A new study from the Southern Education Foundation finds that half of all public school students now live in poverty. The poverty measure they use is eligibility for the federal free and reduced-price lunch program, part of the National School Lunch Program. According to study authors:

The latest NCES [U.S. Department of Education's National Center for Education Statistics] data indicates that 48 percent of all public school children across the nation were eligible for free or reduced lunch in 2011. The rate of low income students in the South was 53 percent – the highest rate among the regions of the nation. For the first time in recent history, at least half of the public school students in the West were low income. In 2010 the rate was 51 percent. In 2011, it remained 50 percent of all public school children. The Midwest had the next highest rate, 44 percent, and the Northeast had a rate of 40 percent. (p. 3)

Reaction to the report has been swift—and harsh. Diane Ravitch, for example, calls such findings a national shame. She blames various education reform efforts of the past few decades for draining resources from cash-strapped government schools. Her "solution" is more money for the schooling system.

To see how well the more money plan has worked out, see these handy charts from the Cato Institute’s Andrew Coulson. Spoiler alert: reading, math, and science scores for American 17-year-olds have remained essentially flat since 1970, but over the same period the cost of K-12 schooling for those students has increased 200 percent in real terms, from $55,000 to $165,000 each.

But let’s take a closer look at the measure the Southern Education Foundation uses to gauge child poverty. Turns out it’s as faulty as the notion that more money means better results from the public schooling system.

The percentage of school children receiving free and reduced-price meals has increased nearly five-fold from 15 percent in 1969 to more than 68 percent in 2012, according to the USDA’s Food and Nutrition Service. Yet over the same period the percentages of families with school age children living below the poverty level grew from 11 percent to 18 percent, according to the Census Bureau (Table 4).

This means the percentage of children receiving federally subsidized meals now outpaces actual poverty rates by four to one.

A leading reason for these inflated figures is that the federal meal program is bloated with fraud—as recent examples in Philadelphia, New Jersey, and Chicago illustrate. Lying about family income is rampant, and income verification is minimal at best.

No child should go hungry, but this kind of abuse is the legacy of the sort of Great Society thinking that’s left us no better off than 50 years ago.

If we can’t trust the government to provide accurate statistics, why should parents entrust their children to government-run schools?

Maybe that’s why today, more than 245,000 low-income, special needs, and other disadvantaged students nationwide are attending non-government schools of their parents’ choice thanks to 32 scholarship programs enacted in 16 states and the District of Columbia.

The most scientifically rigorous research spanning decades shows that parental choice in education improves the academic outcomes of the very students government schooling boosters say are un-teachable (including higher standardized test scores, graduation rates, college attendance and completion rates)—at a fraction of the cost.

Critics such as Ravitch insist such programs are an expensive distraction because they’re not expansive enough to help all students. Talk about a self-fulfilling prophecy.

In spite of detractors’ best efforts to litigate, limit, and de-fund parental choice programs, student enrollment has grown more than four-fold in the past decade alone. Trends such as this one have status-quo defenders hanging on by an intellectual thread.

Maybe that’s why they’re so eager to grasp any statistics that come their way—no matter how empty and inflated they may be.