In the current eCAFE, we allow users to create their own questions to add to their surveys. A number of problems arose:
1.) Users didn't bother to check if the question they wanted was already in the set of previously defined questions, so they ended up creating ones that already existed, except now they can't be used for comparison to other department/campus instructors.
2.) Users created the questions, but forgot to add them to their surveys, resulting in confusion when they got their results and the questions were absent.
3.) Users created vaguely worded, multi-part, and/or confusing questions.
I want to get rid of user-defined questions. Only a small percentage (7.5%) of all participating instructors have used this feature, and many of them fall into category 1. The problems and confusion don't seem to be worth keeping it around. Instead, I propose that people can submit questions to be vetted by an organization (current CAFE staff?) and added to the general set as appropriate.
Subscribe to:
Post Comments (Atom)
2 comments:
In the paper version, how are user-defined questions handled? Are they allowed to just put it in?
I like the idea of being able to submit the question to the test bank - then it would be cool if the "system" could reflect back similar questions and if not on the list, it would automatically submit to the eCAFE advisory group.
I like your idea regarding the system suggesting similar questions, but that's going to have to be a future feature. I'm afraid language analysis isn't my strong suit, so I'll need to look into it. In the meantime, I think we should have a form where the question initially comes to me, and if I think it has merit (it's not a duplicate), I can forward it to the designated group of people.
I don't know what happens in the paper version, so I'll have to wait for C or J to provide the answer.
Post a Comment