Important terms to understand
Proxy measures for Free and Reduced Lunch Program data
Under the Federal Free and Reduced Lunch Program (FRLP), children from families with incomes at or below 130% of the Federal Poverty Level (FPL) are eligible for free school meals, while children from families with incomes between 130 to 185% FPL qualify for reduced-price meals and can be charged no more than 40 cents per lunch. Although FRLP eligibility is based solely on income, which is only one element of overall socioeconomic status, FRLP participation has been a common proxy for the socioeconomic status (SES) of an individual student, and collective rates of FRLP participation have been a common proxy for the disadvantaged status of a school’s community or division’s region, in research and other data uses. However, one of the key provisions of the Healthy, Hunger-Free Kids Act (HHFKA) of 2010 was the Community Eligibility Provision (CEP), which provides an opportunity for schools and local educational agencies (LEAs) in high poverty areas to provide free breakfast and lunch to all students without the burden of collecting and processing school meal applications for free and reduced price meals, but also without the generation of student level data on FRLP eligibility. In Virginia, school divisions that meet the eligibility criteria outlined below may apply to participate in the CEP for one or more schools, group of schools, or division-wide:
• Participate in both the National School Lunch Program (NSLP) and School Breakfast Program (SBP);
• Have an identified student percentage (ISP) of 40 percent or greater as of April 1, 2017. The 40 percent threshold may be determined for an individual school, a group of schools within the school division, or for all schools collectively (division-wide). Grouping schools allows some schools to be below the 40 percent threshold as long as the aggregate percentage of the group meets at least 40 percent);
• Agree to provide both breakfast and lunch at no charge to all students, and if the federal reimbursement is not sufficient to cover the meal costs, the LEA agrees to cover any meal cost beyond the reimbursement rate;
• Submit an application to participate in CEP and receive approval from VDOE.
As participation in the CEP has become more widespread in Virginia, barriers to food access have diminished, which is of paramount importance. However, a side-effect is that the inventory of FRLP participation data on individual students (and, thus, FRLP participation rates for schools or divisions) has eroded, prompting the need for other measures of economic hardship or SES.
The Forum Guide to Alternative Measures of Socioeconomic Status in Education Data Systems, by the National Forum on Education Statistics, provides relevant information on eight alternatives to FRLP eligibility data as a proxy for individual student and family SES, as individual measures or in certain combinations with one another. Chapter 3 of the Guide provides information on each SES proxy measure, including a description of the measure and its granularity (individual or aggregate level), advantages, challenges, verifiability, and usage limitations, along with an example of common use in actual education agencies, in cases where an example is available.
Virginia has another alternative measure for economic hardship, VDOE’s Disadvantaged Status Flag, which identifies a student as economically disadvantaged if they meet any one of the following criteria: eligible for FRLPs, receives TANF, eligible for Medicaid, or identified as either Migrant or experiencing Homelessness. It is important to underscore that a student only has to meet one of those four criteria, but each of those four criteria has a different % FPL threshold. Among the Disadvantaged Status criteria, FRLP eligibility represents the upper income limit for getting the VDOE flag, since it can range from 130% for free meals up to 185% FPL for reduced meals, whereas the general limit for Medicaid eligibility is 133% FPL, and participation in TANF is generally limited to 100% FPL. Therefore, with de-identified aggregate data, all you can assume is that, as a group, all DS flagged students are ≤ 185% FPL.
Since many early childhood organizations that are focused on school readiness cast a wider net when defining “at risk” or “disadvantaged” (i.e., use the < 200% FPL eligibility criteria of the Virginia Preschool Initiative), VDOE Disadvantaged Status data provide a conservative proxy (≤ 185% FPL versus < 200% FPL) for gauging the proportion of children 0-5 living within each school’s community who are at risk.
 The term “socioeconomic status” can be defined broadly as one’s access to financial, social, cultural, and human capital resources (NCES, 2012).
Margins of Error
Spotlight on the U.S. Census American Survey Data Estimates: The American Community Survey (ACS) of the U.S. Census Bureau is a nationwide survey that collects and produces information on social, economic, housing, and demographic characteristics about the nation’s population each year. While the full Census is conducted only every 10 years, the ACS is conducted annually. Every year, the Census Bureau contacts over 3.5 million households across the country to participate in the ACS (sent to approximately 295,000 addresses monthly), making it the largest household survey that the Census Bureau administers. Because the ACS only surveys a sub-set of the entire U.S. population, the statistics that are extrapolated from survey findings for an entire locality’s population (U.S., state, county or city) are only considered estimates.
ACS data, such as poverty levels by age, can be helpful in gauging the scope of need for certain early childhood services. However, because ACS poverty data are based on a sample, they are subject to sampling variability. The degree of uncertainty for an estimate arising from sampling variability is represented through the use of a margin of error. A margin of error is the difference between an estimated number and its upper or lower confidence bounds. Adding the margin of error to the estimate (for an upper bound) and subtracting the margin of error from the estimate (for a lower bound) creates a range of possible values for which there is a specified level of confidence in an estimate’s accuracy, with the level of confidence indicated as a percent. For example, a 90% confidence level in a margin of error can be interpreted roughly as meaning that there is a 90% probability that the interval defined by the estimate minus the margin of error and the estimate plus the margin of error (the lower and upper confidence bounds) contains the true value.
The table above on a sample city helps illustrate how to translate margins of error for Federal Poverty Level (FPL) data: