Educational reform always provokes controversy and arguments. I honestly started engaging in healthy discussions (read arguments) about educational reform when I was in my first education theory class in college. Most education reform arguments come down to whether the current paradigm is really broken, and whether the new paradigm is enough better to be worth the trouble replacing the old paradigm. Most of these arguments have fuzzy data at best to show for either side of the argument.
In my experience, these arguments tend to ride heavily on the past educational experiences of those involved in the arguments. The problem with this approach is that it assumes the two (or three or four) argue-ers are all equivalent learners. It assumes that all learners will thrive in the environment in which the argue-er thrived. It assumes that all learners have a mental processing intake system which acquires and stores information in a similar manner. It assumes that all learners are motivated to learn by the same motivations that drove the argue-er. No wonder these arguments are typically never resolved with the one party spontaneously saying, "Wow, you're right, and I was wrong all along. Thank you!".
Why is this? I think it is because we often make the assumption that all learners are equivalent in every aspect of acquiring, storing, retrieving, and applying knowledge. This makes it easier for us to create what little data we have as our current model for getting data on effectiveness of educational models. A p-value is not so useful if the entire cohort you are studying is a ill-defined mass of goo. Unfortunately, that is exactly what we have as our substrate, an ill-defined mass of goo.
What do I mean by this? Take neuroscience education in medical school as an example. First, there are obvious background differences - people with advanced degrees in neuroscience mingle in the class with those who have no idea what the frontal lobe is all about. Second, the way people learn is different. When I was a resident, I liked to see a few patients, and then take time right then to look up a bunch of stuff about those patients. I had friends who would rather be slammed with as many patients in a shift as they could find, as they felt they learned better in the doing. Some people like learning large concepts, and then going into details, and others like learning the details first, and then piecing them together later into a larger whole. Some people like to focus on one system or organ at a time, and some people like to have multiple courses concurrently running, so there is more time to absorb the information from each course. Some people love concept maps. Personally, I've never been able to get my head around why they are so great. I'm more of an outline guy. With these differences, we are trying to measure and argue over substrate that is an ill-defined mass of goo.
I'm not saying there are not basic learning theory principles which can be universal. I am saying the application of those basic learning theories is sometimes more wibbly wobbly than the ed-heads like to let on in their arguments. It could be that this multiple choice test on whether education reform is needed is not really a multiple choice test. It's an essay test. And there are multiple right answers as long as you can justify your answer. And everybody hated those tests...