Re: [Phys-L] thoughts on how science is done

["John Clement" <clement@hal-pc.org>]

Apprenticeship certainly would fit the bill, but just lectures absolutely do
not.  Studio courses have been shown to improve attitudes on the MPEX, so
that is a start.  There is evidence that working at science can be helpful,
and then a summarizing lecture has been shown to affect further improvement.
The usual lecture/problem solving/verification labs sequence is fairly
bankrupt here.  The big problem is that students have already been taught
nonsense and have fixed paradigms about how science works.  These beliefs
are often resistant to change even in the face of evidence.

There have been studies of belief vs exposure to evidence and they do not
bode well.  Exposure to evidence does not change beliefs.  Indeed exposure
to evidence tends to harden beliefs as then people know which evidence to
pick to bolster their opinions.  Beliefs correspond generally to what people
in their religious/political/social groups think, and not to evidence.
Science suffers from this problem.  There is a recent NYTimes article about
this effect which is well known in psychology.

People who disagree with a particular scientific model will often think that
the evidence was deliberately falsified, or they will equate it with
something they don't like such as atheism.  This is not helped by the usual
verification labs which have random errors which mislead students.  I find
it amazing that apparently educated people in science such as engineers,
MDs, dentists... Will fall prey to this type of thing.  The truth is that
they have been educated as technicians in prescriptive courses and not as
scientists.  Science works by consensus where scientists are continually
looking at models and experiments to see if they agree.  Falsification of
evidence is eventually exposed, so it is generally not worth while.  The one
exception seems to be medicine/biology where the monetary stakes are so high
that some try to beat the odds.  The tobacco companies were moderately
successful at obfuscation, but eventually the research was too definite to
be ignored.  This also happens to people who are looking at scientific
models outside of their area of expertise.  I know a geologist who believes
in the standard geological model and cosmology, but who calls evolution
nonsense because of his religious outlook.  Notice that this corresponds to
the educational experiment which showed U shaped learning.  The experiment
showed that students changed their ideas when confronted by a new situation
and actually regressed, just as a geologist can regress to a non-scientific
point of view outside his field.

At the present juncture we have the problem that a large organized segment
of the population rejects scientific evidence, while they embrace the
magical technology that science made possible.

So what is the solution?  One hint comes in the way we can change students'
ideas about the "fixed talents" paradigm.  Instead of lecturing, a reading
followed by writing a letter to another student explaining how talents and
intelligence can be changed has an effect on beliefs.  It provides an
alternate way of thinking which can be used when confronted with a
difficulty.  Students have to be pushed to engage with the concepts.
Changing classes from information giving, to actually doing science would
also be helpful.  Ultimately as long as our society has a whole segment
which rejects science, our task is much more difficult.  This was not true
in the 1950s, so it may not be a permanent problem.  Even among scientists
there are individuals who reject sciences outside of their domain, and this
is certainly not helpful.  I know a department chair in physical science who
rejects any evidence from psychology or cognitive science.  We also need to
recognize that scientists are not priests who have the only truth and the
only method of figuring out things.  When you look at what scientists do, it
is basically similar to what scholars in other pursuits do.  Econonomists,
archeologists, and literature majors... all use methods similar to physical
scientists.  So the scientific method should not be taught as revealed
theology.  Unfortunately it is still taught that way in most schools.  I
have tried to tell science teachers that the usual scientific method is
nonsense, but they reply it is "a scientific method", and they don't seem to
understand that it prevents students from developing more realistic views.

So far education seems to be stuck in the mode of doing more of the same
which doesn't work well with regards to teaching how science works.  If
something doesn't work why do more of it?  Perhaps it is the streak problem.
You see someone who does succeed so you think that the method is working.
It is like streaks of success in sports.  A player with a streak is
promoted, but statistical analysis shows that the streaks were basically
random.  Humans are sensative to scant evidence and they manufacture beliefs
based on random correlations.  This caused the autism/vaccine belief
reinforced by falsified research.  This would seem to be because hunter
gatherers had greater success when they hunted near to one success.  In
other words you might find more apples near a fallen one.  Evolution has
programmed this into our brains.  But this type of behavior is not
productive in random games.  When a game has anti-correlation where a given
number depresses its probability, people never discover this effect.  In an
anti-correlation game it is possible to win by betting against previous
numbers, but there is no winning strategy in a purely random game.  Notice
that this streak effect causes funny conclusions when students are presented
with data that has random variances.

John M. Clement
Houston, TX


> > I leave it as a question:  Does anybody have any good 
> suggestions for 
> > how to teach people how science is done?