The Seattle Times’ Project Homeless is funded by BECU, Campion Foundation, the Paul G. Allen Family Foundation, Raikes Foundation, Seattle Foundation and the University of Washington. The Seattle Times maintains editorial control over Project Homeless content.

This article was produced in partnership with the Center for Public Integrity, The Seattle Times, Street Sense Media and WAMU/DCist.

A Center for Public Integrity analysis of federal education data suggests roughly 300,000 students entitled to essential rights reserved for homeless students have slipped through the cracks around the country, unidentified by the school districts mandated to help them. 

Some 2,400 districts — from regions synonymous with economic hardship to big cities and prosperous suburbs — did not report having even one homeless student despite levels of financial need that make those figures improbable.

In Washington, just over 30,000 students are identified as homeless each school year — nearly 3% of the state’s total student population. The Center for Public Integrity estimates that as many as 2,000 additional students experience housing instability in Washington each school year but go unrecognized by their districts.

And many more districts are likely undercounting the number of homeless students they do identify. In nearly half of states, tallies of student homelessness bear no relationship with poverty, a sign of just how inconsistent the identification of kids with unstable housing can be.

Advertising

The reasons include a federal law so little-known that people charged with implementing it often fail to follow the rules; nearly nonexistent enforcement of the law by federal and state governments; and funding so meager that districts have little incentive to survey whether students have stable housing.

“It’s a largely invisible population,” said Barbara Duffield, executive director of SchoolHouse Connection, a Washington, D.C.-based nonprofit focused on homeless education. “The national conversation on homelessness is focused on single adults who are very visible in large urban areas. It is not focused on children, youth and families. It is not focused on education.”

The Seattle Times’ Project Homeless is partnering with the Center for Public Integrity for an in-depth look at students who are homeless and the federal law that ensures their access to education. This collaboration will also examine Washington’s graduation gap and discipline rates between students who are housed and unhoused, as well as why Washington receives the least amount of federal funding per homeless student in the country.

Losing a home can be a critical turning point in a child’s life. That’s why schools are required to provide extra support. 

Nationwide, homeless students graduate at lower rates than average, blunting their opportunities for stable jobs and increasing the risk of continued housing insecurity in adulthood.

The gap is often stark: In 18 states, graduation rates for students who experienced homelessness lagged more than 20 percentage points behind the overall rate in both 2017 and 2018.

Advertising

The academic cost is not equally shared. 

If you think you might qualify as a homeless student because you have no housing, are staying with family or friends indefinitely or temporarily, or have another form of unstable housing, you could be eligible for benefits under the McKinney-Vento Act. Learn more about your rights.

Black and Latino children experience homelessness at disproportionate rates, Public Integrity’s analysis showed. Nationally, Black students were 15% of public school enrollment but 27% of homeless students in 2019-20. In 36 states and Washington, D.C., the rate of homelessness among Black students was at least twice the rate of all other students that year. 

American Indian or Alaska Native students were also overrepresented nationally, as were students with disabilities.

Until recently, it was not clear from federal records which students were hit hardest by housing instability. Data disclosed in U.S. Department of Education reports revealed nothing about the race or ethnicity of students recognized by their school districts as homeless.

That changed in the 2019-20 school year when the federal government for the first time made public the race and ethnicity breakdowns for individual school districts. The pattern that emerged is a story of the country’s sharp inequities, which put some families at far higher risk of homelessness than others.

The McKinney-Vento Homeless Assistance Act, first enacted in 1987 and expanded in 2001, requires that districts take specific actions to help unstably housed students complete school. Districts must waive enrollment requirements, such as immunization forms, that could keep kids out of the classroom. They must refer families to health care and housing services. And they must provide transportation so children can remain in the school they attended before they became homeless, even if they’re now outside the attendance boundaries.

Advertising

But just because districts must take these steps doesn’t mean they do.

“School is stability”

For months, Beth Petersen paid acquaintances to take her son to school — money she sorely needed.

They’d lost their apartment in Riverside County, California, her son bouncing between relatives and friends while she hotel-hopped. As hard as she tried to keep the 13-year-old at his school, they finally had to switch districts.

Petersen’s son had a right to free transportation — and to remain in the school he attended at the time he lost permanent housing.

But no one told Petersen that.

“They should have been sending a bus for him. … He’s missed so much school I can’t believe it,” Petersen said. “And school is stability.”

Eventually, the two found housing outside the Temecula Valley Unified School District her son had attended for years in Southern California. He switched districts, keeping up with the schoolwork but struggling to make friends.

Sponsored

Then a friend of Petersen’s who works at a charter school told her that her son had the right to reenroll in the Temecula Valley schools because the McKinney-Vento law allows students to stay in the same school they attended before becoming homeless.

In early September, Petersen moved with her son into a two-bedroom apartment — still outside the district boundaries — paid for by a homeless prevention organization and shared with another family. Under federal law, her son is considered homeless because they live in transitional housing.

Petersen reenrolled her son in Temecula Valley Unified but problems persisted. She said she pleaded with the district for weeks, trying to secure bus rides for the teenager. The district never responded to her emails, she said. He ultimately missed a month of classes, Petersen estimated, because she could not afford to continue paying acquaintances to transport her son every day.

The California Department of Education intervened in late September to ensure her son received transportation.

“This has been a teachable moment for the district and there are protocols and … barriers that have been removed to ensure the law is met,” an employee at the state agency wrote Petersen in an email.

A statement provided by Temecula Valley Unified in response to detailed questions regarding the Petersens said the district “does everything in its power to support our McKinney-Vento families experiencing homelessness” and has “highly responsive site and district teams,” but declined to comment further.

Advertising

Schools are a key component for homeless-student support. But federal funding for that work amounted to about $60 per identified homeless student nationwide before the pandemic. 

That’s a fraction of what school districts actually spend to support homeless students, according to a recent study by the Learning Policy Institute, a nonpartisan research group. The four districts profiled by LPI spent between $128 and $556 per homeless student identified. In two of those districts, McKinney-Vento subgrants accounted for less than 14 cents on every dollar the district spent on homeless education programs.

And that’s the districts awarded federal grants. Most get nothing. 

How we did it

Public Integrity used a statistical modeling technique called simple linear regression to measure the strength of the association between the percent of students identified as homeless and, separately, three measures used to approximate the incidence of economic disadvantage or poverty: 

– The percent of students eligible for free or reduced-price meals

– The percent of school-age children under the poverty line

– The percent of school-age children in households that are under 50% of the poverty line. 

We used federal data aggregated to the level of school districts and similar educational agencies, composing separate models by school year and state. We fit models for each state and the District of Columbia where there was sufficient data in the 2018-19 and 2019-20 school years.

We considered that a model showed a link between a variable we tested and homelessness if the model accounted for at least 20% of the variation in rates of homelessness and if the probability of coincidence driving results at least as extreme was relatively small. Twenty-four states failed this test on each of the three measures of economic disadvantage.

We assumed districts not included in federal data identified no homeless students. Districts may occasionally be left out in error. But we think our count is conservative in another way. That’s because there are additional districts that specifically told the Department of Education they have no homeless students, but the agency categorized them with districts reporting a low number of students and suppressed those figures. 

For more details on our analysis, read our white paper.

Until a temporary funding influx during the pandemic, only 1 in 4 districts nationwide received dedicated funding.

Even in states that receive hundreds of dollars per student, the money does not stretch far, experts said. And it’s definitely not enough to provide long-term assistance for students without stable housing.

Advertising

“There [are] more and more students in crisis, and the districts are not really getting more and more resources to help,” said Katie Meyer Scott, a Seattle-area senior youth attorney with the National Homelessness Law Center. “It comes down to resources rather than any kind of bad intent. The lack of investment in our schools over time is obviously hitting homeless students even harder.”

Maria Foscarinis, the founder of the National Homelessness Law Center and one of McKinney-Vento’s primary architects, said the law’s writers knew it was inadequate and planned to follow it with homeless prevention programs and housing. But they faced stiff resistance. 

Hidden homelessness

It might seem like common sense to assume that where more children experience poverty, more will experience homelessness, too.

But that’s not what the data from school districts shows. One of the most surprising patterns we found is that reported homelessness among students didn’t mirror poverty in 24 states.

The finding runs counter to a growing body of empirical evidence supporting the connection between poverty and housing instability. Children born below 50% of the poverty line had a higher probability of eviction than higher-income peers, lower-income households are more likely to experience forced mobility, and renters who are forced to move end up in higher-poverty neighborhoods than renters who move voluntarily.

“There should be a stronger relationship between homelessness and poverty,” said Jennifer Erb-Downward, director of housing stability programs and policy initiatives at the University of Michigan’s Poverty Solutions, “and the fact that there’s not supports that there’s underidentification taking place.”

Advertising

Districts can tell teachers and staff to look for common signs of housing instability among students — fatigue, unmet health needs, marked changes in behavior. But those aren’t always apparent.

If they’re following the law, districts will survey families so they can self-identify as homeless. But some parents fear that acknowledging their housing struggles could prompt the government to take their kids away.

And then there’s the gulf between what people commonly think of as homeless and the more expansive definition Congress uses for students. Living in a shelter, on the streets, in a vehicle or in a motel paid for by the government or a charitable organization are included, but that’s not all.

More than 70% of children eligible for services were forced by economic need to move out of their homes — with or without their family — and in with relatives or friends, a practice the U.S. Department of Education defines as “doubled up.”

Data on student homelessness is collected by districts and funneled to the federal government by states, which can choose to leave out any districts that did not report having any homeless students. Our data adds those excluded districts back. We assume they identified no homeless students, since they’re not in federal data.

Public Integrity’s analysis focused on noncharter districts in the 2018-19 and 2019-20 school years. In addition to comparing poverty and reported homelessness, we applied a common benchmark used by education researchers and some public education officials — that 1 of every 20 students eligible for free or reduced-price lunches experience homelessness under the federal definition.

Advertising

In each school year we analyzed, more than 8,000 districts did not meet the 1-in-20 guideline.

It’s possible that some school districts genuinely have fewer homeless students than this benchmark predicts. But multiple researchers told us that they see the 1-in-20 threshold as a conservative estimate.

J.J. Cutuli, a senior research scientist at Nemours Children’s Health System, said the analysis bolsters the anecdotal experiences of school district staff, shelter personnel and people who’ve lived through periods of homelessness.

“You’re giving us a clue as to the magnitude of this problem,” he said.

An uphill battle for help

The federal government, state education departments and families have few options to hold districts accountable if they fail to properly identify or provide assistance for students experiencing homelessness.

The U.S. Department of Education delegates enforcement to states. States where school districts fail to follow the law are subject to increased monitoring, but the federal agency would not say how often that happens. A spokesperson said only that the agency “engages in monitoring and compliance activities that can include investigating alleged noncompliance.”

Advertising

Public Integrity reviewed dozens of lawsuits in which families and advocacy groups alleged that school districts denied students rights that are guaranteed under the federal McKinney-Vento law.

Families experiencing homelessness have sometimes prevailed in their standoffs with education agencies, winning reforms like agreements to train school personnel in the law and, in one case, a toll-free number for parents and children to contact with questions about their rights.

As an extreme last resort, the U.S. Department of Education can cut funding — a step officials are loath to take because that would ultimately harm the very students the agency wanted to help. The agency said it has never penalized a state in this manner.

A 2014 investigation by the Government Accountability Office found that eight of the 20 school districts its staff interviewed acknowledged they had problems identifying homeless students. The watchdog agency found that the U.S. Department of Education had “no plan to ensure adequate oversight of all states,” with similar gaps in state monitoring of school districts.

State audits in California, Washington and New York have also made the case that many school districts fail to identify a significant number of students who qualify for the rights guaranteed under federal law. Advocacy groups and researchers, too, have surfaced examples.

The traditional level of funding to support homelessness has left many districts struggling to fulfill the law’s requirements.

Advertising

In April, 92 members of the U.S. House of Representatives signed a “Dear Colleague” letter, urging the chair and ranking member of the House Education Committee to renew the higher pandemic levels in funding, $800 million in total, for the fiscal year that started Oct. 1. That represents 1% of the federal education budget and would be money well spent, they argued.

“Investing in a young person’s life will enable them to avoid chronic homelessness, intergenerational cycles of poverty, and pervasive instances of trauma,” concluded the letter, whose signatories include Democratic U.S. Reps. Suzan DelBene, Pramila Jayapal, Rick Larsen and Adam Smith.

Budget bills from both chambers of Congress requested boosts in the program budget that are far short of what the House members requested. Federal budget negotiations will likely resume in December.

Temecula Valley Unified, the district Beth Petersen’s son attends, received $56,000 to serve homeless students through the American Rescue Plan — about $470 per homeless student identified. District staff did not respond to questions regarding funding for homeless education programs. State financial records for the several years before the American Rescue Plan show the district received nothing. 

Early on a Monday morning in October, Petersen sat at the kitchen table in her shared apartment, applying makeup under the glare of a bowl-shaped ceiling light. Her son emerged from the bathroom, barefoot but otherwise dressed for school.

“Do not miss the bus coming home or we will be up a creek,” she said as the pair walked outside, the air crisp as morning haze yielded to blue sky. 

At 7:02 a.m., a yellow school bus turned the corner. It slowed to a stop before them, the fruits of Petersen’s long struggle to make the promise of the McKinney-Vento law a reality. 

Amy DiPierro and Corey Mitchell are journalists with the Center for Public Integrity, a nonprofit newsroom that investigates inequality.