Path: utzoo!utgpu!tmsoft!mason
From: mason@tmsoft.UUCP (Dave Mason)
Newsgroups: comp.edu
Subject: Re: Ph.D.'s and Teaching
Message-ID: <261@tmsoft.UUCP>
Date: 16 Jan 88 04:13:01 GMT
References: <2144@uvacs.CS.VIRGINIA.EDU> <115@mccc.UUCP> <3469@umn-cs.cs.umn.edu> <429@sdcc15.UUCP>
Reply-To: mason@tmsoft.UUCP (Dave Mason)
Followup-To: comp.edu
Organization: Ryerson Polytechnical Institute, Toronto, Ontario
Lines: 74

I thought I'd throw my $.02 worth into this discussion.  It turns out
to be a rather long $.02 worth, but the gist of it is: it's not clear
to me that a teacher education course is going to help.

I've been teaching Applied Computer Science for 6.5 years at Ryerson,
an undergraduate university/polytechnic.  Normally I teach 3rd & 4th
year students.  (I'm teaching 1st & 2nd this year, but that's a whole
other story for another posting, another day.)

I'm a good, but not great, teacher.  I have difficulty understanding
why my students don't understand my statements/explanations.  Despite
this I get a lot of respect and appreciation from my students.  I see
my job as to challenge the students, to push them where they wouldn't
otherwise go (and judging by the moans and groans, I accomplish this :-).
I run very interactive classes: in my compiler, graphics, and
operating systems courses, I typically have 50-60 students in a
lecture section, but insist on 15-20 per lab section so that the
students can feel more comfortable to ask the questions that spark the
learning: I set the challenges, provide the basic underpinnings of the
facts & drag them kicking & screaming into asking the questions that
I'm there to answer to fill in the gaps.  This lets the student build
the internal models that are most appropriate for them.  I am
constantly on the lookout for incorrect model building (i.e. the
student asks x, I answer y, the student asks z, which lets me see that
they are building the wrong model around the facts y, so I go back &
correct their model).  This is harder for the student... much easier
to accept a model that someone else has structured; but also much
easier to forget the foreign model... and once they've built their own
model, they SHOULD have better intuition about the problem.

  I frequently am in the situation where there are no questions, but I'm
sure it's because no one has a clue what's going on.  My experience is
that this is invariably a case of peer pressure: no one wants to look
stupid (though it's often claimed to be that they don't know where to
start, I've rarely found this to be the case...they just think the
question they want to ask is too embarassingly trivial/simple).  (This
is why I have the small lab sections...fewer people means less peer
pressure.)  My usual response is to say something like: "If you have a
question, ask it cuz there's probably several other people who have the
same question, and are just too shy to ask."  This usually gets a
question, which leads to another & we make some progress. 

I have no formal training in teaching, but the first couple of
years there were some workshops that I attended, although I don't
think they were particularly helpful.  They were taught by someone who
used all the pedagogically sound principles, like `outline, describe,
summary', lots of spiffy overheads, etc., and I thought they were
PRETTY BORING.  I like a VERY loose teaching style... maybe some notes,
or a few minutes before class (often in the halls on the way to class :-)
thinking about what I want to cover, then a willingness to go on a
tangent if it seems helpful in class.  When this doesn't work, it leads
to disorganised classes, and wasted time.  When it does work, it
produces effective, interesting, alive presentations.  (The trick of
course being to have more of the latter classes than the former. :-)

  Today for example, I had a fairly (unusually :-) solid idea of what I
wanted to talk about, and how I was going to present it.  Somebody
asked a question (looking for an example), and after about 15 seconds
thinking, I structured the lecture around an example I had just thought
of which, I feel, made for a much better class than the one I had
planned.  A couple of times I have taken notes that students have taken
down in class, and photocopied them, and taught from those notes the
following year (or 2, or 3, or ...).  This is very seductive, as it is
MUCH easier to do, but the resulting lectures seem very stiff, and
rigid, and, well, ... LECTURES, rather than classes.  Even the classes
that had seemed so bright and alive the first time, seem dryer when
taught from the notes.

Well, I've rambled rather more than I intended.  If you have comments
(yay or nay) about my teaching (as biasedly described here), I'd
appreciate mail.  (And if you thought this was boring, I'm sorry, but
writing it probably helped my teaching a bit, and reading it may help
someone else).
	../Dave