This year the Massachusetts Department of Elementary and Secondary Education (DESE) changed the way they determine which students are low-income. Under the old method, any student qualifying for free or reduced price lunch (for a family of four, income less than about $44,000) was considered low income. With many Mass. districts now offering free lunch to all students, this data is no longer available state-wide. As a result, the state is now determining which students are "economically disadvantaged" by matching enrollment lists to data on families receiving public assistance, mainly food stamps.

As a result of this change, many schools and districts have seen the number of students considered low income drop significantly. The Globe covered the change in an article and editorial. In the article, Revere superintendent Paul Dakin speculates that immigrant families may be undercounted because they are less likely to be using the benefits they're eligible for. I decided to look at the data to see if it's consistent with this theory.

I compared the percentage of students considered low income during the 2013-14 school year with those considered economically disadvantaged in 2014-15. The data covered every public and charter school in the state. DESE doesn't break out children in immigrant families, but it does track students whose first language is not English (FLNE). That's not a perfect proxy for immigrants, but it should track fairly closely. I ranked the schools by percentage of FLNE students in 2014-15 and compared schools in the top third to those in the bottom third.

The schools in the top third averaged 40.9% FLNE students and saw a decline of 32.4% in the rate of students considered low income. The bottom third averaged just 1.5% FLNE and had a decline of 15.2% in low income students.

What does this mean? We can't say conclusively that immigrant students are being undercounted at a higher rate than non-immigrant students, but that certainly appears to be the case. The Food Research & Action Center estimates that about 20% of eligible families do not receive food stamp benefits. It does seem likely that this rate would be higher among immigrants and lower among non-immigrants.

It's likely that there are other reasons that the numbers have declined. It's possible that some children whose families are receiving benefits weren't matched to their school enrollment records. It also appears that food stamp benefits require lower incomes than the free/reduced price lunch program.

Clearly we need a better way to determine which students are low income. This is a problem that DESE should address sooner rather than later as it affects school and district funding levels and is used in determining whether schools are meeting goals for high need students.