[LON-CAPA-users] Custom vs. Numeric Responses

[Mills, Douglas G]

Hi Stefan,

Thank you. That is helpful. To clarify on your first point though, we're
having them enter results from their lab experiments, so they shouldn't be
entering "wrong" answers -- just "their" answers.

I've gotten further down the path in the other direction though of using
customresponse and doing my own evaluation of significant figures. I still
have the problem though of not knowing how to tweak "return SIG_FAIL" so
that it correctly reflects too few vs. too many significant figures. Default
is "enter fewer digits".  I know how to test for too few or too many, just
not how to tell Lon-Capa which one it is. Is there a parameter or alternate
to "return SIG_FAIL" that specifies too few characters?

Doug


On 2/28/12 4:44 PM, "st.bisitz at ostfalia.de" <st.bisitz at ostfalia.de> wrote:

> Hi Doug,
> 
>> However, what I¹m not getting with numeric response which I do with
>> customresponse,
>> is the ability to echo the student¹s answer back to them by setting
>> ³displayanswer²
>> to be equal to their submission.
> 
> This is fine for correct answers by the students but would be a bad idea for
> - no student's answer and the problem is after answer open date -> nothing is
> shown as correct answer
> - incorrect student's answer and the problem is after answer open date ->
> incorrect anser is shown to be the correct one
> 
> So, better don't use it like that.
> 
> 
>> I¹d like to [...] display the student¹s answer back to them in
>> numericresponse.
> 
> You could use the aftersolved block combined with &submission().
> This would display the programmed correct answer from the response block and
> additionally display the student's answer (likely a text block starting with
> "Your answer was: ...")
> 
> Hope this helps,
> Stefan Bisitz
> 
> 
> 
> On 2012-02-28 21:37:58, Mills, Douglas G wrote:
>> Hi All,
>> 
>> I just want to be sure I¹m not overlooking something... I want to work with
>> student-inputted values so expect to use a customresponse for that. However,
>> I¹d also like to verify that they¹re using the correct number of significant
>> figures, which is trivial with the numericresponse type.  The numericresponse
>> works for my purposes in that I know that the student input values in this
>> case should be 50 +/- 2 so setting the tolerance to 4% takes care of it.
>> However, what I¹m not getting with numeric response which I do with
>> customresponse, is the ability to echo the student¹s answer back to them by
>> setting ³displayanswer² to be equal to their submission.
>> 
>> Bottom Line: I¹d like to either check significant figures in my
>> customresponse OR display the student¹s answer back to them in
>> numericresponse.
>> 
>> I have been working on checking number of sig figs in PERL for the
>> customresponse and am getting close, but related question there ‹ is there  a
>> way when returning SIG_FAIL to specify whether there are too many or too few
>> significant figures?  Seems to by default report ³too few² and I don¹t know
>> how to specify it should report ³too many².
>> 
>> Thank so all in advance.
>> 
>> Doug
>> 
>> Douglas Mills
>> Director of Instructional Technologies
>> Department of Chemistry
>> University of Illinois
>> dmills at illinois.edu
>> (217) 244-5739
>> 
>> _______________________________________________
>> LON-CAPA-users mailing list
>> LON-CAPA-users at mail.lon-capa.org
>> http://mail.lon-capa.org/mailman/listinfo/lon-capa-users
> _______________________________________________
> LON-CAPA-users mailing list
> LON-CAPA-users at mail.lon-capa.org
> http://mail.lon-capa.org/mailman/listinfo/lon-capa-users


Doug

Douglas Mills
Director of Instructional Technologies
Department of Chemistry
University of Illinois
dmills at illinois.edu
(217) 244-5739