Re: [Phys-L] Indicators of quality teaching (Was:MOOC:EdxOffers Mechanics course by Prof.Walter Lewin)

[Richard Tarara <rtarara@saintmarys.edu>]

On 6/23/2013 9:55 PM, John Denker wrote:
> On 06/23/2013 03:38 PM, Bruce Sherwood wrote in part:
>
> > The gains are most definitely meaningful; different curricula lead to
> > significant differences in posttest scores:
> Please explain what "meaningful" means.
>    a) Looking at the narrow context, a "meaningful" score might
>     be synonymous with statistical significance.
>    b) Looking at the slightly broader context, a "meaningful"
>     score might be an "indicator of quality teaching".
>    c) Something else???
>
>
> Note that I am exceeeedingly skeptical of (b).  If somebody
> wants to argue in favor of (b), they need to actually make the
> argument.  There are those who take FCI gain as the axiomatic
> definition of "quality teaching" ... but this approach is
> never going to persuade me.
>
> So, what is a "meaningful" gain, and what does it mean?
>
> _______________________________________________
> Forum for Physics Educators
> Phys-l@phys-l.org
> http://www.phys-l.org/mailman/listinfo/phys-l

I won't argue for (b) but might suggest some reasons that people can 
'fool' themselves into thinking there is more significance to their 
'gain' scores.

The  'different curricula' can amount to 'teaching to the test' in 
subtle (or overt) ways.  The FCI requires a pretty deep understanding of 
Newton's Laws although Master's and PhD level physicists mostly find it 
trivial.  A standard curriculum/book spends a week at best on this 
topic.  Any curriculum that spends more time and more focus is teaching 
_more_ to the test than a more traditional one.  The Thornton test, as I 
recall, has a lot of graphical analysis built in (which graph describes 
the given motion type of question).  I would think that as a pre-test 
many students would find that KIND of question unfamiliar and therefore 
do badly..partly due to there lack of experience with such.  Now if the 
curriculum in question actually spends considerable time with this kind 
of analysis (and I'm not saying it shouldn't) then post-test gains could 
be expected to be high.  In both cases the content may be quite 
different from a traditional course.  OK--students do better so maybe 
the different content course is better.....but, my concern is for what 
is NOT covered in order to spend the additional time and effort in 
certain content areas.  [Yes the old breadth versus depth debate, but we 
do pass some of these students on to the next course where there might 
be expectations of coverage (engineering programs for example).]

rwt (who does spend more time on Newton's Laws, gets good gains with the 
FCI, but doesn't see that as a validation of my teaching 
techniques...only my decision to spend the more time.)

--
Richard Tarara
Professor of Physics
Saint Mary's College

free Physics educational software
www.saintmarys.edu/~rtarara/software.html