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] kinematics objectives



Definitely half baked. While a teacher constructed test can be pretty good,
the FCI is the only way you can compare to other teachers, as the scores
have been published. What is really needed is good gain. Or you could use
the FMCE as both tests track almost perfectly. I have the graphs to show
it.

As to problem solving we know quite well that students can be pretty good
problem solvers while at the same time they lack the concepts. The really
superb problem solvers have to use the concepts. So problem solving is not
a good test. If you want to combine the two use the MBT. Actually students
want to memorize procedures for problem solving and that is what happens in
the conventional classes.

Mazur showed that when he emphasized concepts that scores went up on his
previous problem solving final. Actually the scores went up for the lower
students, while the upper students kept even. This may actually have some
saturation in scores. Notice that he no longer emphasizes problem solving
in class. Mazur also made his emphasis on concepts very plain by giving the
students a stiff conceptual test as the first test. That way they were sent
a strong message that the concepts are important.

The most important skill leading to problem solving is being able to analyze
problems in terms of concepts. Conceptual based problem solving is what
physicist do, and most physics and math courses teach just simple algorithms
for problem solving. The conceptual based problem solving is emphasized by
the Minds on Physics texts and an independent evaluator found that they
promoted "expert like" problem solving.

Another possibility is to use the Lawson test. But don't expect huge gains.
Modest gains can be had in a single semester, while it took Shayer & Adey 2
years of interventions to show large gains. Conventional classes show ZERO
gain.

John M. Clement
Houston, TX



I understand the issue. You need something, but the FCI is
perfectly terrible for this purpose. Even the FCI authors do
not recommend using it in this way.

.......................

Here is a partially-baked suggestion: Make your own test,
perhaps based on end-of-chapter problems from the text.
Choose ones that actually require some reasoning. Choose
ones that deal with interesting, worthwhile topics.

Note that students are quick to realize that isolated concepts
are essentially worthless, so they not motivated to learn.
In contrast, skills that lead to solving interesting problems
are seen as worth learning.

Showing gain on an interesting test is the easiest thing in
the world. Most of the incoming students will get a zero, so
things can only go up from there. (Of course in Holmdel
there will be a few wise-guys who show up on the first day
having already read the entire text and perhaps most of
Feynman volume I, so you need to account for that
somehow....)