Re: Lectures and written communications

[John Clement <clement@HAL-PC.ORG>]

> > -----Original Message-----
> > From: John Clement [mailto:clement@HAL-PC.ORG]
> > Sent: Wednesday, May 22, 2002 8:48 PM
> >
> > > There has been an interesting thread on this Phys-L network
> > > concerning the efficiency of  the lecture method and other
> > > methods of teaching.  Don Simanek claims that one major
> > > problem is that students have not learned how to listen, take
> > > notes, and then respond.
> >
> > My question about Don's claim is "How do you know that?".  Anecdotal
> > evidence is very suspect.  As one grows older your memories
> > actually change
> > and what you think was true when you were young is not a good
> > indicator of
> > what actually happened.  Since using some standard tests has
> > only recently
> > been a feature of physics education, we have little
> > comparable data from
> > earlier times.  We do not have direct evidence if the lecture
> > system is less
> > effective now or if it has always been equally ineffective.
>
> I can't speak to the research nor if the quality of students has changed,
> but I do think that a common thread in all the research studies is that
> effective methods are ones that promote
> self-evaluation/meta-cognition/formative-assessment/etc.  This goes
> hand-in-hand with "learning how to listen".  If so, students who already
> have this ability to self-evaluate/etc. would do better than others with
> lecture-based presentations.  This seems like a straightforward research
> project.  Does anyone know if this has been tested?  If students who have
> this ability do better with lectures and if a particular methodology
> develops this ability, it seems that we would then have evidence of the
> "benefit" of the particular methodology (beyond a particular evaluation
> instrument).  This can address the question posed last month on this list
> <http://lists.nau.edu/cgi-bin/wa?A2=ind0204&L=phys-l&F=&S=&P=55155>.
>
> ____________________________________________
> Robert Cohen; rcohen@po-box.esu.edu; http://www.esu.edu/~bbq
> Physics, East Stroudsburg Univ., E. Stroudsburg, PA 18301

Absolutely right.  Some of this type of research has been conducted for
years and reported in science education journals.  I do not recall any
studies that specifically target the question I think you are asking.  I
think you are hypothesizing that
self-evaluation/meta-cognition/formative-assessment (self) enable students
to learn from lectures "equally as well as from other approaches".  The
other approach would generally be a learning cycle approach, as that is the
general method common to virtually all reformed approaches.  From what I
have read, so far virtually all research that compares a conventional
lecture approach with a learning cycle approach favors the latter. The
learning cycle approach actually increases scores on Piagetian style tests
which correlate with the ability to learn difficult concepts.  This latter
fact argues that the learning cycle cultivates the sort of skills you
mentioned.  This has been done in biology as well as physics classes, but I
have not noticed as much in chemistry. The papers by Anton Lawson, Renner,
and Karplus are all good sources.

There is one piece of research by Miehl.  He increased the student passing
rate in a physics class from 50% to 100% as compared to a control group that
took the same tests.  This was done by analyzing student difficulties and
providing targeted worksheets.  Along with this, students were explicitly
taught problem solving methods.  However meta-cognitive strategies were
ignored.  Others who used his method have found that students are generally
not developing their own strategies.  It is clear that students were trained
better, but it was not clear that they had better understanding.

Incidentally the question in the referenced post implied that it is easy to
target a specific test and to raise the scores on it.  While this may be
true of simple fact based tests, it is not true of tests like the FCI/FMCE
(The F words again) or the TUG-K as long as students are not told the
questions and answers beforehand.  Joe Redish tried his best with lectures
and scores did not rise.  However, after certain labs they rose
significantly.  I have personally gotten feedback from students who took my
computer science course.  I use learning cycle and meta-cognitive
strategies, with practically no lecture.  They said that they used the same
techniques on the SAT and had a large rise in scores (40 and 100 points).
Yes, this is anecdotal evidence, but I have no other good evaluations and my
sample is very small.

Knowing how to listen can be valuable, but if the student does not have the
frame of reference necessary for comprehension, they will not be able to
make sense of the communication.  The learning cycle helps provide the
context by exploration prior to term definition.  Then finally application
extends the learning to other contexts.  The term definition phase would be
the lecture or book reading, with the exploration lab as the first phase,
and problem solving or student planned experiments as the last.

As to whether it is a straight forward research project, that is an open
question until it is tried.  The skills necessary to do such projects
require just as much experience and preparation as any physics experiment.
There is already data that shows that these skills accelerate all types of
learning.  I personally have data that shows this.  Richard Hake has said
that PER is actually more difficult than physics experiments.

John M. Clement
Houston, TX