Nationwide, schools of concentrated poverty have poor academic outcomes and limited educational opportunities. Our local schools are no different. Education reform is a priority. But are we basing our reform efforts on accumulated evidence or simply on stories that tug on our emotions?

The students pictured below attended a hyper-segregated school in one of the lowest performing districts in the nation, Baltimore. Yet they were all high-performing. Despite what decades of school research describes – that segregated, nonwhite classrooms in low-income areas are typically low-performing – my experience as the teacher of this class was different.

We all have narratives about education. Pop culture also introduces powerful characters such as Lean on Me’s Joe Louis Clark. They reinforce the narrative that any kid can learn if only the principal or teacher cares enough. Such narratives are powerful, but they pick and choose among facts. In all this, money is a common theme. In Charlotte, the debate generally divides into: Low-performing schools have too few resources, which causes the achievement gap. Or, low-performing schools have the resources but they’re not well allocated, taking resources away from high-performing schools in more affluent neighborhoods.

Neither tells a full story based on evidence. Research makes clear that per-pupil spending (the amount of money spent to educate a child) is a complicated issue. It shows that per-pupil expenditure and achievement have a relationship, but not a direct correlation.

You don’t need to spend $19,000 on every student, the way Washington, D.C., does, to have good educational outcomes. That’s the highest per-pupil expenditure in the nation, and it’s coupled with some of the lowest outcomes. Charlotte-Mecklenburg Schools spends about $8,500 per pupil, with better results.

In Charlotte, the most expensive schools are segregated schools of poverty. Due to the challenges of those environments, to teach the students requires additional staff and smaller class sizes. Even with additional money, racially and socioeconomically segregated schools continue to show minimal growth, if any. This is not necessarily due to poor facilities, teaching or leadership – although high-poverty schools can suffer from those things – but due to the challenges of putting so many extremely high-need students into the same building.

In Charlotte-Mecklenburg, 42 percent of schools are hyper-segregated, with 90 percent or more of the students of one race – the most expensive kind of school to run, and the most difficult in which to increase student achievement. Yet, here and across the nation increasing the number of racially diverse schools is rarely discussed as a viable school reform.

The only time the black/white achievement gap narrowed significantly was during the height of desegregation efforts in the 1970s and ’80s. Evidence suggests that school diversity is an effective tool to improve educational outcomes, and the federal government has declared this to be a “compelling interest” for our nation.

My experience as a teacher showed me that a student can be poor, black, speak a different language at home and also do well in school. But my experience as a school administrator and researcher points to diverse schools as better for kids and for communities.

Why is there such a disconnect between evidence and policy? Problems are complex and rarely solved with simple solutions. We humans like to think we make decisions based on data and logic, but our normal habit is to develop a quick belief about a situation and then seek information to reinforce our belief. We want data to confirm what we already “know” is true. Researchers call this confirmation bias, and all of us are guilty.

Educational research literature is surprisingly consistent: Diverse schools are better schools for a community. But nationwide most school reforms are geared toward making high-poverty schools work despite evidence that such schools are expensive, incredibly challenging and typically have poor academic and social outcomes.

Everyone can play a role in trying to connect evidence to policy and minimize confirmation bias.

Demand data-based decision-making. Decisions should be based on evidence, not anecdote.

Want diverse schools in Charlotte? Demand a diverse school for your child, but don’t blame the school system. Charlotte-Mecklenburg Schools fought against the court case that ended the mandatory school desegregation plan the system was under, 1971-2001. Want to do even more for diverse schools? Support housing initiatives to enable high- and lower-income families to live in mixed-income neighborhoods. Choose to live and work in diverse areas.

Seek information to challenge your belief system. Constructively engage with people who have dramatically different ideas.

In Charlotte, we don’t have to look far for a model of a community that worked together to achieve something many people said was impossible. Not too long ago, our local politicians and business leaders crossed party lines and sent their children to the schools that were making headlines.

Although our school system was not perfect, in 1984 CMS was held up as a national model of what a community could do when people came together and went past their biases in order to do what worked, for kids and the community.

That was the year I entered kindergarten. I was the beneficiary of diverse schools for 13 years. My hope is that one day soon, our community will move toward community building rather than community dividing, and will make wide-scale policy decisions based on what we know – rather than what we think to be true.