Stephen Black wrote:
> My esteemed colleague, Stuart Mckelvie wrote:
> >
> > We are still in what some would call the Dark Ages. We submit
> > percentage grades for students, not letters.
>
> Some might call it the Dark Ages, but not me. I'd call it
> sensible. Letter grades and grade point averages are among the
> dumbest inventions of academic bureaucracy. Percentage grades are
> easily calculated, are readily understood, and mean exactly the
> same thing at any institution. I can't say the same about the
> opposition, which is why those who use them (practically everyone
> else, apparently) have to spend so much time explaining them.
I have always thought having a fixed relationship between % and letter
grades was silly because of the "90% of what?" issue. Hitting a baseball 30%
of the time is great and landing an airplane 90% of the time correctly is not
great. Getting 50% correct on a True/False is chance, but 50% on an essay
(making a few assumptions) is half way between chance and perfection. Raw
percentages need conceptual guidance. This is where I disagree with Stephen
about the letter grades because they provide guidance, especially if
accompanied by the customary descriptors, which act as conceptual anchors.
Then you can adjust the percentages up or down to bring percentages, letter
grades, and common sense into harmony.
i.e. I have in my head a picture of what A (outstanding) is. This is a
student who is capable of carrying on an intelligent conversation about
course material, a student I would eagerly write a great letter of
recommendation for. I also have a picture of the F, someone whose ability to
engage in intelligent conversation about the subject matter has not changed
one whit as the result of the course. Then I have a rough idea of the B, C
and D points between those extremes. Having a variety of conversations with
students during the semester helps me adjust the % so they coincide with
these anchors. In general, if I use this scheme (similiar to Stuart
McKelvie's): A>=80%, B>=70%, C>=60%, D>=50%, F<50%, the percentages align
themselves such that those students who I thought were outstanding by talking
with them end up with As in the course, and those that blew the course off
failed. I suppose the letter grades are unnecessary if you had a scheme that
said "make 90 - 100% outstanding, 80 - 89.999% good, .." but ABC(D)F is
simplier to encode, and falls on the correct side of the magical number 7.
Last year I was on our Academic Standards Committee and we struggled with
differences between departments in the grades. Psych, Math, Philosophy and
Natural Sciences (Bio, Chem ..) had the lowest grades (10 to 20% A if I
remember), while Education, Music, Theatre (and maybe Art) had the highest
(pushing 80% A and B). The entire issue was delicate, as some people thought
publishing the departmental differences would cause one department to
criticise another (actually, we do a good job criticising each other without
data was my argument for releasing the data!).
We have talked over the years about putting in Z ot T scores next to grades,
these based on average performance within that particular class. So, a
student's transcript would resemble:
letter Z
Art 101 B- -1.2
Bio 101 B+ .8
Psy 101 A 1.2
T scores would be better because on a transcript those pesky decimals and -
signs make it difficult to decode, and they might disappear when photocopied.
But either T or Zs would allow grades based on criteria (letter grades) as
well as relative criteria (other students in the class). To date the college
has not done this yet, though.
--
------------------------------------------------------------------
John W. Kulig [EMAIL PROTECTED]
Department of Psychology http://oz.plymouth.edu/~kulig
Plymouth State College tel: (603) 535-2468
Plymouth NH USA 03264 fax: (603) 535-2412
------------------------------------------------------------------
"Kane to kal� ke r�chto sto yal�."
(Do a good deed and cast it to the sea)
Ancient Greek saying
------------------------------------------------------------------
---
You are currently subscribed to tips as: [EMAIL PROTECTED]
To unsubscribe send a blank email to [EMAIL PROTECTED]
