[Phys-l] AP on GPA (the other AP)

[John Denker <jsd@av8n.com>]

AP (by which I mean the Associated Press) put out a story on grades,
grade inflation, et cetera:
  http://www.washingtonpost.com/wp-dyn/content/article/2006/11/18/AR2006111800473_pf.html

Much of the article is anecdotal, but there were also some factoids
and figures.

Some snippets:

> "We're seeing 30, 40 valedictorians at a high school because they don't want to create these 
> distinctions between students"

> The average high school GPA increased from 2.68 to 2.94 between 1990 and 2000, according to a 
> federal study. Almost 23 percent of college freshmen in 2005 reported their average grade in high
>  school was an A or better, according to a national survey by UCLA's Higher Education Research 
> Institute. In 1975, the percentage was about half that.


> Zalasky's GPA is nearly an A minus, and yet he ranks only about in the middle of his senior class
> ..... Last year, he even considered transferring out of his
> highly competitive public school, to some place where his grades would look better.

Now that's a scary anecdote.  It's scary on multiple levels:
 -- I suspect the report is misleading;  it doesn't say how /seriously/
  the kid considered transferring, or why he finally decided not to.  It
  also doesn't say how deeply the reporter had to dig to find a student
  who was willing to say something sensational.
 -- Most of the article is about grade inflation, and this anecdote says
  "something" when it reports that a B+ is a middle-of-the-herd grade.
  However ...
 -- At another level, this makes a deeper point about competitiveness.
  Class rank (at a particular school) is insensitive to what we usually
  call grade inflation.  That is, even if there were "more" grade inflation
  or "less" grade inflation, this kid's class rank would still be in the
  middle of the herd.  Therefore, this anecdote can be interpreted as
  something else, namely a would-be effort to inflate /class rank/ by
  moving to a less competitive school.

  Class rank inflation ... now that's a scary thought.  OTOH there are,
  of course, ways to detect and correct for this sort of thing.

As I've said before, I don't know what grades (or class rankings) mean,
but in any case school is supposed to be about /learning/.  Grades etc.
"should" somehow support the goal of learning.  If a middle-of-the-herd
student is tempted to transfer to a less-competitive school, where he
would almost certainly learn less, then the incentive system is badly
broken.

Incentives for students are generally necessary.  Learning has long-term
value, whereas students are notoriously hypersensitive to short-term
considerations.  This has been a problem for (at least) the last 2300
years, if legend is to be believed:
  http://pirate.shu.edu/~wachsmut/ira/history/euclid.html

Such incentives will never be perfect, but still we should do the best
we can, and we should always be alert for breakdowns and/or opportunities
for improvement.

> From the keen-grasp-of-the-obvious department:  We need to figure out what
we care about, and then design grades, standardized test scores, etc. to
reflect what we care about.

As for me, I care about more than one thing, so it seems verrry unlikely
that any one-dimensional metric such as GPA or class rank will ever tell
me what I want to know.