[LON-CAPA-users] order of bubbles in multiple choice test

[lon-capa-users@mail.lon-capa.org]

Paul,
     You can do what Guy suggests.	
I like to also make the problem a radiobuttonresponse problem rather than
leave it as numericalresponse.
Then the prog will randomize the order for you. This is superficial, in the
sense that the same number is still the answer, but at least it is not
necessarily "C". 

Since you already have the answer (and wrong answers) calculated in the perl
script, just delete the numericalresponse block, insert a
radiobuttonresponse block and put the answer variables into the foils.

A little more tedious than just using a numericalresponse problem, but it is
not hard.	

What is harder, is making sure that none of your calculated wrong answers
ever coincide with each other or with the correct answer.

Note also, in a numericalresponse type exam question, that if your tolerance
parameter is not small, then it is possible for two or more apparently
different choices to be marked correct, because they fall within the
tolerance.

This kind of situation is avoided if you make it radiobuttonresponse.

The downside is that radiobuttonresponse foils always seem to get a linefeed
between them in the printout and your exam printout will require more paper.

Cheers,
Ray

On Sun, 18 Dec 2005 09:10:50 +0200 lon-capa-users@mail.lon-capa.org wrote:
> 
> You need to enter at least one random wrong answer in the range of the 
> wrong answers you devised. Look at 
> http://help.loncapa.org/cgi-bin/fom?file=349 for an example.
> 
>     Guy.
> 
> ----- Original Message ----- 
> From: "ciskep" <ciskep@mio.k12.mi.us>
> To: <lon-capa-users@mail.lon-capa.org>
> Sent: Sunday, December 18, 2005 4:15 AM
> Subject: [LON-CAPA-users] order of bubbles in multiple choice test
> 
> 
> >
> >
> > I want to use several of my lon-capa problems as part of a multiple 
> > choice exam.
> > I am adding smart incorrect options to them.  All is going well but 
> > when I test
> > them the answers are put in order from least to greatest.  Since the 
> > answers
> > have been chosen based on what I think likely incorrect answers will 
> > be, the
> > correct answer usually ends up in the same spot.  So I'm thinking 
> > that all of
> > this is pretty useless if in the end, the correct answer is always 
> > C. I am
> > guessing that there must be a way to tell loncapa to randomize the 
> > answers and
> > not always rank them from low to high.  Does anyone know what it is?
> >
> > Thanks
> > Paul Ciske
> > Mio AuSable High Schol
> >
> > _______________________________________________
> > LON-CAPA-users mailing list
> > LON-CAPA-users@mail.lon-capa.org
> > http://mail.lon-capa.org/mailman/listinfo/lon-capa-users
> > 
> 
> _______________________________________________
> LON-CAPA-users mailing list
> LON-CAPA-users@mail.lon-capa.org
> http://mail.lon-capa.org/mailman/listinfo/lon-capa-users
> 


Ray Batchelor
Senior Lecturer
Department of Chemistry
Simon Fraser University
Burnaby, BC
V5A 1S6

604-291-5635