The School Choice State Flying Under the Radar

Welcoming the newest member of the 100,000 Club

Quietly, North Carolina has grown into one of the largest school choice states in the country. Preliminary counts show North Carolina eclipsing 100,000 students participating in its voucher or ESA programs. Prior to this year, the only other states in the 100,000 Club were Florida and Ohio. (Arizona would count if you included their tax-credit scholarship programs.)

At the time of this writing, the North Carolina State Education Assistance Authority (NCSEAA) reports just under 99,000 students are receiving Opportunity Scholarship vouchers in 2025-26. Based on past experience, I expect final totals at the end of the year to tip over the 100,000 mark. This number triples the number of students enrolled in the program two years ago, when the program was targeted to households below 175% of the free and reduced price lunch income limit. The bulk of that growth occurred in the first year of the program’s expansion, but its rate of growth in its second year of universal eligibility remains substantial.

In addition to the Opportunity Scholarships, North Carolina has an ESA program targeted to students with disabilities. About 5,500 students are using an ESA in fall 2025, growing more than 10% for the second straight year.

As someone who has spent a lot of hours trying to track down descriptive data on school choice programs, I appreciate North Carolina’s reporting practices. (My colleague Colyn did a nice overview of data reporting across several states here.) For both programs, you can track participation by grade, county, race, ethnicity, and renewal status.

Where are the school choice kids?

I want to focus on the geography of choice participation here, as one simple extra step can provide some additional insight. Wake and Mecklenburg counties, the homes of Raleigh and Charlotte, respectively, have the highest raw numbers of participating students. You can see the raw numbers in the interactive table below.

For the full table of county-level numbers, go to this Infogram page.

NCSEAA offers a heat map showing where choice students are concentrated, and the map closely reflects population density. Taking county population into account, however, suggests that population density doesn’t appear to affect the likelihood of a family taking up one of North Carolina’s choice programs. The map below showcases choice participation as a ratio of public school enrollment in 2024-25 (the most recently available school-year data). Hover your cursor over a county to get a sense of the concentration of students using choice programs there.

For the interactive map, go to this Infogram page.

The major metro counties have participation rates around 7% to 8% the equivalent of public school enrollment. I use this ratio as a proxy for take-up rates as North Carolina does not publicly report charter school participation by county, and while the federal Department of Education reports county-level charter school enrollment as provided by states, it lags too much to be useful for me here.

Many counties have similar rates. Notably, though, some of the highest concentrations of choice participation are in some of the rural coastal counties. That said, some of the lowest participation rates are also in some rural coastal counties. The Appalachian counties compose the region with the lowest rates.

Altogether, there appears to be no detectable relationship between urbanicity and choice participation rates, at least on a county level.

Digging a little deeper with Census data

To test this impression, I considered two Census measures of urbanicity from the U.S. Census Bureau: the county-level Rural-Urban Continuum Codes, and the percentage of a county’s land that the Census classifies as urban. I calculated simple linear regressions between each of these measures and the choice participation rate, just to see if there was any hint of “something there” that I was missing.

Sure enough, both calculations produced essentially nil results, though I’d caution against generalizing these very basic statistical explorations beyond the North Carolina context. We’re just looking at 100 counties, after all.

I don’t know nearly enough about North Carolina to posit any reasons that choice participation is distributed the way it is. I don’t know what’s going on in the coastal counties that is not happening in the western counties, and I don’t know what’s preventing the low-rate coastal counties from joining their high-rate neighbors.

I can say, however, that I’m not that surprised rural participation is as high as it is. The Brookings Institution calculated in 2017 that 69% of families in rural areas lived within ten miles of a private school. I don’t know what a focus on North Carolina would have looked like, and I’m not exactly sure how that number might have changed in the last eight or nine years, but it’s a sign that more people are proximate to more schooling options than we often assume.

This analysis doesn’t even account for all the non-tuition, non-brick-and-mortar education services an ESA could deliver. And it’s quite possible that the private education sector will grow, as a survey from the John Locke Foundation posits.

North Carolina’s diverse geography provides some very interesting research opportunities, and it’s a state that demands attention as its choice programs continue to grow.