The disappointing results of the Duncan administration’s controversial School Improvement Grants exemplify the fundamental flaw of test-driven school “reform.” The preliminary analysis of the SIG’s first year indicated that some schools produced double digit gains in math and/or reading test scores, that more produced single digit gains, and one third saw declines in student performance.
A quarter of those schools showed a reversal of the decline in test scores that had been occurring before the infusion of millions of dollars. Another quarter had been improving however, and saw a reversal of their previous gains. It could be argued that SIG benefitted elementary schools, but not high schools or, especially, middle schools.
These findings are consistent with the conclusions of the National Academies of Science’s blue ribbon panel, which explained that test-driven reforms may pressure some schools to become more productive while encouraging others to increase primitive teach-to-the-test and push out the harder-to-educate children. They are also consistent with the research of Paul Tough, who showed that the contemporary reform movement has achieved some successes, especially in schools that can exclude the most troubled children, but that it has probably worsened the lives of children in our toughest schools.
That leads to the metaphorical question of what parent would continue an experimental treatment that significantly helped one of his children, perhaps produced some benefit for another, while damaging a third? Why do we keep trying risky experiments that are bound to harm some children as they may or may not improve outcomes for others?