The biggest problem with NCLB is that we've been training students to do well on multiple choice tests. A skill they will have little to no use for once they leave high school. Every test in real life is open book, free response.
In my core-level geology classes at the local uni, I give take-home tests only. They do have multiple-choice questions, but not the kind requiring rote memorization of trivia. Plus some analytical problem exercises. Students have a week to complete them. I don't care if they get together and try to figure things out. My advice is "use all your resources", and if that includes each other, I'm fine with that. They have to remember that they might all get things wrong, as well as get things right, and both happen on these exams. I've never had any problem with working a class this way.
My experience is that students have a love-hate relationships with these exams. They like the absence of time pressure, but don't much like the difficulty of analytical thought required for them. Because they are conditioned to think of education in terms of vomiting rote-memory answers, stuff they won't remember the moment they walk out of the classroom after the exam.
I consider that testing should be a learning experience as well as an evaluatory one. I give a difficult mapping problem on one of my exams, and part of it involves figuring out a map scale, when none is provided. It's a map of the local area. There are a couple of ways to do figure it out, one rather subtle way on the map itself. But, when I go over the exam in class later, which I always do, I point out another one that never seems obvious: Go find another map that
does have a scale on it. I know for certain that such a map exists at the University library, and is easy to find. You could even figure it out from an on-line city street map.
But, to my unending wonder and applause, the first time I did this, when I asked people who had figured the problem out, and in what manner, I had a woman in the back of the classroom timidly raise her hand, and tell me she had gone to the local U.S. Geological Survey office and asked somebody about it. I think she thought that might somehow be cheating, but how good was that? I actually got a student to take the time and effort to go to the local U.S.G.S. office and talk to a real live geologist, to find out something. I can't think of a better result than that.
In-class multiple-choice exams is something I was really good at in high school and college. It's a skill. I'm good at trivia, and that helps enormously. But my son, who is extraordinarily smart (IQ once measured at 140), is horrid at these kinds of exams, because his mind just doesn't process information in the way that makes these exams easy for other people.
The whole point being that doing well on multiple-choice exams, under time pressure, measures . . . how good you are at doing well on multiple choice exams, under time pressure. And damn little else of lasting value.
caw