The past eight years have marked an unprecedented push to expand and use federal data systems, both in education and across the federal government. As education watchers await the Trump administration, there has been little clarity and some concern about the future of key education studies.
“With this new administration, there are so many things to keep your eye on,” said Laura Speer, the Annie E. Casey Foundation’s associate director for policy reform and advocacy, “and this [federal education data issue] is one of those things that can completely fly under the radar—and before you know it, some critical things can be lost.”
Some of the biggest ongoing federal studies have in recent years faced budget cutbacks and criticism, particularly by Republican members of the House of Representatives. Efforts to curtail those studies—or in some cases, stop them completely—have so far been blocked by the Senate and the Obama administration. So far President-elect Donald Trump and his nominee for Secretary of Education, Betsy DeVos, have given no indication of support for large-scale federal studies or education research more generally in their public statements or on the transition website, but they have voiced concern over invasions of privacy.
Federal longitudinal studies have become ever more prevalent in education policymaking in the last decade, and researchers worry cutbacks to data collection, even those conducted by other agencies, could have a strong impact on education programs and budgets going forward. Here are three examples:
American Community Survey
Budget: $230.9 million, fiscal year 2016
Who collects it? U.S. Census Bureau, Commerce Department
What is it? This is the second-largest study administered by the Census Bureau, after the dicennial census. It is a mandatory long-form survey given to more than 3.5 million households each year.
What is it used for? The data are used to distribute more than $400 billion in federal dollars each year, including Title I and other U.S. Department of Education funding. State, tribal, and local governments use the data to plan school sites and other services. Business and nonprofit groups also use ACS data to target distribution and targeting of services. The Education Department’s Common Core of Data uses ACS data for a variety of measures, including demographics, student poverty beyond free and reduced-price lunch, and information on public virtual and brick-and-mortar charter schools. Researchers also use these data.
Civil Rights Data Collection
Budget: $400,000, fiscal year 2016
Who collects it? U.S. Department of Education
What is it? This is a biennial survey of key demographic, discipline, and academic data from all public school districts and schools in the country. It has been collected in some form since 1968; the 2015-16 survey data is being gathered now.
What is it used for? The Education Department’s office for civil rights uses the data to identify overall trends and local disparities in education for students with disabilities, and those of particular ethnic or racial groups, genders, or language backgrounds, as well as to investigate complaints of discrimination in particular sites. Lawyers and groups interested in civil rights issues also rely on the data as the most comprehensive sample of such trends in the country. The Education Department also uses the data to identify teachers’ experience levels and distribution, school finance issues, and emerging trends in education technology use.
National Assessment of Educational Progress
Budget: $149 million, fiscal year 2016
Who collects it? U.S. Department of Education
What is it? Often dubbed the “Nation’s Report Card,” this is a set of assessments given to nation- and statewide samples of U.S. students at grades 4, 8, and 12 in mathematics, reading, science, writing, civics, economics, technology and engineering, geography, history, and the arts. Different subject-matter assessments are given on different time frames and to different grade ranges, but math and reading are the most frequent.
What is it used for? NAEP remains the only common academic assessment across all states and is used to compare how students of different grades and demographic backgrounds are achieving academically in various subjects. It has also been statistically linked to the Trends in International Math and Science Study, allowing states to get a more nuanced understanding of how their students compare to peers worldwide. It includes background surveys that help researchers connect practices in school and at home to students’ longterm achievement.
“Ideological differences usually don’t matter all that much when it comes to education research—there seldom seems to be argument about doing sound research to find out what works and disseminating that which we know evidence supports,” said Max McConkey, the communications director of the research group WestEd. “But the incoming administration appears to be unlike any other I’ve ever seen, so all bets are off, for now.”
Building on Legacy
The past decade has seen a major expansion of both federal education data and the infrastructure to use it in policy and program design, spurred under both the Republican Bush and Democratic Obama administrations. The number of regional educational research coalitions like the Chicago Consortium for School Research have expanded from a few to several dozen, in part thanks to federal grants, and federal support also helped every state develop a longitudinal student-data system to track individual student data from kindergarten through high school graduation.
The Obama administration also used administrative rules and evidence requirements in competitive grant programs like Investing in Innovation, or i3, to push the use of data and evidence in program and policy development. Both the evidence standards and the competitive grant program have since been codified in the Every Student Succeeds Act, and the newly contracted regional education laboratories are focused on building more local research and data alliances.
“I think that research and practice being brought together in new ways has staying power in the field. I’d be very surprised if that changed,” said Ruth Curran Neild, who recently stepped down as the designated director of the Institute of Education Sciences, the Education Department’s research agency. “At IES, there’s a lot of strength at the everyday civil-servant level. ... Now, could someone come in and change that? Yeah,” Neild said, but, “those cultures can last if they’re really built in.”
Yet the data and evidence requirements in ESSA are only strong if states are willing to implement them and the Education Department under DeVos is willing to enforce them, said Daniel Losen, the director of the Center for Civil Rights Remedies, part of the Civil Rights Project at the University of California, Los Angeles.
“There’s information in the statute, but enforcement of reporting to the public has always been a struggle,” he said. “How do you respond to schools and districts that thumb their noses at reporting requirements? ... I am concerned that this administration will not be supportive of research generally and data collections in particular. It’s easier to say whatever you want if there is no counternarrative that has evidence to support it.”
Reporting Concerns
Enforcing data reporting will be a particular problem for the Education Department’s nearly 50-year-old Civil Rights Data Collection, experts say.
Civil rights data significantly expanded in both scope and use under the Obama administration, providing in-depth demographic, discipline-related, and academic information on every public school and district in the country. Beyond being used for investigating reported educational inequities, the survey is also now the main source of more basic school-level information, such as teacher certification and chronic absenteeism.
But “everyone would identify OCR as potentially at risk,” said Michele McLaughlin, the president of the Knowledge Alliance, which advocates for education research groups. For most of the civil rights collection’s nearly 50-year history, it included only a representative sample of districts from each state. In 2000, and then from 2011-12 onward, the collection expanded to every public school in the country that spends at least half of each day on educational services. This includes charter and magnet schools, juvenile justice facilities, virtual schools, and alternative schools for students with disabilities—but it does not include public schools on tribal lands and military bases or private schools.
The omission of private schools is a concern for Losen. Trump made expansion of private-school voucher programs the center of his campaign platform for education, and “from a data perspective, there’s a huge problem for civil rights laws and data requirements for private schools,” Losen said. “Will they have to report on civil rights or teacher certifications?”
Education watchers are keeping the closest tabs on the biggest federal data collections: the Census Bureau’s 2020 decennial survey and the American Community Survey, a more detailed study by the bureau of 3.5 million households a year.
“Certainly the ACS is one of the main areas where there is potential for there to be problems, because there have been threats to it practically every single year,” said Laura Speer, the Annie E. Casey Foundation’s associate director of policy reform and advocacy.
Trump’s designated budget director—who would also oversee the Census Bureau—is U.S. Rep. John Michael “Mick” Mulvaney, a Republican from South Carolina and a fiscal hawk who repeatedly voted and advocated for making the ACS voluntary or eliminating it completely in favor of privatizing federal Census data collection. The House has passed such amendments in three of the last four years, though they were always scuttled in the Senate.
While Speer doubts that the ACS would be fully privatized, “the real, more likely threat is to have it be no longer mandatory. That would have a major impact on the reliability of the data,” she said. The Census Bureau has estimated that making the survey voluntary would reduce participation by 20 percent, increase annual costs by $35 million to $90 million a year, and reduce its accuracy, particularly in rural areas and communities of color.
That could hit census education data particularly hard, said Speer, because children are already more likely to live in households that are often undercounted in the decennial census, including highly mobile or impoverished families or those in racial or linguistic minority groups.
“Because we don’t do the [decennial] census long-form [survey] anymore, the data in the ACS has become the go-to place to get information at a local level. If we were to no longer have the ACS, the ripple effects within government would be substantial,” said Speer, who manages the Casey Foundation’s Kids Count Data Book, which uses both census surveys for research and to target grants. “For us, it’s a really critical resource.”
Budget Issues
Moreover, three years ahead of the 2020 census, the bureau is still operating on funding near 2010 levels. Congress’s continuing resolution for the federal government, for example, provides $240 million less than requested for the census.
“The census is a finely tuned watch with a 10-year funding and planning cycle, and it’s been fairly dramatically underfunded for several years,” said Phil Sparks, a co-director of the Census Project, which tracks policies and use on the surveys. (Sparks is not related to this article’s author.)
Census data, Sparks said, “are the cornerstone of education policy and funding. It is these numbers that empower everything from Head Start to higher education.”
Speer agreed, noting “the lack of support for the budget could become very problematic very quickly.”
McLaughlin, of the Knowledge Alliance, said leadership on behalf of federal education data likely will have to come from the states and Congress.
“This is where [Republican House Speaker Paul Ryan] meets Trump,” she said. “Ryan has always been into better data, better outcomes, more accountability. With the new administration coming in, we’ll have to see where it stands.”