What the Data Won't Tell You
Last year, from August to June, I sat in the background of Room 121 at Bancroft Elementary School in Washington, DC, and watched as co-teachers Rebecca Schmidt and Rebecca Lebowitz succeeded and struggled to reach the sixty 3rd graders in their charge.
I was there to research a new book about a year in the life of two DC schools - one district, one charter. My hope for the project is to put a human face on the modern landscape of school choice (I also spent time with several families in search of the first school for their children), and to give readers a clearer sense of the obstacles and opportunities our nation's teachers face in their efforts to help all children learn.
It was with great interest, then, that I received an email from the school ratings site greatschools.org with the headline, "What do test scores at Bancroft Elementary really mean?"
I clicked on the link and was taken to a graphical comparison of the school's reading and math scores from the past four years, along with the state average for 2012. Underneath the state score, I read this: "56 students were tested at this school in 2012."
56 students, I thought. I know those students. I know every one of them. I know their strengths and weaknesses. I know what they wrote as their hopes and dreams at the beginning of the year. I know when they got in trouble, and what for, and I know exactly how much growth they demonstrated, individually and as a group, over the course of the school year.
Because I was there, I know two other things: Ms. Lebowitz and Ms. Schmidt are great teachers; and their students showed remarkable progress, made all the more remarkable by the fact that just 5% of them - or three out of sixty - were reading at grade level at the start of the year.
I mention all this because I want to ask you something: if you were a parent shopping for schools, and you read this report, would you send their child there?
To be clear, I'm not an opponent of data-driven decision making: indeed, even John Dewey believed data (which he understood through its original meaning in Latin as "something given") was the only way adults could make informed decisions that would support the development of the children in their charge.
What I oppose is narrowly defined data-driven decision making. So while I applaud Great Schools for rounding out their school profiles with information that goes beyond the school's test scores, I also know that for policymakers, "achievement" remains the benchmark of successful reform. The same is true for too many parents out there, navigating the nascent and chaotic marketplace of school choice in cities like DC. And emails like the one I received today serve as uncomfortable reminders of how little we really know, and actively seek to know, about the classrooms that help raise our children.
We can do better.