Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] Physics job opening in Texas for 2008-09

Hi all-
We seem to have wandered far from my point. To the extent that I am reponsible, I apologize, while, in passing, I suggest that Professor Clement might benefit from investigating how the API prepares and field tests its multiple coice questions. These strike me as similar to the FCI, with similar distractors. Old AP exams are available, and readers can compare the performance of their students on FCI and old AP multiple choice tests (clearly not a waste of students' time - and much superior to exchangge of polemics).
My point, again: Experimental results are subject to uncertainty, both statistical and systematic. Without a discussion of systematic uncertainty, a result is meaningless. Members of any discipline that is, in fact, research based, will demonstrate their awareness of this fact.

Regards,
Jack



On Wed, 23 Jul 2008, John Clement wrote:


Hi all-
John seems to mistake what it means to demonstrate that a
proposition is true. By way of analogy, not too irrelefvant, consider
Newton's laws. Nobody has exhibited an experiment, done with falling
weights and pulleys, that violate Newton's laws. Does that mean that
Newton's laws are true in general? Clearly it does not. It means that
they correctly describe the results of experiments performed in a limited
set of circumstances.
One must view the context of the experiments from which to draw
general conclusions. And John looks to a limited set of experiments.

He is also mistaken when he says, "if you choose to disbelieve
these studies..."
It is not a queswtion of belief or disbelief. It is a question,
probably quantitive - a mode that John repeatedly skips - of the range of
validity of the studies that have been done.
And, yes, there is a mass of data waiting to be mined, namely,
the AP Physics Exams. The multiple choice sections of these exams are
not terribly different from the FCI. The difference is that a student
takes the AP only once, so there is no chance of being coached on the
answers.

There is a huge difference between AP questions and the FCI. The FCI
questions and similar conceptual tests are designed by iterations. One uses
interview protocols, and searches for the appropriate distractors. Each
question has been field tested. I don't think the AP has a similar
approach. Also the distractors in the FCI are so powerful that you can tell
students the answers and they still get them wrong. You can give students a
review sheet which essentially tells them the answers, but of course not in
exactly the same words, and they still get them wrong. Belive me, I have
tried it. Hestenes will tell you the same thing. The AP is more akin to a
teacher made test than to a researched conceptual evaluation.

But the research which is being questioned has nothing to do with the AP.
It is research which shows that student attitudes towards science become
poorer with each science course taken. Now please explain how the range of
validity of these studies invalidates the conclusion? Be specific. How
about a specific example of the MPEX studies by Joe Redish. Before judging
the research read it and come up with a specific criticism.

There was obviously no implication that a choice of career was like catching
a disease. That is a complete misinterpretation. My analogy was to point
out the absurdity of saying that good physicists have been produced "proves"
that conventional education is OK. This is like saying that because you
know people in good health who smoke that smoking is OK. Just because the
system has worked up to a point, does not mean that it is the best way to
educate students.

This began because the question originally was asked whether one could
demonstrate that the research based instruction produced better physicists
or engineers. But the point that there is not yet enough statistics was
completely ignored. I do not know what percentage of courses are taught
correctly using IE techniques, but from what I have seen it is very small,
probably less than a few percent. And the number of HS students who go into
either engineering or science is also probably less than a few percent. So
even if IE doubled either the quality of quantity of technical people, the
effect could not be detected.

The goal of research based teaching is to improve education overall for both
average and superior students. This should improve both the quality and
quantity of people trained in science. But again to see an effect on
national tests you need to have more teachers who are well trained in the
pedagogy.

There is another barrier to showing an effect. If only one course is IE and
subsequent courses are conventional, the effect of IE would be swamped. It
takes time for students to buy into the idea that they have to do the
thinking rather being spoon fed by a lecture.

John M. Clement
Houston, TX


I think that John's medical analogy fails on two counts. (1) I
wan't aware that the discussion is about producing students who want to
become physicists, and (2) it remains to be shown that the process of a
young person's choice of career is like catching a disease.
It also fails because there is nothing in the educational process
that is analogous to exposure to an identifiable chemical substance, nor
is there an identifiable outcome analogous to death.
Regards,
Jack



On Mon, 21 Jul 2008, John Clement wrote:

I basically read every study I can get my hands on, and I have not seen
any
that show something different. Then there is the results of Joe
Redish's
MPEX research. He found that physics courses generally caused a
reduction
in attitudes towards physics to more novice attitudes. The only
exception
to this were studio style courses such as Workshop Physics.

I am willing to read any studies which refute the existing research on
student attitudes towards physics specifically and science in general.
An
it is possible to establish such a general principle, specifically for a
specific case namely conventionally taught courses. Notice there is an
exception in the MPEX research.

Now of course the research shows the result in the average. But since
there
is a spread, a few students buck the average, and if they do it often
enough
they end up as physicists. So the science courses degrade on the
average,
and act a filter passing through favored few.

Now if you wish to disbelieve these studies, fine, find some published
studies which refute the existing ones. And by all means read the
studies
in JRST, TPT, and AJP.

Doing educational studies is very much like medicine. Long before
causal
mechanisms were found, we knew from studies that tobacco smoking caused
lung
cancer and emphysema. But one could still point to individuals who
never
got these diseases. Did the fact that a significant fraction ended up
surviving into ripe old age mean that cigarettes were not harmful? Does
the
fact that a miniscule fraction of students end up as physicists mean
that
the current system is the best?

By the logic in the previous post the medical studies can not logically
establish general principles either.

John M. Clement
Houston, TX


Hi all-
John Clement writes (in part):
The studies have shown that a liking of science goes down with each
science
course taken. One thing that is known is that students are humans are
more enthusiastic about something if they have success in doing it,
and
are usually less enthusiastic if they just beat their head against the
<wall.
____________________________________
I contend that it is a logical impossiblity for any set of studies
to establish sucdh a general principle. The interpretation of the
studies
referred to must take into account the time, place and manner of each
study. In experimental science this is part of the determination of
the
"systematic uncertainty" of the study.
And if, as sometimes happens, it is claimed that there is no
variation among the "studies", than that is as much a defect as
excessive
variation that hides the possibility of a clear conclusion. Measuring
intstruments must show a necessary minimum of fluctuation.
Also, before relying on conclusions from such "studies", the
proponent of the conclusions is expected, in most fields of science, to
make a diligent search for counterexamples to those conclusions. See,
e.g., Feynman's discussion of premonitions in the book (I think this is
the one) "The Pleasure of Finding Things Out".
Regards,
Jacdk

--
"Trust me. I have a lot of experience at this."
General Custer's unremembered message to his men,
just before leading them into the Little Big Horn Valley



_______________________________________________
Forum for Physics Educators
Phys-l@carnot.physics.buffalo.edu
https://carnot.physics.buffalo.edu/mailman/listinfo/phys-l


_______________________________________________
Forum for Physics Educators
Phys-l@carnot.physics.buffalo.edu
https://carnot.physics.buffalo.edu/mailman/listinfo/phys-l


--
"Trust me. I have a lot of experience at this."
General Custer's unremembered message to his men,
just before leading them into the Little Big Horn Valley



_______________________________________________
Forum for Physics Educators
Phys-l@carnot.physics.buffalo.edu
https://carnot.physics.buffalo.edu/mailman/listinfo/phys-l


_______________________________________________
Forum for Physics Educators
Phys-l@carnot.physics.buffalo.edu
https://carnot.physics.buffalo.edu/mailman/listinfo/phys-l


--
"Trust me. I have a lot of experience at this."
General Custer's unremembered message to his men,
just before leading them into the Little Big Horn Valley