<p>There’s a huge difference between reporting data that would be in one institution’s database (e.g., a student’s college grades, and whether he needed remedial instruction, as well as his high school, would all be in the college’s database), and matching up granular information from two different institutions (e.g., college performance vs. high school curriculum). The latter would be much more powerful, but essentially requires a lot of individual tracking. My understanding about initiatives such as Colorado’s is that eventually the state or independent researchers – who would not have access to names, just numbers – would be able to analyze the data across institutions, not that high school administrators would have access to individual information about how their graduates did in college.</p>
<p>Not that I have so much trouble with that, but I sense lots of people would.</p>
<p>As for the study in the AP book, I think it stops one level short of what I think would be important. It’s not a question of tracking, or academic background. In many schools, the students with worse academic backgrounds and little prior science who suddenly turn up in AP science courses are students whom the teachers have identified as “diamonds in the rough”. That’s certainly how it worked in my kids’ school. And it’s not so surprising that those students would later outperform their peers who were not so identified and encouraged.</p>