Series: Lessons Lost
How Federal Schools Are Failing Native American Students
This article was produced for ProPublica’s Local Reporting Network in partnership with The Arizona Republic. Sign up for Dispatches to get stories like this one as soon as they are published.
For years, federal law has required all school systems to publicly report how well they help children learn. But the federal government’s own Bureau of Indian Education has failed to do so, despite repeated warnings about the quality of education Native American children receive in its schools.
Now, a comprehensive analysis of school performance data by Stanford University’s Educational Opportunity Project, The Arizona Republic and ProPublica reveals for the first time bright spots and problem areas facing an agency that oversees more than 180 schools and dormitories across Indian Country.
The analysis of nearly 200,000 standardized test scores taken over a nine-year period shows BIE schools have produced mixed results.
On the one hand, while the BIE has drawn criticism for providing a substandard education, the analysis found above-average learning rates for students in the BIE’s classrooms. Compared to Native students who attended nearby public schools, BIE students advanced at slightly faster rates each year. And BIE students’ average test scores improved by almost half a grade level between 2009 and 2018, a period in which the agency launched a vast reform effort aimed at shifting more control over schools to tribes.
But the analysis also revealed an achievement gap that even the BIE’s highest-performing schools cannot close. BIE students performed more than two grade levels below the national average on standardized tests. And compared to Native students attending the nearest public school district, the BIE students still remained nearly one-third of a grade level behind.
The results suggest that the pace of learning in nearly all BIE schools wasn’t rapid enough to compensate for centuries of disinvestment in tribal communities and families or a lack of early childhood learning opportunities, which contribute to many BIE students being behind schedule academically by the time standardized testing begins in the third grade.
“We have kindergarteners that are two years behind,” said Charles Cuny Jr., superintendent of Little Wound School, a tribally controlled school on the Pine Ridge Reservation in South Dakota. “The question is, how are you two years behind in kindergarten?”
Glendora Lopez, who lives in the Hopi village of Kykotsmovi, in Arizona, recognized the importance of early childhood education. She applied early to get two of her children coveted spots in their local Head Start, a federally funded program designed to help low-income children prepare for school.
Then came time to pick an elementary school. There were four within driving distance of the family’s home. Lopez knew their reputations, but couldn’t be certain of their performance because each was overseen by the BIE. Like most parents in communities the agency serves, Lopez had no way to look up the Hopi schools’ test scores. Nobody gave the schools public letter grades as Arizona does for its public schools.
She’d planned to stick with the local school in Kykotsmovi. Then a family member said good things about Second Mesa Day School, a few miles down the road, which had a new building and several teachers Lopez knew. To her, Second Mesa felt like a good school.
Eight years later, Lopez said she is confident she made the right choice. One son finished the sixth grade and moved on to middle school, and she had two children still attending the school. Lopez said all of them are strong students. “I’m fortunate that things worked out for my kids,” she said.
While Lopez’s children thrived, the analysis presents a complicated picture of Second Mesa’s performance overall.
Its students’ test scores were similar to those in other Hopi schools — a little more than two grade levels below the national average — but the school appeared to have done a much better job helping those students learn. Second Mesa test scores grew by about 1.2 grade levels each year, according to the data, suggesting students absorbed more than a full grade level of learning in a single school year. That’s by far the highest mark of the five Hopi elementary schools for which enough data was available to reliably estimate growth.
The data also shows Second Mesa’s overall performance declined over time. Despite its students’ strong year-to-year growth, the school’s’ overall average test scores fell by more than 1.5 grade levels over the decade included in the analysis. The drop suggested that each year’s class of third-graders entered the year a little further behind than the class before it, though all classes learned slightly faster than average.
Lopez was not surprised to hear the school showed strong growth in test results as students advanced. But the slowly declining test scores left her uneasy.
Maybe her kids were being left behind, Lopez wondered aloud. “But I have nothing to compare it to, that’s the thing.”
Analyzing BIE Schools
This analysis, the product of a yearlong collaboration between the news organizations and Stanford, provides the most comprehensive look yet at the performance of BIE schools.
It examined about 193,000 standardized test scores from students in 3rd through 8th grade, starting in the 2008-09 school year. The analysis did not include scores from nearby high schools, or for the school years ending in 2013, 2014, 2015 or 2018, years for which the BIE did not submit scores to the U.S. Department of Education in time for inclusion in its EDFacts database. Data from the 2018-19 school year was not available at the time of the analysis, and most students did not take standardized exams in 2020 because of the pandemic. Students living in BIE-funded dormitories, who typically attend nearby public school districts, were not counted as BIE students.
The news organizations provided BIE and the Department of the Interior, to which the BIE belongs, with a summary of the analysis. Neither the agency nor DOI commented on the analysis.
In a brief statement, a DOI spokesperson said, “Our leadership continue to work to identify areas where students, teachers, and faculty would benefit from increased resources and support. We look forward to working with Congress and Tribal communities to elevate this issue and make progress.”
The analysis is centered on two key metrics: proficiency, or students’ grades on standardized tests; and growth, which measures how fast their test scores improve. The analysis also calculated the change in individual schools’ average test scores over time. The three measurements are independent of one another, and research has found little correlation between student proficiency and growth.
Stanford describes its proficiency scores as a measure of “educational opportunity.” Proficiency is highly correlated to factors outside a school’s control, like socioeconomic status, while growth measures a student’s progress from one year to the next, regardless of where they started. That makes growth a better estimate of school quality, experts say, as it measures how much a student gains from a year in the classroom. A student who gains more than one grade level in a year is considered to be learning faster than expected.
All told, there are 183 BIE schools and dormitories enrolling about 45,000 students — about 10% of Native students nationwide — many of whom live in rural communities with few educational alternatives. The analysis could reliably calculate proficiency for 141 schools, but could only estimate growth for 92 schools because data for some schools was incomplete.
Of the schools for which growth could be calculated, more than 40% recorded learning rates significantly above the national average. At about one-third of schools, growth closely tracked the national average. About 20% of schools posted slower-than-average growth.
“Something good is happening, but it’s not happening enough,” said Paige Kowalski, executive vice president of the Data Quality Campaign, which advocates for better usage of educational data.
Students at Tiospa Zina Tribal School, on the land of the Sisseton Wahpeton Oyate in South Dakota, learned at a rate about 5% below the national average, meaning that each school year the average Tiospa Zina student fell a little further behind their peers.
Over time, those gaps can add up. In South Dakota, for example, a fourth-grade student is expected to be able to solve problems using multiplication and division and draw inferences while reading. Students two years behind, in second grade, are still mastering addition and subtraction problems and demonstrating they understand key details of texts, like characters and settings.
The grade levels used in the analysis refer to the national average score of students in each year, and do not necessarily correspond with academic standards, which vary state to state.
Sherry Johnson, the tribe’s education director, knew higher growth was possible for Sisseton Wahpeton Oyate children. Students at the tribe’s other BIE-funded school, Enemy Swim Day School, learned nearly 22% more each year than the national average.
Nadine Eastman, the Enemy Swim superintendent, said the difference was due in part to attendance rates at each school. Enemy Swim students had far fewer absences than their counterparts at Tiospa Zina, she said, meaning they had more time in class to learn and grow.
Enemy Swim’s staff considered the school one of the best in the BIE, Eastman said.
But the average Enemy Swim student was still far behind. Despite the high growth, the average student tested more than 2.5 grade levels below average.
“We don’t understand why we can’t make headway on that proficiency,” Eastman said, shaking her head. “We think we should.”
Tiospa Zina administrators did not respond to requests for comment.
Students in more than 70% of the BIE schools analyzed performed more than two grade levels behind the national average, the analysis showed.
Test scores, however, must be examined in context. Tribal education advocates emphasize that standardized exams do not capture the full scope of what many communities value in education, including tribal civics, language and traditions. And research by the Educational Opportunity Project has found a strong correlation between income and test scores: The wealthier a school district, the more likely its students will perform above average on standardized tests.
Families in poorer districts, meanwhile, often don’t have the same access to early childhood education, or the same amount of time or resources for learning opportunities outside of school. That typically translates to lower test scores.
Many BIE schools are located in communities that are among the poorest in the nation, a result of the United States’ historic attempts to destroy tribal nations and current federal policies and budgets that have stifled economic development. Because the federal government holds reservation land in trust, for example, tribal governments cannot collect property taxes, which are a vital source of program funding for cities and towns across America.
Today, more than 42% of families with children on the Navajo Nation live below the poverty line, according to census data. That’s nearly triple the statewide Arizona rate.
“Socioeconomic status is a really powerful predictor,” said Sean Reardon, a Stanford University sociologist who studies the relationship between inequality and education and directs the Educational Opportunity Project. “And Native Americans are the poorest group of major racial and ethnic groups in the United States, so that then translates into fewer resources in the home and fewer educational opportunities for kids early on.
“Not because their parents aren’t interested in educational opportunities, but because economic circumstances make their ability to provide those more limited.”
“Always Playing Catchup”
The BIE is one of many federal programs tasked with repairing centuries of harm done to tribal communities. The agency traces its roots to the United States’ system of boarding schools, which weaponized education in an attempt to stamp out tribal culture.
In its mission statement, the agency promises to provide quality educational opportunities “from early childhood through life.” For decades, it has failed to do so. The U.S. Department of Education has said for nine consecutive years that BIE’s special education program “needs intervention,” and the agency went years without an approved plan to hold schools accountable for their performance, as required by federal law.
The agency is now in the midst of a vast reform effort aimed at fixing problems in its school system, repairing its facilities and allowing tribes greater control of their schools. Its accountability plan was approved by the Education Department in January — nearly four years after the BIE’s initial deadline.
The new analysis found signs of progress. But the results also suggest that even when BIE teachers help students learn and grow, the system’s youngest students face a staggering achievement gap.
“They’re having to catch up constantly,” said Johnson, the Sisseton Wahpeton education director. “That tells me that they’re not ready to learn, not ready to start school at that point in time.”
Reardon, of Stanford’s Educational Opportunity Project, said, “If we always get them to kindergarten behind where they should be, then they’re always playing catch-up. And the schools are always playing catch-up. And sometimes they don’t have the resources or the capacity to do that.”
Experts and lawmakers said the early achievement gap likely reflects the lack of investment in tribal communities and families. Native children face a greater likelihood to encounter adverse childhood experiences like poverty or family instability, which have been shown to affect children’s academic development, as well as a shortage of quality early childhood education opportunities.
Access to child care has been identified as a growing national problem, spurred by sky-high costs for parents and the inequitable supply of childcare centers across the nation. But it is likely intensified in the rural communities served by the BIE, many of which have higher rates of poverty and lack the variety of early childhood programs found in cities, such as Mommy and Me classes or private preschools.
“Many of the tribal communities I’ve worked with have Head Start programs, but they don’t have early pre-K programs because they don’t have the resources for it,” said U.S. Rep. Teresa Leger Fernandez, a New Mexico Democrat who chairs the House Subcommittee for Indigenous Peoples of the United States. “The BIE is not providing those. Unless the tribal government has the resources to provide that, nobody else is providing that.”
Glendora Lopez saw firsthand the benefits, and the limitations, of Head Start on the Hopi reservation. By the time her children were 5 years old, Lopez said, they’d mastered colors, practiced writing their names and grown used to learning every day. Then they transitioned smoothly into kindergarten, already comfortable in the classroom.
But she also knew that those opportunities were limited. Aside from four Head Start centers and a tribally operated day care, Lopez said, the Hopi reservation had few early childhood options. The scarcity made each spot in Head Start more valuable.
“There’s a lot of waiting sometimes to get in,” she said.
The Hopi Tribe’s Head Start program did not reply to a message asking about its enrollment capacity.
A 2020 article published in the Encyclopedia of Infant and Early Childhood Development found that children who attended tribally operated Head Start programs showed significant gains in both reading and math — though a substantial portion of them still scored well below their peers nationwide.
It concluded that while it was difficult to know the true need for Head Start services in tribal communities, “it is reasonable to think that the available Head Start funds fall short of meeting the need and demand for these services” among all tribes.
The report notes that at the time it was published, there were 573 federally recognized tribes in the United States. But the section of Head Start dedicated to Native children included only 145 grantees.
A 2018 report by the federal Administration for Children and Families found that in one three-month span, just 21% of young Native children nationwide attended a nursery or preschool program.
Nationwide, about half of all 3-year-olds and two-thirds of all 4-year-olds are enrolled in some type of preschool program, according to the National Institute for Early Education Research.
Research by Stanford professor Francis A. Pearman and his colleagues suggests the benefits of early childhood programs are more likely to last when communities and school systems offer both access to preschool and a high-quality elementary school with quality teachers.
“You ensure that it’s high-quality,” Pearman said. “But you also ensure that where children end up in the years after that are also high-quality.”
Rep. Raúl Grijalva, an Arizona Democrat who chairs the House Natural Resources Committee that oversees tribal issues and also sits on the Committee on Education and Labor, said the BIE — which already funds many tribal colleges alongside its K-12 schools — should expand into early childhood, creating a unified education system that guides children from infancy to adulthood.
“I think when you consolidate, that helps a lot — then you can start looking at Native American education in totality, and that includes early childhood,” said Grijalva, who noted such changes would likely have to come from the Interior Department. “I think you need a whole new attitude about how to put all this together and make it a continuum, that it’s education, period, from early childhood to post-secondary.”
The BIE’s Strategic Direction document lists “High-Quality, Early Childhood Education” as its first area of focus. The agency received $21 million in the 2021 fiscal year to operate early childhood and family development programs — less than 2% of its $1.2 billion budget.
President Joe Biden’s budget request for the fiscal year 2022 would increase that figure to nearly $25 million.
The BIE uses that money to run school-level Family and Child Education programs, which bring together parents and children for integrated learning. Known as FACE, the program has proven successful but is not available in all BIE schools. In the 2018-19 academic year, only 49 BIE schools operated FACE programs. Together they enrolled about 4,200 children and adults.
A 2017 evaluation of FACE found that 150 families could not get spots and were put on waiting lists.
In its Strategic Direction, which was published in 2018, the BIE said it planned to develop a broader preschool program. The agency aimed to build a comprehensive plan for the program in the Strategic Direction’s second year.
An agency spokesperson did not respond to questions about the status of that plan.
No Way to Compare Schools’ Performance
For years, evaluating or comparing BIE school performance has been difficult. The problem stems from the BIE’s yearslong failure to comply with the nation’s preeminent education law, the Every Student Succeeds Act, which requires states and the BIE to create customized plans to hold schools accountable for their performance. The agency missed the deadline to have plans implemented by the 2017-18 school year.
The effects of this failure have been widespread and damaging to an education system that, according to an Arizona Republic and ProPublica investigation published in August, had not heeded repeated warnings it was failing to provide students with an adequate education: BIE leaders can’t tell whether years of reforms have sparked academic growth across the system, principals don’t know how their schools compare to others and the public doesn’t have access to the detailed information that’s available for every other public school.
Without a uniform plan to measure school performance, the BIE continued to require its schools to use the standardized exam offered by the states where they’re located. The agency has schools in 23 states, nearly all of which use different assessments. Scores cannot be easily compared, preventing even the BIE’s leadership from comparing schools across state lines or analyzing the system as a whole.
“You couldn’t compare, it’s very hard,” Charles “Monty” Roessel, a Navajo citizen who led the BIE from 2013 to 2016, said in an interview last summer. “It became hard to look at your effectiveness, and then to target, based on your analysis, where you’re going to put your funding.”
Kim Thomas, Second Mesa’s first-year principal, said she had no way to know whether her students’ scores topped those of other Hopi schools.
“It’s really hard to even gauge how well we are doing” in comparison to other schools, she said. “I’ve never been shown another school’s data. We’ve never looked at it as a tribe, to see how our schools are performing.”
This lack of comparable data has also made it difficult for the BIE to accurately identify the schools most in need of assistance.
The agency acknowledged in a 2018 presentation that it was “impossible for BIE to use a statistically valid method to rank schools” because its schools do not take the same standardized test.
When asked in a 2019 interview if student performance had improved systemwide following years of administrative reforms, BIE Director Tony Dearman said he didn’t have the data to provide an answer, because of the wide variety of assessments given to students.
The agency has since finalized its accountability plan and announced that it will switch to a single assessment for all schools. But full accountability will have to wait. In May, the Education Department waived the BIE’s obligation to identify schools in need of improvement for the 2020-21 school year.
The BIE is still required to publish its schools’ test scores. It hasn’t done so since the 2015-16 school year.
“We’re Going to Raise Them Up”
Many Farms, a lakeside community of about 3,000 people on the Arizona side of the Navajo Nation, presents parents with a rare and vexing opportunity: a choice of two elementary schools. Many Farms Public School and Many Farms Community School sit on opposite sides of Highway 191, separated by less than two miles: one a member of the Chinle Unified School District, the other a tribally controlled BIE school.
At first glance, they’re nearly identical. But the public school received a C letter grade from the Arizona Department of Education, while neither the BIE school’s nor the agency’s websites have published similar information.
The analysis found a distinction.
Students enrolled at the public school recorded slightly higher test scores, on average about 0.7 grade levels higher than their peers across the street. But in the BIE school, students learned faster — about 13% of a grade level more per year. Both differences were statistically significant.
Bill Wachunas, the superintendent at Dilcon Community School on the Navajo Nation, said he constantly competes for enrollment with surrounding schools.
Wachunas argued that the students who enrolled at the nearby Holbrook Unified School District, who according to the analysis scored higher on standardized tests, entered at a higher academic level than Dilcon’s youngest students. It was difficult to compare the schools, but Wachunas always suspected Dilcon’s approach, with an emphasis on academic interventions and high expectations for student behavior, was effective. The analysis, which estimated that Dilcon students learned about 18% of a grade level more per year than Native students at other district schools, affirmed that belief.
“If your child is low at our school, we’re going to raise them up,” he said.