"perhaps we need to take a scientific approach to figuring out how to take a scientific approach to education."
That's part of what I meant by "aggressive". I've actually been involved in some education research too and was generally unimpressed by the end result, which I would define as "impact on the education given to real students". (Ultimately, isn't education research a very practical thing? In the end, abstract research results are irrelevant.) Partially, the problem is the same as the one nutritionists face; it's a complicated system resistant to the sort of reductionist analysis that science really thrives on. All any given study can do is nibble around the edges, with no guarantee that the edge in question has any particular relevance.
On the other hand, there are massive structural problems too. I don't see much evidence that any of this science is actually feeding back into schooling itself; it seems very detached from the process, whereas schooling remains dominated by political concerns and people just sort of spouting theories off. It was certainly that way when I was a kid 20 years ago (I clearly recall being on the receiving end of educational theories that when I checked up on them ten years later never had anything remotely resembling a scientific background, it was just some educator's pet theory), and I'm not seeing a lot of evidence that has changed much. Schools are now too busy teaching to the state mandated tests to have any room to try anything new that might fail. The feedback loop between science and schooling seems to be almost entirely severed. Please do correct me if I'm wrong on this. But in all the endless debates about school I hear I sure don't here much science getting cited.
Note that for both issues, I have no beef with the researchers doing real science.
Perhaps, at the risk of offending the scientists who wrote the initial post, what we need is a more engineering approach to schooling. Engineering of the sort that engineers do when they have to run ahead of the science a bit, but still produce results. Certainly looking at it from a computer programmer's point of view, I see a system that needs to be rolled back to step one of the engineering process. Now, you might think I mean "burn it down and start from scratch", but I actually mean, it's time to sit down and do requirements analysis again. Do we even have a clear sense of what we really want from schooling? I'm not even sure we have an answer to that question, and from an engineering standpoint that indicates to me we're not starting on a firm foundation.