http://www.boston.com/news/globe/editorial_opinion/editorials/articles/2005/04/03/grading_schools/GLOBE EDITORIAL
Grading schools
April 3, 2005
WITH THE No Child Left Behind Act, President Bush promised to accomplish something that should have been done long ago: turn public schools into temples of opportunity for every student who walks through the doors. Bush has called for a hard-nosed look at the numbers, from standardized test scores to on-time graduation rates. The danger is ending up in an ocean of data with no plan for getting to shore -- for making sure that students have reached academic proficiency by the federal deadline of 2014.
Now the Broad Foundation and the Gates Foundation have given $45 million to fund SchoolMatters, a free website that crunches the numbers to discover and disseminate the best school policies and practices. A flashy option would be to have top 10 lists like the ones that rank the nation's colleges. But the website, which was developed by Standard & Poor's, correctly says this would be a ''serious misuse of the data." So there are no best-of-the-West or top-urban-scorers lists. Instead, the site puts schools in context, looking at a mix of demographics such as geography, spending, and numbers of poor students. Educators can take a look at their own schools or at schools with similar traits and ask: What are my colleagues doing that could improve my school? The site also calculates spending-to-outcome ratios, which can indicate how effectively money is being spent.
Educators in Michigan and Pennsylvania have used the website to examine personnel costs, assess student performance, and modify teaching methods. Expanding SchoolMatters to cover all states is a two-year experiment in venture philanthropy. The premise is that transparent, accessible data can spark public debate and, ultimately, progress. The site can also be used by parents, legislators, and students -- an attempt to democratize access to the data and increase the number of people who might develop sound ideas for education reform.
The website must be used with care. Because SchoolMatters is new, it will take time to make sure that the site has the best statistics available for each state. And because of states' varying policies and programs, straight numerical comparisons aren't always possible. The bottom line: Numbers and statistical models can't tell the whole story. To get the best out of the website, users also have to do the legwork of looking at the qualitative factors that make schools fail or thrive. SchoolMatters must succeed at two jobs: helping improve education and helping the public manage and master the information age's massive streams of data.