🚀 go-pugleaf

In article <1993Jan12.163525.9122@mcshub.dcss.mcmaster.ca> kenm@maccs.mcmaster.ca (...Jose) writes:

>	McMaster University is having to abandon its currently-supported
>computerized-testing system because we are doing away with the VAX/VMS
>systems that it is run on.

>	Before we run out an buy a new testing package or invest the time
>and money into recoding/porting our old system, we are considering whether
>we should even provide this service.

>	I would greatly appreciate information from faculty, students and
>staff from other universities and colleges about how on-line testing (ie.
>test written and scored on computers, for non-computer courses) is used.
>Any suggestions of MS-DOS- or Unix-based packages (commercial, shareware, 
>in house) would also be very welcome.

I would advocate not only abolishing anything which smacks of multiple-
choice questions.  I know very well why it is being done, but to use the
old adage, anything worth doing is worth doing well, and computerized
testing does not measure anything worth measuring.  Sometimes there 
might be no alternative, but I have seen no situation involving grades
in a university where this is the case.

Now if one can test ingenuity and understanding how to attack a problem,
rather than looking at the answer, this could be a place where computers
are a useful and valuable tool.  I can see a computer program checking
the correctness of a formal proof, and not issuing a grade (other than
absolutely correct) but informing the grader of possible or actual errors.
But this kind of package does not exist.
-- 
Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399
Phone: (317)494-6054
hrubin@snap.stat.purdue.edu (Internet, bitnet)  
{purdue,pur-ee}!snap.stat!hrubin(UUCP)