NJ auditor: Free-lunch errors skew aid

Students’ poverty rates are estimated by eligibility for a free or reduced-price lunch.  But many ineligible students are getting a lunch discount, concludes a report by New Jersey’s auditor. “School districts have little incentive to question applications because a higher participation rate also increases their state aid,” the report  charges.

“There is a significant error rate,” state auditor Stephen M. Eells said of the school lunch database. “It’s not accurate by a long shot, and I don’t think we should be using it to determine state aid.”

New Jersey gives schools an extra $4,700 to $5,700 per free or reduced-cost lunch recipient.

For 2011-12, the income limit for a family of four will be $29,055 for the free meal program and $41,348 for the reduced-cost meal.

A state audit in 2009-10 found at least 37 percent of lunch participants were ineligible or produced no supporting documentation. A random sampling in 10 districts found 23 percent were ineligible and another 24 percent could not be verified because they did not provide Social Security numbers. I’ll assume most of the 24 percent are here illegally; they probably do have low incomes.

Using lunch-program participation to generate poverty rates has many critics. For one thing, free lunch applications go way down in high school, apparently because even low-income students can’t stand the food. Using lunch data may overestimate poverty in elementary school and underestimate poverty in high school.


About Joanne


  1. Richard Aubrey says:

    I knew lying about the incidence of poverty was a useful technique among those who wanted to reproach American society. I figured it was a good deal for the poverty pimps.
    Didn’t know it helped with school budgets, too.
    Still learning from public education after all these years.

  2. CarolineSF says:

    I’ve been a member of the San Francisco Unified School District Student Nutrition & Physical Activity Committee for eight years and am thus fairly familiar with the situation regarding meal applications.

    What mechanism within a school district would be deployed to question or challenge meal applications? At least here in SFUSD, there’s no department or staff qualified or enabled to poke into family finances, and of course the potential for charges of intimidation and discrimination would be sky-high. Use common sense.

    The big issues in high school are that it’s universally uncool to eat in the cafeteria and that teens are far less likely to be compliant about bringing home and returning school paperwork (as any parent of a normal teen will tell you). (There’s no reason high schoolers would like the food less than elementary schoolers do.)

  3. Using the free-lunch numbers (which are largely unverified) to determine US and state aid to schools gives them a huge incentive to cheat.

  4. Verification: Oh…I don’t know….how about a W-2 or a tax return from the previous year. Really, would it be that hard to check?

  5. An easily verifiable indicator of the fact that most of the kids aren’t all that in need of the free lunch is watching at lunchtime (which, as a monitor, I’ve had to). A substantial number eat none or only a small portion of the lunch. They spend $2-5 on junk food and drink. As far as I’m concerned, that’s not hunger, and that’s not poverty.

  6. Knowledgeable people inside and outside school systems have been aware of this issue for a long time. While W-2”s or tax returns would be useful in verifying some incomes, that system would leave out the most vulnerable children, whose parents or guardians have only off-the-books income or no income at all, or who drift from household to household. It’s a bad system but no-one has come up with a better one, given that the goal is to make sure children who need food, get some. Or at least try. I have to admit that it’s to the point where I sympathize with the objections above.

  7. Michael E. Lopez says:

    Caroline saith:

    (There’s no reason high schoolers would like the food less than elementary schoolers do.)

    Actually, I disagree. Many adults would patently refuse (absent actual starvation) to eat some of the lunches I’ve seen in the LAUSD; they were a culinary visual horror and, having sampled a few, they didn’t taste much better than they looked.

    Little kids, however, have no taste. That’s not to say they don’t have taste buds, or that those taste buds don’t work — but they don’t work as well because the wiring isn’t complete. Tasting well is a learned skill, much like walking or running or understanding the motion of objects behind one another.

    Once you get to high school, you’re getting better at it. You’ve also tasted more stuff and have had a chance to differentiate, so you know what you’re missing.

    That said, I agree with you: the main issue is social pressure. But you might ask, “From whence comes the genesis of this social pressure?” In other words, why is it uncool to eat in the cafeteria? There’s probably a very complex answer involving lots of moving parts, but food quality might be a part of it.

    That said, the food in my high school cafeteria (which I rarely ate) was significantly better than the stuff we were fed in elementary school. We had an open campus though, so it was Taco Bell for me: 59 cent bean burritos!

  8. Soapbox0916 says:

    I can only really speak to my local situation, but this is what I do for a living (with my STEM degree LOL). Even though I work for local government, I am still very much a scientist at heart and I am very much an anal-retentive data geek, so I have a pretty good feel for data for my local area. A big part of my job is data collection and data quality. I really am the main person that deals with homeless, poor, poverty, low income data for my local area. I have to do lots of reports.

    I work with many agencies that will accept eligibility for low income based programs determined by free school lunch (many do not accept reduced due to high need) in order to qualify for their programs such as weekend backpack programs and free tutoring. Families can also qualify for these extremely low income targeted programs on direct income verification too, and we find that many families even with grade school children don’t want to apply for the free/reduced school lunch program for various reasons. I am very certain for my local area that the number of families that don’t claim a free/lunch that could be eligible for a free/reduced lunch still far exceed the number of cheaters.

    I don’t want to get into any specifics, but we have fraud measures and quality checks in place for my local area. The way a few people usually try to cheat, is that the families really do qualify income-wise, but they try to double dip services. We share data across many agencies and collect data each time, so this is a lot harder for them to do now. I think many would be amazed by the data that we can collect.

    However there are still limitations. The reason free/reduced lunch is used to estimate low income and poverty is because many families have already done the paperwork, and if they qualify for free/reduced lunch, then they qualify for the criteria for low income targeted services. Finding out income for a family can be really invasive and unless the family is very willing, it is especially difficult to do and also meet confidentially rules. Each type of government funding source has different rules for confidentially that must be met and it all very confusing, so while free/reduced lunch may be a crude estimate, it is the easiest way to get an estimate.

    For my local area, I would estimate that K-5 is still underestimated (despite any cheaters), I am certain that 6-8 is underestimated even more, and high school is underestimated to the point that we typically rely on other data sources. You are right about high school and I had not really thought about it until now, but we find free/reduced lunch numbers be so low and inaccurate for high school that we won’t even use it for data. It is almost a given that many in high school won’t eat the food or do the paperwork.

    Personally I won’t use free/reduced lunch data beyond 5th grade level.

  9. Stacy in NJ says:

    I grew up on welfare and was eligible for free lunches my entire k-12 years. As an elementary aged kid, I ate lunch with my friends in the lunchroom and thought little of who paid for it. As I moved into junior high, I became aware of the stigma attached to receiving aid and prevented my mom, who was aided by school employees, from applying for free lunch. By the time I reached high school I was working several jobs and could afford to purchase food from the vending machine, which I did daily. Plus, the food sucked. Now, probably only 10-20% of the kids in my school district were eligible for aid, so there was a stigma attached. In schools were the majority of students are receiving aid, I imagine that is less of an issue.

  10. Deirdre Mundy says:

    We have 6 people in our family— We could be making close to 55K a year and qualify for free lunch?????


    Good grief! Of course, my daughter has food allergies, and so couldn’t eat it. And we home school, so lunch is usually leftovers from dinner.

    Maybe part of the issue is how we define ‘poor.’ and how we don’t change the eligibility depending on location! For instance, I could see 55K being ‘poor’ in NYC—but out here in the semi-rural midwest? It’s solidly middle class. (Our food and housing are much less expensive out here.)

    I know families who could afford to feed their kids lunch, but get the free or reduced lunch anyway because they’re eligible and it saves time and money.

    The system is broken— even ‘eligible” kids are often not facing any sort of nutritional risk—sometimes they just come from a single income family where mom stays at home—and therefore has time to prepare inexpensive yet healthy meals.

  11. Soapbox0916 says:

    Free lunch is 130% of the national poverty level. For my local area, most programs that I deal with will only accept eligibility for free lunch as an estimate for low-income. The demand is so high for certain programs, that even free lunch eligibility is not enough. There are other criteria that are in play as well. However, Reduced lunch is too high for a decent estimate for my local area.

    Not to confuse things even more, but I think things are already very confused. Low-income is more generous term than poverty. Different government programs use different terms unfortunately, and Department of Education is really different sometimes. Part of the confusion is that some people use low income and poverty interchangeably, when they are really distinctly different. I think it would be more accurate to use the term low income instead of poverty. We really are talking about low income. But the definition of low-income is tied to area median income with HUD, and DOE does their own thing. I have to deal with all these various definitions and different government agencies on a daily basis.

    I actually do face families that are starving or have really poor nutrition due to not having enough money for food. Is it everyone that could qualify? Absolutely not. But, there are so many factors that come into play, that income is really only part of it, the income limits are only meant to be rough guidelines. Someone near poverty with a higher income on paper may also have higher debt or higher housing costs than someone else. Locally, we have other criteria in place.

    The other question is if the school system is the best place to feed kids? Probably not, but school systems are a constant across America. Kids come to schools. The kids are there. This is less about the schools themselves, and more about schools being convenient. Locally, we have a food bank that I personally think could do a much better job, but this food bank only operates in my local area. Trying to piece meal services through various agencies across the nation would be a nightmare. However, Feeding America might be the national agency that could pull it off. Still I get why schools are easier.

    I have to kind of disagree with the system being broken in terms of data, this is all meant to be estimates. The abuse is in relying too much on this particular data for school budgets and stuff that it was not ever meant to be relied upon, but I don’t think this means it is broken, just needs to be scaled back, and more types of data should be used instead.

  12. Roger Sweeny says:

    In the Atlanta Cheating Scandal thread, CarolineSF said,

    Campbell’s law is an adage developed by Donald T. Campbell:[1]

    “The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor.”

    The social science principle of Campbell’s law is sometimes used to point out the negative consequences of high-stakes testing in U.S. classrooms.

    Caroline then seems to argue that Campbell’s Law indicates we should get rid of high-stakes tests because, when the stakes are high enough, “Cheating would be a perfectly rational response.”

    Does that mean we should get rid of free and reduced-price lunch programs because, at $4700 or $5700 a pop, “Cheating would be a perfectly rational response.”?

  13. I don’t think it’s incredibly intrusive to do random auditing of families whose children receive free or reduced price lunches the way the IRS does random auditing of those families receiving the Earned Income Tax Credit. Requiring everyone to provide full documentation probably wouldn’t be cost-effective as the processing costs would likely outweigh the savings from catching cheaters. Having families who apply be aware that they could face an audit, however, would likely deter at least some of the cheaters at a much lower cost to the government.