Hi

On 5 Feb 2003, Radford Neal wrote:
> In article <[EMAIL PROTECTED]>,
> Gus Gassmann  <[EMAIL PROTECTED]> wrote:
> 
> >When I was a TA at the University of British Columbia there was a
> >(modified) percentage system in use. (Modified means that the grade
> >reported was weighted by the number of credits.) Even though they
> >did not use letter grades, a score of 80% or higher meant "First class
> >standing", which was useful for scholarship eligibility. Instructors were
> >actively discouraged from assigning 78 or 79%. It was either 77 or 80.
> 
> This is approximately the case at the University of Toronto as well
> (which in typical U of T fashion assigns and uses _both_ numeric and
> letter grades, with 80 being the threshold for an A-).  Nothing dire
> happens to me when I give out a grade of 79 (as I regularly do), but
> there is indeed official "encouragement" not to.
> 
> In my opinion, this verges on dishonesty.  It seems to be motivated by
> the desire to avoid complaints from students who get a 79 and think
> that some small increase in a mark on some component would put them
> over the boundary line.  Such students can indeed be annoying, but
> avoiding the problem by lying to them about the location of the
> boundary line is not the proper solution.

I routinely raise grades just below boundaries to the next higher
level, not for institutional reasons, but rather for what I think
are valid statistical/measurement reasons (biased somewhat
perhaps by a desire to err on the side of the students).  

Student performance on tests and other assignments provide sample
information about some underlying population parameter that we
wish to estimate.  If the population parameter for a student is
80 or better, there is still some probability that their sample
mean will be 79 (say) because of sampling and other sources of
measurement error.  There is a much lower probability that their
sample mean will be 78 (say).  So it makes sense to me to raise
the 79.

Or if you thought about a confidence interval around obtained
grades, the likelihood that the interval around 79 overlaps with
80 is much higher than that the interval around 78 overlaps with 
80.

So I see the practice as simply some acknowledgement that our
measures involve error (perhaps unlike jumping between roof-tops,
as mentioned in a previous post).

Best wishes
Jim

============================================================================
James M. Clark                          (204) 786-9757
Department of Psychology                (204) 774-4134 Fax
University of Winnipeg                  4L05D
Winnipeg, Manitoba  R3B 2E9             [EMAIL PROTECTED]
CANADA                                  http://www.uwinnipeg.ca/~clark
============================================================================

.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to