[Building Sakai] Issues with the Tasks, Tests and Surveys tool

[Karen Tsao]

Hi Gerwood,

Can I take a look at your data. Can you please run some queries for me?

1. Get the published assessment id. You will need this for the following
queries.
select publishedassessmentid from sam_publishedassessment_t where title =
:published_assessment_title

2. Get the submission records for the two students:
select * from sam_assessmentgrading_t ag, sam_publishedassessment_t pa
where ag.publishedassessmentid = pa.id
and pa.id = :published_assessment_id
and agentid in (:first_student_id, :second_student_id)

3. Get the answer records:
select * from sam_itemgrading_t
where assessmentgradingid in (
select assessmentgradingid from sam_assessmentgrading_t ag,
sam_publishedassessment_t pa
where ag.publishedassessmentid = pa.id
and pa.id = :published_assessment_id
and agentid in (:first_student_id, :second_student_id))

4. To get the results of your other email (student answers of the second
part):
select * from sam_itemgrading_t
where publisheditemid in (
select pi.itemid from sam_publisheditem_t pi, sam_publishedsection_t ps,
sam_publishedassessment_t pa
where pi.sectionid = ps.sectionid
and ps.assessmentid = pa.id
and pa.id = :published_assessment_id
and ps.sequence = 2
and agentid in (:first_student_id, :second_student_id))

Is this a timed assessment? A linear assessment? Is the second part a random
draw part? Also, I would like to know what Sakai version do you use?

If you need other queries, please feel free to let me know.

Thanks,
Karen


On Tue, May 5, 2009 at 4:33 PM, Gerwood Stewart <gstewar8 at une.edu.au> wrote:

>  We are having a problem with the tasks test and surveys tool and I’d like
> to know if anyone else has been encountering it?
>
> We got a large quiz that was setup incorrectly (because we assumed
> something incorrect about the way one of the question types worked) as a
> result in the second part of the quiz was incorrect and needed to be graded
> manually.
>
> The first part of the quiz marked correctly and was worth 20 points. (as a
> side note it would be good to see totals for each part).
> The second part was out of 10 and consisted of only one question. This was
> the part setup incorrectly.
>
> With several students we have tried adjusting their marks by changing the
> mark attributed to the correct mark and the following happened:
>
> The student had a score of 14 for the first part.
> The autoscore for the second was 8.
> So the auto mark was 22.
> The student go 4 not 8 correct in the second part so we adjusted the mark
> attributed to 4
> This should be fine. Except.
>
> The final score now shows 22.
> Removing the manual mark and leaving the autoscore show a result of 26
> Adding 0 as the mark for part 2 gives a score of 18 (which is the correct
> score for the manual remark).
>
> This seemed particularly odd so I removed ALL the auto scores setting them
> to 0 and the total score show 4.0 as the result.
>
> It would appear that there is a phantom 4 points appearing.
>
> We tried a second student.
>
> They had a score of 19 for the first part.
> A score of 8 for the second part.
> The autoscore was 27
> The correct score for part 2 was 5 .
> Adjusting the score to 5 shows the score as still being 27
> The correct score should be 24.
> Setting the part 2 score to 0 shows the score as 22.
> It would appear that this result is out by 3 marks.
>
> It appears that there is some form of bug. In both cases the scores have
> ‘phantom’ marks by the amount that the second score is out.
>
> I think this relates to how the scores are changed but I’m not 100% sure.
>
> Any ideas? Does anyone know of a bug that would match this behaviour?
>
> Gerwood
>
> _______________________________________________
> sakai-dev mailing list
> sakai-dev at collab.sakaiproject.org
> http://collab.sakaiproject.org/mailman/listinfo/sakai-dev
>
> TO UNSUBSCRIBE: send email to
> sakai-dev-unsubscribe at collab.sakaiproject.org with a subject of
> "unsubscribe"
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://collab.sakaiproject.org/pipermail/sakai-dev/attachments/20090505/6aad4afa/attachment.html