Researchers rely on district-level English-learner data to craft reports and propose policy on the state and national level. The problem is that states may not always report the data the same way鈥攁nd sometimes it goes missing.
In at least 28 states, more than 1 in 5 districts have no information reported for the past three years of available data. Of the more than 6,600 districts that enrolled more than 1,000 students for the past ten years, 21.6 percent were missing ELL data for at least one of those years, according to an Education Week analysis of the federal database known as the Common Core of Data. The nation鈥檚 13,500 school districts are required to report those numbers annually to their state departments of education, which then provide the data to the National Center for Education Statistics, the keeper of the Common Core of Data.
The Department of Education created a in 2018 that showed the percent change in the numbers of English-learners in all the nation鈥檚 school districts between 2009-10 and 2014-15. Big swaths of Mississippi, Illinois, and Maine had missing data, along with other scattered districts.
鈥淭here鈥檚 so much missing that it鈥檚 going to be misleading because we don鈥檛 know if there鈥檚 a pattern as to why some are missing,鈥� Julie Sugarman, a senior policy analyst for preK-12 education at the Migration Policy Institute, said, 鈥渓ike, if we鈥檙e missing all of the small districts or we鈥檙e missing all of the districts that have some other characteristic.鈥�
The Education Department gathers information on the number of English-learners through many different surveys and data-collection programs. In the Common Core database, the year-to-year omissions can be glaring. The 4,000-student Brawley Elementary school district in Southern California, for example, is missing data on the number of ELL students for three years in the last decade. The number of English-language-learner students that were listed range from 1,200 to 1,700. In the 2017-18 school year, the most recent with data available, just over 1,300 of the district鈥檚 students were English-language learners.
In 2016-17, however, that data was not reported at all.
Brawley鈥檚 superintendent, Richard Rundhaug, who is new to the district, said he could not explain the discrepancy.
On the opposite coast, more than 30 English-language-learner students numbered among the 2,500 students in the New Jersey Caldwell-West Caldwell district in 2016-17, but there was no data for 2017-18. When it is reported, the district鈥檚 number of ELLs has been relatively stable except for the 2013-14 school year, when only one English-learner is listed.
Superintendent James Heinegg told Education Week the district currently has 68 ELL students鈥攁nd that the numbers are collected during the district鈥檚 annual enrollment count on October 15 and reported to the state.
While the NCES data is not the only information the federal Education Department collects about English-learners, it is important.
Data about English-language learners is also collected at the school level through the Civil Rights Data Collection. The utility of that data is limited, though, because it cannot be reliably aggregated up to the district level, as it will double-count students who transferred between schools in the same district.
Without the NCES data, researchers would have to request the information from each individual school district. That can be difficult, according to David Lai, a special projects manager for the Council of Great City 69传媒, because some districts hesitate to give out this data out of concern for students鈥� privacy.
Fortunately for districts, the NCES data collection is not tied directly to the federal funding that goes to states for the education of English-learner students. The federal government allocates states money based on data from the American Community Survey of the U.S. Census Bureau. States then distribute the funding to districts, using their own criteria. Colorado鈥檚 department of education, for example, uses the number of English-language learners that districts report in the annual October 15 student count to distribute its funds.
鈥淲ith the new census this year the counting is going to be really important to get the funding,鈥� said Matt Weyer, a senior policy analyst at the Education Commission of the States.
Further complicating matters, districts鈥� definitions of who is an English-language learner鈥攁nd whether and how long they remain in that category and continue to receive services鈥攙ary widely.
鈥淭hey just have different figures for how many [English-learners] they have in all of these different reporting mechanisms,鈥� said Sugarman of the Migration Policy Institute, 鈥渁nd it鈥檚 just completely infuriating.鈥�