But on the testing end of things, perhaps the models provided by the AICC are the best easy models available. Expository writing, interviewing is the only real way I know of to test the depth of retention and comprehension of what a student has learned.
Perhaps an answer is to create the tools by which the student must create the test themselves, rather than take it, using the models you cite. Then you will be assured they have at least known that material long enough to create it, and in contradiction to the wrong answers they provide, which should provide a context that would imply some real comprehension. How to evaluate this would pose another problem.
Mark
On Aug 11, 2004, at 9:53 AM, Richard Gaskin wrote:
Marian Petrides wrote:
Not only in teaching programming but in designing custom educational courseware. Who wants the student to have ONLY simple multiple-guess questions to work with?
Life doesn't come with a series of four exclusive-or questions tattooed across it, so why give student this unrealistic view of the real world, when a little work in Rev will permit far more challenging interactivity?
Agreed wholeheartedly. Education-related work was the largest single set of tasks folks did with HyperCard, and for all the tools that have come out since there remains an unaddressed gap which may be an ideal focus for DreamCard.
But moving beyond simple questions models like multiple choice is difficult. The AICC courseware interoperability standard describes almost a dozen question models, but most are variants of "choose one", "choose many", "closest match", etc., sometimes enlived by using drag-and-drop as the mechanism for applying the answer but not substantially different from what gets tested with a simple multiple choice in terms of truer assessment of what's been learned.
The challenge is to find more open-ended question models which can still be assessed by the computer. For example, the most open-ended question is an essay, but I sure don't want to write the routine that scores essays. :)
What sorts of enhanced question models do you think would be ideal for computer-based learning?
-- Richard Gaskin Fourth World Media Corporation ___________________________________________________________ [EMAIL PROTECTED] http://www.FourthWorld.com _______________________________________________ use-revolution mailing list [EMAIL PROTECTED] http://lists.runrev.com/mailman/listinfo/use-revolution
_______________________________________________ use-revolution mailing list [EMAIL PROTECTED] http://lists.runrev.com/mailman/listinfo/use-revolution
