What is it about education that causes otherwise disciplined people from business and philanthropy to forget the very lessons that made them successful? Focus on the core business, measure results, and change course when something isn’t working. Yet in education there is a recurring temptation to broaden the mission, multiply the objectives, and assume that addressing more of the problems surrounding children will eventually produce better academic results.
The experience of two enormously ambitious philanthropic efforts suggests otherwise. Walter Annenberg’s $500 million Challenge and Priscilla Chan’s more recent Primary School failed in very different ways but they offer the same lesson: Focus matters—and, while schools can and should address barriers preventing children from learning, the organizing purpose of any good school must be academic achievement.
The Problems of Scale and Scope
In 1993, Annenberg’s massive gift to improve American public education was the largest of its kind. With the $600 million it leveraged from other sources, the Challenge ultimately invested $1.1 billion in the nation’s public schools. But public schools were already spending more than $230 billion a year. Even this extraordinary act of philanthropy amounted to barely two-tenths of one percent of a single year’s public-school spending—and it was spread across 24,000 schools, dozens of networks, innumerable strategies, and several years. In Chicago, $49.2 million in Annenberg funds was tripled with matching money and reached hundreds of schools. Yet a final evaluation found that the participating schools saw no statistically significant achievement gains.
The lesson was neither that philanthropy could not improve education nor that Annenberg had spent too little. It was that spreading resources across thousands of schools employing numerous reform strategies made it exceedingly difficult to determine what worked.
A more concentrated investment offered a different possibility: build a specific model, invest enough to give it a genuine test, measure the results, and, if it succeeds, give public schools something worth replicating. Three decades later, Priscilla Chan and Mark Zuckerberg pursued something much closer to that approach. As The Wall Street Journal recently chronicled, The Primary School operated for a decade with more than $125 million in support from the Chan Zuckerberg Initiative before closing. Unlike Annenberg, it concentrated substantial resources on a defined number of disadvantaged children, providing an unusually well-funded test of a particular approach to improving their lives and educational outcomes.
The Primary School opened in East Palo Alto in 2016 based on an entirely understandable premise. Children living in poverty do not leave their other problems at the schoolhouse door. Health issues affect attendance and concentration; housing instability affects families; hunger and mental health matter. The school therefore set out to integrate three systems that usually operate separately: education, health, and family well-being.
As it grew one grade at a time from preschool to sixth grade, the breadth of The Primary School’s mission was remarkable. It coordinated medical care; conducted vision, hearing, and dental screenings; provided mental-health support; assigned parent wellness coaches; and connected families with food, health insurance, housing, diapers, clothing, and legal assistance. It offered financial-wellness programs, expanded dental services to toddlers, and developed systems for sharing information between educators and health-care providers. These efforts produced real results. In 2022–23, 90 percent of students received annual wellness checks, 80 percent received dental screenings, and 90 percent had vision screenings. The following year, every East Bay student identified as needing follow-up dental care received it.
The Primary School’s health and family-service gains were real but not educational accomplishments.
These were real but not educational accomplishments. The Primary School built an impressive network of health and family services, but it struggled with the most basic academic task: teaching children to read.
Getting the Basics Wrong
The Primary School initially adopted a whole-language approach to reading, encouraging children to use pictures and context to identify unfamiliar words rather than systematically decoding letters and sounds. The school’s own documents acknowledge that students’ literacy skills were not developing fast enough. When experienced school administrator Katherine Carter arrived in 2019 to examine academic performance, she was surprised the school was not using science-based reading instruction. In 2019–20, four years after opening, The Primary School began what it called a “fundamental shift” toward structured phonics and the science of reading.
Even as The Primary School built an extensive network of health and family services, its approach to reading remained unsettled. In 2022–23, more than 20 teachers took additional Science of Reading training to “refocus” on the five pillars of reading instruction. The school added two teachers to provide small-group reading and math instruction to fourth and fifth graders who were far below grade level. The following year, it was still developing reading-intervention systems and expanding instruction in foundational reading skills.
In math, the picture was no better. Seven years after opening, The Primary School was still working toward what it called “curricular stability”—developing a “math instructional vision,” adopting an upper-elementary and middle-school curriculum, and piloting MAP assessments. Its 2022–23 annual report highlighted that 65 percent of students in grades 3–5 met projected math “growth,” but omitted the percentage performing at grade level. Its public materials set no ambitious academic goals, such as Algebra I by eighth grade or a trajectory toward calculus in high school. The evidence does not establish poor math instruction. But seven years into an extraordinarily well-funded experiment, the school had neither defined nor demonstrated exceptional mathematical achievement.
The record raises questions about curricular and teacher quality. We cannot know whether The Primary School began with the caliber of teachers and instructional leadership needed to produce exceptional results. But years after opening, it was still investing in science-of-reading training, instructional leadership, academic interventions, and small-group remediation. It also created pathways for employees to earn bachelor’s and master’s degrees, teaching credentials, and early-childhood qualifications. Those programs had value, particularly for an organization committed to recruiting from the community. But with more than $125 million behind the experiment, shouldn’t the starting point have been an extraordinary corps of accomplished teachers and school leaders armed with the strongest possible instructional model?
The academic results make those questions more pressing. The Primary School did make progress, particularly after changing its approach to reading. Its students also came from disadvantaged backgrounds, and there is no randomized comparison that tells us precisely how the same children would have performed elsewhere. In 2022–23, however, seven years after the school opened, just 46 percent of its K–2 students were at or above grade level in literacy. Among students in grades 3–5, 58 percent met projected literacy-growth targets and 65 percent met projected math-growth targets.
The following year brought some improvement: the share of first graders reading at or above grade level rose by 34 percentage points, while the share of second through sixth graders performing far below grade level declined by 15 points. But among third- through sixth graders, still fewer than half—47 percent—met their reading-growth targets. Without an appropriate control group, we can’t determine precisely how the children would have done elsewhere. But after seven years, the school’s broader record is difficult to ignore. Moreover, the school was still training teachers in the science of reading, building out its math program, and remediating students far below grade level. It was still developing reading interventions. Its academic results showed mild improvement. But, for all its resources and ambition, The Primary School never hit its stride on basic academics.
Being Clear About the Mission
The school’s strategic priorities help explain why. Its five-year plan identified five: helping children and families learn, grow, and be healthy; expanding its reach; codifying and disseminating its model; embedding diversity, equity, and inclusion in its programs and culture; and strengthening the organization. The Primary School later described DEIB (Diversity, Equity, Inclusion and belonging) as a “top priority,” focusing on staff demographics and promotions, enabling employees to bring their “authentic selves” to work, sharing decision-making power with the community, and incorporating identity, social-emotional development, and DEI into its programs.
The question is not whether these objectives had value. Inclusion, for example, can be an important educational value. But such objectives should support a school’s paramount goal—academic excellence—not displace it.
At The Primary School, DEIB remained a “top priority” even as the school was still trying to train teachers in reading instruction and develop intervention systems. It continued to be a top priority even when it was obvious, years in, that the school wasn’t getting the academic fundamentals right or delivering the breakthrough results it promised. Money was abundant; management attention, teacher time, and organizational capacity were not. Schools can pursue many worthy objectives, but they cannot make everything a priority. Something must come first.
Schools can pursue many worthy objectives, but they cannot make everything a priority. Something must come first.
Geoffrey Canada’s Harlem Children’s Zone started from much the same premise as The Primary School: children growing up in poverty face obstacles that do not begin or end at the schoolhouse door. Canada responded with early-childhood and parenting programs, health services, and other supports for children and families. The difference was what he put at the center of that network: a school intensely focused on academics.
Putting Academics at the Center
Promise Academy is an unusually intensive academic institution. Its school year lasts 210 days rather than the conventional 180, with eight hours of instruction a day, Saturday school, an optional after-school program, and average classes of about 20 students. The schools use frequent assessments to identify weaknesses, differentiate instruction for individual students, and provide additional help to struggling students. The curriculum combined original material with Core Knowledge. Canada doesn’t merely give children more time in school. He organizes instructional time, assessment, differentiation, and remediation around academic performance, then surrounds the school with health and social supports.
Economists Will Dobbie and Roland Fryer were able to evaluate Promise Academy with unusual rigor because oversubscription meant admission was determined by lottery. They found that middle-school students gained 0.2 standard deviations per year in mathematics—enough, by their estimates, to close the Black-white math achievement gap by ninth grade. At the elementary level, the estimated gains were 0.2 standard deviations per year in both mathematics and English language arts, enough to close the racial achievement gap in both subjects by third grade.
Dobbie and Fryer then asked a question that is directly relevant here: How much of that academic success came from Harlem Children’s Zone’s extensive social and community programs? They first compared Promise Academy students living inside the Zone with students at the same schools who lived outside it and were therefore much less likely to receive those services. In middle-school math, the estimated annual effects were 0.201 standard deviations for students inside the Zone and 0.241 for those outside—a difference that was not statistically significant. Nor did the researchers find significant inside-versus-outside differences in mathematics, English language arts, or absences.
Their second test looked at older siblings of Promise Academy students. They could benefit from many of the same family and community supports—including food, transportation assistance, and guidance for parents—but did not attend Promise Academy themselves. Dobbie and Fryer found no detectable test-score gains among them. They cautioned that these estimates were imprecise, but the two tests pointed in the same direction and their conclusion was striking: the community and family programs themselves appeared to have little or no effect on academic achievement. High-quality schools appeared sufficient to produce large academic gains among poor children; the community programs appeared neither necessary nor sufficient to produce them.
The researchers did not identify which elements of Promise Academy produced those results. They pointed to possibilities including teacher quality, the combination of school policies, and data-driven remediation. But their analysis indicated that the academic gains came principally from the school, not the elaborate network of services surrounding it.
Promise Academy’s health and social supports served—they did not supplant—the academic mission.
That distinction matters for philanthropists interested in education. This is not a choice between caring about the whole child and caring about academics. Canada recognized the same health, family, and social barriers that motivated The Primary School and invested heavily in addressing them. But he put an academically demanding school at the center. The supports served the academic mission; they did not rival or supplant the mission.
What Schools Are For
A child who cannot see the board needs glasses. A child whose asthma keeps him out of school needs medical treatment. A child who arrives hungry needs breakfast. A family facing eviction may need housing assistance. Schools serving disadvantaged children cannot ignore these realities. Removing barriers to learning can be essential to academic success.
But removing barriers is not the same as defining a school’s purpose. A school must teach children exceptionally well: to read fluently and with understanding; to write clearly and persuasively; to master arithmetic, algebra, and geometry; to understand science, history, and civics; and to evaluate evidence, construct an argument, and think independently. When students fall behind, schools must identify why and intervene rather than move them along without mastering the fundamentals.
There is nothing narrow about focusing on a school’s academic mission. For a child who enters kindergarten well behind more affluent peers, mastering core academic subjects can change the trajectory of a life.
There is nothing narrow about that mission. For a child who enters kindergarten well behind more affluent peers, mastering these subjects can change the trajectory of a life. Delivering that education consistently for disadvantaged children is among the hardest and most consequential tasks in education.
Annenberg spread its money too broadly and lacked measurable goals. The Primary School concentrated its resources on relatively few children but blurred its mission with too many “top” priorities. As a result, it never hit its stride on basic academics. Geoffrey Canada showed another way. Harlem Children’s Zone addressed many of the same barriers but put Promise Academy—and its intensive academic program—at the center.
Schools can and should address barriers to learning, but academic achievement must be their organizing purpose. Everything else must be subordinate to learning, with social supports serving the academic mission, not competing with it. For philanthropists, the lesson is simple: before adding much-needed social assistance, build a school that educates children extraordinarily well.
