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] Evaluation tests



On 12/17/2013 03:01 PM, Philip Keller wrote:
But I am not sure that is our role to make these predictions. If a
student has a destination in mind and physics is one of the
requirements along the way, then despite what my (admittedly fuzzy)
prediction model would say, my first instinct is to say come on in
and be prepared to work hard.

Wiser words were never spoken.

I learned that in the first course I ever taught. There was one
student who by any objective measure reeeeally did not belong in
the course. I told her the only way to pass the course would be
to get an infinite amount of help and spend an infinite amount of
time, to the point where she flunked all her other classes. I
encouraged her in unsubtle ways to drop the course, and trust me,
I can be unsubtle when I want to be.

She believed everything I said, except the part about dropping the
course. She worked on the course 24x7 and wrangled a couple of
guys into helping her. She did in fact flunk all her other classes.
However, she had a reason for taking the class, and it was actually
a pretty good reason. I talked to her undergraduate advisor about
it, and he was fine with it. He said something to the effect that
she doesn't need an undergraduate advisor, she needs a graduate
thesis advisor. I was reminded of the saying, you don't have to
be crazy to be here, but it helps.

===

I say the placement test does have predictive value, if we are
careful how we describe the predictions. We can predict how
much time and effort and help somebody is going to need. If
they understand the costs and risks and are motivated to take
the course anyway, they might do great. This is of course the
exception not the rule. My point is that the system needs to
be set up to handle the exceptional cases.