Path: utzoo!mnetor!uunet!lll-winken!lll-lcc!ames!mailrus!tut.cis.ohio-state.edu!osu-cis!att-cb!clyde!watmath!water!ljdickey
From: ljdickey@water.waterloo.edu (Lee Dickey)
Newsgroups: comp.edu
Subject: Re: cruelty to undergrads
Message-ID: <1502@water.waterloo.edu>
Date: 5 Apr 88 16:34:50 GMT
References: <8470@eleazar.Dartmouth.EDU> <359@jc3b21.UUCP> <524@mccc.UUCP>
Reply-To: ljdickey@water.waterloo.edu (Lee Dickey)
Organization: U. of Waterloo, Ontario
Lines: 21

In article <524@mccc.UUCP> pjh@mccc.UUCP (Peter J. Holsberg) writes:

] I grade each exam one problem at a time, noting the errors in each. 
] After a problem is graded, I then go back and determine the value of
] each error, and subtract them from the points for that problem.  I very
] rarely have two students graded differently.  (Yes, it takes forever, so
] I put lots of multiple-choice/fill-in-the-blanks/even T-F questions to
] keep grading time reasonable.)

I use a scheme similar to this.  I print one question per page and
leave plenty of space for the work.  (I find it hard to think of fair
multiple choice or TF questions.)

When marking one page, I take a sample of 20 or so papers and read that
question on each of the sample papers and prepare my marking scheme.
Then I mark that question on all papers.

-- 
 L. J. Dickey, Faculty of Mathematics, University of Waterloo.
	ljdickey@waterloo.edu
	ljdickey@WATDCS.UWaterloo.ca	ljdickey@water.BITNET
	ljdickey@water.UUCP	or	...!uunet!water!ljdickey