Protestors march in front of the New York State Education Department building in Albany on Monday. (Tim Roske)

Greg Richmond is president and chief executive of the National Association of Charter School Authorizers.

The national conversation around how and why to test kids in public schools has gone astray. Loud and persistent voices have been decrying a culture of “test and punish,” which they say hurts teachers, stresses kids and compromises creativity in the classroom. I’d join them in their chorus if what they were saying was true. But it isn’t.

We don’t have a “test and punish” culture in our schools, we have a “test and improve” model that has produced dramatic results for poor and minority students in recent years. As a nation, we need to continue to test and improve.

Charter schools are one part of our public school system where “test and improve” has delivered good results for students. Charter schools and their authorizers use test data to continuously monitor student learning and take actions to improve. Charter operators that consistently deliver better outcomes for children often open additional schools. Those that consistently fail children risk closure by their authorizers.

This culture of “test and improve” is working. Students attending charter schools in urban areas receive the equivalent of 40 days of additional learning per year in math and 28 additional days per year in reading. In the past two years, our researchers have determined, charter school authorizers used school performance data to approve 944 new, quality charter schools and to close 416 schools that have persistently failed. As a result of those actions, 452,000 students are attending better schools. That’s not punishment, that’s improvement.

“Test and improve” is not limited to charter schools, of course; it has been happening throughout our public schools since the passage of the No Child Left Behind act in 2001. NCLB is reviled by those who decry standardized tests, yet the act has been working. The reading scores of 13-year-olds increased more in the first eight years of testing after NCLB than they did in the 28 years before NCLB. In math, the scores of 13-year-olds increased roughly twice as fast after NCLB as before.

Tests don’t punish, they inform. With data from tests, educators and communities can take actions they deem necessary to improve the quality of our schools.

Yet right now, Congress is considering policies that threaten these successful “test and improve” policies. On the right, advocates for vouchers and the free market are pushing for testing loopholes such as opt-out provisions and doling out federal money in block grants with no performance requirements whatsoever. On the left, teachers unions are attacking the provisions of NCLB that require failing schools to take actions to improve.

Ironically, if it moves away from successful “test and improve” policies, Congress would in fact be creating a system that punishes students. After all, if students aren’t tested and schools don’t improve, it is students who will be punished. Students will be punished when a school fails to teach them to read and write. Students will be punished when a school fails to prepare them for success in college, jobs and life.

Kids shouldn’t have to go to schools that fail them year after year. As Congress debates whether to reauthorize NCLB, let’s not forget that our nation needs quality testing data to make well-informed decisions about how well all public schools are working for our children.