Why are so many DCPS schools listed as 99% low-income? It’s not necessarily because they are.

Publicly available education data shows that many DCPS schools have a 99% poverty rate. But that figure is based on an average and doesn’t reflect the actual number of poor kids at any particular DCPS school.

In the past, each DCPS school counted how many of its students were eligible for free and reduced-price meals (FARMs). Last year, DCPS began participating in a federal program that allows some schools to provide free lunches to all their students without determining how many are actually eligible.

The new method reduces the administrative burden on schools and allows more poor kids to get free meals. But it’s made it harder to figure out how many poor kids there actually are, and it lumps an undetermined number of higher-income kids into the low-income category. That makes it harder to track the academic progress of low-income kids.

The new method, called the Community Eligibility Option (CEO), is a federal program available to any school district that includes at least one school where 40% or more of students can be identified as needy through certain measures. Those measures include whether they’re receiving food stamps or other federal welfare benefits, are homeless, or in foster care.

The formula then multiplies that “direct-certified” percentage by a certain factor to get the percentage of kids eligible for FARMs. FARMs includes not only students on welfare, but also students whose family income is equal to up to 185% of the federal poverty guidelines. The FARMs figure is commonly used as a proxy for low-income status.

Charter schools can also participate in the CEO program, but generally they only have one or perhaps a few campuses. DCPS chose a group of 75 of its schools for the CEO program, and the average rate of direct-certified students for the group as a whole is 62%.

Under the CEO formula, that average gets multiplied by 1.6, and the result is 99%. And that explains why DCPS school profiles and school equity reports show the same percentage of FARMs students at many schools: 99.

Wide variation in underlying percentages

But according to the Office of the State Superintendent of Education, which administers the FARMs program, there’s actually a wide variation in the percentage of direct-certified students within the group of 75 schools in the CEO program. The number ranges from about 38% to 85%.

If you multiply 38% by 1.6, the result is much lower than 99%—60.8%, to be exact. So almost 40% of the students at that school are being classified as low-income when in fact they may not be.

Officials at DCPS and the Office of the State Superintendent of Education (OSSE) have the direct-certified percentages for individual schools but declined to release them. They did not provide a reason.

Dr. Sandra Schlicker, Deputy Superintendent at OSSE, pointed out that the previous method of determining FARMs eligibility often resulted in undercounting poor students. Parents didn’t always bother to fill out the FARMs application, even if they were eligible, and school administrators were powerless to provide free meals to students who clearly needed them.

Even the direct-certified percentages may understate the extent of need, especially at schools with a large immigrant population. Undocumented immigrant families aren’t eligible for the federal programs that often trigger certification. Grouping those schools with others that have a higher percentage of direct-certified kids enables them to provide free lunch to all students.

Beyond that, the CEO program clearly has other benefits. It relieves schools of the burden of dealing with the paperwork required under the old method. Under the new method, OSSE simply compares lists of children who are receiving federal benefits or are otherwise eligible for direct-certified status against lists of students at the CEO schools. And including all students at a school in the lunch program avoids stigmatizing low-income children.

Measuring the achievement gap

But education policy activists and researchers say the new method has made it hard to determine the size of the achievement gap between low-income and higher-income children.

Local and national standardized test scores compare subgroups using the FARMs percentage to determine the performance of low-income students. School reports also separate out suspension and expulsion rates for FARMs students.

HyeSook Chung, executive director of DC Action for Children, said in an email that “any evidence of low-income student performance in DC now comes with an asterisk, because there are likely more than a few non-low-income students being lumped into that category.”

It’s also become difficult to compare test scores and other data to information from previous years when FARMs kids were counted differently, she said.

The DC Council passed a bill last year providing additional funds for “at-risk” students. That designation could be used as a proxy for low income in the future, but, like the direct-certified designation, it’s limited to a fraction of those counted under FARMs, only about 30% of students. The FARMs category has included more like 70% of them in the past.

Other problems with the method

The CEO program also makes it hard to track demographic changes that may be taking place at some DCPS schools, since there’s no way of knowing whether more affluent kids are beginning to attend them.

And some critics of high-achieving charter schools that serve high-poverty populations have suggested that their success is due to lower FARMs rates than comparable DCPS schools, say 80% rather than 99%. But if a DCPS school is in the CEO program, it could actually have a FARMs rate significantly lower than the 99% figure attached to it.

This is a problem that will soon become more widespread. The federal government has been phasing the CEO program in gradually, with only 10 states and DC participating this year. But next year it will be open to all qualified schools and school districts nationwide.

It’s great that more kids who need free meals are getting them, and school administrators are undoubtedly relieved that they no longer have to spend time identifying low-income students. But if we’re going to understand how poor kids are actually doing in DC, and elsewhere, we’re going to need to come up with some more accurate way of determining who they are.

At the very least, DCPS and OSSE could literally add an asterisk when they list a school as having a 99% FARMs rate, and use the footnote to let the public know just how it was calculated.