This is a very good statement.
Alan
"Robert J. MacG. Dawson" wrote:
>
> Dennis Roberts wrote:
>
> > however, BY FAR, most of the course work is taken up by doing these
> > assignments ... perhaps 100 hours for assignments and, only 5 for the
> > tests. it would be almost unethical to say that because of the lack of
> > variability in grades on the assignments and, the fact that they CAN (and
> > probably do ... though rarely do students know one another in this course)
> > get help ... that assignments should only count say ... 5/10 percent of
> > the course. this would be totally unfair to the students.
>
> I disagree. By that argument, the 100m sprint event at the Olympics is
> colossally unfair because the athletes spend years training for it and
> only get "graded" on a few tens of seconds. Thus, we ought to decide the
> medals on career averages including training runs, only holding the
> actual Olympic race if those results were close enough that one race
> might change it.
>
> Fairness *to*the*students* requires that they get significant benefit
> if possible from the work that they do. Fairness does not necessarily
> demand that they be evaluated at all; there are many learning situations
> in which the student is not evaluated at all.
>
> If the student is evaluated, the primary reason is usually to indicate
> what the student has learned from the course. (I wrote "what they have
> learned", not "how hard they worked attempting to learn it", as this is
> still the expectation of future instructors, employers, etc.) Fairness
> demands only that the evaluation reflect what it claims to reflect to a
> reasonable degree of accuracy - both in the sense that a student who
> learns the material gets a good grade, and in the negative sense that a
> student who gets a good grade knows the material.
>
> If the tests cover approximately the same material as the assignments
> (yours surely do??) then they may fulfil the requirement of fairness
> *to*the*students* as well as the assignments. It is not a matter of
> being "fair to the coursework". As all the students do the same
> assignments and write the same tests, it balances out for the students.
>
> Of course, perfect fairness is unattainable, or attainable only at
> infinite cost in terms of other desiderata. ("Well, it's so clean, sir!"
> "It's certainly uncontaminated by cheese!") As soon as any other goals
> are acknowledged, small sacrifices in fairness must be made. So yes,
> putting much weight on the exam does penalize the student who happens to
> work harder (deliberately or otherwise) on the assignments than on
> studying for the exams; and putting much weight on the assignments
> penalizes the student who "gets it together" late in the term.
>
> -Robert Dawson
> .
> .
> =================================================================
> Instructions for joining and leaving this list, remarks about the
> problem of INAPPROPRIATE MESSAGES, and archives are available at:
> . http://jse.stat.ncsu.edu/ .
> =================================================================
--
Alan McLean ([EMAIL PROTECTED])
Department of Econometrics and Business Statistics
Monash University, Caulfield Campus, Melbourne
Tel: +61 03 9903 2102 Fax: +61 03 9903 2007
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
