Driving for Equity -- in What?

Count me among those who applaud the new guidance from the U.S. Department of Education on enforcing fair access to critical educational resources. For far too long and in far too many places, the deck has been stacked against students of color: fewer experienced teachers; less instructional time; fewer teaching resources; antiquated technology. It's long past time that we turn these patterns around and deliver on the American promise.

But unless the secretary of education takes advantage of the opportunity he will have later this month to reverse an ill-fated decision he made three years ago, there is great risk that schools and districts will get the wrong message: that low achievement for some groups of children doesn't matter so long as resources are equal. Not a legacy, I suspect, that either the secretary or his boss wants.

A little history:

Even the most fervent critics of the No Child Left Behind (NCLB) law think it did one thing right by pushing schools to improve the achievement of every group of children they served. But the law's mechanisms for making all children matter weren't very nuanced. Schools that failed by an inch to meet their goals for one group of students -- say, students with disabilities -- got the same label and, over time, were subject to the same consequences as schools that were habitually failing all of their students by a mile. This encouraged states and districts to spread their available resources over a lot of their schools instead of concentrating them on the ones that needed the most help.

Secretary Arne Duncan sought to correct the problems of NCLB in 2011 when he decided to issue broad waivers exempting states from many of the law's requirements. States were told to focus their energies on fixing their lowest performing, "priority" schools -- those in the bottom 5 percent -- as well as the 10 percent of schools with the largest achievement gaps. That's a good thing; a lot more resources and attention are now going to schools that really need help.

But along the way, Duncan also did something that keeps me and a lot of others in the education and civil rights communities awake at night: He took away the pressure on the other 85 percent of schools to improve achievement among their low-income students, students of color, students with disabilities, and English-learners. Yes, states still have to set improvement goals for those groups of students. But meeting those goals, and actually accelerating the learning of these students, doesn't have to count at all in school rating systems.

As public policy, that just doesn't make sense. After all, the federal government is sending billions of dollars to local schools to aid in educating these very groups of children. To not even ask that they learn more as a result is unconscionable, especially in view of the long history in most states of under-educating these very same children.

But as our team at Ed Trust found in a new analysis, that is exactly what Duncan's waiver policy did. Under new accountability systems, schools are getting top ratings despite low performance among some groups of students. In fact, the differences are so large that top-rated schools often perform similarly for their low-income students and students of color as middling to low-rated schools do for their white and higher income peers.

In Minnesota, for example, low-income children in top-rated schools are performing about the same as higher income children attending schools identified for intervention. The same is true in Florida for black children in "A" schools -- they are performing below white children in "C" schools -- and Latino children in "B" schools are performing about the same as white children in "D" schools.

In Kentucky, the pattern is much the same: Black children in "distinguished" schools are performing about the same as white children in "needs improvement" schools. Yet despite these patterns, top-rated schools in most states are under no pressure whatsoever to improve achievement among their low-income students or students of color in order to keep their ratings.

Fortunately, Duncan and his boss have one last chance to put this right. Later this month, when they issue guidance for the renewal of NCLB state waivers, they can make a little tweak to the requirements that could, over time, help make a big difference in the lives of children. They don't have to ask states to gut their accountability systems and start again; states are already setting progress goals for these groups of children. All the secretary has to do is ask them to make achieving progress against those goals matter in school ratings.

This is critically important. When low-income children and children of color didn't matter in state accountability systems, state and local policymakers and educators were given a pass to ignore their needs. That a Republican administration would try to change that by making them, in effect, the only thing that matters in school ratings, only to have this administration make them not matter at all, feels just plain surreal. That's certainly not a legacy that either the secretary or the president should want to have.