Pharyngula just got back from the Society for Developmental Biology meeting, and he's writing a lot of neat stuff about it (and I'm sure there's more on the way). One thing that caught my eye was his discussion of one of the education forums, including a good discussion of the in-class response systems I've written about before.
The presenter that Pharyngula discusses got relatively unexpected answers to two questions he asked using the in-class response system. Instead of following up these unexpected answers, the presenter just continued on with the presentation, which caused some audience grumbling (though in his defense he may have had a tight time constraint). This is a good illustration of just how much flexibility is required when using in-class response systems: as an instructor you need to be willing to throw out your planned lecture so you can discuss the results of questions you've asked. If you don't, it's likely the students will learn that their answers are meaningless, and you'll probably stop getting thoughtful responses. And, after all, the whole point of the system is to get feedback on how the students are doing, so using the system while ignoring the data seems self-defeating.