When I was at Agassi Prep in Las Vegas, we piloted programs to use data in our assessment of students. Even though the kids still lagged far behind many of their peers from across the City in mostly non-minority schools, the students made great gains in a period of less than four years.
I must admit that I do not recommend that quantitative data alone is solely the answer but it certainly helps. It helps when trying to move the very bottom quartile kids from where they were to the next quartile up. It also works when trying to move the second two quartiles to the top two, but less well. However, it didn't work as well when trying to move the top kids up more. Why? It seems that instruction was aimed at the middle and lower half of the class rather than to the very top achievers. Often, there is even a negative return on data mining investments because the bright kids in a class want to be engaged in a different way.
What does this mean?
The question asking and other inquiry-based methods of learning help students more than putting kids into buckets (or quartiles). Yet, "sit and get" kinds of exercises are meant to help a large majority of students get through the gates that a particular school holds.
What are your experiences in using data successfully to drive student instruction?
