I tried to post this comment at the Ecosys wiki. For some reason, computer wouldn't let me. So I post it here.
I think this problem is easier to clarify if we look at using the lens of Time and Space. In that framework, one can argue that traditional education had little to do with teaching or learning, but in fact was a way to time train agricultural workers to enable them to integrate into a mass industrial society. I think there is enough research out there to suggest that the value add of education has been negligible once the family income and educational level of parents are held constant.
This has not previously been seen as a problem because in the past there were an abundance of jobs that didn't require higher order thinking. It is presenting as a problem today because financial resources in the States are becoming limited, and the cost of maintaining the edu-industrial complex is unsustainable. It is presenting as a political problem (USA) because it is only since the 1960's that educating traditional under class parts of the population has been moving to the top of the political agenda.
The emerging reality is that productive workers no longer have to be bounded by the same time and space constraints. With this new found possibility, there is a push from the bottom - most especially from parents - to control the time of education. I think it's fair to say that growth of homeschooling is evidence of this push.
To better contextualize the stakeholders I think it helps to consider that they are driven by proximate incentives that allow them to save time. Mostly they are too busy, being busy to focus on the system problems at hand. People in general tend to do what is easiest for them to do in the service of managing going forward risk. What is often under appreciated is that the risk of "wasting time" is an always on proximate dis incentive to doing anything they have not previously done.
The potential power of "documented" IEP is that the documentation itself creates long term focus. There are a number of examples from the world of clinical medicine of the power of the list to get to significant reductions in ER medical mistakes. I bet the same would work in education.
As pointed out above the challenge is Time for documentation. The most promising avenue is on line learning programs for well defined skill acquisition. It frees the teacher from the data entry function and creates real time data on the progress of student learning.
My own feeling is that the only things we should measure are things that can be reliably measured in real time. In addition to the results from on line activity, my thought is that the best things to measure are attendance and homework compliance. They are surrogates for "engagement." They are also the best early warning signals for dropping out.
