[benkahuna]

Middle school grades don't matter, at least for most colleges. The people I knew that did the best in middle school were hardly the best in at my high school and later in college.

Grades are a superficial indicator of success, learning, and study habits. They're even worse as a predictor of future success in education and later in careers.

Your experiment could easily result in kids achieving better grades by means other than actually studying hard and learning, such as cheating. I'm not going to say cheating is valueless though I don't care for it. It may mean someone is smart in non-conventional way or knows how to work a system, some of this skills of resourcesfulness could be useful later in life, however, I think playing by the rules for the most part (unless you're a pioneer in a field) leads to the most consistent long term success.

I think the idea of your experiment (small investment for bigger returns, helping out kids with trouble in school, etc.), but I think the means you mention using to achieve that success and your definition of success (though you obviously have to choose something operationalized and grades are not useless in that regard) are flawed.