Re: [Phys-l] explanatory and response variables (was calibration )

[John Denker <jsd@av8n.com>]

On 08/10/2007 12:39 AM, Barbara Bay wrote:

> The websites for all of the applicable CA state standards:
> All Math < http://www.cde.ca.gov/be/st/ss/mthmain.asp >
> Algebra 1 < http://www.cde.ca.gov/be/st/ss/mthalgebra1.asp >
> All 9-12 Sciences < http://www.cde.ca.gov/be/st/ss/scmain.asp >
> Physics < http://www.cde.ca.gov/be/st/ss/scphysics.asp >
> Investigation and Experimentation < http://www.cde.ca.gov/be/st/ss/scinvestigation.asp >

That's very useful.  Thanks.

> .... the terms "horizontal axis" and "vertical axis" are not used in
> any Math standard from kindergarten on up. 

I stand corrected.  

> .... But the terms "horizontal
> axis" and "vertical axis" are used on the Math sample test questions
> (and the real test questions). Which is, again, unfair.

It only seems unfair.  It's not really unfair for it to be on the
test if everybody knows it's going to be on the test.  It only 
seems unfair if you buy into the theory that the test is controlled 
by the standard ... which is not the case.

I once had a conversation with one of the poo-bahs responsible
for the California standards.  He kept telling me how lofty the
standards were.  I kept telling him I didn't care.  "The /test/ 
is the only thing that matters, and the test is lousy.  The 
teachers are going to teach to the test, no matter what the 
standards say.  If you really believe the standards are driving 
the test, then your standards are lousy, because they produce 
a lousy test.  And on the other hand, if the standards are not 
driving the test,  it doesn't matter what the standards say.  
So don't tell me about the lofty standards;  tell me how you're 
going to straighten out the test."

> I think your definition of "controlled" was too narrow for an
> informal discussion. But, "controlled" was not my word. I used the
> word change, which is wrong too, if we are going to be picky. I
> prefer the phrase "first variable measured" which would correspond to
> the first value in an ordered pair. For example, time is not really
> "controlled" (students can't make it go backwards or speed up or slow
> down), but many times it is used as the independent variable simply
> because it is usually easier to measure. The students "control" the
> time data only in the respect that they choose the time increments at
> which to measure the other variable which may, or may not, be
> dependent on the elapsed time.

I cannot figure out what that is trying to say.  If we define 
"independent" in terms of the pre-existing notion of "control", 
and then redefine "control" to mean "not really controlled" then 
I for one have no idea what is being said.  I'm not trying to be
obtuse;  I have carried out some pretty serious experiments
over the years, and I have a pretty good idea what "independence"
means and what "control" means in the real world.

The minor point remains:  The typical high-school textbook
abuses these terms.

The much larger point remains:  No matter what terminology 
is used, the typical high-school textbook says things about
investigational procedure that are just not correct, just 
wildly inconsistent with how things are done in the real 
world.

I chose the Hertzsprung-Russell diagram as my first illustration of 
this point, because astronomy is the quintessential /observational/
science.  Astronomers don't "control" the stars;  they only
observe them.  There are no "controlled" variables on the H-R
diagram.  And that's not the only category of example;  don't
get me started about the notion that experimenters should vary
at most one variable at a time;  this is discussed at
   http://www.av8n.com/physics/scientific-methods.htm#sec-doe
Thirdly, any feedback loop -- even the thermostat in my house -- 
will reverse the roles of what is being "controlled" versus what 
is "measured".  I /control/ the temperature setting on the front 
of the thermostat, while the thermostat /measures/ the temperature 
and /controls/ the valves and motors.  It is absurd to ask what is 
being "controlled" here;  there are multiple notions of "control",
and using the word "control" without qualification, as if its
meaning were a_priori obvious, is a mistake.  It is the sort of 
mistake that could only be made by someone with no experience 
actually doing experiments.

Yeah, the standards tell me I have to have one "single independent
variable"
  http://www.cde.ca.gov/be/st/ss/scgrade5.asp
Well, given the choice between the standards and reality, I'm going 
to believe reality and disbelieve the standards.

Books are important, and tests are important, but we are not yet 
pawns of the totalitarian state, even though things seem to be 
heading down that road.  The way it is supposed to work is that
we figure out -- by means of phys-l discussions or otherwise -- 
what is the right answer, and then we get them to fix the test.  
Fifty years ago, the PSSC folks figured out (in advance) that 
kids who took PSSC physics would get terrible scores on the SAT 
as it existed at the time.  So they got the ETS guys to make a 
special PSSC version of the SAT.

To repeat:  Yes, we should discuss the standards and the tests.
But that should be the start of the discussion, not the end.