Re: Arbitrary Grades?????

["Rauber, Joel Phys" <RAUBERJ@mg.sdstate.edu>]

Comments and opinions (solicited?)

We give 40-50% of the lab grade based on showing up and performing the lab, 
meaning the student must show up and take data and turn in a raw data sheet. 
 The other 50-60% of the lab grade is based on a laboratory write-up turned 
in a few days after the lab.  It is ideally graded on a basis organization, 
analysis of data, conclusions and the answering of questions that appear in 
the student's laboratory manual.  How a lecture instructor uses the lab grad 
is up to the instructor.  It is rather typical here that they count anywhere 
from 10-20% of the total course grade.  (There is no separate lab grade 
given to the registrar).

> Regarding:  Grading Lab Reports

> 1.  I try to assign some questions with the lab experiment.  These
> can then be graded and help to determine the overall grade.

> 2.  One interesting idea that I have not tried yet (but might in the
> future) is to give a lab final exam.  In this system, the lab grade
> comes from (1) completion of the experiments, and (2) the lab
> final.  The individual experiments are graded perhaps only on
> completeness, thus avoiding the problem of how to grade the
> individual reports.

I'm rather philosophically opposed to grading a lab on mere completion.  The 
lab grade then becomes a certificate of attendance.  Why is there a problem 
of grading individual reports.  The only problem I'm aware of is my own 
laziness; I hate doing it.  But it is no worse a problem than an English 
professor grading a short term-paper.

> Does anyone else have any comments on lab finals?

In my intro physics course we had a lab final, it consisted of 4 stations 
where we actually had to perform some task ("experiement") , we worked in 
groups of two or three (I forget which).  I rather enjoyed enjoyed it as a 
student, although we were pressed for time to do the 4 stations in 3 hours 
(I think).  I've heard of other institutions that have a lab final which is 
more like taking a test, rather than performing a task.  I think either 
approach has validity and provides some reason for learning procedures in 
individual labs.

Joel