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] Reading PER literature

Well Shayer & Adey compared using standard tests, and not school grades, so
there is no effect due to different school grading scales. Yes, school
grades would generally lead to a wide spread.

When I met Philip Adey, he pointed out that people want to deny their
results that basically all schools were doing the same things. People want
to believe their schools are the better ones despite lack of evidence for
this.

A similar situation occurs in medicine, and also repair shops. It is nearly
impossible to get accurate information about which doctors/hospitals or
repair shops are more reliable. At least in a repair shop, you have only
lost money for shoddy work. Most decisions have to be made on the basis of
recommendations, which are not usually based on good statistical samples.
In the case of products there are testing groups such as Consumers which
provides carefully analyzed data over time.

One of the great advances that has been provided by PER is the availability
of research based concept evaluations which can be compared across classes
and schools. With the concept of normalized gain it is possible to factor
out some of the effect of prior study, and by also correlating with the
Lawson test it is possible to factor out one of the hidden variables. So in
the end it is possible to compare teaching by different instructors in a
much more accurate fashion.

If this capability is not used by all of us to try to improve physics and
science education, then we may end up with an outside agency measuring our
students and whipping us the way teachers in public primary and secondary
schools are being whipped by high stakes testing.

John M. Clement
Houston, TX


I, of course, should not have equated grades w/goodness. Rephrased:
Students from some schools uniformly when transferring obtain better
grades at the receiving school, from others worse, and finally, about
the same from a third set. Instead it indicates camparringly differing
grade stds. (or in/de flation). The preceding, I suspect would lead to
a very thick line for a similar study.

bc

John Clement wrote:

But grades are not a necessarily a good indicator of how well the school
is
educating the students. It may be an indicator of grade inflation or
deflation.

There is some very revealing data from Shayer & Adey. They looked at the
intake scores for students at the beginning of what would be middle
school
and then looked at the output scores on the English national exam at what
would be the end of HS. When you plot output vs input you find that all
schools fell on the same line. So essentially all schools tested were
doing
the same thing and the only reason for good output was good input GIGO.

But when Thinking Science was implemented the output scores for the
experimental group moved to be significantly above the line. In many
cases
schools below average moved up to above average, and sometimes superior.
So
it is possible to do a better job.

I have not seem similar data comparisons for American schools, but I
suspect
that they will be somewhat similar in that GIGO will be the rule. In
other
words the superior schools have students with high SES.

John M. Clement
Houston, TX



Not quite -- transfers give a good clue as to the quality of schools. I
learned this when I was discussing someone from shall remain nameless.
The office mgr. said students from that school always received lower
grades at "our campus".

bc

John Clement wrote:



One interpretation is that there is a problem with education in their
countries. It could be that the prestigious schools there are full, or


the


social climate here is alluring.

As to superiority of American schools, this is often a self fulfilling
prophesy. When a school is considered superior people send their
budding
geniuses there and viola, it has better test scores and superior


graduates.


My kids school is precisely one of those. Schools ratings are actually
completely independent of the education. They don't do good pre and
posttesting to see if the graduates have "learned" more. It is the old
computer adage GIGO "garbage in garbage out" or in this case genius in,
genious out.

John M. Clement
Houston, TX




| > Keep talking about how more and more physics & science grad
students
| are
| > from other countries.
| > Fine. Do something about it. Teach differently.
|
| Hmmm. But many of them study here, so what does THAT say?
| --- That says that our schools are good. It doesn't say that there
is
| no room for improvement in teaching methods.
|

There are, of course, other interpretations of that data that are
likely
to be of equal value.
_____________________________





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