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

["LaMontagne, Bob" <RLAMONT@providence.edu>]

I often teach a physics lab for students who are Chem and Bio majors.
One of our labs is the traditional free fall experiment where a ball
falls an arbitrary distance and a set of switches start and stop a
timer. We make various linear, semilog, and log-log plots to bring out
the nature of constant acceleration. I usually specify plotting h vs t,
log h vs log t, etc. These students have been drilled by their other lab
instructors in their majors to use dependent and independent in
referring to the axes. They become frustrated because I refuse to call
the axes by those names. One can adjust the apparatus for a given fall
time or for a given distance of fall. I fail to see how quantities like
log h or log t are either dependent or independent. They are simply
"related".

My other pet peeve is the long time spent in Chem labs at my institution
on significant figures. I have an exercise in one of my physics labs
where the students measure the length of 100 matchsticks and form a mean
and standard deviation. They then plot the distribution of the lengths.
They form the mean length by adding all the lengths and then dividing by
100. Because of their training in Chem, they then get rid of any digits
beyond the significant figures in their individual length measurements.
This skews all their individual deviations - resulting in a meaningless
standard deviation. It also produces a graph of length distributions
that is highly lopsided and unrecognizable as following a bell shaped
curve (which the lengths of wooden matches actually do in practice.)I
have them go back and keep all the digits until all their computations
are completed and the results then follow the expected form.

Whether in class or in lab, I never mention nor use significant figures.
I do, however, have the students perform exercises that relate the
errors in making measurements to the uncertainty in the accuracy of the
results.

Bob at PC

> -----Original Message-----
> From: phys-l-bounces@carnot.physics.buffalo.edu [mailto:phys-l-
> bounces@carnot.physics.buffalo.edu] On Behalf Of Polvani, Donald G.
> Sent: Wednesday, August 08, 2007 3:15 PM
> To: Forum for Physics Educators
> Subject: Re: [Phys-l] explanatory and response variables (was
calibration
> )
>
> Ludwig Kowalski wrote:
>
> > It seems we must first agree on the operation definition of the
phrase
> " well established mathematical term."
> > Here is my suggestion: Randomly select 20 calculus textbooks in a
math
> library. A word or phrase is "a well established mathematical
> > term" if it is found in at least 80% of textbooks. The textbooks
should
> be up to 60 years old. Why up to 60? Because I was introduced to
> these
> two terms about 60 years ago. They were in Polish -- zmeinna zalezhna
> and zmienna niezalezhna.
>
> Why not just take a poll of Phys-l members?  Although we are not
> professional mathematicians, most of us are familiar with "well
> established mathematical terms".  I, for one, think that "independent
> and dependent variables" are well established mathematical terms.
These
> terms appeared in many texts I learned mathematics from in high
school,
> college, and graduate school (although graduate school was 40 years
> ago).
>
> Has usage changed in recent times?
>
> Don Polvani
> Northrop Grumman Corp.
> Undersea Systems
> Annapolis, MD
>
> _______________________________________________
> Forum for Physics Educators
> Phys-l@carnot.physics.buffalo.edu
> https://carnot.physics.buffalo.edu/mailman/listinfo/phys-l