MoodleRoom's proposed Outcomes changes

Re: MoodleRoom's proposed Outcomes changes

by Tim Hunt -
Number of replies: 1
Picture of Core developers Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Peer reviewers Picture of Plugin developers

I can't help thinking that your last paragraph rather makes my case for me.

Another point: Are you going to get Microsoft to change the format of .doc files, so that when a teacher uploads a .doc to the course, it automatically gets assigned to the right outcome? I didn't think so. Quiz questions are similar.

Outcome mapping is not inherently part of a learning object, although it is metadata associated with it, and associating metadata with things is still a weakness in Moodle. Moodlerooms may not want to completely solve atbitrary metadata on things as part of this development, but we could try to store the data in a way that assumes that is coming eventually.

All this should be entirely compatible with getting the Outcomes UI you want. I agree that your use case (Teacher imports Examview files, builds quiz from the questions, has outcomes automatically set up (or not at their choice) is well worth having.

On the subject of metadata, does IMS LOM (http://www.imsglobal.org/metadata/index.html) have an encoding for outcomes mappings? As usual IMS standards are sufficiently incomprehensible that I was not able to tell quickly.

In reply to Tim Hunt

Re: MoodleRoom's proposed Outcomes changes

by Phill Miller -

Tim, 

Not sure I understand the point on the Microsoft document.  

I guess that I need to add a new use case (or three) about sharing of questions that are mapped to outcomes.  I've been working under the assumption that we would want users to be able to share and collaborate on questions without collaborating on the whole quiz.  So, in the simplest use case (which I can add to my use case document), two secondary math teachers are wanting to give Algebra quizzes.  One wants to give a quiz of 100 questions (he's a mean, mean teacher), and the other wants to give a quiz of 10 questions.  They should be able to use a shared question (if they both were to have access to it).  

This is the way that other LMS systems have implemented it (I know that ANGEL and Blackboard do it this way, because I wrote the spec and implemented it for ANGEL, and Blackboard basically did it the same way a couple years later).  Not saying that we should it because of that, but the workflow has been ironed out over the past 5 years in those systems.  

I guess in the end, I'll do in the spec whatever the users that we focus group with the spec next week tell us to do.  From a technical perspective, I don't necessarily mind how it gets implemented.  In the end, you own Quiz, so I'll defer to your judgment.  I just want to make sure that the end users get to weigh in before we make any conclusions.

re: IMS - I know that IMS LOM is not really being used anywhere that I know of.  The key standards would be IMS Common Cartridge.  The newest version of IMS CC includes an Outcome ID, which is mapped at the question level in the assessment and in the "Question Bank".  I will have to look at how we implemented Common Catridge in Moodle to see what we are exporting to the "Question Bank" item in a Common Cartridge.  We may not be exporting anything at this point into that particular field.  I know that we have not implemented the most recent version of the CC spec which has outcomes, so I included a use case for that.  If we are exporting into the Question Bank element of a CC, we should include the Identifier there.  

Unfortunately, today, the CC ONLY has an Outcome ID, so unless you have a shared set of outcomes with matching IDs, this is not terribly helpful.  Of course, you could export it and import it into the same moodle, but you can just use Moodle Backups to do that. The reason that it was implemented that way is that the publishers do all agree on the same set of outcomes (in this case, it's the Academic Benchmarks outcomes, or in a few cases, it is their own internal set of outcomes), so it works for the publishers.  It doesn't really work for the individual institution that wants to implement outcomes, however.  I'll be at the IMS meeting in May, and I can try to dive deeper into them.  

Thanks for spending time on this.  I'm really enjoying the dialog.  Getting this stuff hashed out before we start coding is key, and Kris and Mark are itching to get started.  Glad to see that there's been a lot of interaction on this spec already!  

Thanks! 

Phill

Average of ratings: Useful (1)