Teaching Old Dogs New Tricks.....

.....What Would It Take to get the Start of the Semester Add-Drop Period Down from 10 Days to 5 Days?

In my next class session we'll spend some time on Paul David's famous paper on QWERTY.  The paper initially goes through the history of the typewriter.  In the nineteenth century the mechanism was slow and clunky.  Having one key depressed before the previous key returned to its original position would cause the keys to jam.  The inventor of the qwerty keyboard designed it as a way to slow the typist down, to reduce the number of key jams.  At the time of the invention, it was an efficiency enhancing development, which is why it took hold. But then the typewriter mechanism improved. This was followed by electric typewriters, some of which no longer had keys at all. Then there were word processors and personal computers.  Eventually there was no reason whatsoever to slow people down with typing.  Yet the qwerty keyboard has persisted, in spite of some efforts to unseat it by keyboards that are more sensible for encouraging rapid and accurate typing.

I think my own situation is typical and explains why we're locked into qwerty.  I learned touch typing in the summer after 10th grade (1970), attending a secretarial school in Flushing for that very purpose.  At the time I learned on a manual typewriter.  But those skills have survived to the present and are evident to me as I compose this blog post. I surely have a preference now for a qwerty keyboard.  I would hate any other system.  That would require me to practice hunt and peck and slow down the flow of the writing substantially.  This is why I don't like to text on my phone, even with a qwerty keyboard.  It's painful to input that way, at least for me.

The qwerty example is the best known example of lock-in to a practice that is no longer efficient, but it is certainly not the only example, not by a long shot.  The university offers several other examples. In class we will consider the lecture and ask whether it remains an efficient form of instruction or if it no longer is yet persists because of lock-in.  My views on this are somewhat eclectic.  I believe that lecture still makes sense as a mean of discourse when researchers are communicating with other researchers in the same field about their recent work.  I also believe lecture makes sense, in other settings that are closer to a general interest presentation when the audience is reasonably large, though then you might say the person is giving a talk. The first case, especially, means that much doctoral education will be in the form of lecture, so the students can prepare both to give talks about their own research, and to understand the obligation of members of the audience.  Reading a professional working paper requires diligence and keeping up with the literature.  It is effortful.  Having gone through this rigor as a doctoral student, is it any wonder that new assistant professors lecture?  It is what they were trained to do.

But for teaching undergraduates, especially when not in a large class setting, is lecture the best way for them to learn?  There are several different angles on this.  One is that making online micro-lectures is now a viable option and further, the videos can be made broadly available, so students elsewhere can view them if they so desire.  As an example, my profarvan channel on YouTube has recently gone over 250,000 views, the vast majority of which is from other than my own students.  Second, the online micro lectures can be coupled with online assessments that encourage the students to pay attention while watching the video.  Third, the research on learning shows "active learning methods" superior to lecture for most students.  Then there is my personal preference, Socratic dialog.  I like it because the sequence of question and response is the closest thing I know to thinking, and what I'm trying for in the classroom is to model thinking for the students. More than a dozen years ago, Barbara Ganley and I debated these issues on our respective blogs.  (Barbara's post that I was responding to can be found here, but it takes a while to load.)  At the time I had yet to experience the-post retirement teaching that I've now been doing for many years, where around then when I did teach it was students in the Campus Honors Program.  My sense of the student has changed as a result of this more recent teaching.  And I'm more in Barbara's camp now then I was then.

Another example of lock-in at the university that is usually given is the academic calendar, which treats the summer in an asymmetric fashion.  Once upon a time students needed to go home to tend to the harvest.   That need is no longer present, but the academic calendar persists as if it still were there. The first 10 days of the semester is the add-drop period.  While I don't know this for a fact, I do remember back to when students had to line up at the Armory to register and the process was all done on paper.  Instructors would eventually get a paper based copy of their course roster.  Adds and drops after that were printed on manila cards.  I'm guessing that the 10 days were needed then. But now course registration is done online, yet we still have these first 10 days.   Is that too an example of lock-in.  Or does it survive now simply out of institutional inertia?

Before trying to answer that, I want to note the example of Google, which has taken the approach to kill applications it had been supporting, even if the applications are popular with current users.  Google does this when the application no longer fits into their overall business plan.  Examples include Google Reader, iGoogle, and most recently Google+.  Even though I was a pretty avid user of the first two of these services, I didn't jump off the Google bandwagon after those services were cancelled.  So the lockin, if it existed in this case, might not be to a particular service or way of doing business.  Alternatively, the company might take some near term hits when it does cancel such a service, users who relied on the service are indeed upset by the cancellation, but the company can endure that and those users eventually do adjust to the new circumstance, perhaps more quickly than they had anticipated.

Now I want to consider the first 10 days of the semester and the add-drop period, where students can add or drop a class without getting permission from an instructor or an advisor.  In theory, students are not  supposed to game the system by registering for more classes than they plan to take.  I informally polled my students on that issue in my class on Tuesday, where I asked, to be cute and so my students didn't feel they needed to lie about themselves, do they know of other students who hoard classes like this, where they plan to drop at least one of them during the semester?  I only eyeballed the results, but it seems that a vast majority present raised their hands, indicating they were aware of such gaming of the system.   Then I explained that the Econ Department also apparently gamed the system.  On August 15, 11 days before the start of the semester, I did a tally of all the 400-level classes in Economics based on the listings in the online Class Schedule. At that time there were no openings whatsoever for undergraduates, while there were 11 cross-listed graduate sections that had some openings.  (I think there was 27 offerings overall.)   A week or so later, there were many openings for undergraduates in 400-level classes.  Apparently the department had held back seats for late arriving students who had yet to register, perhaps sensible under the circumstances.   Then the free-for-all began.

Because I use a class Website that is not on a campus supported application, I email the students ahead of time with the url for the site and some other information specific to each student.  (Thank you, mail merge.)  As my class is Tues and Thurs, I always make this mailing on Sunday.  I downloaded the roster for my class the day before.  At the time I had 37 students registered for my class.  Since then, I've had 14 students drop and 18 students add the class, and in one case a student dropped but then added it again later. This amount of churn is unbelievable. Is it really necessary?

From my perspective, instructors are definitely not locked into the current 10-day arrangement.  I don't speak for other instructors but I imagine that most of them would echo the sentiments in this rhyme, written back in 2011. You don't want students adding your class who have missed the opening session(s).  Catch up is hard and it places an added burden on the instructor.

This point makes it necessary to distinguish switching discussion sections for the same large lecture class, either because the new section is at a preferred time or because the student wants a different TA, from what I was describing with the 400-level classes.  In the case of switching discussion sections in a large lecture class there isn't the catching up I mentioned in the previous paragraph and the section switching probably should be accommodated, even after the 10 days are over.  Indeed, at the U of I students can add a class after the 10 days with the permission of the instructor.  So there maybe be enough flex in the system as is to handle this case.

Missing the first two weeks of class in an entirely new course is a different matter. Such adds do happen and I'm fearful now that I will experience some of them this time around.  It is a downer to consider the possibility.

A real study of late student add-drop behavior would entail interviewing many of the students who make such course changes to understand their situation and their motives.  Did they find some other course they were registered for too difficult or the instructor too rigid?  Do they have a friend taking the course they add late who gives them the skinny on the new course, something they couldn't get from the available information to all students (course syllabi)?  Did they make some bad mistake in their previous class and would prefer not to be accountable for that so want to switch classes to start with a new slate?  Or was it just sloth and they couldn't figure out the right courses to register for the previous semester under the normal scheme where students register for classes?  Who knows?  Some data gathering might shed some light on this.

Maybe an efficiency argument can be made for enabling adds through day 10.  (The last day students can drop a class is much later than that.) But I doubt it.  Further,  there clearly are a lot of bad things that result from the churn.  Number one of these is that instructors have to adjust their teaching during the first two weeks of the semester so that late adding students can catch up.  Students who are in the class on day one are done a disservice by this practice.

Once the appropriate study has been done, the issue of shortening the duration of the add-drop period becomes a matter of campus politics.  I wouldn't want to speculate on whether it's possible to do something in this dimension or not.  But I suspect that if another peer institution were to do this, that would make it much easier for us to do it here.  The arguments I've been making are not specific to my campus.  Somebody should have a crack at this, please.