if you can’t define specific learning objectives that list what the student is expected to be able to do when they have finished the module, then you certainly can’t determine whether or not students have understood the material.
Instead of attending lectures, the student studies written texts and works out exercises that are specified in the module. When the student feels ready, they present themselves to a proctor and take a mastery exam to demonstrate whether or not they have met the learning objectives for the module. In order to pass the mastery exam, the student must demonstrate virtually perfect performance, and must show that they have no conceptual errors. In addition, they must exhibit few careless errors, since these might indicate an imperfect understanding of the material. When the student successfully demonstrates mastery, they can move on to the next module. If the student fails the mastery exam, they can repeat the mastery exam as many times as necessary to demonstrate mastery.
There is no set schedule for taking mastery exams, and the program is entirely self-paced, recognizing that students don’t all learn at the same rate. Some zip through the modules at a rapid pace, whereas others move more slowly and take more time in fully understanding the material. A student could move through a course as quickly or as slowly as they choose.
I was involved in such a program back in the day when I was teaching at Research Intensive Technological Institute. I found that not all students could adapt to such a self-paced instructional mode. Procrastination was the big problem. Since there are no formal lectures, the student often feels that they have complete freedom to goof off as they please, only to find out that the course really has a deadline and that they need to finish the modules within a set amount of time. In order to be successful in a self-paced mastery-based program, the student needs to be relatively mature, must be a good scheduler of their time, and must be able to avoid distractions and must not yield to the temptation to procrastinate.
The faculty and the administration were not entirely happy with the self-paced mastery concept. A lot of faculty felt threatened by the whole program, fearing that their formal lectures could be easily replaced by a set of modules and mastery exams administered by a bunch of graduate-student proctors, placing their jobs at risk. The administration found it difficult to deal with the grading system used by the self-paced mastery course. In conventional lecture-based courses, there is typically a bell-shaped distribution of grades, under which there are relatively few As and Fs, with most grades being Cs. However, in a self-paced mastery course, either a student demonstrated mastery of the course or they didn’t, and there were a lot of As and a lot of Fs, and no Cs. This made the administration rather nervous, and sometimes accused the self-paced course of giving out too many easy As.
I happen to be the very happy product of an experimental course based on mastery rather than just competency. It didn't last long because it was expensive (good tutorials required top faculty rather than grad students who were good at mopping up after a lecturer in a recitation section), but it was a spectacular way of learning.
Anonymous is right about it being a reprise of Programmed Learning (which I also experienced), but far more flexible and responsive when done with computers that have the right heuristics to adapt to the mistakes made by the student.