From G.H.Smith at leeds.ac.uk Mon Feb 8 18:42:18 2010 From: G.H.Smith at leeds.ac.uk (Graham Smith) Date: Mon Feb 8 18:42:55 2010 Subject: [QTI] Modal feedback -- a better way to do this? In-Reply-To: <20100131135555.GA15587@sponvoid.ecs.soton.ac.uk> References: <20100128160259.GA2274@teee> <20100130123559.k608c76zko4okkkw@webmail.leeds.ac.uk> <20100131135555.GA15587@sponvoid.ecs.soton.ac.uk> Message-ID: <20100208184218.tj0az9xgqo08g4w0@webmail.leeds.ac.uk> Hi: Attached is one way of doing what you want. It is relatively concise, but needs some tidying up to get the feedback display into a better order. I leave that to you. Graham Quoting Bart Nagel : > At 2010-01-30 12:35:59 +0000, Graham Smith wrote: >> In your XML example you have assigned 'null' for some outcome >> values, on the assumption I think that the rendering engine would >> treat 'null' as 'false'. I hesitate to say that this is incorrect >> -the specs are silent on the issue- but I think this assumption is >> dangerous. In JAssess at least null values in this position are >> rejected, and the result (in this case) is that your feedback with >> 'hide' attributes is never shown. > > Understood -- I'll no longer assume that. > >> I attach a version of the XML which is more concise and follows more >> usual practise. >> Because of the unusual way you have arranged the XML, I am not quite >> sure that I have interpreted what you intend correctly. But perhaps >> the attached version will point the way for you to achieve what you >> want. > > Hmm, the way you've done it wouldn't work if it was a multiple > response question. > > I'm attaching such a multiple response question (it's still making the > bad assumption I was making before, though). The feedback is pretty > useless here but it should hopefully get across what I want to do. > Each matching piece of feedback is shown, so if the first two choices > are picked the candidate is given the feedback for choice 1 picked, > choice 2 picked, choice 3 not picked, choice 4 not picked, choice 5 > not picked and choice 6 not picked. Not all of these bits of feedback > might exist in a given question but that's the idea. > > So in this case if the candidate picks "big", "blue" and "extremely", > any feedback for each of those having been picked is shown, and also > any feedback for the unpicked ones /not/ being picked. So here the > candidate would be told > > Right -- "big" is an adjective. > Right -- "blue" is an adjective. > Wrong -- "extremely" is an adverb. > You missed "cubic". > > Hope this makes sense. > > --bart > > -- > bart nagel > e: bjn@ecs.soton.ac.uk Skype: trem-b > w: ecs.soton.ac.uk/people/bjn Aim: bartremby > XMPP: tremby@jabber.org MSN: msn@tremby.net > -------------- next part -------------- A non-text attachment was scrubbed... Name: AdjectivesG.xml Type: text/xml Size: 7562 bytes Desc: not available Url : /public/ims-qti/attachments/20100208/1cf12708/AdjectivesG.xml From bjn at ecs.soton.ac.uk Thu Feb 11 12:14:02 2010 From: bjn at ecs.soton.ac.uk (Bart Nagel) Date: Thu Feb 11 12:14:59 2010 Subject: [QTI] Modal feedback -- a better way to do this? In-Reply-To: <20100208184218.tj0az9xgqo08g4w0@webmail.leeds.ac.uk> References: <20100128160259.GA2274@teee> <20100130123559.k608c76zko4okkkw@webmail.leeds.ac.uk> <20100131135555.GA15587@sponvoid.ecs.soton.ac.uk> <20100208184218.tj0az9xgqo08g4w0@webmail.leeds.ac.uk> Message-ID: <20100211121402.GF7240@sponvoid.ecs.soton.ac.uk> At 2010-02-08 18:42:18 +0000, Graham Smith wrote: > Attached is one way of doing what you want. It is relatively > concise, but needs some tidying up to get the feedback display into > a better order. I leave that to you. Wonderful -- with a couple of modifications it works just as I want it. And yes, it's much more concise -- lovely. Thanks --bart -- bart nagel e: bjn@ecs.soton.ac.uk Skype: trem-b w: ecs.soton.ac.uk/people/bjn Aim: bartremby XMPP: tremby@jabber.org MSN: msn@tremby.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : /public/ims-qti/attachments/20100211/5f280bf3/attachment.bin From bjn at ecs.soton.ac.uk Thu Feb 11 18:44:08 2010 From: bjn at ecs.soton.ac.uk (Bart Nagel) Date: Thu Feb 11 18:44:58 2010 Subject: [QTI] Eqiat version 0.1 Message-ID: <20100211184408.GA29846@sponvoid.ecs.soton.ac.uk> I've just released the first version of Eqiat, an easy QTI item authoring tool. It is web based and aims to make simple question types easy to author. It's licensed under the Creative Commons "Attribution non-commercial share alike" licence. Only a few question types are supported at present but it's pretty easy to add more -- just write one PHP class file and no other changes to the code should be necessary. I don't have a permanent publicly viewable demo server yet but my testing server is temporarily visible to the world at so please have a play and send any comments my way. There's an issue tracker at so any bug reports should go there, please. It's available to download at -- both read-only git access and package downloads. Eqiat's written in PHP but calls a small Java program to validate the QTI (validate is available at but is also included as a subtree within Eqiat). It calls a remote instance of QTIEngine to provide previews of the items -- it uses the demo at qtiengine.qtitools.org by default. (Note that there's an outstanding issue whereby QTIEngine will throw error pages at you if you have a QTIEngine session cookie -- just clear all your session cookies or just that particular one and you should be good to go. Haven't figured out a workaround to this yet.) More information is in the README file. So please let me know if you have any comments or suggestions and report any issues you might find. I've only briefly tested on any browser other than Firefox 3.5 but things seem fine in Safari, Arora, Opera, Chrome and even my little goes on IE 7 and 6 seemed to go OK... Patches are welcome too, of course. --bart nagel -- bart nagel e: bjn@ecs.soton.ac.uk Skype: trem-b w: ecs.soton.ac.uk/people/bjn Aim: bartremby XMPP: tremby@jabber.org MSN: msn@tremby.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : /public/ims-qti/attachments/20100211/2278a28f/attachment.bin From Schepens at evident.nl Fri Feb 12 16:31:40 2010 From: Schepens at evident.nl (Koen Schepens) Date: Fri Feb 12 16:32:02 2010 Subject: [QTI] Multiple correct answers when cardinality = Ordered Message-ID: Hello, This is my first post to this mailinglist, actually to any mailinglist ?, so I hope I'm using it the right way. Our customer uses an orderInteraction for a question that enables users to drag items in a certain order. However, instead of one correct order, there are two. E.g. order [a,b,c,d] is correct as well as [a,c,b,d]. I cannot find a way to handle this in QTI (since there can only be zero or one correctResponse). Also, for budgetary reasons, there will be no use of the responseProcessing class. Thanks in advance! Koen Schepens Software Engineer Evident Interactive BV Koen Schepens Software Ontwikkelaar Schepens@evident.nl Evident Interactive B.V. Kanaalweg 14 C-D F: 030 298 50 51 From R.Bacon at surrey.ac.uk Mon Feb 15 17:44:29 2010 From: R.Bacon at surrey.ac.uk (R.Bacon@surrey.ac.uk) Date: Tue Feb 23 12:48:12 2010 Subject: [QTI] CAA conference References: Message-ID: <798670B35DD7C940921AFAE38CC7939301FFB73C@EVS-EC1-NODE1.surrey.ac.uk> Apologies for any cross postings. Please see attached call. Dick. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/ms-tnef Size: 53338 bytes Desc: not available Url : /public/ims-qti/attachments/20100215/d004005a/attachment-0001.bin From Srikar_Doddi at kaplan.com Fri Feb 19 15:44:08 2010 From: Srikar_Doddi at kaplan.com (Srikar Doddi) Date: Tue Feb 23 12:48:12 2010 Subject: [QTI] meta data in QTI Message-ID: <8547A5ECC02ED1479EE587D4E99E8DC608AB2AE5@EXCOATTV2.kaplaninc.com> Hi, I am not sure if this is an appropriate question for this group. Is there a separate mailing list for IMS meta data standard? I have the following questions: I would like to tag not only my items but within my items (sub-items like stem, stimulus, answer choices, media assets used inside the item) which are QTI format as follows: Meta-data --> Item level Meta-data ? stimulus level within an item Meta-data? stem level within an item Meta-data? image inside the stem level Meta-data?answer choices Etc. etc., IMS Meta-data seems to rely on LOM and it does not seem to have placeholders for sub-item level. Thanks, Srikar -------------- next part -------------- An HTML attachment was scrubbed... URL: /public/ims-qti/attachments/20100219/bd1c18ae/attachment.html From P.R.Neve at kingston.ac.uk Fri Feb 26 17:18:17 2010 From: P.R.Neve at kingston.ac.uk (Neve, Paul R) Date: Fri Feb 26 17:18:33 2010 Subject: [QTI] Spectatus - QTI 2.1 test authoring tool Message-ID: Dear all, I am pleased to announce the first public release of Spectatus, a QTI 2.1 test authoring tool, released under the New BSD Licence. Spectatus is a Java desktop application and can be installed via Java Web Start from the URL: http://aqurate.kingston.ac.uk/spectatus/webstart/spectatus.jnlp The intention was to provide QTI content authors on the FETLAR project with a simple, user friendly means of assembling questions into test content packages. As such, Spectatus is intended to interface with a number of other components used within FETLAR (most of which will probably be familiar to those on this mailing list, e.g. Minibix, QTIEngine/MathAssessEngine, Mathqurate and so on). However, you do not need to use those components in order to author tests with it. In terms of test authoring, I have tried to keep it simple, and make no assumptions about question content or subject-specific extensions to QTI. At least in theory, a test constructed in Spectatus should render in any tool that accepts a correctly packaged QTI 2.1 assessmentTest, assuming that the questions that comprise it do not use any extensions beyond those supported by the renderer. A brief overview document outlining the application's use can be downloaded fromhttp://aqurate.kingston.ac.uk/files/spectatus.pdf. This gives a user-level overview of the application's use, and some technical notes including source and compilation details for those so inclined. Feedback and thoughts are most welcome. Kind regards, Paul Neve Senior E-Learning Software Developer/Researcher CISM, Kingston University PS: I say that you don't *need* to use components such as Minibix, MathAssessEngine, etc. However, I would be happy to assist anyone who wishes to take a closer look at them. We now have a number of options within FETLAR that allow such components to be used with a minimum of configuration and IT-related effort, so those of you who might have been put off in the past might find it worthwhile to revisit them. The "useful links" page at the back of the overview document gives some more information, and please get in touch if you'd like some help to go further. This email has been scanned for all viruses by the MessageLabs Email Security System. -------------- next part -------------- An HTML attachment was scrubbed... URL: /public/ims-qti/attachments/20100226/41a7ae95/attachment.html