Educational Choice

Foreword

Few decisions shape a child’s life more than where and how they are educated. American families today have more options than at any point in the country’s history: traditional neighborhood public schools, magnet schools within public districts, public charter schools with varying degrees of independence, private schools both religious and secular, and homeschooling in its many forms. In many states, families can now choose among these options with substantial public funding following the child — through vouchers, tax-credit scholarships, or education savings accounts. The landscape is more fluid than it has been in generations, and it continues to change.

For individual families, this abundance of choice can be disorienting. Each model has advocates who describe it as the obvious right answer and critics who describe it as deeply flawed. The tone of public debate about schooling is often more heated than the evidence warrants, and claims made by advocates and critics alike frequently outrun what research can actually establish. A parent trying to decide where to send a particular child — a child with specific strengths, specific needs, and a specific family situation — is not well served by ideological argument. What they need is honest description of what the different options involve, what is actually known about how they work, and what considerations matter most for their specific circumstances.

This guide is an attempt to provide that description. It treats each of the four main models — public, private, charter, and homeschool — on its own terms, presenting the strongest case for it and the strongest case against it. It surveys the research evidence where evidence exists, identifies where it is stronger or weaker, and is candid about the questions research cannot settle. It does not recommend one model over another as a universal answer, because the evidence does not support doing so. For most families, the better question is which model fits this particular child, in this particular community, at this particular time — and that question the guide tries to help with directly.

The guide is non-partisan. Education policy in the United States is now deeply entangled with political alignment in ways it was not a generation ago, but the practical question of what works for specific children is less ideological than the public debate might suggest. Good teachers, engaged parents, strong curricula, and reasonable discipline matter in every kind of school; they can all be found or missed in every kind of school. The guide tries to keep focus on the practical and the empirical rather than the polemical.

A note on limits. This guide cannot tell you whether a specific school is good. That judgment requires visiting, asking questions, and thinking about your particular child. The guide offers frameworks for what to look for and what questions to ask, but the actual evaluation of a specific school is local and specific. Similarly, the guide cannot tell you what your family’s values are. Different families, looking at the same set of options and the same evidence, will reach different conclusions because they weigh things differently: academic rigor versus child development breadth; religious formation versus neutrality; community belonging versus individual fit; parental involvement versus institutional expertise. The guide tries to lay out the considerations honestly and leaves the weighing to readers.

The guide is organized in six parts. Part I describes the current landscape, how schools are funded and governed, and what research can and cannot tell us. Parts II through V treat each of the four main models in turn, covering structure, the cases for and against, and what to look for when evaluating specific schools. Part VI addresses vouchers and related public policy, special-needs and gifted students, the factors that actually drive educational outcomes, and a framework for making the decision for an individual child. Appendices provide a glossary, resources, and references for further reading.

For most families, the choice is not a one-time event but an ongoing series of decisions that may lead to different answers for different children, or for the same child in different years. The goal of this guide is to make those decisions better informed, rather than to make them for you.

PART ONE

The Landscape

Who goes where, how schools are funded and governed, and what research can and cannot tell us

CHAPTER 1

The American Educational Landscape Today

The United States educates roughly 54 million K–12 students, distributed across four main settings: traditional public schools, public charter schools, private schools, and homeschooling. The proportions have shifted modestly over the past two decades but public schools still dominate in aggregate. Understanding the relative scale of each option is a useful starting point before examining any of them individually.

Enrollment by the numbers

According to the most recent National Center for Education Statistics (NCES) data, traditional public schools enroll the large majority of American students — roughly 45 to 46 million students in the 2023–2024 school year, or approximately 84 to 85 percent of the total. Public charter schools — which are technically public but operated under different governance rules — enrolled approximately 3.7 million students, roughly 7 percent of the total. Private schools enrolled about 4.7 million students, roughly 8 to 9 percent of the total (a figure that had been relatively stable for the previous decade, though with shifts in composition). And roughly 3 to 4 percent of students — an estimated 2 to 3 million — were being homeschooled as of the 2022–2023 school year, approximately double the pre-pandemic share.

These proportions vary enormously by state and community. In some states — Hawaii and the District of Columbia have high private school shares (around 19 percent), and Arizona has the highest charter school share (around 17 percent of its public school students) — the options beyond traditional public schools are substantial. In others, particularly rural areas and some states without charter laws, families have few practical alternatives to their assigned neighborhood public school.

What changed during and after COVID-19

The 2020–2021 disruption to schooling produced the most significant shifts in enrollment patterns in decades. Public school enrollment fell by roughly 1.3 million students between the start of the pandemic and the 2023–2024 school year — a decline of about 2.5 percent. Homeschooling roughly doubled: from approximately 3.7 percent of K–12 students before the pandemic to 5.2 percent by the 2022–2023 school year, according to NCES household surveys. Private school enrollment, which had been declining for decades, stabilized and in some regions grew. Charter school enrollment continued its long-term upward trajectory.

Some of these shifts represent permanent changes; some may partially reverse. But the post-pandemic environment, combined with expanded state school-choice programs (covered in Chapter 16), has produced an educational landscape in which families are more actively choosing among options than they were a decade ago. The choice question is therefore more practically relevant for more families than at any time in recent memory.

Composition of the private sector

Private schools in the United States are diverse. Roughly 77 percent of private school students attend religiously affiliated schools; 23 percent attend secular private schools (including independent day schools, Montessori programs, Waldorf schools, and various specialized academies). Among religious schools, Catholic schools are the largest category, enrolling about 35 percent of all private school students. Other Christian schools (evangelical, mainline Protestant, and other Christian traditions) together enroll roughly 30 to 35 percent. Jewish schools, Islamic schools, and schools of other faiths enroll smaller but significant numbers, particularly in cities with large communities of those faiths.

Private school enrollment has been declining for most of the past half century, from a peak share of about 13 percent in the 1960s (when Catholic schools alone educated more than 12 percent of American children). The decline has been driven primarily by the shrinking of Catholic schools — a combination of demographic change, rising costs, and decreased religious vocations among teachers — and has only partially been offset by growth in evangelical Christian schools and secular independent schools. Private school tuition, which averaged around $13,000 per year for secular schools and substantially less for religious schools as of 2024, places these options out of reach for many middle-class families absent scholarship aid or public subsidies.

The rise of charter schools

Charter schools are a relatively recent innovation. The first charter school in the United States opened in Minnesota in 1992. As of 2024, roughly 45 states and the District of Columbia authorize charter schools. Approximately 3.7 million students attend them, up from essentially zero 30 years ago. The growth has been concentrated in urban areas: cities like New Orleans (essentially all charter), Detroit, Philadelphia, Washington DC, and Newark have substantial charter sectors, while rural areas have relatively few.

Charter schools are structurally diverse. Some operate as stand-alone schools founded by teachers or community groups; others operate as networks (charter management organizations, or CMOs) that run dozens or hundreds of schools. Some are nonprofit; a smaller number are for-profit (though for-profit involvement has declined under recent policy pressures). Their pedagogical approaches range from back-to-basics models (KIPP, Success Academy) to progressive approaches (High Tech High) to specialized arts or STEM programs to classical or Montessori curricula. Generalizations about “charter schools” as a monolithic category are therefore usually misleading; the within-sector variation is as large as the between-sector variation.

Homeschooling as a moving target

Homeschooling is the most difficult of the four options to measure precisely, because states vary substantially in how and whether they track homeschooled students. Estimates from NCES surveys, state-level records, and academic researchers converge on roughly 2 to 3.5 million students homeschooled as of 2024–2025 — between about 4 and 7 percent of K–12 students. The demographics have shifted significantly over time: historically associated with religious families (particularly conservative Christian families in the 1980s and 1990s), homeschooling now includes substantial numbers of secular families, racial minorities, academic enrichment-focused families, and families whose children have specific needs that conventional schooling does not serve well. The Census Household Pulse surveys during the 2020–2021 school year found homeschooling rates among Black families approaching and in some regions exceeding rates among white families, reversing a decades-long demographic pattern.

The rural vs. urban divide

A point rarely emphasized in debates about school choice: the practical options available to families depend enormously on where they live. A family in a dense urban area typically has multiple public schools within easy commute, several charter schools, perhaps a magnet or specialized public option, a range of private schools including religious and secular, and the option of homeschooling with in-person co-ops and enrichment programs. A family in a rural area may have exactly one public school district within practical reach, perhaps one or two private schools (often religious), no charter schools, and few homeschool co-ops. For many rural families, “school choice” in practical terms means the quality of the single available school, not selecting among options. This geographic variation means policy debates framed as if all families face the same choice set are usually misleading.

The landscape in numbers

Roughly 54 million K–12 students in the United States are distributed across four main settings: traditional public schools (about 85 percent), public charter schools (about 7 percent), private schools (about 8–9 percent, with Catholic and other religious schools predominating), and homeschooling (about 4–7 percent). Shares have shifted since the pandemic, with public enrollment declining modestly and homeschooling roughly doubling. Within each category, variation is substantial: charter schools range from classical academies to back-to-basics networks to progressive programs; private schools include religious, secular, and specialized programs; homeschooling includes religious families, secular academic enrichment families, and families whose children have specific needs. Geographic availability varies enormously; urban families typically have many options, while many rural families have few. The subsequent chapters examine each model in turn.

What to read or watch next

• National Center for Education Statistics, The Condition of Education (annual). The authoritative federal source for enrollment data, trends by sector, and demographic breakdowns. nces.ed.gov.
• Pew Research Center, Facts about public, private and charter schools in the US (updated periodically). Accessible summary of enrollment patterns, with state-by-state comparisons.
• Thomas Dee, Where the Kids Went (2023 and subsequent papers). Stanford professor tracking post-pandemic enrollment shifts, including movements among sectors.
• EdChoice, edchoice.org. School-choice advocacy organization that publishes extensive data on programs state by state; useful as a reference while acknowledging its advocacy stance.

CHAPTER 2

How Schools Are Funded and Governed

To understand why schools differ so much — and why the same state can have affluent suburban public schools and struggling urban public schools a few miles apart — it helps to understand how American schooling is funded and governed. The system is unusually decentralized compared with most developed countries: most funding comes from state and local sources rather than the federal government, and most operational decisions are made at the local level. This chapter describes the basic structure.

The federal, state, and local division

Funding for public elementary and secondary education in the United States totals roughly $800 billion per year, split roughly into these proportions: about 8 to 10 percent from the federal government, about 45 to 50 percent from state governments, and about 40 to 45 percent from local sources (primarily property taxes). The federal share is small and concentrated in specific programs — Title I (for schools serving low-income students), IDEA (for students with disabilities), school nutrition programs, and similar targeted grants. The large majority of funding decisions and operational control are state and local.

State governments set educational standards, license teachers, design accountability systems, distribute state aid, and in many states provide a substantial share of total funding. Local school districts — typically governed by elected school boards — hire staff, operate buildings, select curricula within state frameworks, manage day-to-day operations, and raise local revenue (mostly through property taxes). There are roughly 13,000 school districts in the United States, ranging from enormous systems like New York City (over a million students) to rural districts with fewer than 100 students.

Why property taxes produce unequal schools

Because local funding comes primarily from property taxes, the amount of revenue a district can raise depends heavily on local property values. A district with high-value homes and commercial property can raise substantial revenue at modest tax rates; a district with low property values may tax itself heavily and still raise comparatively little. State-level equalization formulas exist to partially offset this — states typically provide more aid per student to lower-wealth districts — but the equalization is rarely complete. Per-pupil spending in the United States ranges from under $9,000 to over $30,000 across districts, and much of this variation reflects local wealth rather than educational need or effort.

This structural fact underlies much of the inequality in American public schooling. It is not primarily that wealthy districts spend more because they care more or are better managed; they spend more because they can. When commentators note that a particular urban district spends more per pupil than a particular suburban district (which is sometimes true, because urban districts often receive substantial state and federal aid for concentrated poverty), the comparison often misses the broader pattern: on average, high-poverty districts have less effective educational resources — teacher experience, curriculum stability, facility quality — than lower-poverty districts, even when nominal spending is comparable. Money alone does not guarantee outcomes, but structural funding inequalities are a real feature of the system.

How the four sectors are funded differently

Each of the four main schooling sectors operates under a distinctive funding model:

• Traditional public schools. Funded by the combination of federal, state, and local sources described above. Per-pupil spending averages roughly $16,000–$17,000 nationally, with substantial variation by state and district. Operate under direct control of locally elected school boards and state-imposed regulations.
• Public charter schools. Also publicly funded, typically receiving per-pupil allocations from the state and local sources where the student would have attended traditional public school. In most states, charters receive somewhat less per student than traditional publics because they often do not receive the local facilities portion of public funding; they must fund buildings and capital from operating revenue, private grants, or separate state allocations. Charter schools are accountable to their authorizer (often a state agency, university, or school district) rather than to an elected school board.
• Private schools. Funded primarily through tuition paid by families, sometimes supplemented by endowment earnings, donations, and (at religious schools) subsidies from sponsoring religious institutions. Private school tuition varies enormously: parish Catholic elementary schools average around $5,000–$8,000 per year in 2024–2025; non-denominational Christian schools often $8,000–$15,000; elite independent day schools $25,000–$60,000; the most expensive boarding schools over $70,000. Most private schools offer financial aid for some families. Private schools are generally not subject to state standards or accountability frameworks (though they are subject to basic health, safety, and civil rights laws).
• Homeschooling. Funded entirely by families (in most states) or, in states with education savings accounts or similar programs, partially funded by the state. Direct costs to homeschooling families vary from modest ($500–$2,000 per year for materials and selective paid classes) to substantial ($5,000–$15,000 per year for families that purchase full curricula, enroll in co-ops, and pay for extensive extracurriculars and tutors). The much larger implicit cost for many homeschooling families is the forgone income of the parent (usually, but not always, the mother) who serves as primary educator.

School choice programs: letting funding follow the child

Over the past 30 years, and particularly since 2021, a growing number of states have created programs that allow some portion of public funding to follow children to non-public schools. These programs take several forms:

• Vouchers. Direct state payments for private school tuition. The oldest and most straightforward form; Wisconsin’s Milwaukee program, the first major modern voucher program, was launched in 1990.
• Tax-credit scholarships. States provide tax credits to donors (individuals or businesses) who fund scholarship programs for private school tuition. A way of channeling funds through private intermediaries rather than as direct state payments.
• Education Savings Accounts (ESAs). States deposit a portion of per-pupil funding into accounts that families can use for a wider range of educational expenses — private school tuition, tutoring, curriculum materials, therapy for special-needs students, educational technology, and so on. The most flexible form of choice program and the one growing most rapidly.
• Tax deductions or credits. Some states allow parents to deduct or claim tax credits for private school tuition or homeschool expenses; these are typically smaller benefits than direct vouchers or ESAs.

As of early 2026, more than 30 states operate some form of private-school choice program, and approximately 13 states have enacted programs with universal or near-universal eligibility (meaning all or nearly all students in the state can qualify regardless of income). This is a substantial change from a decade ago, when universal eligibility existed in no state; the shift has been concentrated in the 2021–2025 period. Chapter 16 examines these programs in detail.

Who governs each type of school

Governance structures differ across the sectors in ways that affect how schools operate day to day:

• Public schools. Ultimate accountability to elected school boards (local level) and state education agencies. Teachers are employees of the district; curriculum must align with state standards; testing and reporting requirements are substantial. Parents can vote for school board candidates and can attend open meetings, though individual parental influence over policy is typically modest. Parental involvement in individual classrooms varies widely.
• Public charter schools. Governed by a nonprofit board specific to the school or network, accountable to a state-level authorizer that granted the charter. Subject to a specified set of state requirements — typically including testing and civil rights protections — but exempt from many regulations that apply to traditional public schools. Can be closed by the authorizer for underperformance or non-compliance; this happens to roughly 3 to 5 percent of charter schools per year nationally.
• Private schools. Governance varies by type. Independent schools typically have boards of trustees. Religious schools may be governed by the sponsoring religious body, by the local congregation, or by a separate board. Parent involvement in governance is highly variable — some schools have strong parent boards, some are essentially closed to parental input.
• Homeschooling. Governed by the individual family, within whatever state regulations apply. State regulation varies from minimal (many southern and western states require essentially nothing beyond filing a notice of intent) to substantial (New York requires detailed curriculum filings, quarterly reports, and annual testing).

Structure drives behavior

American K–12 education is funded primarily from state and local sources, with roughly 8–10 percent federal. Local property-tax funding produces substantial per-pupil spending variation across districts, and this structural fact underlies much of the inequality between wealthy and poor districts. Each sector has a distinct funding model: public schools through the state/local combination, charters through state per-pupil allocations (often somewhat less than traditional publics), private schools primarily through tuition, and homeschooling primarily through family resources (though ESA programs are changing this). Governance differs in parallel: elected boards for public schools, authorizer-accountable boards for charters, varied structures for private schools, and individual families for homeschools. These structural differences shape how schools actually operate — what they can and cannot do, who they answer to, and what kinds of changes are feasible. Understanding the structure is the foundation for understanding everything else about the sectors.

What to read or watch next

• National Center for Education Statistics, Digest of Education Statistics (annual). Detailed tables on revenue sources, per-pupil spending, and governance across sectors.
• Education Commission of the States, ecs.org. Nonpartisan organization tracking state education policy, including funding formulas and school choice programs by state.
• Bruce Baker, Educational Inequality and School Finance (2018). Detailed scholarly treatment of how public school funding works and the inequalities the structure produces.
• EdChoice, “The ABCs of School Choice.” Annual guide to school choice programs by state; useful reference while recognizing the organization’s pro-choice advocacy orientation.

CHAPTER 3

What the Research Can and Cannot Tell Us

Throughout this guide, we will refer to research on educational outcomes — studies that compare, say, charter school students to public school students, or private school students to public school students, or homeschooled students to conventionally schooled students. Before interpreting any of those findings, it is important to understand the methodological challenges specific to this field. Educational research is unusually difficult, for reasons that matter both for reading studies and for being appropriately humble about what is actually known.

The selection problem

The fundamental challenge in comparing schooling models is that students are not randomly assigned to them. Families choose schools based on their own circumstances, values, and judgments; schools have admission criteria (formal or informal); neighborhood sorting produces non-random distribution of students across public schools. A simple comparison of “private school students” to “public school students” therefore compares two groups that differ in many ways besides the schools they attend: on parental income, parental education, family structure, student ability, specific interests, and dozens of other variables that affect educational outcomes independently of the school.

This is called the “selection problem” and it bedevils nearly every comparison of schooling models. If private school students outperform public school students on standardized tests, is it because the schools are better, or because the students would have outperformed their counterparts regardless of school? If homeschooled students score higher on college entrance exams, is it because homeschooling produces better academic outcomes, or because the kind of families who homeschool are disproportionately families whose children would have scored well anywhere? Answering such questions requires methods that can isolate the effect of the schooling setting from the effects of everything else.

Methods that partially solve the problem

Several research methods address selection bias with varying effectiveness:

• Lottery-based studies. When more students apply to a school than it has seats, some schools use random lotteries to assign admission. Researchers can then compare students who won the lottery (and enrolled) to students who lost the lottery (and attended other schools). Since winners and losers were, in expectation, identical before the lottery, any differences in later outcomes can plausibly be attributed to the schools. This is the methodological gold standard for evaluating school effects. It is available only for over-subscribed schools that use random lotteries, which limits the settings where it can be applied. It has been most extensively used for charter schools in cities like Boston, New York, and Chicago, where lottery admissions are common.
• Matched-pair and propensity-score studies. Match students in one school type to students with similar characteristics (income, prior test scores, parental education, etc.) in another school type, and compare outcomes. Addresses some selection bias by comparing apparently similar students. Limitation: can only control for observable characteristics; unobservable differences (motivation, family engagement, specific student-school fit) may remain uncontrolled. Better than naive comparisons but weaker than lotteries.
• Virtual control records. Used extensively by CREDO at Stanford for charter school research. Creates “virtual twins” for each charter school student by averaging outcomes of demographically similar students in nearby traditional public schools. Attempts to approximate what charter students would have experienced absent the charter. A refinement of matching approaches; still has limitations when important characteristics are unobserved.
• Regression discontinuity designs. When admission depends on meeting a specific threshold (e.g., a test score cutoff), researchers can compare students just above the cutoff to students just below, since these groups are nearly identical on unobservables. A rigorous method in settings where it applies.
• Natural experiments. Policy changes that move students between sectors (e.g., sudden expansion of voucher programs in specific cities) can occasionally be analyzed as near-experimental. Hurricane Katrina’s effect on New Orleans schools, for instance, produced a quasi-experimental transformation of an entire school system that researchers have analyzed extensively.

What we have good evidence on

Some questions in this area have produced robust evidence:

• Specific urban charter networks. Lottery studies of over-subscribed charter schools in Boston, New York, and Chicago have consistently found substantial positive effects on student achievement, particularly in disadvantaged urban areas. The effects are concentrated in specific schools and networks (KIPP, Success Academy, Uncommon Schools, and similar “no-excuses” operators) and are not a general finding about all charter schools — but within this subset, the evidence is strong. Angrist and colleagues at MIT have been the most prominent researchers in this area.
• Teacher effects. Research on teacher quality by Raj Chetty, Jonah Rockoff, John Friedman, Eric Hanushek, and many others has produced consistent findings that teachers vary substantially in their effectiveness, that the differences affect both short-term learning and long-term life outcomes, and that these effects are larger than almost any other factor under schools’ control. This is a finding about teachers, not about school types — good and poor teachers exist in all sectors.
• Family factors. Decades of research, beginning with the 1966 Coleman Report and continuing through contemporary work, have shown that student outcomes are powerfully shaped by family factors: parental education, family structure, home literacy environment, expectations, and similar variables. This does not mean schools do not matter, but it does mean that comparisons among school types are working on a relatively narrow slice of the total variation in student outcomes.

What the evidence does not settle

Many questions remain unsettled or weakly settled:

• The overall effect of “private schooling” on academic outcomes. Raw comparisons favor private schools, but most of the difference disappears when controlling for family background. Randomized evidence from voucher programs has been mixed: some studies (Milwaukee, some early D.C. cohorts) found modest positive effects; others (more recent Louisiana and Indiana programs) found null or negative effects. The lack of consistent findings likely reflects both methodological variation and genuine variation in program design.
• Homeschooling outcomes. The research base on homeschooling is methodologically weak, in part because homeschooled students are rarely included in large-sample achievement studies and because homeschooling families are highly self-selected. Advocacy organizations (notably the National Home Education Research Institute) report that homeschooled students score above average on standardized tests, but the samples are volunteer-based and the comparisons do not control for family background. Peer-reviewed research by scholars like Joseph Murphy and Robert Kunzman has been more cautious, finding that methodologically rigorous comparisons of homeschooled to conventionally schooled students are scarce and that strong claims about the average outcomes of homeschooling — in either direction — lack solid empirical foundation.
• Long-term outcomes beyond academics. Test scores are measurable but narrow. What about long-term outcomes: college completion, career success, civic participation, marital stability, personal wellbeing? Research on these questions by sector is sparse and often methodologically weak. Some findings suggest Catholic school students have somewhat higher college completion rates conditional on socioeconomic background, and some homeschool researchers report favorable civic outcomes, but these findings come with substantial caveats.

How to read research claims

Readers encountering claims about educational outcomes should keep several questions in mind:

• What comparison group? “Charter school students outperform public school students” means nothing without knowing which charters, which public schools, and whether the comparison adjusts for student background.
• What outcomes? Standardized test scores in a specific subject, graduation rates, college enrollment, or something else? These can move differently.
• What study design? Lottery-based, matched, simple correlation? The first is trustworthy, the last can be misleading.
• What size of effect? An effect size of 0.1 standard deviation — often reported as a statistically significant finding — corresponds to a fairly small practical difference. Educational research is often conducted on very large samples, so even small effects can be statistically significant; the practical importance depends on the size.
• Who paid for the research, and what does the organization do? This is not dispositive — good research comes from many sources — but advocacy organizations on all sides often publish work that selects favorable findings. Academic journals and independent research centers generally produce more rigorous work, though they are not immune to ideological influences.

The honest summary

The honest summary is that educational research can tell us a good deal about some questions — the effects of specific urban charter networks, the magnitude of teacher effects, the role of family background — and relatively little about others, including the overall comparative effect of school type. Anyone claiming that research clearly shows one model is best is overstating what the evidence supports. The subsequent chapters will try to present findings in proportion to the confidence they warrant, acknowledging uncertainty where it exists rather than papering over it. Readers who want more confident answers will find them in advocacy literature on all sides; readers who want accurate ones will have to accept some residual uncertainty about questions where the research does not yet provide clear answers.

Selection bias is the central challenge

Comparing schooling models is methodologically difficult because students are not randomly assigned. Families choose schools based on their circumstances and values, so simple comparisons of students across sectors confound the effect of schools with the effects of everything else about those students. Rigorous methods — lottery-based studies, regression discontinuity, matched-pair designs with careful controls — can partially address this. On some questions, evidence is reasonably strong: specific urban charter networks show substantial positive effects in lottery studies, teacher quality varies substantially and matters enormously, family factors explain large portions of outcome variation. On others, evidence is weaker or mixed: overall effects of private schooling, outcomes of homeschooling, long-term non-academic outcomes across sectors. When reading research claims, ask about comparison groups, outcomes, study design, effect sizes, and source. Accurate readings of the research leave substantial uncertainty about some important questions; anyone claiming otherwise is overselling.

What to read or watch next

• Joshua Angrist and Jörn-Steffen Pischke, Mastering ’Metrics: The Path from Cause to Effect (2014). Accessible introduction to the methods of causal inference in applied economics, with multiple education examples.
• James Coleman et al., Equality of Educational Opportunity (1966). The original “Coleman Report” that documented the importance of family background in educational outcomes. Dated in specifics but the basic findings have been replicated many times.
• Eric Hanushek and ongoing research. Stanford economist’s decades of work on school productivity, teacher effects, and the economic consequences of educational quality.
• What Works Clearinghouse, ies.ed.gov/ncee/wwc. Federal clearinghouse that rates the methodological rigor of education research; useful for distinguishing strong from weak studies on specific interventions.
• Russ Whitehurst, “Don’t Forget Curriculum” and related Brookings papers. Former IES director on what matters in education research and how to interpret it honestly.

PART TWO

Traditional Public Schools

The dominant model, the case for and against it, and what to look for when evaluating one

CHAPTER 4

Public Schools: The Dominant Model

For most American children, the default option is a traditional public school operated by the local school district, attended by the children of the surrounding neighborhood. This system was built up over the nineteenth and twentieth centuries to achieve universal basic education, and for all the criticism it receives, it still performs that task: roughly 45 million students, across every demographic group, region, and family circumstance, are educated through it each year. Before examining the other options, it is worth understanding what the public school system actually is, its variety, and how it has changed.

The common school tradition

The idea of universal, tax-funded public schooling in the United States is most closely associated with Horace Mann and the “common school” movement of the 1830s and 1840s. The argument was that a self-governing republic required an educated citizenry, that education should not be reserved for those who could pay for it, and that bringing together children of different backgrounds into a common institution would both educate them and build social cohesion. The system expanded enormously over the subsequent century, compulsory attendance laws spread state by state, and by the early twentieth century public schooling was the dominant mode of education for American children — a position it has retained ever since.

The common-school ideal continues to shape how public schools are typically evaluated, both by defenders and critics. Public schools aim to educate all students in their geographic area regardless of ability, family circumstances, or interests; they are expected to handle students who would be excluded from private schools or who would struggle in the structured environment of many charter schools; they are expected to provide a broadly civic education, not one focused on specific religious or ideological perspectives; and they are expected to do all this with whatever funding the local and state political processes provide them. Complaints about public school performance often miss the extent to which the sector is performing a broader and more demanding mission than any of the alternatives.

Structure and variety

Public schools are not monolithic. Within a single state, public schools vary enormously by:

• Funding level. Per-pupil spending in New York State, for instance, ranges from under $15,000 to over $40,000 across districts, with wealthy suburban districts spending vastly more than rural or urban districts.
• Student body. Wealthy suburban schools with college-educated parents present different educational environments than urban schools with concentrated poverty or rural schools with dispersed populations.
• Academic programs. Some public high schools offer dozens of Advanced Placement courses, International Baccalaureate programs, and specialized tracks; others offer a basic curriculum with few advanced options. Some elementary schools have rich arts, music, and language programs; others have been stripped down to reading, math, and test preparation.
• School culture and discipline. Schools with strong leadership, engaged faculty, and supportive parents often maintain coherent educational environments; schools with high turnover, chronic behavioral issues, and weak leadership can fail to do so. These differences are often larger between schools within the same district than between districts or between public and private sectors.
• Specialty programs within the public sector. Many districts operate magnet schools, career and technical education (CTE) centers, gifted programs, language immersion schools, and other specialty options alongside their neighborhood schools. These are addressed in Chapter 12.

The generalization that “public schools are failing” or, conversely, that “public schools are doing fine” is therefore usually too broad to be useful. Some public schools are excellent by any measure; some are genuinely struggling; most are somewhere in between. The question for a specific family is typically not about public schools as a category but about the specific public schools available to their child.

What public schools do that others do not

Several features of traditional public schools distinguish them from the alternatives:

• Universal admission. Public schools accept every student who lives within their attendance zone. They cannot turn away students for poor academic records, behavioral histories, disabilities, language status, or any other reason (with narrow exceptions for severely disruptive behavior). This is a substantial civic function and an expensive one.
• Special education at scale. Public schools educate the large majority of students with significant disabilities. Federal law (IDEA) requires public districts to provide appropriate services and accommodations for students with identified special needs, and the public school sector has infrastructure for this — resource rooms, therapists, specialists, transportation — that other sectors often lack. For students with serious disabilities, public schools are often the most substantively equipped option.
• English language learners. Public schools educate the overwhelming majority of students learning English as an additional language, including recent immigrants and refugee children with varied educational backgrounds.
• Ancillary services. School meals (including free and reduced-price meal programs for low-income students), transportation, school health services, social work support, after-school programs, and athletic opportunities are more extensively developed in the public sector than in alternatives, particularly for students from lower-income families.
• Democratic accountability. Public school policy is made by elected school boards, subject to open-meetings laws, with parents and community members able to participate in decision-making (imperfectly, but in a form that does not exist for private schools).

Where public schools have changed

Public schools in 2026 are different from public schools in 1990 or 1960 in several important respects. A few changes that matter for families evaluating options:

• Standardized testing and accountability. The No Child Left Behind Act (2001) and its successor the Every Student Succeeds Act (2015) substantially increased standardized testing, accountability frameworks, and the public reporting of school performance data. Supporters credit this with raising standards, particularly for historically underserved students; critics credit it with narrowing curriculum, increasing teacher burnout, and pushing test preparation over substantive learning.
• Curriculum content changes. Public schools have undergone substantial changes in both how and what they teach over recent decades. Some changes (such as the Common Core standards adopted by most states in the early 2010s and subsequently modified or replaced in many) have been contested; others (increased emphasis on STEM, changes in English language arts pedagogy, growth of social-emotional learning) have happened more gradually but significantly. How specific public schools approach contested curricular areas — history and social studies in particular, but also sex education, gender and sexuality, and discussion of current controversies — varies substantially.
• Teaching profession changes. Teacher compensation, working conditions, and professional status have evolved in complex ways. Teachers are better paid in real terms than a generation ago but less so relative to comparable professions. Preparation and certification requirements have risen, but so have concerns about teacher quality and turnover, particularly in high-poverty schools. Post-pandemic teacher shortages have been particularly acute in some regions and subjects.
• Behavioral and cultural changes. Many teachers and parents describe a change in the discipline climate and behavioral expectations of schools over the past decade, with effects that differ across schools and regions. Some of this is a story about specific policy changes (restorative justice approaches, school suspension policy); some is about broader cultural shifts. Families evaluating a specific public school will want to assess the discipline and behavioral environment directly.
• Post-pandemic academic effects. The 2020–2022 period produced substantial learning loss, particularly in math, concentrated in lower-performing and higher-poverty schools. Recovery has been partial and uneven. A family evaluating a school in 2026 should understand where the school is in this recovery.

The spectrum of public school experience

Given all this variety, what can a family expect from a “typical” American public school? There is no typical public school. A family whose assigned public elementary is the highly rated school in an affluent suburb experiences something quite different from a family whose assigned public high school is a struggling urban comprehensive. Both are public schools. Both are real parts of the American educational system. The rest of this part of the guide tries to help families understand what they actually have, and how to evaluate it.

The public sector is diverse and performs a specific mission

Traditional public schools educate roughly 85 percent of American children and perform a civic mission that the other sectors do not: educating every child in a geographic area, including those with significant disabilities, English language learners, and children whose families lack the resources to navigate alternatives. Public schools vary enormously in funding, student body, program offerings, culture, and performance — differences often larger within a given state than between the public sector and alternatives. Public schools have changed substantially over the past two decades, with increased testing and accountability, contested curriculum shifts, changes in teaching profession and discipline, and significant pandemic-related effects. For most families evaluating options, the right question is not whether “public schools” are good but whether the specific public schools they can access are a good fit for their specific child.

What to read or watch next

• Diane Ravitch, The Death and Life of the Great American School System (revised ed., 2016). Former education official’s influential defense of public schools and critique of privatization approaches. Opinionated but serious.
• Dana Goldstein, The Teacher Wars (2014). History of the American teaching profession and how public school teaching has evolved over two centuries.
• Jal Mehta, The Allure of Order (2013). Analytical history of school reform efforts and their limited effects, from a Harvard Graduate School of Education professor.
• David Labaree, Someone Has to Fail: The Zero-Sum Game of Public Schooling (2010). Stanford professor’s treatment of the structural realities of American public education.

CHAPTER 5

The Case For and Against Public Schools

Debates about public schooling are often framed as if one side is obviously right. They are not. There are serious arguments for the public school system and serious arguments for reform or alternatives, and most thoughtful observers — including many inside public schools — recognize merit on both sides. This chapter presents the strongest version of each case, not as neutral mediation but as a fair attempt to help families think about what they value when evaluating this option.

The case for public schools

The case for the traditional public school system rests on several interconnected claims:

• Universal access and civic mission. Public schools educate every child, including those the alternatives do not want or cannot serve. They function as the one American institution that meets, educates, and shapes virtually all the country’s children. Weakening them, the argument goes, weakens the civic function of schools to produce shared citizens across differences of class, race, and background. Children from different walks of life being educated together has value that schools of any single demographic — religious, class-based, or otherwise — cannot replicate.
• Quality where it exists. American public schools include some of the best K–12 institutions in the world. Top suburban high schools, magnet schools like Thomas Jefferson or Stuyvesant, and many ordinary public schools in engaged communities provide excellent education. The stereotype of universally failing public schools is a caricature that obscures the genuine quality the sector produces across much of the country.
• Infrastructure and services. Public schools have infrastructure for special education, English language learner services, counseling, nutrition, transportation, and athletic and extracurricular programs that private schools and homeschools typically cannot match. For students with significant disabilities or complex needs, the public sector’s scale and professional capacity is often a substantial advantage.
• Stability and predictability. Public schools have existed in most communities for generations and are unlikely to disappear. Private schools can close; charter schools can lose their authorization; homeschool situations can become unworkable. A family that moves, changes circumstances, or has to shift plans typically has a public school option available on short notice. This institutional durability is underappreciated.
• Democratic accountability. Public schools are governed by elected boards, subject to open meetings laws, required to publish financial and performance data, and open to parent participation in formal processes. Decisions can be contested through ordinary democratic means. Private schools answer to their boards or religious authorities; homeschools answer only to families. The public sector is unique in its democratic structure.
• Equity of access. Even in states with expanded choice programs, families without time, networks, transportation, or administrative capacity have difficulty navigating alternatives. Public schools admit students automatically. A strong public system protects families who, for whatever reason, cannot effectively shop for educational options.
• The data on what works, where it works. Countries with excellent educational outcomes — Finland, Singapore, Japan, Canada, much of Germany — all educate the large majority of their children in public schools. None has achieved high aggregate outcomes primarily through privatization. The U.S. public sector has specific problems, but the international evidence does not support the claim that private systems systematically outperform good public ones.

The case against (or for reform of) public schools

The case for reform or alternatives to the traditional public school system is also serious, resting on several arguments:

• Chronic underperformance of many urban and rural schools. National Assessment of Educational Progress (NAEP) data shows that roughly two-thirds of American fourth-graders are not “proficient” in reading by federal standards; gaps by race and income remain large and have been persistent for decades. Specific urban districts have particularly poor outcomes — Detroit and Baltimore have produced high school graduates with fourth-grade reading levels in significant numbers — and the structural responses have been, on the whole, ineffective. Families who see this up close reasonably conclude that something other than the status quo is needed for their children.
• Monopoly effects. Critics argue that when public schools face no meaningful competition, there is less pressure to improve. Parents who are dissatisfied with a public school historically had either the option of paying for private school (available to few) or of accepting the situation. Choice programs, the argument goes, create pressure that can benefit both families who leave and those who stay.
• Curricular and cultural concerns. Families with specific values — religious, philosophical, pedagogical — may find public schools poor fits. Public schools in a given state teach what state standards require, within the constraints of local decisions; for many families, that content is acceptable, but for families with views at odds with local or state curriculum choices, the mismatch can be substantial. This is a recurring source of conflict in both conservative and progressive directions; which way it cuts depends on the specific community.
• One-size-fits-all limitations. Public schools are structurally designed for the typical learner and have limited capacity to accommodate children whose needs, interests, or abilities diverge significantly from the typical. Gifted programs exist in some districts but are thin in many; specialized programming for intense interests is rare; children who need significant curricular acceleration or deceleration often find public schools unable to provide it. This is a structural feature of operating at scale for heterogeneous populations, not necessarily a failing, but it can matter enormously for specific children.
• Teacher labor issues. Public schools are typically unionized and subject to collective bargaining agreements that, critics argue, make it difficult to reward excellent teachers, remove poor teachers, or restructure operations in response to evidence about what works. Defenders respond that teacher unions have protected teachers from arbitrary administrative action and that collective voice has produced better teaching conditions. The debate is empirically complicated.
• Social and behavioral concerns. Some families’ concerns about public schools reflect specific issues — bullying, discipline issues, peer group effects, exposure to materials or content they find objectionable. These concerns are sometimes overstated and sometimes understated depending on source, but they are real for specific families and schools.

What the evidence actually suggests

Research on public school performance does not produce a simple verdict. A few things seem well established:

• Public school quality varies enormously. Some public schools produce excellent outcomes; others produce poor outcomes; most are somewhere between. The variation within the public sector is larger than the variation between the public sector and any of the alternatives on average.
• Money helps but is not sufficient. Research on school spending (by Jackson, Johnson, Persico, and others) has found that higher per-pupil spending does tend to improve outcomes, particularly for low-income students. But the effects are moderate, and money spent on specific things (early childhood, teacher quality, extended learning time) tends to have larger effects than general spending increases.
• Family factors dominate. As the Coleman Report established in 1966 and subsequent research has confirmed, family background (parental education, family structure, home resources) explains more variation in student outcomes than schools do. This does not mean schools do not matter; it means that the swing between a good and bad school is smaller than the swing between a supportive and unsupportive family environment.
• Teacher quality matters enormously. Within any given school, the variation between effective and ineffective teachers is large. Schools that attract and retain effective teachers produce better outcomes. Teacher effects matter more than most school-structural factors.

What this means for families

For a family evaluating whether their local public school is right for their child, the generalizations about public schools as a category are less useful than specific evidence about the specific school. An excellent public school may be the best option available to a family, far better than mediocre private alternatives; a struggling public school may genuinely not serve a child well, and alternatives may be worth considering. The next chapter addresses how to evaluate a specific public school.

Both cases have real weight

The case for public schools rests on their civic mission, their genuine quality across much of the country, their infrastructure for special education and similar services, their stability, and their democratic accountability. The case against (or for reform) rests on the chronic underperformance of many specific schools and districts, monopoly effects in the absence of competition, cultural and curricular concerns for families with divergent values, structural limitations for non-typical learners, and specific institutional issues. What the evidence actually shows is that public schools vary enormously, that money helps but is not the dominant factor, that family background explains more variation than schools do, and that teacher quality matters substantially. For an individual family, the general question of whether public schools are good matters less than whether the specific public school their child would attend is good. The next chapter addresses how to figure that out.

What to read or watch next

• Diane Ravitch, The Death and Life of the Great American School System (revised ed., 2016). The strongest defense of public schools against what the author views as privatization drift.
• Eric Hanushek and Paul Peterson (eds.), Endangering Prosperity (2013). A reform-oriented analysis of American educational performance compared internationally.
• Jonathan Kozol, Savage Inequalities (1991) and The Shame of the Nation (2005). Classic accounts of inequality within the public school sector.
• C. Kirabo Jackson, Rucker Johnson, and Claudia Persico, “The Effects of School Spending on Educational and Economic Outcomes,” Quarterly Journal of Economics (2016). Influential recent research on the effects of school funding.
• National Assessment of Educational Progress (NAEP), nationsreportcard.gov. The authoritative federal data source on American K–12 academic outcomes, with results by state, district, and demographic group.

CHAPTER 6

Evaluating a Public School

For most families, the practical question is not whether public schools in general are good but whether the specific public school their child would attend is a reasonable fit. Because public school quality varies so much, evaluating the specific school is essential. This chapter lays out what to look for, where to get information, and how to interpret what you find. Much of this framework applies to evaluating any school, but the specifics here are tailored to the public sector.

Sources of information

Several sources provide useful information about specific public schools:

• State report cards and accountability data. Every state publishes report cards for its public schools, typically including standardized test scores, graduation rates, attendance, disciplinary data, teacher characteristics, and demographic breakdowns. These are authoritative and free. The state education department’s website is the starting point.
• National Assessment of Educational Progress (NAEP). Federal testing produces state-level data on fourth and eighth grade reading and math, and for large districts the “Trial Urban District Assessment” (TUDA) produces district-level data. Useful for understanding how a district compares nationally, though not at the individual-school level.
• GreatSchools.org and similar aggregators. Commercial sites that rate public schools based on test scores, growth measures, and equity indicators. Useful as starting points but should be understood as reflecting specific methodological choices; their ratings can shift meaningfully with methodology changes, and some have been criticized for over-weighting raw test scores relative to growth or equity metrics.
• School websites and district publications. Directly from the source. Provide information about programs, staff, schedules, extracurriculars, and handbooks. Treat with some skepticism (schools naturally present themselves favorably) but useful for specifics.
• School visits and observations. Most public schools permit visits, at least during school tours or open houses, and increasingly allow prospective parents to observe classes or meet with administrators. Nothing substitutes for seeing a school in operation.
• Current parents and recent graduates. Talking to families whose children currently attend, or who recently did, provides information data cannot: what daily life at the school is like, what the administration is like to work with, how specific kinds of students fare. Neighborhood networks, local Facebook groups, and mutual acquaintances can yield useful conversations.

What the test scores mean and don’t

Test scores are the most commonly cited school quality indicator, and they do contain useful information, but they should be read carefully:

• Raw test scores largely reflect who attends the school. Schools with high-income, college-educated parents tend to have high test scores; schools with lower-income students tend to have lower test scores. This is mostly about the students, not the school. A high-scoring school in an affluent suburb may not be doing a better job than a modestly-scoring school in a different context; it may simply be serving students who would score well in any school.
• Growth measures are more informative. Some state report cards and some rating systems include “growth” or “value-added” measures — how much students’ scores improved over a year relative to expectations based on prior scores. These attempt to isolate the school’s contribution and are typically more informative about school quality than raw scores. When available, prioritize growth measures.
• Look at demographic subgroups. A school’s average may hide significant variation by subgroup. Does the school serve English language learners well? Students with disabilities? Students from families with different income backgrounds? The subgroup data often tell the more important story about whether a school is working for students like yours.
• Proficiency rates can mislead. Reporting that “80 percent of students are proficient” tells you the share above a threshold but not how close those below are to it, or how far above the top students are. Averages and distributions give a fuller picture.

Non-academic factors

Test scores are only one dimension. Other factors that affect whether a school is a good fit:

• School leadership. The principal’s quality is consistently among the most important factors in school effectiveness. A school with a strong principal can often overcome other challenges; a school with a weak principal struggles even with favorable conditions. How long has the principal been in place? What is their background? How do they talk about the school when you meet them?
• Teacher stability. High teacher turnover is generally a negative signal. Schools where experienced teachers stay long-term tend to have more coherent cultures and more consistent instructional practices. Ask about average teacher tenure and turnover rates.
• Discipline climate. A school does not need to be draconian to be orderly, but students cannot learn in chaos. Observe during a school day if possible: What happens in hallways between classes? How do teachers handle transitions? Is the atmosphere calm and purposeful, or disorganized and loud?
• Specific programs that matter for your child. If your child is advanced in math, does the school have appropriate options? If your child has dyslexia, what reading instruction is used? If your child is interested in music, is there a program? Schools that look similar on aggregate measures can differ dramatically in specific programs that matter for specific kids.
• Parent engagement. Schools with engaged parents tend to perform better, and for many families the degree of parent engagement matters to their experience. What kinds of parent involvement does the school expect or welcome?
• Physical facilities. Modern, well-maintained facilities are not essential for good education, but crumbling facilities can signal larger problems with district management. Is the building clean and functional? Are there adequate resources — working technology, appropriate books, reasonable class sizes?

Questions worth asking

When visiting a school or speaking with administrators, useful questions include:

• How does the school serve students who are behind grade level, and how does it serve students who are ahead? The answer reveals whether the school has actual mechanisms or just aspirations.
• What happens for students who struggle behaviorally or emotionally? The answer reveals the school’s approach to discipline and student support.
• What is the school’s approach to homework and parental involvement? Reveals expectations and culture.
• How does communication with parents work? A school where parents hear from teachers regularly has different dynamics than one where communication is minimal.
• What have been the school’s biggest changes or challenges in the past few years? How administrators answer reveals both what has actually happened and how they think about it.

District-level considerations

The specific school matters, but so does the district it operates within. Strong districts tend to have strong schools; weak districts tend to struggle even in their better buildings. District-level considerations include:

• Superintendent stability. Frequent superintendent turnover (more than once every three or four years on average) is typically a sign of governance or political difficulty that affects schools.
• Financial health. Districts under financial stress often cannot sustain program quality. State report cards typically include financial data.
• Middle and high school pipeline. If you are choosing an elementary school, look ahead to where your child will continue. Great elementary schools feeding into struggling middle schools complicate long-term planning.
• Magnet and choice options within the district. Many districts offer internal choice options — magnet schools, specialty programs, gifted programs — that may be better fits than the neighborhood school. Chapter 12 addresses these.

Evaluate specific schools, not categories

Because public schools vary so much, the question for a family is whether the specific school their child would attend is a reasonable fit. Use state report cards, NAEP data, commercial aggregators (as starting points), direct school communications, and conversations with current parents. Look beyond raw test scores (which mostly reflect who attends) to growth measures, subgroup data, and non-academic factors — leadership, teacher stability, discipline climate, specific programs, parent engagement, and facilities. Ask substantive questions on school visits. Consider the district’s health and the pipeline of schools your child would attend over time. No school is perfect; the question is whether a given school is a good enough fit given the alternatives available. The next parts of this guide treat those alternatives.

What to read or watch next

• GreatSchools.org. Commercial school-rating aggregator with data on most American public schools; useful as a starting point if treated critically.
• U.S. News & World Report, Best High Schools Rankings (annual). High school rankings based on academic metrics; one of the most-cited commercial rankings.
• Your state’s department of education website. Authoritative school and district report cards, typically including test scores, demographics, and finances. Usually easily found by searching “[state name] school report card.”
• Nation’s Report Card, nationsreportcard.gov. NAEP data including, for large urban districts, TUDA district-level results comparing to peers nationally.
• Local parent groups and neighborhood networks. Conversations with current parents provide information that data cannot — the texture of daily life at a school — and are often the most useful supplement to official data.

PART THREE

Private Schools

Independent and religious schools, what the evidence does and does not show, and how to evaluate a specific one

CHAPTER 7

Private Schools: Independent, Religious, Specialized

Private schools educate roughly 4.7 million American children, about 8 to 9 percent of K–12 students. The category is diverse: it includes parish Catholic schools serving working-class neighborhoods, elite independent academies with $60,000 tuitions, evangelical Christian schools with explicit religious missions, Montessori schools following a specific pedagogical tradition, Jewish day schools, Islamic schools, Waldorf schools, military academies, and dozens of other types. Understanding what “private school” means requires understanding the variety within the category, and that variety is substantially larger than for any of the other main options.

Religious schools

Roughly 77 percent of private school students attend religiously affiliated schools. The composition:

• Catholic schools. Approximately 35 percent of all private school enrollment, historically the largest single category. Parish elementary schools, diocesan high schools, and schools run by specific religious orders (Jesuits, Dominicans, Christian Brothers) form the core of the sector. Catholic school enrollment has declined substantially from its 1965 peak — when roughly 5.2 million students attended Catholic schools — to about 1.6 million today, a decline driven by the costs of lay teaching staff (replacing religious sisters and brothers), demographic shifts, and competition from other options. Many Catholic schools remain strong institutions; others have closed or merged. Tuition at Catholic parish elementary schools averages $5,000–$8,000; high schools typically $12,000–$22,000.
• Evangelical and non-denominational Christian schools. Roughly 15–20 percent of private school enrollment, concentrated in the South and Midwest, with substantial growth since the 1970s. Range from small church-based academies to large regional schools. Vary substantially in curriculum, academic rigor, and religious emphasis. Some emphasize strong academics alongside religious formation; others are more focused on religious and cultural formation. Tuition typically $6,000–$15,000.
• Mainline Protestant schools. Episcopal, Lutheran, Presbyterian, Methodist, and other traditional Protestant schools. Often academically rigorous, some highly selective. Together with Catholic and evangelical schools, account for the large majority of religious private school enrollment.
• Jewish day schools. Small but significant sector, particularly in cities with large Jewish populations. Range from Orthodox yeshivot to Conservative and Reform day schools to pluralistic community schools. Academic programs generally strong; Jewish studies and Hebrew language instruction combined with secular academics. Tuition varies widely but typically substantial.
• Islamic schools. Growing sector serving American Muslim families, particularly in cities with large Muslim populations. Academic programs combined with Islamic studies and Arabic language instruction.
• Other religious traditions. Quaker schools (often known for progressive academic cultures and peace education), Seventh-day Adventist schools (nationwide network), and smaller sectors serving Hindu, Buddhist, and other communities.

Secular independent schools

The remaining roughly 23 percent of private school enrollment is at secular (non-religious) schools. These are particularly concentrated in the Northeast, major metropolitan areas, and certain Southern and Western cities. Types include:

• Traditional independent day schools. The classic “private school” of American imagination: an academically rigorous institution, often with a long history, serving a predominantly professional-class clientele. Tuition at established day schools in major cities typically ranges from $35,000 to $65,000 per year for upper grades. Many offer significant financial aid. Admission is selective and often involves interviews and testing.
• Boarding schools. A smaller but distinctive segment, mostly in the Northeast, with tuitions now commonly above $70,000 per year including room and board. Historically associated with New England and elite circles, though with substantially more diverse student bodies than in past generations. About 250 boarding schools in the country, enrolling perhaps 50,000–60,000 students total.
• Montessori schools. Schools following the pedagogical tradition developed by Maria Montessori in the early twentieth century, emphasizing self-directed learning in carefully prepared environments with multi-age classrooms. The American Montessori Society and the Association Montessori Internationale accredit schools. Montessori is most commonly available for preschool and elementary; fewer secondary programs exist. Tuition varies widely.
• Waldorf schools. Schools following the pedagogical tradition developed by Rudolf Steiner in the early twentieth century, emphasizing arts integration, delayed academic pressure in early years, and a particular developmental approach. About 130 Waldorf schools in North America; distinctive culture that fits some families well and others not at all.
• Progressive schools. Schools in the tradition of John Dewey and progressive education — emphasizing project-based learning, student voice, democratic classroom practices, and experiential education. Range from well-established institutions to newer experiments.
• Specialized schools. Schools focused on specific populations or purposes: schools for students with dyslexia (Landmark, Carroll, the Lab School), schools for gifted students, performing arts schools, sports-focused schools, and others serving specific niches.
• Military academies. A small but distinctive category, typically secondary-level, emphasizing structure, discipline, and leadership development.

How private schools are different from public schools, structurally

Private schools differ from public schools in several structural ways that affect what they can and cannot do:

• Admission control. Private schools can decline admission to any applicant for any reason, with narrow exceptions under civil rights law. They can require testing, interviews, essays, recommendations, and assessments of academic history. This allows them to shape their student bodies in ways public schools cannot. It also means they can exclude students who would struggle in their environment, including many with significant disabilities or behavioral difficulties.
• Curriculum autonomy. Private schools generally set their own curriculum, within minimal state requirements. They can teach religiously or secularly, include or exclude specific content, and design programs that would not be possible in public schools. This is both an advantage (for fit with family values and educational priorities) and a concern (for consistency and quality control).
• Teacher hiring flexibility. Private schools are not bound by state teacher certification requirements in most cases. They can hire subject-matter experts without pedagogical credentials, which sometimes produces stronger subject expertise at the cost of pedagogical training. Compensation, hiring, and dismissal are at the school’s discretion.
• Financial sustainability depends on enrollment. Private schools must attract enough tuition-paying students to cover their costs. This creates market discipline that public schools do not face, but it also makes schools vulnerable: enrollment declines can force closures, and financial pressure can affect program decisions. Some storied private schools have closed in the past decade; others remain robust.
• Accountability to customers, not voters. Private schools answer to families paying tuition, to their boards, and (for religious schools) to their sponsoring institutions. They do not answer to elected officials or the general public. This produces different incentives and different cultures.

Financial aid and accessibility

Published tuition is not what many families actually pay. Most established private schools offer financial aid — typically covering the difference between tuition and what the family can afford to pay based on a need analysis — to enrolled students who apply for it. The extent varies dramatically: some elite day and boarding schools have very large endowments and meet full demonstrated need, with some families paying nothing; many Catholic and smaller schools have limited aid resources. Families interested in private schools should inquire specifically about financial aid and not be deterred by sticker prices alone.

Public scholarship programs are also expanding access to private schooling. Vouchers, tax-credit scholarships, and education savings accounts (covered in Chapter 16) now provide meaningful subsidies in many states. In states with universal ESA programs, many private schools have adjusted tuition structures or enrollment practices to accommodate ESA-funded students. Catholic schools in particular have historically enrolled many students on tuition aid, scholarship funds, or sliding-scale tuition structures designed for working-class communities.

A category this diverse cannot be generalized

The private school sector includes Catholic parish schools at one end and $70,000 boarding schools at the other, religious schools of many traditions, secular independent schools, and specialized programs. Approximately 77 percent of private school students attend religiously affiliated schools; Catholic schools are the largest single category at about 35 percent of all private school enrollment. Structural differences from public schools include admission control, curriculum autonomy, hiring flexibility, dependence on enrollment for financial sustainability, and accountability to families and boards rather than to voters. Tuition ranges from modest (parish Catholic: $5,000–$8,000) to substantial (elite day and boarding: $35,000–$75,000+), but many schools offer financial aid. Generalizations about “private schools” as a category are rarely useful; the important questions are about specific schools in specific contexts.

What to read or watch next

• Anthony Bryk, Valerie Lee, and Peter Holland, Catholic Schools and the Common Good (1993). Landmark sociological study of Catholic schools, including why they appear to produce stronger outcomes for disadvantaged students than comparable public schools.
• Nicole Stelle Garnett, Lost Classroom, Lost Community: Catholic Schools’ Importance in Urban America (2014). Notre Dame law professor on the civic and educational contributions of Catholic schools.
• Nancy Koehn and National Association of Independent Schools data. NAIS (nais.org) is the trade association for independent schools and publishes annual data on the sector.
• American Montessori Society (amshq.org) and Association Montessori Internationale (amiusa.org). The two main accrediting bodies for Montessori programs in the United States; useful for identifying schools following the method authentically.

CHAPTER 8

The Case For and Against Private Schools

As with public schools, there are serious arguments on both sides of the private school question. This chapter lays out the case families are typically making when they choose private education, the case against, and what the research actually says. As before, the aim is honest presentation rather than advocacy.

The case for private schools

Families choose private schools for varied reasons, and different reasons dominate different choices:

• Academic quality and rigor. Many private schools, particularly selective independent schools and strong Catholic schools, have reputations for academic rigor that some families see as superior to local public options. Class sizes are often smaller; expectations for student work often higher; teachers often chosen for subject expertise. Whether any specific private school actually delivers on this depends on the specific school.
• Values and religious formation. For many families, particularly those choosing religious schools, the formation of a child within a specific religious and moral framework is a primary reason. A Catholic family may want their children educated in the Catholic tradition; an evangelical family may want biblical worldview integrated across subjects; a Jewish family may want Hebrew language and Jewish studies alongside secular academics. Public schools, by design, cannot provide religious formation; private schools can make it central.
• Character and culture. Private schools can cultivate distinctive cultures — academically serious, religiously devout, athletically competitive, artistically serious, or otherwise — in ways that public schools, with their broader missions, typically cannot. A family that wants a specific kind of environment for their child can often find it in a private school more readily than a public one.
• Peer environment. Critics of this motivation describe it as class-selection, and sometimes it is. But many families do prioritize the peer group their children will spend years with, and private schools can offer more control over peer composition. A family concerned about behavioral or academic norms at the local public school may see private school as a way to ensure their children grow up with different peer influences.
• Individual attention. Many private schools have lower student-teacher ratios than public schools and can provide more individualized instruction. Families with children who would benefit from closer attention — whether because of academic needs, social needs, or other factors — sometimes find private schools can provide this more readily.
• Escape from specific public school problems. For some families, the choice is about specific dissatisfaction with the available public option — its academic quality, its discipline climate, its curricular content, or other factors. The private school is not chosen in the abstract but specifically because the public option is not working.
• Specialized programs. Children with specific needs, interests, or learning styles sometimes find private schools better suited. Dyslexic students at schools specifically designed for them; gifted students at schools with appropriately advanced programs; arts-focused students at conservatory-style programs; students on the autism spectrum at programs designed for them. The variety of specialized private schools means specific needs can sometimes be met better than in a public system designed for general students.

The case against private schools

The arguments against private schooling, or for caution about it, also have weight:

• Cost and access. Private schools are expensive — often prohibitively so — and the systems of financial aid, while real, do not make private schooling accessible to most middle-class families without substantial sacrifice. The effective result is that private schools often function as institutions for families with substantial means, which has implications for both the schools and the society. For many families who could technically afford a private school’s tuition, doing so requires significant financial tradeoffs in other areas.
• Selection effects are often confused with school effects. Many private schools produce better outcomes than their local public alternatives, but much of this difference reflects who attends rather than what the school does. When studies control for family background, the differences often shrink substantially. A family choosing a private school should be clear-eyed about whether they are paying for the school’s effect or simply for the effect of being surrounded by peers from similar families.
• Limited services for certain students. Private schools can and do decline admission to students they do not wish to serve. Students with significant disabilities, severe behavioral difficulties, or complex learning issues are often not well served by private schools, which lack the infrastructure or, in some cases, the mission to serve them well. For these students, public school typically provides substantially more.
• Variable quality. The private sector is not uniformly high quality. Some private schools are genuinely excellent; others are mediocre or worse; still others have specific strengths and weaknesses that matter differently for different students. The reputation of “private school” can mask substantial variation, and families should be cautious about assuming quality simply because a school charges tuition.
• Civic and social concerns. Critics argue that when families who could support public schools withdraw to private alternatives, they deprive those schools of voice, advocacy, and the benefit of their children’s presence. This concern has particular force in communities where the flight to private schools has been associated with racial and economic segregation. Defenders respond that families have a first duty to their own children and that the civic argument, while real, cannot require families to sacrifice their children’s education for systemic concerns.
• Insularity. Some private schools, particularly specific religious and elite institutions, produce students with narrow peer groups that may not prepare them well for the broader society. The degree to which this is a concern depends on the specific school and on the family’s priorities.
• Lack of accountability mechanisms. Private schools are not subject to the public accountability measures of public schools. This can be a feature (they can experiment, maintain distinctive missions, avoid the pathologies of bureaucratic accountability) but also a limitation — they can underperform academically without the structured scrutiny public schools face, and specific problems (from academic mediocrity to abuse scandals) have sometimes gone undetected longer in private settings.

What the research shows

Research on private school outcomes has produced mixed and contested findings:

• The “Catholic school effect.” A body of research beginning with James Coleman’s 1982 work and continuing through Anthony Bryk, Valerie Lee, and Peter Holland’s 1993 book Catholic Schools and the Common Good has found that Catholic schools produce somewhat better outcomes than comparable public schools for disadvantaged students, controlling for family background. The proposed mechanisms include a coherent moral mission, a relatively narrow curriculum focused on core academics, and strong expectations for all students. More recent research has been more mixed; some studies find the effect has diminished, and some question whether selection effects were fully controlled. Still, the Catholic school effect is one of the better-documented findings in the private school research literature.
• Voucher program research. Studies of voucher programs — where families receive public subsidies to attend private schools — have produced mixed results. Early studies of Milwaukee and the D.C. Opportunity Scholarship Program found modest positive effects on some measures. More recent studies of newer programs (Louisiana Scholarship Program, Indiana Choice Scholarship Program) have found null to negative effects on test scores. It is not clear whether these different findings reflect different program designs, different student populations, different private schools participating, or methodological variations. The cumulative evidence does not clearly favor or oppose private school effects.
• Long-term outcomes. Some research suggests private school attendance is associated with higher college attendance and completion rates, controlling for observables. But observables rarely capture everything, and the family factors driving private school attendance also drive college attendance. Causal claims are harder to sustain than correlational ones.
• Non-academic outcomes. Research on civic, moral, and other non-academic outcomes by sector is sparse and often methodologically weak. Some findings suggest religious school students report stronger religious practice and some measures of civic participation as adults, but these findings are confounded by family background.

The honest synthesis

The research does not support strong claims that private schooling produces better academic outcomes than public schooling on average, once family background is controlled. It also does not refute such claims; the evidence is too mixed to settle the question. What the research more clearly supports: specific types of private schools (notably Catholic schools serving disadvantaged students) produce outcomes somewhat better than comparable public schools; the private sector is diverse, and generalizations are often misleading; for individual students in individual situations, private schools can be clearly better than alternatives, and for others clearly worse. Families should make the choice based on the specific school and specific child, not on generalizations about the sector.

The decision is specific, not categorical

Families choose private schools for academic quality, religious or values formation, specific cultures and peer environments, individual attention, dissatisfaction with public options, and specialized programs. The arguments against include cost and access, the frequent confusion of selection effects with school effects, limited services for certain students, variable quality within the sector, civic concerns, insularity, and limited public accountability. Research has produced mixed findings: Catholic schools appear to produce somewhat better outcomes for disadvantaged students controlling for background; voucher program studies have produced inconsistent results; long-term outcome research is weakened by selection effects. The honest conclusion is that private schools are not systematically better or worse than public schools on average, but specific private schools can be clearly better or worse than specific alternatives for specific children. The decision should be about the specific school and the specific child.

What to read or watch next

• Anthony Bryk, Valerie Lee, and Peter Holland, Catholic Schools and the Common Good (1993). The most influential scholarly treatment of the distinctive features of Catholic schools.
• James S. Coleman, Thomas Hoffer, and Sally Kilgore, High School Achievement (1982). Foundational empirical study comparing public and Catholic high schools that launched decades of subsequent research.
• Patrick Wolf and colleagues, evaluations of the D.C. Opportunity Scholarship Program. A series of Institute of Education Sciences reports on one of the most-studied voucher programs, with mixed findings over time.
• Mark Dynarski, Ann Nichols-Barrer, Philip Gleason, and others, voucher program evaluations. Mathematica and similar contractor evaluations of the Louisiana Scholarship Program and other state voucher initiatives.
• Sarah Theule Lubienski and Christopher Lubienski, The Public School Advantage (2013). Contrarian analysis arguing that, controlling for student background, public schools outperform private schools — contested but worth reading for balance.

CHAPTER 9

Evaluating a Private School

Because private schools vary even more than public schools, evaluating specific private schools carefully is essential. Families considering private education face a more asymmetric information environment than with public schools: there is less standardized reporting, less public comparison data, and more variation in what schools actually do versus what they claim. This chapter lays out how to evaluate a private school, how to see past marketing, and what questions to ask.

Accreditation and basic credibility

A first filter is accreditation. Established private schools are typically accredited by regional accreditors (such as the Middle States Association, New England Association of Schools and Colleges, Western Association of Schools and Colleges), national associations (National Association of Independent Schools for secular independent schools, National Catholic Educational Association for Catholic schools, Association of Christian Schools International for evangelical schools), or pedagogically specific bodies (American Montessori Society, Association Montessori Internationale, Association of Waldorf Schools of North America).

Accreditation is a baseline quality indicator; it confirms the school has been evaluated by external reviewers and meets certain standards. Not all reputable schools are accredited (some deliberately avoid accreditation for philosophical or practical reasons), but a school that is neither accredited nor part of a recognized tradition with alternative quality signals deserves extra scrutiny.

What a good school tour reveals

Most private schools welcome prospective families for tours, open houses, and sometimes classroom observations. These visits are a critical part of evaluation. What to look for:

• Classrooms in operation. If possible, observe actual classes rather than just sit in a closed meeting. What does instruction look like? Are students engaged? Is the work substantive? Are teachers skilled at what they are doing?
• Teacher quality and credentials. Private schools are not required to hire state-certified teachers and often do not. The question becomes whether the teachers they hire are actually good — by subject expertise, by teaching skill, by commitment to students. Ask about teacher backgrounds, how long teachers typically stay, and how the school supports teachers’ professional growth.
• Curriculum clarity. What is actually taught, at what grade, using what materials? A school with a coherent curriculum can answer this specifically; a school with an incoherent one typically cannot. Ask to see sample coursework, curriculum maps, or syllabi. This matters particularly for specialized schools (classical, Montessori, Waldorf, religious) where the curriculum is a central part of what the school offers.
• Student culture and behavior. How do students interact with each other and with adults? Is the environment orderly without being oppressive? Do students seem engaged with their education or going through the motions?
• Leadership quality. The head of school (or principal, or headmaster) matters enormously. How long have they been in the role? What is their vision? How do teachers and parents talk about their leadership?
• Facilities and resources. Are they appropriate for the school’s mission and price? Elite tuition does not require palatial facilities, but sustained neglect of maintenance often signals deeper issues.

Questions worth asking

Productive questions for private school administrators include:

• What kind of student thrives here, and what kind does not? A school that can answer honestly — rather than claiming to serve all students equally well — is displaying self-awareness. Most schools have particular strengths and limitations.
• How do you handle students who struggle academically or behaviorally? Reveals the school’s actual capabilities for support and whether it has infrastructure beyond rhetoric.
• What does your homework and workload look like by grade? A concrete answer reveals expectations and pace; a vague answer may suggest incoherent practices.
• What percentage of students who enroll in [grade X] are still enrolled at graduation? Attrition patterns reveal something about student and family satisfaction over time.
• How do you handle disagreements between parents and teachers? Reveals the culture around family involvement.
• What are your main challenges right now, and what are you working on? A school with no problems is a school with no self-knowledge. Honest answers signal seriousness.
• What do graduates typically go on to do? For high schools, college placement matters; but look at the range of outcomes, not just the top. What happens to the median student, not just the best ones?

Warning signs

Some patterns should trigger caution:

• High leadership turnover. If the head of school or many key leaders have changed in the past few years, ask why. Frequent turnover often signals governance or cultural difficulties.
• Financial distress. Schools in serious financial difficulty sometimes face reductions in programming, staff cuts, or even closure. Tax filings (Form 990 for nonprofit schools) are public; financial warning signs include deficit operations for several years, unusual reliance on large gifts, or board tensions about finances. Most schools are fine; serious distress is worth noticing.
• Enrollment declines. Schools that have been losing students for several consecutive years may face decisions that affect current students (program cuts, facility consolidations, tuition increases). Stable or growing enrollment is generally a better sign.
• Disciplinary or safety concerns in recent news. Independent news coverage of specific concerns (abuse allegations, financial malfeasance, significant academic or behavioral incidents) deserves attention, particularly if the school’s response seems inadequate.
• Limited transparency. Schools that resist sharing basic information (curriculum, teacher credentials, financial status, accreditation) deserve extra scrutiny. Legitimate privacy about specific matters is normal; stonewalling on basic questions is not.
• Extreme claims. Schools that claim dramatic superiority over alternatives — test-score outperformance, college placement, character formation — often overstate. Look at the actual data behind the claims.

Fit beyond quality

A school can be excellent and still not be right for a particular child. Fit considerations include:

• Religious alignment. For religious schools, does the family actually share the tradition and practice the school expects? Schools vary: some expect active religious participation, some are more pro forma about it. A mismatch between family religious life and school religious expectations can produce friction for students.
• Pedagogical fit. Does the school’s approach match how the child learns? A child who thrives in structured environments may struggle at a progressive school; a child who needs creative outlets may struggle at a highly traditional one; a child needing significant support may struggle at a school without appropriate infrastructure.
• Social fit. Is the student body one where the child will feel comfortable, or will they be the only student of their background? Children can thrive in many social environments, but being a significant outlier can be difficult, particularly in smaller schools.
• Family resources for the relationship. Private school tuition is one cost; the broader relationship involves time (parent meetings, volunteer expectations, events), money (fundraising expectations, additional fees), and social investment. Families should consider whether they can sustain the total relationship, not just the tuition.

Specific evaluation for a specific fit

Private schools vary enormously and require careful individual evaluation. Start with accreditation as a baseline filter, then visit for actual observations of classrooms in operation, teacher quality, curriculum coherence, student culture, and leadership. Ask substantive questions: what kind of student thrives, how do you handle struggling students, what is the workload, what is the attrition pattern, what are your current challenges. Watch for warning signs: leadership turnover, financial distress, enrollment declines, recent problematic news coverage, limited transparency, or extreme claims. Consider fit beyond quality — religious alignment, pedagogical match, social environment, and family capacity for the total relationship. A private school that would be wrong for your family does not become right by being good; a school that fits well can be genuinely transformative for a particular child.

What to read or watch next

• National Association of Independent Schools, nais.org. Searchable database of member schools with published data, plus resources for families on the evaluation process.
• National Catholic Educational Association, ncea.org. Resource for Catholic school information including accreditation status and performance metrics.
• Association of Christian Schools International, acsi.org. Umbrella organization for evangelical Christian schools, with accreditation information.
• Niche.com and Private School Review. Commercial aggregators for private school information; use as starting points and verify specifics independently.
• Form 990 filings (for nonprofit schools), available at GuideStar/Candid or ProPublica Nonprofit Explorer. Public financial disclosure that can reveal schools’ financial health.

PART FOUR

Charter and Magnet Schools

The experiment, what the research actually shows, and specialized public options within and beyond the charter sector

CHAPTER 10

Charter Schools: The Experiment

Charter schools are the most significant structural innovation in American public education in the past half century. They did not exist before 1992; today they enroll roughly 3.7 million students in 45 states and the District of Columbia. They are publicly funded, tuition-free, and open to all students — like traditional public schools — but they operate under different governance rules that give them more flexibility and, in principle, more accountability. They are also the most intensively researched sector of American K–12 education, because their lottery-based admissions have made rigorous study possible. This chapter describes what they are and how they work; Chapter 11 examines the research.

What a charter school is

A charter school is a public school operated under a charter — a contract with a state-authorized body (typically a state education agency, a university, or occasionally a local school district) that specifies what the school will do, how it will be held accountable, and on what terms it can be closed. The charter allows the school flexibility on many operational matters: it can set its own curriculum (within state frameworks), hire teachers without traditional public-school constraints, organize its schedule differently (longer school days, longer school years, different grade structures), and operate without being part of a local school district’s collective bargaining agreement.

In exchange for this flexibility, charter schools must meet the accountability terms of their charter. Most are evaluated on student academic performance against specified goals; schools that fail to meet these goals can have their charter revoked and be closed. This is a substantial accountability lever — unlike traditional public schools, which almost never close for academic reasons, charters that underperform can and do get closed, typically at a rate of 2 to 5 percent per year. Conversely, successful charters can often expand, replicate, or form networks.

The history

The charter idea originated in the late 1980s and early 1990s from several sources: teacher-union thinking about giving teachers more autonomy to innovate (Albert Shanker, then president of the American Federation of Teachers, was an early advocate); reformers looking for alternatives to the rigidities of traditional districts; and political coalitions spanning left and right that found common ground in the idea of public schools with more autonomy. Minnesota passed the first charter law in 1991, and the first charter school opened there in 1992. California followed in 1992, and other states gradually enacted their own charter laws.

Early charter schools were diverse experiments. Some were progressive, some traditional; some teacher-founded, some founded by community groups or educational entrepreneurs; some small, some envisioned as networks from the start. Growth accelerated in the 2000s with the rise of charter management organizations (CMOs) — networks like KIPP (founded 1994), Success Academy (founded 2006), Uncommon Schools (founded 1997), IDEA Public Schools (founded 2000), and Harmony (founded 1999) — that applied a more standardized approach to charter operation and expanded aggressively.

The federal government encouraged charter expansion under multiple administrations. The George W. Bush administration promoted charters as part of the No Child Left Behind framework; the Obama administration’s Race to the Top program explicitly incentivized states to lift caps on charter schools; various federal grant programs have supported charter school growth. State politics have produced substantial variation: some states (Arizona, Florida, Colorado, North Carolina) have extensive charter sectors; others (including most of the Northeast outside of New York and Massachusetts) have more limited sectors; and seven states still do not allow charter schools at all as of 2026.

Variety within the sector

Charter schools are structurally diverse in ways that resist generalization:

• Stand-alone schools. Single schools founded by individual teachers, parent groups, or community organizations, operating independently without affiliation with a larger network. Approximately 60 to 70 percent of charter schools operate this way.
• Networks (Charter Management Organizations, or CMOs). Nonprofit organizations that operate multiple charter schools under a common model and governance structure. Large CMOs include KIPP (roughly 250 schools), Uncommon Schools (roughly 55 schools in the Northeast), Success Academy (roughly 50 schools in New York), IDEA (roughly 140 schools in Texas and other states), Harmony (roughly 60 schools), and Rocketship (roughly 30 schools). CMO schools tend to be more standardized and often focus on college preparation for disadvantaged students.
• For-profit management organizations. A smaller category of charter schools managed by for-profit companies. Once a significant segment, particularly virtual charters; has declined in size and reputation amid regulatory changes and performance concerns. Some states have restricted or banned for-profit charter operators.
• Virtual charter schools. Online schools delivering instruction remotely, serving students who for various reasons cannot or do not want to attend brick-and-mortar schools. Performance has been consistently weaker than brick-and-mortar alternatives in multiple studies, and the sector has been the source of substantial policy concerns. A minority of the total charter sector but a high-profile segment.
• Pedagogical variety. Across charter schools of any structure, pedagogy varies dramatically. “No-excuses” charter models (Success Academy, KIPP’s earlier model, Uncommon Schools) emphasize highly structured environments, rigorous academics, and explicit behavioral norms. Progressive charter schools (High Tech High, Expeditionary Learning) emphasize project-based learning, student voice, and experiential approaches. Classical academies (Great Hearts, Hillsdale-affiliated schools) teach a traditional Western humanities curriculum. Montessori, Waldorf, and other pedagogically specific models have charter versions. STEM-focused charters, arts-focused charters, language immersion charters, and others all exist.

Admissions and the lottery system

Charter schools must be open to all students (within whatever geographic area they serve), without selective admission criteria like test scores or grades. When more students apply than the school has seats — which is common in good charters — admission is determined by random lottery. This is the structural feature that has made rigorous research possible: lottery winners and lottery losers are, in expectation, identical before the lottery, so later differences can be attributed to the school.

Lotteries vary in their specifics. Most give weight to factors like having a sibling already at the school, living in a specified geographic zone, or being a child of a school employee. Some include preferences for low-income or previously disadvantaged students. But after these weighted categories are filled, remaining seats are assigned randomly from the applicant pool. This structure is central to why charter school research is methodologically stronger than private school research: random assignment exists naturally.

Accountability and closure

Charter schools’ accountability differs from traditional publics. Their charter typically runs for a defined term (commonly 5 years, sometimes 3 or 10) and is renewable based on performance. Authorizers — the bodies that granted the charter — conduct reviews at renewal time and can deny renewal for academic, financial, or governance failures. In addition, many charters include annual performance benchmarks; serious failures can trigger non-renewal or revocation outside the normal cycle.

The result is that charter schools close at meaningful rates. A 2019 Stanford CREDO analysis found that approximately 16 percent of charter schools opened between 1997 and 2015 had closed or been closed by the end of the study period, with closures accelerating for schools that performed worst. This is far higher than the public school closure rate. For families, this has implications: a charter school is not a guaranteed long-term destination. Most do continue, but some close; families should think about contingency plans.

How charters fit into the broader picture

Charter schools occupy an unusual position in the American educational landscape. Technically public — tuition-free, open to all students, publicly funded, subject to civil rights laws — but operationally more similar to private schools in some ways (independent governance, curriculum freedom, fewer labor constraints). They are simultaneously celebrated by their supporters as a transformative innovation for urban education and criticized by their opponents as a drain on traditional public schools and a step toward privatization.

For individual families, the practical questions are simpler: Is there a good charter school available to my child? What does it actually do? Does it fit? Chapter 11 addresses what the research shows about charter outcomes; Chapter 12 addresses the related but distinct category of magnet schools and other specialized public options that families should consider alongside charters.

A structural innovation, not a pedagogical one

Charter schools are publicly funded, tuition-free, and open to all students, but operate under different governance rules than traditional public schools. Created by state laws beginning in 1991, they have grown to enroll roughly 3.7 million students across 45 states. The sector is diverse: stand-alone schools and networks (CMOs), nonprofit and for-profit operators, virtual and brick-and-mortar, with pedagogical approaches ranging from structured “no-excuses” models to progressive and classical and Montessori programs. Admission is by open lottery when oversubscribed, which has enabled rigorous research. Accountability comes through authorizer review, with meaningful closure rates (roughly 16 percent over 18 years in one major study). For families, the practical question is not about charters as a category but about specific charters available in their area and whether those specific schools fit their specific children.

What to read or watch next

• Richard Whitmire, The Founders: Inside the Revolution to Invent (and Reinvent) America’s Best Charter Schools (2016). Journalistic history of the charter movement’s rise, focused on the founders of major networks.
• National Alliance for Public Charter Schools, publiccharters.org. Industry association that publishes extensive data on charter enrollment, schools, and state laws.
• National Association of Charter School Authorizers, qualitycharters.org. Professional association for charter authorizers, with resources on oversight, accountability, and closure decisions.
• Ember Reichgott Junge, Zero Chance of Passage (2012). First-person account of the creation of the first charter school law in Minnesota in 1991, by the state senator who carried the legislation.

CHAPTER 11

The Research on Charter Outcomes

The charter sector has generated more rigorous research on its educational effects than perhaps any other part of American K–12 education. Lottery-based admissions have allowed researchers to compare students who won charter admission lotteries to students who lost them — groups that are, in expectation, identical before the lottery, so that later differences can be attributed to the schools. The resulting body of research is substantial. This chapter summarizes what it shows and what it does not.

The short answer

Charter schools, on average across the country, produce academic outcomes roughly similar to those of traditional public schools — with substantial variation above and below that average. Some charter schools produce dramatically better outcomes than their comparison group; others produce dramatically worse. The distribution of outcomes within the charter sector is wider than within the traditional public sector. That is, the best charters are better than the best traditional publics (by some measures) and the worst charters are worse than the worst traditional publics. The sector-wide average can be close to zero or modestly positive depending on which schools are included and how the comparison is done.

This headline finding has held, with some evolution, across repeated studies over fifteen years. The most comprehensive source is CREDO — the Center for Research on Education Outcomes at Stanford University — which has published three major national studies, in 2009, 2013, and 2023. The 2009 study found charter schools on average producing worse outcomes than traditional publics; the 2013 study found roughly neutral average outcomes; the 2023 study (As a Matter of Fact: The National Charter School Study III) found charter students outperforming traditional public students by about 16 extra days of learning in reading and 6 extra days in math per year, based on data from 31 states. The trend is of gradual average improvement, with substantial variation by state, sector (stand-alone vs. network), and student population.

What explains the variation

The average masks large underlying variation, and understanding the variation is more useful than citing the average. Several patterns have been reasonably well established:

• Urban charters outperform rural/suburban charters. Charter schools in dense urban areas, particularly those serving disadvantaged populations, have produced the largest positive effects in the research. Charter schools in rural and suburban areas have produced smaller or null effects on average. This is partly because the counterfactual differs: urban disadvantaged students’ alternative is often a struggling urban public school, while suburban students’ alternative is often a successful suburban public school.
• Network schools (CMOs) outperform stand-alone charters. On average, charter schools operated by established networks produce stronger outcomes than stand-alone charters. This may reflect network advantages in curriculum, operations, teacher training, and knowledge transfer; it may also reflect network selection effects. The CREDO studies have consistently found this pattern.
• “No-excuses” urban networks have produced the largest effects. Lottery studies of specific networks and schools — Boston charters (Joshua Angrist and MIT colleagues), New York charters (Sarah Cohodes and others), KIPP schools (Mathematica evaluations), Success Academy, Democracy Prep — have found effects of 0.2 to 0.4 standard deviations per year or more for some populations. These are large effects in educational research, translating to several months of additional learning per year. The specific models — longer school days and years, structured behavioral expectations, intensive teacher support, focus on academic basics — appear to contribute to these effects, though isolating which specific practices matter most has been difficult.
• Virtual charters underperform substantially. A 2015 CREDO study found virtual charter students performing dramatically worse than traditional public school students, losing the equivalent of 72 days of reading learning and 180 days of math learning per year. Subsequent research has largely confirmed this pattern. The virtual charter sector is a category where the evidence strongly points to worse outcomes on average.
• Demographic subgroup effects. CREDO and other studies have found charter schools producing larger positive effects for disadvantaged students — Black and Hispanic students, students in poverty, English language learners — than for white or affluent students. The 2023 CREDO study found these patterns persisted, with Black students gaining approximately 35 extra days of learning per year in reading at charters versus traditional publics, while white students saw smaller gains.

Specific standout findings

A few specific findings are worth highlighting because they are well-established and have particularly strong methodological foundations:

• Boston’s oversubscribed charter middle schools and high schools. Joshua Angrist and colleagues at MIT have conducted multiple lottery-based studies of Boston charter schools. Findings consistently show large positive effects — often 0.3 standard deviations per year in math and English language arts. These effects persist into later academic outcomes, including higher SAT scores and increased college enrollment.
• New York charter lottery studies. Research by Sarah Cohodes and others on New York City charters, particularly Success Academy and similar networks, has found comparable large effects. Students admitted by lottery to high-performing New York charters substantially outperform their lottery-loser peers.
• Post-Katrina New Orleans. The transformation of New Orleans schools into a predominantly charter-based system after Hurricane Katrina provided a natural experiment that Doug Harris and colleagues at Tulane have studied extensively. Findings generally indicate substantial improvement in student outcomes, though interpreting the results requires attention to other concurrent changes.
• KIPP schools. Multiple Mathematica Policy Research evaluations of KIPP have found substantial positive effects on student achievement, both in the early middle school years where KIPP has historically focused and in the elementary and high school grades as the network has expanded.

Critiques and caveats

The charter research literature has its critics, and legitimate methodological concerns deserve attention:

• Lottery studies apply only to oversubscribed schools. By design, lottery-based research studies schools where demand exceeds supply. These are disproportionately the better charter schools. Results from these studies may not generalize to charter schools that are not oversubscribed, which may have different effects — including, potentially, effects that are neutral or negative.
• Matched-pair studies have known limitations. The CREDO studies and other non-lottery research rely on matching techniques that control only for observable student characteristics. Unobserved differences (family engagement, student motivation, specific circumstances) may remain uncontrolled. Critics, including the National Education Policy Center at the University of Colorado Boulder, have argued that CREDO’s reported effects are small enough to be within the range of what methodological imperfections could produce.
• Attrition in charter schools. Charters have sometimes been criticized for higher student attrition than comparable traditional publics — meaning students who leave during the school year are replaced less frequently or not at all. Research on this has been mixed; some critics argue charters “counsel out” struggling students, while defenders argue attrition rates are comparable once demographics are controlled.
• Effects on neighboring traditional publics. Charter expansion can affect nearby traditional public schools through enrollment effects (loss of students and funding), peer effects (if higher-motivated families exit), and competitive effects (if charters push traditional publics to improve). Research on these spillover effects has been mixed, with some studies finding positive competitive effects on traditional publics and others finding negative enrollment effects.
• Long-term outcomes beyond test scores. Research has increasingly examined outcomes beyond test scores — college enrollment, college completion, civic outcomes. Several studies of the best urban charters (Angrist and colleagues on Boston; Cohodes and others on New York) have found positive effects on college enrollment, though effects on college completion and long-term earnings are still being studied.

What it means for families

The research provides useful guidance for families considering charters:

• Specific schools matter more than “charter” as a category. Since the variation within the charter sector is so large, a family should evaluate specific charters rather than drawing conclusions from sector-wide averages.
• Urban network charters with track records have the strongest evidence. If a family is choosing among urban charter options, schools in networks with documented track records of results — KIPP, Success Academy, Uncommon Schools, Democracy Prep, Achievement First, Aspire, IDEA — have stronger evidence than random local charters. This is not a guarantee for any specific school or child but is useful as a general indicator.
• Virtual charters deserve extra scrutiny. The research on virtual charter outcomes is clear and negative on average. Families considering virtual options should have specific reasons for doing so and should be clear-eyed about the evidence.
• Pedagogy should be evaluated directly. Some families thrive in “no-excuses” environments; others would find them oppressive. Some children fit progressive approaches; others need more structure. The data on which charter models work best on average should inform decisions but not dictate them; fit matters.

Average effects modest, variation large

Charter schools on average produce academic outcomes roughly similar to or modestly better than traditional public schools, with substantial variation. CREDO’s 2023 study found typical charter students gaining about 16 extra days of reading and 6 extra days of math learning per year compared with traditional public students, with larger gains for disadvantaged populations. The variation within the charter sector is wider than within the traditional public sector. Urban charters, network-operated charters, and specific “no-excuses” models have produced the largest documented effects in lottery studies (0.2–0.4 standard deviations per year for some Boston and New York schools). Virtual charters have consistently underperformed substantially. The research supports evaluating specific schools rather than the category; the variation is too large for sector-wide generalizations to be useful. For families choosing among urban charters, schools in networks with documented track records have stronger evidence than random local charters, though specific fit remains important.

What to read or watch next

• CREDO, As a Matter of Fact: The National Charter School Study III (2023) and earlier studies at credo.stanford.edu. The most comprehensive national research on charter outcomes; critical readers should also read methodological critiques.
• Joshua Angrist et al., Boston charter school lottery studies. A series of papers in economics journals examining specific Boston charter schools using lottery-based designs. Methodologically the gold standard.
• Mathematica Policy Research evaluations of KIPP. Independent contractor evaluations of the KIPP network at various points in its development.
• National Education Policy Center (nepc.colorado.edu). Academic center at the University of Colorado Boulder that publishes critical reviews of education research, including charter school claims. Useful counterweight to advocacy-oriented sources.
• Sarah Cohodes and related New York charter research. Columbia economist’s work on New York City charter schools using rigorous methods.

CHAPTER 12

Magnet Schools and Other Specialized Public Options

Within the public school sector, families in many districts have options beyond the neighborhood school. Magnet schools, specialized academies, gifted programs, exam schools, career and technical education (CTE) schools, and other programs exist within the traditional public sector, and they can be particularly strong options for specific students. They are often overlooked in debates about “public vs. private vs. charter” because they blur the category — they are public, but they are not typical public schools. This chapter describes the main types and how to access them.

Magnet schools

Magnet schools are public schools with specialized curricula or themes — STEM, performing arts, international studies, Montessori, dual-language immersion, and so on — that draw students from across a district rather than just the neighborhood. Most magnet schools are in urban districts and were originally created in the 1960s and 1970s as voluntary integration tools, allowing districts to draw diverse students by offering distinctive programs. There are approximately 3,000 magnet schools in the United States, enrolling over 2 million students. Some are highly selective; others use lotteries; a few use a combination.

Magnet schools can be excellent options for families in districts that have them. They offer specialized programming that most neighborhood schools do not, operate within the public sector (no tuition), and often have strong academic outcomes. The best magnet schools in major cities — schools like Thomas Jefferson High School in Fairfax County, Stuyvesant and Bronx Science in New York, Boston Latin School — are among the strongest high schools in the country by academic metrics, and admission is correspondingly competitive.

Access to magnet schools depends on the district. Some districts’ magnets use academic criteria (grades, test scores, or entrance exams); some use lotteries weighted by specific priorities; some use holistic applications. Families interested in magnet options should investigate their district’s specific processes well in advance of application deadlines, which are often in the fall of the year before enrollment.

Exam schools and selective academic programs

A subset of magnet schools, sometimes called “exam schools,” uses competitive academic admission — typically an entrance exam, GPA, and sometimes interviews or teacher recommendations. These schools are typically high schools serving academically advanced students. They have long histories in some cities: New York’s Stuyvesant dates to 1904; Boston Latin to 1635; Chicago’s Walter Payton to 2000. They consistently produce strong academic outcomes and send large shares of students to selective colleges.

Exam schools have been controversial in recent years because of demographic disparities in who gains admission. In several cities, admission policies have been modified or contested, particularly regarding how admissions decisions treat specific demographic groups. Policies continue to shift; families interested in these schools should understand current admissions practices and their likely evolution.

Gifted and talented programs

Many districts offer gifted and talented programs, either as pull-out programs within ordinary schools (students receive specialized instruction for part of the day) or as dedicated classrooms or schools. Gifted programs vary enormously in rigor, availability, and identification practices. Some districts have extensive gifted offerings through high school; others have minimal programs; still others have eliminated gifted programs entirely in the name of equity. For families with academically advanced children, investigating what gifted programming the district offers — and how students are identified — is essential.

Identification practices matter. Some districts use standardized tests alone (often the Naglieri, Wechsler, or OLSAT); some combine tests with teacher recommendations; some require parent referrals. Some tests have been criticized for demographic biases; various districts have modified approaches in response. If a child might qualify for gifted programming, understanding the identification process and ensuring the child has the opportunity to be assessed is worth attention.

Career and technical education (CTE) schools

Career and Technical Education schools, sometimes called career academies or vocational-technical schools, offer specialized programming in specific career fields: automotive technology, culinary arts, healthcare, information technology, agriculture, construction trades, and many others. They have undergone substantial revitalization in recent years; the old stereotype of vocational education as a dead end has been substantially replaced in many districts by modern programs that combine strong academics with career preparation, often including industry certifications, internships, and articulated credit toward two-year and four-year college.

For students with specific career interests or who prefer applied learning alongside academics, CTE schools can be excellent options. Some CTE schools are highly selective; others admit all comers. Outcomes research on modern CTE is encouraging: programs that combine rigorous academics with quality career preparation show positive effects on high school graduation, earnings, and (in some studies) college completion. The old narrative of vocational education as second-rate is not supported by evidence on modern well-designed CTE programs.

Dual language and immersion programs

Dual language and immersion programs teach academic content in two languages, typically English and Spanish, though Mandarin, French, German, and other languages are offered in various districts. Students become bilingual over the course of the program. Research on outcomes has been generally positive: students in well-implemented dual language programs often reach or exceed the academic achievement of students in English-only programs by later grades, with the added benefit of bilingualism.

These programs are particularly common in districts with large Spanish-speaking populations but have expanded substantially in districts motivated by the cognitive and economic value of bilingualism. Programs vary in their specific structures — some start with 90 percent of instruction in the non-English language and gradually shift; some use 50/50 models; some are language-immersion within otherwise standard schools. The variety means families should investigate the specific model used.

Open enrollment within districts

Many districts allow some form of intra-district choice — students can apply to attend public schools other than the neighborhood school, often subject to space availability and transportation considerations. This is different from the specialized programs above; it is the ability to attend a regular public school that is not the one assigned by address. Processes vary: some districts have formal lottery-based choice systems; others allow individual transfers when space permits. Families interested in a specific public school that is not their assigned school should investigate whether intra-district transfer is possible.

Inter-district open enrollment

Some states allow students to attend public schools in districts other than the one where they reside, either at all times or under specific conditions. Rules vary substantially: some states are fully open-enrollment (a student can attend any public district in the state with space), some allow it with restrictions, and some do not allow inter-district transfers at all. Where available, this can be a way to access a different district’s schools without changing residence. Transportation is usually the family’s responsibility, which limits practical use in some circumstances.

Why these options matter

For many families, specialized public options are the best combination of quality, fit, and cost. They offer programming often unavailable in neighborhood schools, at no tuition, within the public sector’s accountability and services infrastructure. They are particularly valuable for:

• Academically advanced students. Gifted programs, exam schools, and selective magnets provide the pace and depth that many bright students need but most neighborhood schools cannot provide.
• Students with specific interests. Performing arts magnets, STEM academies, international baccalaureate programs, and CTE schools serve students with focused passions better than general schools.
• Students whose neighborhood school is a poor fit. Magnet and open-enrollment options may provide access to better schools within the public sector before families consider private alternatives.

The main caution: these options require active choice and often early application. Families unaware of the options, or who do not apply in time, often end up in neighborhood schools that may not be the best fit. Districts vary substantially in how well they communicate these options; families should take initiative to investigate what is available rather than assume they will be informed.

Public options beyond the neighborhood school

Magnet schools, exam schools, gifted and talented programs, career and technical education (CTE) schools, dual language programs, and intra- and inter-district open enrollment provide specialized options within the public sector. These often combine the quality of private or charter options with public-sector accessibility (no tuition, stronger support services, broader accountability). Availability varies enormously by district; access often requires early application and navigation of specific admission processes. For academically advanced students, students with focused interests, and students whose neighborhood school is a poor fit, these options can be excellent and are often underutilized because families do not know about them. Families should investigate the full range of public options available before concluding the neighborhood school is their only public choice.

What to read or watch next

• Magnet Schools of America, magnet.edu. Organization representing magnet school programs, with directory information and national data.
• National Association for Gifted Children, nagc.org. Resources on gifted identification, programming, and state-by-state variation in gifted education policies.
• Association for Career and Technical Education, acteonline.org. Information on CTE programs, outcomes research, and specific state implementations.
• Chester Finn Jr. and Jessica Hockett, Exam Schools: Inside America’s Most Selective Public High Schools (2012). Survey of selective exam schools in the United States, their practices, and outcomes.
• Your local school district’s choice office. The single best source for understanding what specific options are available locally, application deadlines, and admission processes.

PART FIVE

Homeschooling

The fastest-growing sector, its variety and evidence, and the practical realities of teaching at home

CHAPTER 13

Homeschooling: Growth and Variety

Homeschooling is the option that has changed most in both size and character over the past decade. Before the COVID-19 pandemic, homeschooling was a distinct but relatively small sector — roughly 3 to 4 percent of American K–12 students — associated in public perception primarily with religiously motivated families, particularly conservative Christian families. The pandemic nearly doubled the homeschooling population, and the post-pandemic rate has settled at roughly 5 to 7 percent depending on how it is measured. The demographic composition has also shifted substantially. This chapter describes what homeschooling actually looks like in 2026, the varieties of approaches families use, and the legal and regulatory environment in which it operates.

How many, and who

NCES surveys from the 2022–2023 school year found that 5.2 percent of K–12 students were receiving academic instruction at home — a figure that includes both traditional homeschooling and some families using full-time virtual schools who describe themselves as homeschooling. The Johns Hopkins Homeschool Research Lab estimated a similar rate of 5.4 percent for the 2024–2025 year. State-by-state variation is substantial: Alaska, North Carolina, and several other states have homeschooling rates above 6 percent; some urban areas have much lower rates. The total number of homeschooled American children is somewhere between 2 and 3.5 million, depending on how “homeschool” is defined.

The demographic profile has shifted. While religiously motivated families, particularly conservative Christian families, remain the largest single group among homeschoolers, substantial growth has come from:

• Secular families. Families homeschooling for academic, pedagogical, lifestyle, or individual-fit reasons rather than primarily religious ones. Now a large fraction of the homeschool population.
• Black and Hispanic families. Traditionally underrepresented in homeschooling, these populations grew substantially during the pandemic and have continued to grow. Census Household Pulse data from the 2020–2021 period found homeschool rates among Black families reaching roughly 16 percent in some regions and periods, up from under 3 percent pre-pandemic. Reasons cited include dissatisfaction with local public schools, concerns about racial climate, and specific educational goals for their children.
• Academic enrichment families. Families homeschooling to pursue accelerated or distinctive academic programming for children, often combined with participation in outside classes, music lessons, athletics, and social activities.
• Families of children with specific needs. Families whose children have specific educational needs, health concerns, or social challenges that conventional schools do not address well. This includes children on the autism spectrum, children with chronic health conditions, gifted children, twice-exceptional children, and children who have been bullied.
• Families who simply prefer it. Pandemic-era experience of home-based education led some families to discover they liked the flexibility, the time with their children, and the ability to customize education. These families continued homeschooling by choice after conventional schools reopened.

Approaches to homeschooling

Unlike traditional schools, homeschooling encompasses a wide range of pedagogical approaches. Most homeschooling families do not adhere strictly to one model but combine elements that work for their children. The main approaches:

• Traditional / school-at-home. Using formal curricula and structured schedules that approximate a conventional classroom, often with textbooks and workbooks organized by grade. Accessible for families transitioning from conventional school; offers clear progression and accountability. Can feel inflexible for families wanting to take advantage of homeschooling’s adaptability.
• Classical education. Based on the trivium — grammar, logic, and rhetoric stages tied to developmental phases — drawing on classical texts and emphasizing language, history, and the liberal arts. Popularized in modern homeschooling by Susan Wise Bauer and Jessie Wise’s The Well-Trained Mind. Strong emphasis on reading, writing, and critical thinking.
• Charlotte Mason method. Based on the work of British educator Charlotte Mason, emphasizing “living books” (well-written literature) over textbooks, short focused lessons, nature study, habit formation, and narration (students retelling what they have read or heard). Popular across religious and secular communities.
• Unit study approach. Organizing learning around themes or topics rather than separate subjects, integrating history, science, language arts, and art around a common focus (e.g., a unit on ancient Egypt that covers history, geography, writing, art, and relevant science).
• Unschooling. The least structured approach, following the child’s interests and letting learning emerge from everyday life. Based on the work of John Holt and popularized by writers like Grace Llewellyn. Ranges from “relaxed homeschooling with child-led emphasis” to genuinely radical unschooling where children direct essentially all of their education. Results depend heavily on children’s own initiative and family support.
• Eclectic. The most common approach in practice: combining elements from various methods based on what works for the specific family and child. A family might use a traditional math curriculum, a Charlotte Mason approach to literature, unit studies for history, and unschooled science based on the child’s interests.
• Online and virtual programs. An increasingly large segment, ranging from fully online schools (like K12’s Stride programs) to specific online classes (Khan Academy, Outschool, Art of Problem Solving, CTY, Veritas Press, Wilson Hill Academy, and many others) used to supplement family-led instruction. Some families rely heavily on online programs; others use them selectively for specific subjects, particularly high school sciences and foreign languages.

The support ecosystem

Modern homeschooling in the United States is supported by an extensive ecosystem that did not exist in its current form 30 years ago. Elements include:

• Homeschool co-ops. Groups of families who meet regularly, often weekly, for shared classes, activities, or field trips. Co-ops range from informal groups of a few families meeting in homes to large organizations operating near-full-time programs several days per week. They address the common concern about socialization and can provide courses (particularly in subjects like lab sciences, art, or music) that individual families find difficult to teach alone.
• Hybrid / university-model schools. Programs that meet two or three days per week for structured classes, with parents providing instruction at home on the other days. Have proliferated since roughly 2000, offering a middle ground between conventional schooling and traditional homeschooling. Particularly popular in classical and Christian communities.
• Homeschool sports leagues and arts programs. Many communities have developed homeschool-specific athletics (basketball, soccer, volleyball, cross-country), orchestras, theater troupes, and debate teams. In some states, homeschooled students can also participate in public school extracurriculars.
• Microschools. A post-pandemic development: small schools of 5 to 15 students meeting in homes, rented spaces, or shared facilities, often serving families who want something between traditional school and home-only education. Has grown significantly with the expansion of ESA programs, which fund a wider range of educational arrangements.
• Curriculum publishers and vendors. Extensive for-profit and nonprofit offerings. Major publishers include Abeka, BJU Press, Sonlight, Timberdoodle, The Good and the Beautiful, Masterbooks, and many others for religious families; Oak Meadow, Build Your Library, Blossom and Root, and others for secular families; specialized publishers for classical education (Memoria Press, Classical Academic Press), Charlotte Mason (Simply Charlotte Mason, Ambleside Online), and many others.
• Conventions, conferences, and online communities. Annual homeschool conventions in most major states; extensive Facebook groups, Reddit communities, and forums organized by approach, religious affiliation, and specific curriculum choices.

Legal and regulatory environment

Homeschooling is legal in all 50 states, but state regulation varies enormously. The Home School Legal Defense Association (HSLDA) categorizes states by regulation level:

• Low regulation. Approximately 11 states — including Texas, Oklahoma, Missouri, Idaho, Alaska, and several others — require essentially nothing beyond the child not being enrolled in a public or private school. No notice of intent, no curriculum requirements, no testing, no record-keeping obligations to the state. Families can homeschool essentially as they see fit.
• Moderate regulation. Roughly 20 states require some combination of: a notice of intent to homeschool filed with the state or district; evidence of instruction in specific subjects; parent qualifications (typically a high school diploma); periodic assessments or portfolio reviews.
• Higher regulation. Approximately 5 states — most notably New York, Pennsylvania, Massachusetts, Rhode Island, and Vermont — have substantial requirements including quarterly reports, annual assessments or portfolio reviews by certified evaluators, specified curriculum requirements, and sometimes ongoing oversight by local school districts.

Families considering homeschooling should first understand their state’s specific requirements. HSLDA’s website (hslda.org) provides current state-by-state summaries; the Coalition for Responsible Home Education (responsiblehomeschool.org) provides an alternative perspective. Local homeschool support groups are usually the best source for specifics on how state requirements work in practice in a given community.

Homeschooling is diverse and has changed substantially

Homeschooling in 2026 is a large, diverse, and changing sector, representing 5 to 7 percent of American K–12 students. The demographic has broadened substantially beyond its historical base of religiously conservative families to include secular families, growing numbers of Black and Hispanic families, academic enrichment families, and families of children with specific needs. Pedagogical approaches range from structured school-at-home to classical education to Charlotte Mason to unit studies to unschooling, with most families mixing approaches. An extensive support ecosystem — co-ops, hybrid schools, sports leagues, curriculum publishers, online programs, microschools — has developed to support the growing sector. Regulation varies enormously by state, from essentially none in roughly 11 states to substantial oversight in 5 others. Families considering this option should begin by understanding state requirements and connecting with local homeschool networks, which are usually more useful than any online research for understanding what homeschooling looks like in their specific community.

What to read or watch next

• Susan Wise Bauer and Jessie Wise, The Well-Trained Mind (4th ed., 2016). The most influential guide to classical education in modern homeschooling; covers all grade levels with practical recommendations.
• Robert Kunzman and Milton Gaither, Homeschooling: A Comprehensive Survey of the Research (2013) and subsequent updates. Academic review of research on homeschooling from scholars at Indiana University.
• Home School Legal Defense Association, hslda.org. Advocacy and legal resource with current state-by-state regulatory summaries. Conservative Christian orientation; useful for legal information while recognizing its perspective.
• Coalition for Responsible Home Education, responsiblehomeschool.org. Alternative perspective focused on child welfare and reasonable oversight; useful counterweight to advocacy-oriented resources.
• Pamela Price, How to Work and Homeschool, and various modern practical guides. Range of specific approaches and practical how-to books now widely available.

CHAPTER 14

The Case For and Against Homeschooling

Homeschooling inspires particularly strong opinions on both sides. Advocates describe it as producing dramatic academic and character outcomes and as the antidote to a failing system. Critics describe it as risky, insular, and inadequately supervised. As with the other models, the truth is more complex than either pole suggests, and the evidence base is weaker than either side typically acknowledges. This chapter lays out the main arguments and what can and cannot be concluded from research.

The case for homeschooling

Families choose homeschooling for a range of reasons, and the arguments for it have several threads:

• Individualized pacing and content. A homeschool curriculum can be calibrated to the individual child — moving faster in subjects where the child is advanced, slower where the child is struggling. Few conventional schools can provide this degree of individualization; a homeschool can do so as a matter of course. For children whose learning does not match the typical pace — either substantially advanced or significantly behind — this individualization can be transformative.
• Time efficiency. Conventional schools spend substantial time on transitions, group management, and activities that would not be needed for a single child or small group. Most homeschooling families find that 2 to 4 hours per day of focused instruction, depending on age, covers what a full school day would provide — leaving the rest of the day for other activities, independent learning, or free time. This efficiency is a structural feature, not a pedagogical virtue; it reflects the nature of one-on-one or small-group instruction versus classroom instruction of 20–30 students.
• Family relationships and values. Homeschooling allows parents to spend far more time with their children during the formative years. For families who value this — and for religious and philosophical communities where passing on specific traditions is central — homeschooling provides a setting in which family relationships and values are not competing with substantial time in institutions outside the home. Critics may frame this as insularity; supporters frame it as prioritizing the family’s primary role in child development.
• Curriculum control. Parents choose what is taught and how. For families with specific religious, philosophical, or pedagogical commitments — and for families who want to emphasize specific subjects, languages, or traditions not emphasized in conventional schooling — this is a decisive advantage.
• Escape from specific school problems. For children who have had difficulties in conventional schools — bullying, behavioral issues that did not serve them well in structured environments, academic mismatch, social anxiety, difficulties with specific schools’ cultures — homeschooling can offer a fresh start outside the environment that produced the difficulties.
• Time for children’s passions. Children with deep interests — in music, in athletics, in a specific academic area, in a trade — can devote substantially more time to those pursuits in a homeschool setting. Many accomplished young musicians, athletes, and scholars have homeschool backgrounds that allowed for the hours of practice their pursuits required.
• Flexibility. Homeschooling accommodates family situations that conventional schooling does not — non-traditional work schedules, families who travel for work or personal reasons, families with specific medical needs, military families who move frequently. The ability to continue education through disruptions is a significant practical advantage.

The case against homeschooling

The concerns about homeschooling also have substance:

• Quality variation. Homeschool quality depends heavily on the parents — their own education, their available time, their commitment, and the resources they bring to the task. A well-educated, committed parent with time and resources can provide excellent education; a less-prepared or less-engaged parent may provide substantially less than a conventional school. The variation is enormous and, unlike variation in conventional schools, is not systematically measured or addressed.
• Parent qualifications. Most states do not require homeschooling parents to have specific qualifications. Parents can teach subjects they do not know well, particularly at middle and high school levels where content expertise matters. Workarounds exist — online classes, co-ops, tutors — but not all families use them consistently.
• Socialization concerns. Homeschooled children spend substantially more time with their families and less time in large peer groups. Research on homeschool socialization has produced generally favorable findings on measures like self-concept and adult social functioning, but the studies are methodologically weak (discussed below), and the concerns are not entirely unfounded. Certain kinds of social experiences — navigating peer dynamics, working with people chosen by others rather than self, dealing with institutional structures — are harder to replicate in homeschool settings. Co-ops, activities, and community involvement can address this partly but not always fully.
• Limited exposure to diverse perspectives. A child educated primarily at home by parents is, in part, educated in a curated version of the world reflecting the parents’ choices about what to emphasize. For some families, this is the point; for others, it is an unintended side effect. Public schools, whatever their imperfections, bring children into contact with peers and adults of varied backgrounds and perspectives. Homeschool co-ops, extracurriculars, and communities can provide some of this, but the degree varies.
• Transitions to conventional settings. Homeschooled students who later attend conventional high schools or colleges face transitions that can be difficult. Most homeschooled students handle these transitions well, but some — particularly those from more insular homeschool backgrounds — find the adjustment significant. Research on college performance of homeschooled students has been mixed; some studies find them performing on par with or better than conventionally schooled peers, but these studies typically have selection biases (they are based on homeschooled students who chose to attend college).
• Oversight and child welfare concerns. In states with minimal regulation, homeschooled children can be largely invisible to any system outside their family. This creates risks for the small minority of children in families where serious neglect or abuse occurs. The Coalition for Responsible Home Education has documented cases where lack of oversight allowed severe maltreatment to go undetected. Homeschool advocates respond that most states’ child welfare systems handle this appropriately when concerns arise, and that broad regulatory regimes impose costs on the many responsible homeschoolers to address the behavior of a few. The debate over appropriate oversight is ongoing.
• Sacrifice required of the educating parent. The substantial time and energy demands of homeschooling fall disproportionately on one parent, usually the mother. The opportunity cost — in income, career development, and personal autonomy — is real, and the equity implications of the arrangement deserve honest reflection. Families considering homeschooling should be clear-eyed about what it will mean for the adult whose time it will consume.

What the research can and cannot tell us

Research on homeschool outcomes is the weakest body of evidence on any of the four sectors, for reasons discussed in Chapter 3. The specific issues:

• Self-selection of samples. Most homeschool achievement studies rely on voluntary samples of families who choose to participate. These families are not representative of homeschoolers generally; they are disproportionately committed, organized, and confident in their children’s performance. Studies based on these samples substantially overstate average outcomes. Brian Ray and the National Home Education Research Institute (NHERI) have produced the most widely cited research of this kind; the findings typically show homeschooled students substantially above national averages on standardized tests. Critics point out that the samples are not representative and that the studies do not control for family background factors that would produce similar outcomes regardless of schooling setting.
• Absence of comparison groups. Most research on homeschooling does not include rigorous comparison groups of conventionally schooled students matched on family background. Without this, attributing outcomes to the schooling setting rather than to the family doing the schooling is not possible.
• Academic peer-reviewed research is sparse. Researchers outside the advocacy community who have studied homeschooling — Milton Gaither, Robert Kunzman, Joseph Murphy — generally conclude that strong claims about homeschool academic outcomes cannot be sustained with current evidence. The existing peer-reviewed research is too methodologically weak to support either “homeschoolers outperform” or “homeschoolers underperform” as an average claim.
• What we can say. Homeschool families tend to have higher parental education than average, which correlates with good educational outcomes in any setting. Well-supported homeschoolers — children of engaged parents with time and resources — tend to do well. Poorly-supported homeschoolers — children of parents who are not able to provide adequate instruction — may or may not do well, and we have limited data on this because these families are less likely to participate in voluntary research. The distribution of homeschool outcomes is almost certainly wide, with very strong outcomes in some families and weaker outcomes in others.

The honest conclusion

Homeschooling can be an excellent educational choice for children of families with the resources — time, education, and commitment — to do it well. It can also be inadequate when any of those resources is missing. The research does not support sweeping claims in either direction. Families considering homeschooling should evaluate their own specific capacity to carry it out, not rely on optimistic averages drawn from voluntary samples or on alarmist warnings drawn from specific cases of poor outcomes. The next chapter addresses the practical realities more directly.

Evidence is thinner than advocates on either side suggest

The case for homeschooling includes individualized pacing, time efficiency, family relationships and values, curriculum control, escape from specific school problems, time for children’s passions, and flexibility. The case against includes quality variation depending on the parents, limited parent qualifications, socialization concerns, limited exposure to diverse perspectives, transition challenges, oversight and child welfare concerns, and the substantial sacrifice required of the educating parent. Research on homeschool outcomes is methodologically weak: most studies use self-selected voluntary samples without appropriate comparison groups. The best-supported conclusion is that outcomes vary enormously based on family resources and commitment — well-supported homeschoolers tend to do well, while poorly-supported ones may not. Strong claims in either direction overstate what current evidence supports. Families considering homeschooling should evaluate their own specific capacity honestly rather than rely on category-wide claims.

What to read or watch next

• Milton Gaither, Homeschool: An American History (2nd ed., 2017). Historian’s balanced account of homeschooling’s development, with careful treatment of what research does and does not show.
• Robert Kunzman, Write These Laws on Your Children: Inside the World of Conservative Christian Homeschooling (2009). Ethnographic study of conservative Christian homeschooling from an Indiana University researcher.
• Joseph Murphy, Homeschooling in America: Capturing and Assessing the Movement (2012). Vanderbilt scholar’s careful survey of research, with candid discussion of methodological limits.
• Brian D. Ray and the National Home Education Research Institute, nheri.org. The most prolific producer of homeschool achievement research; results should be read alongside methodological critiques because samples are typically self-selected.
• Rachel Coleman and Coalition for Responsible Home Education, responsiblehomeschool.org. Advocacy for reasonable homeschool oversight, with careful documentation of cases where absence of oversight produced harm.

CHAPTER 15

Practical Realities: Time, Money, Socialization

Whether homeschooling will work for a family depends less on the family’s commitment to the idea than on whether the practical arrangements will hold together day by day, year by year. This chapter addresses the practical realities honestly: how much time it takes, how much it costs, how socialization actually works, and what happens at the transition points.

The time commitment

The time demand is the first practical reality, and it is larger than most families realize before starting. Homeschooling requires someone — almost always a parent, typically the mother — to be substantially available during school hours. For elementary children, this often means active teaching of 2 to 4 hours per day, plus planning and preparation, plus supervision of independent work, plus managing co-op commitments, extracurriculars, and household logistics that no longer have the buffer of school hours. The total time commitment typically is more than a part-time job and often approaches a full-time job.

For middle and high school students, direct teaching time often decreases as children take on more independent work and outside classes, but planning, coordination, and oversight remain significant. Families with multiple children of different ages face compounding complexity: different curricula, different schedules, different needs. Many experienced homeschoolers describe the first three to six months as particularly demanding as families find their routines; some describe the experience as easier over time, while others find the cumulative demand grows with children’s advancing levels.

The practical implication is that homeschooling typically requires one parent not to be working full-time outside the home, or to have highly flexible work. Some families make homeschooling work with two full-time working parents — through careful scheduling, cooperative arrangements with other families, heavy use of online classes, or older children taking more independent responsibility — but this is genuinely difficult. Families considering homeschooling should be honest about whether the time is actually available, rather than assuming it will work out.

The financial reality

Direct costs of homeschooling vary enormously based on approach. Ranges for a single child:

• Minimalist approach. Using library resources, free online curricula (Khan Academy, Easy Peasy, various open educational resources), and basic supplies. Annual cost: $200–$800.
• Standard approach. Purchasing a main curriculum package, selected workbooks, and basic supplies. Annual cost: $500–$1,500 per child.
• Enhanced approach. Full curriculum packages, co-op fees, online classes for certain subjects, and modest extracurriculars. Annual cost: $1,500–$4,000 per child.
• Substantial approach. Hybrid school tuition or extensive co-op participation, multiple online classes, outside tutors for subjects like advanced mathematics or foreign languages, significant music or sports instruction. Annual cost: $5,000–$15,000 per child.

For families with multiple children, costs often scale less than linearly because curricula can be passed down and some resources are shared. ESA programs (Chapter 16) have substantially changed the cost calculus in states that have them: a family receiving $6,000–$10,000 per child in ESA funds can access levels of resources (tutors, online classes, specialized curricula) that would have been prohibitive before.

The larger cost for most homeschool families is the opportunity cost of the educating parent’s time. A parent who would otherwise be working at an income of $50,000 per year is effectively spending $50,000 per year on homeschooling even if direct costs are minimal. This does not mean homeschooling is not worthwhile — many families judge it a good investment of that time — but families should recognize the cost clearly rather than assume homeschooling is “free” because tuition is not being paid.

Socialization in practice

“What about socialization?” is the most common question homeschoolers are asked. The short answer is that socialization happens through structured activities rather than through incidental school-day interaction. Common elements of a homeschool child’s social life:

• Co-op or hybrid school attendance. Regular time with other children in a structured setting. One to three days per week is typical.
• Extracurricular activities. Sports teams (either homeschool-specific leagues or community leagues), music lessons and ensembles, art classes, scouts, 4-H, church youth groups, and community programs. Homeschool children often have more time for extracurriculars than conventionally schooled children because their school day is shorter.
• Neighborhood play and friendships. Common particularly for younger children.
• Multi-generational social networks. Homeschool children often have more substantial interaction with adults of various ages than conventionally schooled children — teachers at co-ops and classes, tutors, community members, fellow homeschool parents. This is different from — not necessarily better or worse than — the heavy same-age peer socialization of conventional schools.
• Specific social learning. Handling peer relationships, working with people not chosen by oneself, navigating institutional social structures — these can be more limited in homeschool settings. Intentional cultivation of these experiences through activities, jobs, and community involvement is important.

The research on homeschool socialization outcomes is, as Chapter 14 noted, methodologically weak but generally favorable. Most homeschooled students appear to navigate adult social demands reasonably well. The concerns are real but often overstated; many homeschoolers are genuinely well-socialized, and some are more so than typical conventionally schooled peers. The key for families is active, intentional cultivation of social experiences, not assumption that socialization will happen automatically.

High school considerations

Homeschooling through high school involves specific considerations that do not apply in earlier grades:

• Credit tracking and transcripts. Parents must document coursework in ways colleges and employers can evaluate. This includes course titles, descriptions, grades, and credit hours. HSLDA and various online resources provide templates; many families use platforms designed for this purpose (HomeSchoolTracker, Transcript Pro, or various free alternatives).
• Standardized testing. Homeschooled students need SAT or ACT scores for most colleges. PSAT should be taken in 11th grade for National Merit consideration. AP exams can provide external validation of coursework. CLEP and AP credits can reduce college costs substantially for students who can pass them.
• Dual enrollment. Taking college classes during high school at a local community college or four-year institution. Particularly valuable in states where dual enrollment is subsidized. Provides external academic validation and college-level coursework for subjects that might be difficult to teach at home.
• College admissions. Most American colleges are experienced with homeschool applicants and have developed reasonable processes for evaluating them. Some highly selective institutions place additional emphasis on standardized test scores and external coursework (AP, dual enrollment) when evaluating homeschool applicants. The widely cited concern that homeschooled students are disadvantaged in admissions is, at most colleges, overstated; some institutions actively recruit homeschool applicants.
• Social and practical preparation for college. The transition from homeschool to residential college can involve adjustments that conventionally schooled students have already made — dealing with institutional structures, living with peers, navigating large social groups. Most homeschool students adjust successfully; some find the transition significant.

Who should consider it, and who probably should not

Based on practical considerations, homeschooling is more likely to work for families where:

• One parent can dedicate substantial time to it. Either not working outside the home or working on highly flexible arrangements that accommodate homeschooling’s demands.
• That parent has reasonable academic preparation. Not necessarily advanced degrees, but enough education and intellectual engagement to teach the elementary years confidently and find help for what they cannot teach directly in later years.
• The family has stable housing and enough financial stability. Homeschooling adds demands that families in acute financial or housing stress typically cannot absorb.
• The family has access to a homeschool community. Co-ops, activities, and other families doing the same thing make homeschooling substantially more sustainable. Isolated homeschoolers without community often struggle.
• Children’s temperaments fit home-based learning. Not every child thrives at home with a parent as primary educator. Some children need the peer environment and external structure of conventional schooling; some need the structured adult authority of teachers who are not their parents.

Families without these conditions may still make homeschooling work, and families with them may still find it does not fit. But the baseline of time, resources, community, and child fit matters. Starting homeschooling without any of these is more likely to produce difficulty than success, for both child and family.

Practical sustainability matters more than philosophical commitment

Homeschooling’s time commitment typically requires one parent available through school hours, often amounting to more than a part-time job. Direct costs range from $200–$800 (minimalist) to $5,000–$15,000 (substantial) per child per year, with ESA programs changing the calculation in states that have them. The larger cost is usually the opportunity cost of the educating parent’s forgone income. Socialization happens through deliberate activities — co-ops, extracurriculars, community involvement — rather than through incidental school-day interaction, and most homeschool children adjust well if families cultivate these connections intentionally. High school requires specific preparation: credit tracking, standardized tests, dual enrollment, and intentional college transition planning. Homeschooling is more likely to succeed for families with substantial time, reasonable academic capacity, financial stability, community access, and children whose temperaments fit. Without those foundations, families starting homeschooling face difficulty regardless of commitment.

What to read or watch next

• Susan Wise Bauer, The Well-Trained Mind. Practical chapters on scheduling, grade-by-grade expectations, and navigating high school and college admissions.
• Cathy Duffy, 101 Top Picks for Homeschool Curriculum. Long-running practical review of homeschool curricula across approaches and subjects.
• Debra Bell, The Ultimate Guide to Homeschooling (latest ed.). Comprehensive how-to resource covering practical aspects of daily homeschool life.
• Lee Binz, The Homescholar publications and website. Specialized guidance on homeschool high school, transcripts, and college applications.
• State and local homeschool associations. Typically the best sources for state-specific practical information, including co-op directories, field trip groups, high school support networks, and subject-specific classes available in your area.

PART SIX

Deciding

The policy landscape, students who do not fit the typical mold, what actually matters for outcomes, and a framework for making the decision

CHAPTER 16

Vouchers, ESAs, and the Public Policy Debate

The structure of American school choice has changed faster in the first half of the 2020s than in any equivalent period in the last half century. Universal education savings accounts (ESAs) did not exist anywhere before 2021; they now operate in roughly 13 states with programs open to all or nearly all students. Voucher and tax-credit scholarship programs have expanded in most states that had them and have been added in several that did not. This chapter describes what these programs are, the arguments for and against them, and what evidence exists on their effects.

The four types of programs

Private-school choice programs in the United States take four main forms, sometimes combined within a single state:

• Vouchers. Direct state payments for private school tuition, either paid to the school or to the family with the restriction that it be used for tuition. The oldest form of modern private school choice; Milwaukee’s program (started 1990) was the first major modern voucher program, and states including Ohio, Indiana, Wisconsin, Louisiana, and others have operated voucher programs for decades. As of early 2026, approximately 20 states operate voucher programs.
• Tax-credit scholarships. States provide tax credits to individuals or businesses who donate to scholarship-granting organizations, which then fund private school tuition for eligible students. The mechanism channels the subsidy through private intermediaries rather than as direct state payments. Approximately 25 states have tax-credit scholarship programs.
• Education Savings Accounts (ESAs). States deposit a portion of per-pupil funding into accounts that families can use for a broader range of educational expenses: private school tuition, tutoring, curriculum materials, therapy for special-needs students, educational technology, and so on. More flexible than vouchers and the fastest-growing category. Approximately 18 states have ESA programs as of late 2025, with rapid growth.
• Tax deductions and credits for education expenses. Several states allow parents to deduct or claim tax credits for private school tuition, homeschool expenses, or both. Typically smaller benefits than vouchers or ESAs and more useful for families with tax liability to offset.

The rapid expansion of universal programs

Until 2021, all private-school choice programs in the United States had eligibility limits — typically based on income, district performance, disability status, or similar criteria. The idea that any family could qualify regardless of income was largely unimplemented. West Virginia’s 2021 Hope Scholarship broke this pattern by creating an ESA available to essentially all students. Arizona expanded its existing ESA to universal eligibility in 2022. Between 2021 and 2025, approximately 13 states enacted universal or near-universal ESA or voucher programs. States with new universal or near-universal programs include Arizona, Arkansas, Florida, Indiana, Iowa, Louisiana, Montana, New Hampshire, North Carolina, Ohio, Oklahoma, Tennessee, Texas, Utah, West Virginia, and Wyoming.

Award amounts and specific rules vary. National average ESA awards are approximately $7,500 per year as of 2025. Texas’s 2025 Universal ESA provides $10,000 per student. Tennessee’s 2025 Education Freedom Scholarships provide $7,000. Florida’s long-established programs provide roughly $8,000. Some states offer higher amounts for students with disabilities (North Carolina offers up to $17,000 for students with autism or severe disabilities). Income limits, if any, vary substantially: programs range from fully universal (all students eligible) to priority-based (students from lower-income families or specific categories receive priority but do not exclude others) to restricted (limited to families below specified income thresholds).

The case for school choice programs

Advocates of expansive choice programs make several connected arguments:

• Parental rights and fit. Parents know their children better than state institutions do and should be able to direct their education. School-choice programs give parents the practical ability to choose schools that fit their children’s needs, values, and circumstances, rather than being limited by what they can afford or where they can live.
• Equity of access. Historically, choice among schools has been available primarily to wealthy families (who could pay private tuition) and families who could afford to move to districts with better public schools. Choice programs extend these options to lower-income families, addressing an equity problem that property-tax-funded public schools do not solve.
• Accountability through competition. When schools must attract students rather than receive them by geographic assignment, they face pressure to improve. This pressure, advocates argue, benefits both the schools families choose and the ones they leave, producing system-wide improvement.
• Pluralism and value diversity. A public school system that tries to serve all families can satisfy few fully. Choice programs allow different families to pursue different visions of education — religious, progressive, classical, traditional — without requiring any single vision to prevail in public institutions.
• Innovation. Letting a variety of school models operate and compete is more likely to produce genuine pedagogical improvements than requiring innovation within the constraints of traditional district structures.

The case against expansive choice programs

Critics of broad choice programs, particularly universal ESAs, raise several counter-arguments:

• Effects on public schools. When significant numbers of students leave public schools for private alternatives, public schools lose per-pupil funding but often retain fixed costs (buildings, administration, required programs). This can reduce public schools’ capacity to serve the students who remain — often the students with the fewest options. Critics argue universal programs disproportionately subsidize families already using private schools rather than giving new options to disadvantaged families.
• Accountability gaps. Private schools receiving public funds through voucher or ESA programs are typically subject to less oversight than public schools: less standardized testing, less public reporting, less scrutiny of curriculum and practices. Critics argue that public funding should come with public accountability; advocates argue that regulatory burdens would undermine the distinctive missions private schools serve.
• Effects on civic integration. Public schools, whatever their failings, bring children together across demographic, religious, and class lines. Choice programs that allow families to sort into schools matching their demographics and values may, over time, reduce the cross-cutting experiences that supporters of common public schools value.
• Risk of fraud, waste, and abuse. ESA programs in particular have had documented problems with spending on items ranging from playsets to vague “educational services” with minimal oversight. Auditors in Arizona and other states have found substantial questionable spending in the early years of universal programs. Defenders argue these are growing pains and that public school spending has its own inefficiencies; critics argue the structural differences produce genuine accountability problems.
• Limited evidence of academic benefits. As Chapter 8 noted, research on voucher program outcomes has been mixed at best. Louisiana and Indiana voucher programs showed null-to-negative effects on test scores in early cohorts. Recent results from newer programs are not yet available at scale. Critics argue the political and financial commitment to these programs outpaces the evidence of their effectiveness.
• Effects on private schools themselves. Some private-school communities are ambivalent about large-scale public funding, which often comes with strings attached and which may change the character of schools that must adapt to serve larger, more heterogeneous populations. This is a concern specifically within the private-school sector, not primarily from its critics.

What the research shows on program effects

Research on voucher and ESA program outcomes is growing but limited:

• Older programs. Milwaukee’s long-running voucher program has produced mixed research results over its more than three decades of operation; effects have varied across studies and time periods but have generally been modest in either direction. The D.C. Opportunity Scholarship Program, evaluated by Patrick Wolf and colleagues in multiple Institute of Education Sciences reports, showed modest positive effects on some outcomes (high school graduation rates) and null or negative effects on others.
• Louisiana Scholarship Program. Evaluations by Dynarski, Nichols-Barrer, and others found large negative effects on test scores in early cohorts, which attenuated but remained concerning over time. The findings prompted significant debate about the specific program’s design and whether they generalized to other programs. At minimum, they refute the assumption that vouchers automatically produce academic gains.
• Indiana Choice Scholarship Program. Research has found modestly negative effects on test scores for participating students in some cohorts, though with methodological limitations and uncertainty about generalization.
• Newer ESA programs. Too new for rigorous outcome evaluations. Early data from Arizona’s universal ESA suggests substantial participation and substantial funding of families already using private schools — a pattern that affects fiscal impact but does not directly address academic outcomes.
• General patterns. No clear consensus that choice programs produce systematic academic improvement; also no clear consensus that they produce systematic harm. Effects appear to vary by program design, participating schools, and student populations. Families considering private options through choice programs should rely on evaluation of the specific schools rather than assumptions about programs in general.

Practical considerations for families

For families whose state has a choice program, practical considerations include:

• Verify current program rules. Programs and eligibility criteria change; states frequently modify amounts, deadlines, and eligible uses. The state’s official program website is the authoritative source.
• Understand what is covered. ESA programs typically have approved-vendor lists and eligible-expense categories. Not every school or service is eligible; families should verify coverage before planning.
• Application deadlines. Most programs have specific enrollment windows, often in the spring for the following school year. Missing deadlines can result in losing a year of funding.
• Tax implications. Some programs have tax implications that vary by state and household situation. Consult a tax professional if the amounts are significant.
• Sustainability. Programs are subject to legislative and judicial changes. Families should be cautious about making educational choices that depend entirely on continued program funding, particularly in states where programs face legal challenges.

A fast-changing landscape with mixed evidence

School-choice programs — vouchers, tax-credit scholarships, ESAs, and tax deductions — have expanded rapidly since 2021, with 13 states now having universal or near-universal programs and more than 30 states having some form of program. Arguments for expansion emphasize parental rights, equity of access, accountability through competition, pluralism, and innovation. Arguments against emphasize effects on public schools, accountability gaps, civic integration concerns, fraud and waste risks, and the limited and mixed evidence of academic benefits. Research on program effects is thin, inconsistent, and does not clearly favor either sweeping expansion or categorical opposition. For families, the practical questions are local and specific: what programs are available, what they cover, what schools participate, and whether a particular school is a good fit for a particular child. The policy debate is beyond any one family’s control; the decisions for a specific child are not.

What to read or watch next

• Patrick Wolf and colleagues, Institute of Education Sciences evaluations of the D.C. Opportunity Scholarship Program. The longest-running rigorous evaluation of a modern voucher program; reports spanning multiple cohorts.
• Mark Dynarski, Ann Nichols-Barrer, Philip Gleason, and others, Louisiana Scholarship Program evaluations. Research that documented negative effects in early cohorts and generated significant debate about voucher program design.
• EdChoice, edchoice.org. Advocacy organization that publishes extensive state-by-state program data. Pro-choice orientation; useful for current program details.
• National Coalition for Public Education, ncpecoalition.org. Organization skeptical of voucher expansion; useful counterweight for understanding the critical case.
• Douglas Harris, Charter School City: What the End of Traditional Public Schools in New Orleans Means for American Education (2020). Tulane economist’s analysis of the most comprehensive natural experiment in American school reform, with relevance for the broader debate.

CHAPTER 17

Special Needs, Gifted, and Non-Typical Learners

Much of the analysis in this guide has assumed a roughly typical learner — a student whose pace, interests, and needs fit within the range that any of the four main schooling models could reasonably accommodate. For children whose circumstances fall outside that range — children with significant disabilities, gifted children, children with specific learning differences, children with chronic health conditions, twice-exceptional children, and others — the analysis shifts. Different models have very different strengths and limitations for these populations. This chapter addresses what matters for families whose child does not fit the typical mold.

Students with significant disabilities

Federal law — the Individuals with Disabilities Education Act (IDEA) — requires public school districts to provide a “free appropriate public education” (FAPE) to all students with identified disabilities, in the “least restrictive environment.” This is a substantial legal framework that other sectors do not generally share. The practical implications:

• Public schools typically have the strongest infrastructure. Special education teachers, speech-language pathologists, occupational therapists, physical therapists, school psychologists, and other specialized professionals are typically more available in the public sector than in alternatives. Public schools can mandate and provide services through individualized education programs (IEPs) that specify what the student will receive. The scale of the public system means most districts have accumulated expertise in serving students with various disabilities.
• Charter schools must provide special education. As public schools, charters are subject to IDEA and must provide special education services. In practice, charter schools vary in how well they serve special-needs students; some have built strong special education programs, while others have been criticized for “counseling out” students whose needs they could not meet. Families with significant special needs should investigate specific charter schools’ track records carefully.
• Private schools generally are not required to provide special education. Private schools can decline admission to students with significant disabilities and are not obligated to provide the services public schools must provide. Many private schools have minimal special education capacity. Exceptions exist: specialized private schools designed specifically for students with particular disabilities (dyslexia schools like Landmark, Carroll, and The Lab School; schools for students with autism like Eagle Hill; schools for students with emotional or behavioral needs) can provide specialized services that exceed what most public schools offer, though typically at substantial tuition.
• Homeschooling has a distinctive role. Families homeschooling children with disabilities can customize instruction in ways that can be powerful for students whose needs do not fit institutional structures. But they are responsible for obtaining therapies and services themselves; public special education services are generally not available to homeschoolers in most states. Some states with ESA programs allow funds to be used for specialized therapies, which has made homeschooling more viable for some special-needs families.

For students with significant disabilities, the default advice is often “start with the public school and see what services are actually provided, then consider alternatives if appropriate.” A specific, well-resourced public school special education program is usually the best option available; a public district that is failing to implement services may push families toward alternatives, but those alternatives may or may not provide better outcomes. Specialized private schools for specific disabilities are expensive but sometimes transformative. Homeschooling works for some families with resources and support but often is not sustainable for families dealing with complex needs without the institutional infrastructure public schools provide.

504 plans and students with specific accommodations

Beyond IDEA and special education, Section 504 of the Rehabilitation Act provides accommodations for students with disabilities who do not need specialized instruction but do need adjustments to access education. 504 plans can include accommodations like extended time on tests, preferential seating, frequent breaks, access to specific technology, or modifications for attention-related, anxiety-related, or health-related needs. Schools in all sectors handle 504 plans differently; families with children who have or might have specific accommodation needs should inquire directly about how specific schools implement such plans.

Gifted students

Students significantly above typical academic levels face a different but equally real set of fit challenges. Among the considerations:

• Public school gifted programming varies enormously. Some districts have extensive gifted programming (Oakland, Maryland’s Montgomery County, and Fairfax County VA are among those with extensive programs); others have minimal or no gifted programming. For some districts, gifted programming has been reduced or eliminated in the name of equity; families with gifted children in those districts may find limited support for their children’s needs. Identification practices and thresholds vary. Families should investigate specifically what their district provides and how access works.
• Magnet and exam schools. For academically advanced students, selective public magnets and exam schools can be excellent options where available. They provide peer groups and academic rigor that neighborhood schools often cannot. Chapter 12 addresses these options.
• Charter schools for gifted. Some charter schools focus specifically on accelerated academics. Classical academies and certain STEM-focused charters are often well-suited to academically advanced students. Lottery-based admission makes them accessible without the selectivity of exam schools.
• Private schools for gifted. A smaller category of specialized private schools serves specifically gifted students. These can be excellent fits for children whose needs are not met elsewhere, though they are expensive and concentrated in major metropolitan areas.
• Homeschooling for gifted. A common option for families of academically advanced children, particularly when local schools cannot provide appropriate pace or depth. Online programs through Johns Hopkins Center for Talented Youth (CTY), Stanford Online High School, Art of Problem Solving, and others allow challenging coursework at the student’s actual level. Homeschooling’s flexibility allows students to work at their academic pace rather than being held to grade-level expectations.
• Grade acceleration. A research-supported but underused option: moving students up one or more grades. Research by Karen Rogers, Susan Assouline, and others at the University of Iowa’s Belin-Blank Center has shown grade acceleration to be effective for appropriate students, with generally positive academic and social outcomes. Districts vary enormously in whether and how they accommodate acceleration. Families with children for whom this might be appropriate should investigate specific policies.

Twice-exceptional (2e) students

Twice-exceptional students are those who are both gifted and have a disability or learning difference. They present particular fit challenges because their strengths can mask their struggles and vice versa. A 2e student may be intellectually advanced while struggling with dyslexia, dysgraphia, executive function differences, autism spectrum characteristics, or ADHD. They are commonly underserved in both gifted programs (which may not accommodate their specific needs) and special education programs (which may focus on deficits and underestimate their capabilities).

Options for 2e students depend on specific profile and local availability. Some specialized schools serve 2e students specifically (the Bridges Academy in California, the Quad Preparatory School in New York, others). Some public schools with strong gifted programming can serve 2e students well when identification and support are thoughtful. Homeschooling is commonly chosen by families of 2e students precisely because it can address strengths and differences simultaneously in ways institutional programs often cannot. Expert assessment, by a psychologist familiar with 2e profiles, is often valuable in understanding a specific child’s needs and considering options.

Students with chronic health conditions

Students with chronic health conditions — serious allergies, autoimmune disorders, conditions requiring frequent medical appointments, mental health conditions that affect school attendance, and others — face specific challenges in conventional schooling. Public schools can accommodate many of these through 504 plans, medical support systems, and flexible attendance policies, though the quality of accommodations varies. Private schools may or may not accommodate well depending on the school. Homeschooling and virtual/online schooling are commonly chosen by families whose children cannot consistently attend in-person school; the flexibility to work around medical needs is often decisive.

Students who do not fit for other reasons

Beyond identified categories, there are students whose circumstances simply do not match typical school structures. Student-athletes in intensely demanding sports (figure skating, gymnastics, elite competitive athletics) who need flexibility for training and competition. Students with significant performing arts commitments. Students who have had traumatic experiences in conventional schooling and need a reset. Students who are substantially ahead or behind the pace of their chronological peers. Students in unstable family or housing situations. Each of these can make different schooling options more or less viable.

The general principle: the child’s specific circumstances, not generic categorizations, should guide the decision. A child with a specific need sometimes thrives in a school that would not obviously be the best choice on paper, because a specific teacher, a specific program, or a specific community provides what the child needs. Another child with apparently identical circumstances would not. For families with non-typical children, the decision process is typically more involved and more iterative than for families of typical learners.

Non-typical children need decisions tailored to their specifics

Students with significant disabilities are typically best served by the public school sector’s IDEA-mandated infrastructure, with specialized private schools for specific disabilities being excellent but expensive alternatives; charters vary in capacity; homeschooling is viable with significant family resources and support. Gifted students are served by varied public options (gifted programs, magnets, exam schools), charter options, private options, and homeschooling, with the best fit depending on child and local availability. Twice-exceptional students often require specific expert assessment and tailored approaches that few schools provide well. Students with chronic health conditions often benefit from the flexibility of homeschooling or virtual programs, though accommodations in conventional schools are possible. For any non-typical child, the specific circumstances matter more than category labels; the best option depends on the specific child, specific local options, and specific family resources.

What to read or watch next

• Belin-Blank Center, University of Iowa, belinblank.education.uiowa.edu. Leading research center on gifted education and acceleration; extensive resources including the Iowa Acceleration Scale.
• Karen Rogers, Re-Forming Gifted Education (2002). Research-based guide to appropriate programming for gifted students.
• Susan Assouline, Nicholas Colangelo, Joyce VanTassel-Baska, and Ann Lupkowski-Shoplik, A Nation Empowered: Evidence Trumps the Excuses Holding Back America’s Brightest Students (2015). Comprehensive research review on acceleration and related practices.
• Wrightslaw, wrightslaw.com. Comprehensive resource on special education law, IEPs, 504 plans, and parental rights in navigating the public school special education system.
• Council of Parent Attorneys and Advocates, copaa.org. Professional organization of attorneys and advocates specializing in special education; resources for families navigating disputes.
• 2e Newsletter and related resources for twice-exceptional students. The twice-exceptional community has developed specialized resources that mainstream education sources often do not cover well.

CHAPTER 18

What Actually Matters for Outcomes

Families choosing a school often focus primarily on the type of school — public, private, charter, homeschool. But the research on what actually drives educational outcomes suggests that school type is, on average, a smaller factor than several other things. Understanding what the research has consistently found to matter most can help families direct their attention to the factors most likely to affect their children’s outcomes, and correspondingly to relax about factors that matter less than the public debate suggests. This chapter summarizes what the evidence has actually established.

The Coleman Report and its legacy

The 1966 Coleman Report — Equality of Educational Opportunity, commissioned by Congress under the Civil Rights Act of 1964 — was the first comprehensive national study of what actually produces educational outcomes. Its central finding, still controversial and still supported, was that family background factors explain more variation in student outcomes than school factors do. Parental education, family structure, home resources, and similar variables were larger predictors of student achievement than the characteristics of the schools students attended.

Subsequent research has complicated but not overturned this basic finding. Schools do matter; teacher quality matters substantially; specific programs and practices matter; peer effects matter. But the overall magnitude of school effects relative to family effects has proven remarkably consistent across six decades of research. A child from a supportive, educated family will do reasonably well in most schools; a child from a less-supported background faces challenges that the best school cannot fully offset. This is not a statement about worth or potential; it is a statement about the distribution of outcomes across the population.

The practical implication for a single family is less stark than the aggregate finding might suggest. An individual family has limited control over its general background circumstances but substantial control over what it actually does with its children day to day. Family practices — reading together, conversation, cultural exposure, supporting academic work, maintaining reasonable expectations, providing stability and structure — are among the most powerful educational inputs, and they operate largely independently of the school. A family that is doing these things well will see its children generally thrive regardless of school type; a family that is not doing them well will see its children struggle regardless of school type.

Teacher effects

Within schools, research has consistently found teacher quality to be one of the largest controllable factors in student outcomes. Raj Chetty, Jonah Rockoff, John Friedman, and colleagues have produced influential research showing that teachers vary substantially in their effectiveness — with the difference between a high-value-added teacher and a low-value-added teacher producing effects on student test scores equivalent to roughly three months of additional learning in a single year, and with these differences predicting long-term outcomes including college attendance and earnings.

Eric Hanushek’s decades of research have found similar patterns, with the specific estimate that a student assigned to a high-quality teacher for several consecutive years can largely close gaps that might take many years to close otherwise. The corresponding finding is that being assigned to low-quality teachers for consecutive years produces cumulative disadvantages that are difficult to reverse.

The practical implication: the specific teachers a child has matter more than most structural features of the school. A school with generally strong teachers will produce better outcomes than a school with weaker teachers, across all types. Evaluating a specific school should include evaluating what is known about teacher quality: How long do teachers stay? What is their preparation and ongoing development? How does the school attract and retain effective teachers? How are underperforming teachers addressed? These questions often matter more than broader categorization of the school.

Peer effects

A substantial research literature on peer effects has found that the classmates a student has also affect outcomes. Caroline Hoxby, Scott Carrell, Bruce Sacerdote, and others have documented that students benefit from being in classes with peers who are engaged, prepared, and high-achieving. The effects are modest but real, and they compound over years.

The practical implication is that the peer culture of a school matters, not just its formal programs. A school where students generally take academics seriously, where there are strong norms of preparation and effort, and where positive peer influence reinforces learning tends to produce better outcomes than a school where those norms are weaker — even if the two schools are otherwise similar in resources and teaching quality. This is one reason that specific schools within sectors matter more than sector averages; peer culture varies substantially between schools.

Student engagement and effort

A body of research on non-cognitive factors has documented the importance of student engagement, effort, and persistence — sometimes called grit, conscientiousness, or academic mindset. Research by Angela Duckworth, Paul Tough, and others has established that these factors predict academic and life outcomes beyond what IQ and prior achievement predict, and that they can be developed through appropriate experiences and expectations.

Schools vary in how they cultivate these qualities. Some have cultures of high expectations and support; some have cultures of lower expectations or less individual attention. A child whose engagement is cultivated well by a school — by teachers who notice, expectations that are reasonable but high, activities that engage the child’s interests and develop persistence — will often outperform a similar child in a school with weaker practices in this area.

Curriculum content

What is taught matters, and not all curricula are equal. Research on reading instruction in particular has demonstrated that specific instructional approaches — systematic phonics instruction combined with language-rich experiences — produce dramatically better outcomes than approaches based on whole language or balanced literacy models that neglect systematic phonics. Recent work by Natalie Wexler (The Knowledge Gap), Emily Hanford (American Public Media’s Sold a Story reporting), and others has demonstrated how widespread ineffective reading instruction has contributed to persistent reading difficulties. States and districts are in various stages of adopting more research-aligned reading instruction (“the science of reading”).

Curriculum content in other subjects matters in similar ways. Math curricula vary in their rigor and coherence; history and literature curricula vary in content and approach; science curricula vary in emphasis and depth. A family evaluating a school should investigate what curricula are actually used, not just general claims about quality.

School climate and safety

Students in safe, orderly schools can learn; students in chaotic or unsafe environments have greater difficulty. Research on school climate has consistently found that predictable, positive school environments — clear expectations, consistent discipline, adults who know students, limited disruption to learning time — produce better academic and behavioral outcomes than environments without these features. This is not an argument for severity; it is an argument for basic functional order. Schools achieving this look different in different sectors and communities, but the underlying requirement is consistent.

What matters less than is often claimed

Some factors that receive substantial attention have smaller effects than their prominence suggests:

• Class size. Class size matters at the extremes (classes of 15 vs. 35 make a real difference, particularly in early grades), but within typical ranges (20 vs. 25), the effects are modest. Reducing class size is expensive relative to the effects produced. Teacher quality matters more than moderate differences in class size.
• Technology and facilities. Beyond a functional floor (working buildings, basic resources), additional investment in technology and facilities produces modest effects. Spending that goes to people (teachers, tutors, coaches) tends to produce larger effects than spending that goes to buildings and equipment.
• Name recognition and prestige. The reputation of a school matters for peer group and sometimes for later opportunities, but the specific educational effect of attending a highly reputed school is often smaller than the reputation suggests, because reputation reflects student selection as much as school contribution.
• Specific fads and brand-name programs. Schools and districts frequently adopt new approaches, new curricula, new organizational structures. Most produce modest effects that are difficult to detect against the variation between teachers and classrooms. Skepticism about specific programs claiming transformative effects is usually warranted.

The boring fundamentals matter most

What research has consistently found to matter most for educational outcomes: family background and family practices (home environment, support for learning, expectations, stability); teacher quality (the specific teachers a child has, more than structural features of the school); peer culture (classmates who take academics seriously and provide positive influence); student engagement and persistence (cultivated through appropriate experiences and expectations); curriculum content (particularly reading instruction aligned with research evidence); and basic school climate and safety. What matters less than often claimed: moderate variations in class size, technology and facilities beyond a functional floor, name recognition, and specific new programs. The implication for families is that the type of school matters less than the quality of the specific school and, more fundamentally, than what the family does at home. A family focusing on the things most within its control — its own practices, the specific school’s quality across the factors that matter — will typically achieve better outcomes than a family focused on sector-wide debates.

What to read or watch next

• James S. Coleman et al., Equality of Educational Opportunity (1966). The original Coleman Report. Dated in specifics but the foundational findings on family background have been replicated many times.
• Raj Chetty, John Friedman, and Jonah Rockoff, “Measuring the Impacts of Teachers” (2014) and related papers. Landmark research on teacher value-added and long-term outcomes of students.
• Eric Hanushek’s decades of research at Stanford. Sustained body of work on school productivity, teacher quality, and the economics of educational outcomes.
• Natalie Wexler, The Knowledge Gap (2019). Readable treatment of how specific curriculum choices — particularly in reading and background knowledge — shape student outcomes.
• Emily Hanford, Sold a Story (American Public Media, 2022). Investigative reporting on reading instruction that has influenced state policy changes nationwide.
• Paul Tough, How Children Succeed (2012) and Helping Children Succeed (2016). Accessible treatment of non-cognitive factors and how they develop.

CHAPTER 19

Matching Child to School

Different children, even within the same family, thrive in different environments. A school that works beautifully for one sibling may be wrong for another. This chapter addresses the matching question directly: what to look for in a specific child, how to understand what they need, and how to align school choice with the specific human being rather than with a generic idea of a student.

Understanding your child

Before asking which school is right, it helps to ask what this child is like. Useful dimensions to think about:

• Temperament. Is this child energized or drained by large groups? Does structure help them focus, or does it constrain them? Do they need external accountability to do their work, or are they intrinsically motivated? Do they thrive on routine or need variety? Each of these has implications for school fit.
• Academic profile. Is this child advanced, on pace, or behind? In which subjects? Does the child have specific learning differences or challenges? Is the child comfortable with conventional academic expectations, or do they need accommodation?
• Social profile. Does this child make friends easily or with difficulty? Do they thrive with a few close friends or a wide circle? Are they comfortable in large groups or more comfortable in smaller ones? Do they need support with social skills, or are they skilled social navigators?
• Interests and passions. Does this child have specific deep interests — music, sports, a specific academic area, a craft, a field — that a school could or could not accommodate? How important is protecting time and space for those interests?
• Pace of development. Is this child developmentally ahead of, behind, or on pace with chronological age? Academic age and social-emotional age can differ; both matter for school fit.
• Resilience and support needs. How does this child handle challenge? How much structure and support do they need? How much independence can they handle, and at what age?

None of these are static; children develop, and the right school at one age may not be right at another. But taking a clear-eyed look at where a child is now — rather than where the parent wishes they were or assumes they should be — is the foundation for choosing well.

What the child will need from school

From the understanding of the child, certain school features become more or less important:

• For a child who thrives in structure. Schools with clear expectations, consistent rules, and predictable routines will feel comfortable. Many public schools fit this profile, as do most “no-excuses” charters, structured classical academies, traditional religious schools, and conventional private schools.
• For a child who needs flexibility. Progressive schools, Montessori programs, homeschooling, and some project-based charters tend to accommodate children who do not fit well with rigid structures. The tradeoff is that these environments require more self-direction; children who need external structure may flounder.
• For a child with specific academic strengths. Gifted programs, selective public magnets, specialized private schools, and homeschooling can provide the pace and depth that bright children need. A typical neighborhood school may or may not be able to accommodate, depending on its specific practices.
• For a child with specific challenges. The public sector’s special education infrastructure, specialized private schools for specific disabilities, or carefully designed homeschool arrangements may each be the right answer depending on the specific challenge and local options.
• For a child with deep interests. Specialized public schools (arts, STEM, CTE), private schools with strong programming in specific areas, or homeschool arrangements that protect time for the child’s passion may be better fits than general schools that treat the interest as merely an extracurricular.
• For a child who has been in difficulty. Sometimes the best match is simply a change — from a school where things have gone badly to a different environment that offers a clean start. The specific new environment matters less, in these cases, than whether it is genuinely different from what was not working.

Visiting schools with the child

For children old enough to have opinions — roughly middle elementary and older — visiting candidate schools with the child can be valuable. Children often perceive things adults miss. Their physical and emotional responses to a school environment — comfort or discomfort, interest or boredom, connection with other students or isolation — can provide useful information. This is not decisive (children’s first impressions can be misleading and can be shaped by superficial factors), but it is one input alongside the parents’ evaluation.

For younger children, visits are less diagnostic about the child’s own fit but still useful — both for getting a sense of the school and for easing the transition. Most schools welcome prospective-family visits; those that do not are providing information of their own.

Siblings and family logistics

Families with multiple children face the additional question of whether all siblings should attend the same school. Arguments for keeping siblings together include logistical convenience (one drop-off, coordinated schedules, shared community), sibling relationships, and consistent family experience. Arguments against include that different children may genuinely need different environments, and forcing a fit on a sibling who does not fit can create problems for that child and for the family.

The practical answer is usually some compromise: start with the preference for keeping siblings together, but be willing to make different choices for children whose specific needs clearly differ. Many families end up with different children in different schools at different times, adjusting as needed. This is not a failure of planning; it is responding to children’s differences honestly.

Revisiting decisions

Schooling decisions are not permanent. Children change; schools change; family circumstances change. A decision that was right at age five may not be right at age ten, and a decision that was right at ten may not be right at fifteen. Some families make different choices for different stages — a Montessori preschool, a public elementary, a private middle school, a homeschool high school arrangement — based on what their specific child needs at each stage. Others make a single choice and stay with it. Either approach can work.

Signs that revisiting a decision may be appropriate include: persistent academic struggles that the school is not addressing; social difficulties that are not resolving; behavioral changes that suggest the child is not thriving; the child expressing sustained distress about school (not ordinary daily complaints but sustained unhappiness); significant changes at the school (leadership changes, program reductions, specific concerns); significant changes in the family (moves, changes in available time, financial changes). Not every difficulty warrants changing schools, but sustained patterns should prompt honest reassessment.

Fit is specific to the child, not general

Matching a child to a school begins with understanding the child: their temperament, academic profile, social profile, interests, developmental pace, and support needs. From this understanding, some school features become more or less important. Children who thrive in structure and children who need flexibility typically fit different schools; children with specific academic strengths or challenges, with deep interests, or who have had difficulty in prior schools each benefit from different matches. Visits with the child (for those old enough) provide useful if imperfect input. Siblings often benefit from attending the same school, but fit should not be sacrificed to sibling unity when clearly needed. Schooling decisions are not permanent; revisiting is appropriate when children change, schools change, or circumstances change. The goal is finding what works for this specific child in this specific time, not maintaining consistency for its own sake.

What to read or watch next

• Mary Sheedy Kurcinka, Raising Your Spirited Child (updated eds.). Practical work on understanding children’s temperaments, useful for thinking about fit with different kinds of environments.
• Anthony Esolen, Out of the Ashes and related works on educational fit for specific kinds of children; opinionated but thoughtful about what children need.
• Peter Gray, Free to Learn (2013). Developmental psychologist’s treatment of children’s learning, with implications for how various educational settings fit children’s natural development.
• Parenting and child-development resources from the American Academy of Pediatrics and related mainstream medical and psychological organizations. Generalist resources that address child development and what children need to thrive, useful as a baseline for thinking about school fit.
• Conversations with your child’s current and past teachers. Often the best single source of specific information about your child’s learning style, social profile, and what environments tend to work well.

CHAPTER 20

Making the Decision: A Framework for Families

The purpose of this guide has been to help families think clearly about schooling options. This final chapter pulls the pieces together into a framework for making the decision. It is not a formula; families will weigh considerations differently based on their own values and circumstances. But the sequence of questions is useful even when the answers will differ.

Inventory: what is actually available

Start with what options actually exist in your specific community. Many families never map the full set of options, focusing on the neighborhood public school and one or two alternatives they happen to know about. A reasonably thorough inventory:

• Assigned public schools. The neighborhood school(s) your child would attend by default at each level (elementary, middle, high).
• Other public options within the district. Magnet schools, specialized academies, exam schools, gifted programs, language immersion programs, open enrollment possibilities — as discussed in Chapter 12.
• Public options across district lines. If your state allows inter-district open enrollment, what other districts can your child attend?
• Charter schools. What charter schools operate in your area? What are their pedagogical approaches, admission processes, and track records?
• Private schools, religious. Catholic, Protestant, Jewish, Islamic, or other religious schools. Tuition and aid availability?
• Private schools, secular. Independent day schools, Montessori, Waldorf, progressive, classical, specialty schools. Tuition and aid availability?
• Homeschool resources. Co-ops, hybrid schools, microschools, online options, local networks. What would homeschooling actually look like in your specific community?
• Choice program eligibility. What vouchers, tax-credit scholarships, or ESAs is your family eligible for in your state? How much funding do they provide? What schools or services accept them?

Most families discover through this inventory that they have more options than they realized. Some of the options will be immediately unsuitable; others will be worth investigating seriously. The goal is not to visit every school but to know what exists.

Screen: filter by fundamental fit

From the inventory, screen options based on fundamental fit with your family’s situation and values. Factors that can reasonably eliminate options without deep evaluation:

• Cost. Private schools whose tuition (after any aid or public scholarship funding) would strain the family financially beyond what is reasonable. Be honest about aid possibilities — many families rule out private schools based on sticker prices when aid would have made them affordable — but also honest about what the family can truly sustain.
• Logistics. Schools whose commute or schedule would be unsustainable. A brilliant school that requires two hours of daily driving is often not actually an option for most families.
• Values alignment. Religious schools of traditions your family does not share, or schools whose fundamental approach conflicts with values you consider important. This is a legitimate filter, used honestly.
• Capacity. Homeschooling is not a real option for families without sustainable time to do it; some private schools are not real options for families whose child would need services the school does not provide; etc.

Evaluate: investigate the remaining options seriously

For options that survive the screen, investigate seriously. Use the evaluation frameworks from Chapters 6 (public), 9 (private), and 11 (charter research) to understand specific schools. Visit if possible. Talk with current parents. Ask substantive questions. Read what is available — state report cards, school websites, relevant research, news coverage. The effort at this stage is substantial but worthwhile; the goal is having real knowledge of what the finalists actually are.

Match: align school features with child needs

Bring together what you know about the child (Chapter 19) with what you have learned about the finalist schools. Which school is most likely to meet this specific child’s specific needs? Which has the strongest evidence of serving children like yours well? Which has the teacher quality, peer culture, curriculum, and general environment that would work for your child?

A common trap is to choose based on adult preferences rather than child fit. The school that looks most prestigious on the family Christmas card is not necessarily the school where the child will thrive. The school that the family imagines they would have attended is not necessarily the school their child needs. Stay focused on the child.

Decide: make the call, with appropriate humility

After evaluation and matching, make a decision. Accept that it is made with incomplete information and that any choice has tradeoffs. The goal is not the perfect decision — which does not exist — but a reasonable one, made thoughtfully, based on the available information.

Some reasonable decisions look like: “The neighborhood public school is good enough and logistically straightforward, and our child is likely to do well there; we choose it”; or “The public school has problems we cannot accept, the specific charter we visited is a strong fit, and we will try it”; or “Our local schools cannot serve our child’s specific needs, and we will homeschool with co-op participation for the next few years”; or “The Catholic school aligns with our family’s faith and has an academic program we trust; we will invest in it.” No single answer fits all families.

Revisit: treat the decision as provisional

After deciding, commit fully — showing up, supporting the school, integrating the child into the community, giving the choice time to work. Do not second-guess every bump. But also plan to revisit the decision periodically, particularly at natural transition points (elementary to middle, middle to high) or when circumstances change materially.

Most families will make schooling decisions multiple times over a child’s K–12 years. Different children may attend different schools; the same child may move among sectors as needs change. This is not failure or inconsistency; it is responding honestly to reality. The framework is not a one-time exercise but a repeated one.

What to do if the choice is limited

For many families, particularly in rural areas or in areas without charter or private options, the practical “choice” is limited: the neighborhood public school, perhaps one religious school, perhaps homeschooling if the family can sustain it. For these families, the framework above compresses but still applies. Understand the available options clearly. Know the specific school well. Do what is within family control — home practices, support for learning, advocacy within whatever school is chosen. Most children in limited-choice environments do fine; supportive families and engaged parents are more important than menu size.

The last word

No schooling choice is the whole story of a child’s education. Parents who stay involved — who read with young children, who take interest in older children’s work, who know their children’s teachers and friends, who maintain expectations and provide support — shape their children’s educations more than the choice of school does. This is not a counsel of passivity about schools; schools vary and school choice matters. It is a counsel of perspective. A good school will serve a child better than a poor one; but the family that raises the child will, in most cases, shape the child more than any school does. Families who choose schools thoughtfully and then invest in the day-to-day life of their children’s education are doing what is most likely to matter. That is both the most demanding and the most reassuring conclusion this guide can offer.

A framework, not a formula

Making a schooling decision well involves inventorying what options actually exist locally, screening for fundamental fit (cost, logistics, values, capacity), evaluating remaining finalists seriously, matching school features to the specific child’s needs, making the best available decision with appropriate humility, and revisiting as children and circumstances change. Avoid choosing based on adult prestige over child fit. Accept that no choice is perfect and that most families will make schooling decisions multiple times. For families with limited local choice, understand the options and invest heavily in home practices that matter regardless of school. The most durable insight from research is that family involvement and practices shape children’s educations as much as school choice does; families who choose thoughtfully and then invest in their children’s education day by day are doing the things most likely to matter. The framework is not a formula; it is a sequence of honest questions whose answers differ for every family and every child.

What to read or watch next

• This guide’s earlier chapters. The evaluation frameworks for public (Ch. 6), private (Ch. 9), and charter (Ch. 11) schools, and the treatment of what matters for outcomes (Ch. 18), are meant to be read together as a set.
• Robert Pondiscio, How the Other Half Learns (2019). Thoughtful treatment of urban education with particular focus on Success Academy charter schools; useful for understanding how specific high-performing charters actually operate.
• Amanda Ripley, The Smartest Kids in the World (2013). Accessible comparative analysis of education in different countries, with implications for what matters across systems.
• Alfie Kohn, The Schools Our Children Deserve (1999, still relevant). A progressive voice worth considering for families thinking about broader questions of what school should be.
• E. D. Hirsch Jr., Why Knowledge Matters (2016). A contrasting voice from the knowledge-based curriculum tradition; reading both Kohn and Hirsch is useful for grasping the range of reasonable views about what schools should do.

Appendix A: Glossary of Educational Terms

This glossary covers key terms used throughout the guide and commonly encountered in discussions of K–12 education. It is a reference starting point, not a comprehensive dictionary.

504 Plan. Written plan providing accommodations for students with disabilities who do not need specialized instruction but need adjustments to access education. Named for Section 504 of the Rehabilitation Act.

Accreditation. External evaluation by a recognized accrediting body confirming that a school meets specified quality standards. Different accreditors for different types of schools.

AP (Advanced Placement). College-level courses and exams offered in high schools, administered by the College Board. Passing scores may earn college credit.

Catholic school effect. Research finding that Catholic schools produce somewhat better outcomes than comparable public schools for disadvantaged students, controlling for family background. Documented by Coleman (1982), Bryk, Lee, and Holland (1993), and others.

Charter Management Organization (CMO). Nonprofit organization that operates multiple charter schools under a common model. Examples include KIPP, Uncommon Schools, Success Academy, IDEA.

Charter school. A public school operating under a charter — a contract with a state-authorized body — with more flexibility and different accountability than traditional public schools. Publicly funded, tuition-free, and open to all students via lottery if oversubscribed.

CREDO. Center for Research on Education Outcomes at Stanford University. The leading producer of comparative research on charter school outcomes.

CTE (Career and Technical Education). Specialized programming in career fields such as healthcare, information technology, culinary arts, and the trades. Often available as standalone schools or programs within comprehensive high schools.

Dual enrollment. Program allowing high school students to take college courses for credit, often at a local community college or four-year institution.

ESA (Education Savings Account). State program depositing per-pupil funding into accounts that families can use for various educational expenses including private school tuition, tutoring, curricula, and therapies.

Exam school. Public high school with competitive academic admission, typically including an entrance exam. Examples: Stuyvesant, Boston Latin, Thomas Jefferson.

FAPE (Free Appropriate Public Education). The legal standard under IDEA requiring public school districts to provide appropriate education to all students with identified disabilities at no cost to families.

Gifted and talented program. Specialized programming for academically advanced students; varies enormously by district in availability, identification practices, and rigor.

Hybrid school / University-model school. Program meeting two or three days per week with parents providing instruction at home on the other days. A middle ground between conventional schooling and traditional homeschooling.

IDEA (Individuals with Disabilities Education Act). Federal law requiring public schools to provide appropriate education to students with disabilities, including through IEPs.

IEP (Individualized Education Program). Written plan specifying special education services and accommodations for a student with an identified disability; required under IDEA.

Lottery admission. Random selection process used by oversubscribed charter and magnet schools; the methodological foundation for rigorous research on charter effects.

Magnet school. Public school with specialized curriculum or theme, drawing students from across a district rather than just the neighborhood.

Microschool. Small school of typically 5–15 students, often meeting in homes or small shared spaces. Grew significantly after 2020 with the expansion of ESA programs.

NAEP (National Assessment of Educational Progress). Federal standardized assessment administered periodically to national samples of students; provides state-level and large-district comparative data.

NCES (National Center for Education Statistics). Federal agency that collects and reports data on American education, including enrollment, demographics, and outcomes.

“No-excuses” charter. Charter school model emphasizing structured environments, explicit behavioral expectations, longer school days, intensive academics, and college-preparatory focus. Examples: Success Academy, traditional KIPP, Uncommon Schools.

Proficiency rate. Percentage of students scoring at or above a defined threshold on a standardized test. Contrast with growth measures.

School choice. Broad category of policies allowing families to choose among schools, including public choice (magnets, open enrollment, charters) and private choice (vouchers, tax-credit scholarships, ESAs).

Selection bias. Methodological challenge in educational research arising from non-random assignment of students to schools; the central difficulty in comparing outcomes across schooling models.

Tax-credit scholarship. State program providing tax credits to individuals or businesses who donate to scholarship-granting organizations funding private school tuition for eligible students.

Title I. Federal program providing supplemental funding to schools serving high concentrations of low-income students.

Traditional public school. Public school operated by a local school district, serving students assigned by geographic attendance zones.

Twice-exceptional (2e). Students who are both gifted and have a disability or learning difference. Often underserved by both gifted and special education programs.

Value-added / growth measure. Metric estimating a school’s or teacher’s contribution to student learning by measuring change over time, controlling for prior achievement. More informative about school quality than raw test scores.

Virtual charter school. Online charter school delivering instruction remotely. Research has consistently found weaker outcomes than brick-and-mortar alternatives.

Voucher. Direct state payment for private school tuition.

Appendix B: Quick-Reference Resources

This appendix lists authoritative and useful resources for the topics covered in the guide. These are starting points.

Federal and official data

• National Center for Education Statistics (nces.ed.gov). The federal statistical agency for education. Condition of Education and Digest of Education Statistics are the authoritative annual compilations of K–12 data.
• The Nation’s Report Card (nationsreportcard.gov). NAEP data, including state-level and Trial Urban District Assessment results.
• Institute of Education Sciences (ies.ed.gov). Federal research arm of the Department of Education. Home to the What Works Clearinghouse and many rigorous evaluation reports.
• What Works Clearinghouse (ies.ed.gov/ncee/wwc). Federal clearinghouse rating the methodological rigor of education research on specific interventions.
• Your state’s department of education. Typically the authoritative source for state-level policy, school report cards, and state-specific program details. Accessible at [state name].gov or by searching “[state name] department of education.”

Public school resources

• GreatSchools.org. Commercial school-rating aggregator covering most American public schools. Use as a starting point.
• U.S. News & World Report Best High Schools. Annual high school rankings.
• Parent Teacher Association (pta.org). Largest national parent organization with resources on advocacy and involvement.

Charter school resources

• National Alliance for Public Charter Schools (publiccharters.org). Industry association with national data on charter enrollment and state laws.
• National Association of Charter School Authorizers (qualitycharters.org). Professional association for authorizers; resources on oversight and accountability.
• CREDO (credo.stanford.edu). Stanford research center producing the leading national studies of charter outcomes.
• National Education Policy Center (nepc.colorado.edu). University of Colorado academic center producing critical reviews of education research, including charter claims.

Private school resources

• National Association of Independent Schools (nais.org). Trade association for independent (typically secular) private schools; directory and data.
• National Catholic Educational Association (ncea.org). Resources on Catholic schools including accreditation and performance.
• Association of Christian Schools International (acsi.org). Umbrella organization for evangelical Christian schools.
• American Montessori Society (amshq.org) and Association Montessori Internationale (amiusa.org). Main accrediting bodies for Montessori programs.
• Association of Waldorf Schools of North America (waldorfeducation.org). Accrediting body for Waldorf schools.
• Niche.com and Private School Review. Commercial aggregators of private school information.

Homeschool resources

• Home School Legal Defense Association (hslda.org). Advocacy and legal resource with state-by-state regulatory summaries. Conservative Christian orientation.
• Coalition for Responsible Home Education (responsiblehomeschool.org). Alternative perspective focused on child welfare and reasonable oversight.
• National Home Education Research Institute (nheri.org). Primary producer of homeschool research; read alongside methodological critiques.
• State and local homeschool associations. Usually the best sources for state-specific and local practical information.

School choice policy

• EdChoice (edchoice.org). Pro-choice advocacy organization with extensive state-by-state program data.
• Education Commission of the States (ecs.org). Nonpartisan organization tracking state education policy, including funding and choice programs.
• National Coalition for Public Education (ncpecoalition.org). Organization skeptical of voucher expansion; useful counterweight.
• FutureEd (future-ed.org). Education policy think tank at Georgetown with balanced analysis of current policy developments.

Special needs and non-typical learners

• Wrightslaw (wrightslaw.com). Comprehensive resource on special education law, IEPs, and 504 plans.
• Council of Parent Attorneys and Advocates (copaa.org). Professional organization for special education advocates.
• Belin-Blank Center (belinblank.education.uiowa.edu). Leading research center on gifted education and acceleration.
• National Association for Gifted Children (nagc.org). Resources on gifted identification and programming.

Appendix C: References

Sources consulted in preparing this guide, organized by chapter. Style approximates Chicago Notes-Bibliography.

Chapter 1 — The American Educational Landscape Today

National Center for Education Statistics. The Condition of Education 2024. Washington, DC: U.S. Department of Education, 2024.

National Center for Education Statistics. “A Higher Percentage of K–12 Students Are Receiving Academic Instruction at Home.” Press Release, September 17, 2024.

Pew Research Center. “Facts about Public, Private and Charter Schools in the US.” June 6, 2024.

Dee, Thomas. “Where the Kids Went: Nonpublic Schooling and Demographic Change during the Pandemic Exodus from Public Schools.” Stanford University working paper, 2023.

Chapter 2 — How Schools Are Funded and Governed

Baker, Bruce D. Educational Inequality and School Finance: Why Money Matters for America’s Students. Cambridge, MA: Harvard Education Press, 2018.

Education Commission of the States. School Funding reports. ecs.org.

EdChoice. The ABCs of School Choice. Indianapolis: EdChoice, annual editions.

Chapter 3 — What the Research Can and Cannot Tell Us

Coleman, James S., et al. Equality of Educational Opportunity. Washington, DC: U.S. Department of Health, Education, and Welfare, 1966.

Angrist, Joshua D., and Jörn-Steffen Pischke. Mastering ’Metrics: The Path from Cause to Effect. Princeton, NJ: Princeton University Press, 2014.

Institute of Education Sciences. What Works Clearinghouse. ies.ed.gov/ncee/wwc.

Hanushek, Eric A. Various papers on school productivity and teacher quality. hanushek.stanford.edu.

Chapter 4 — Public Schools: The Dominant Model

Ravitch, Diane. The Death and Life of the Great American School System. Revised ed. New York: Basic Books, 2016.

Goldstein, Dana. The Teacher Wars: A History of America’s Most Embattled Profession. New York: Doubleday, 2014.

Mehta, Jal. The Allure of Order: High Hopes, Dashed Expectations, and the Troubled Quest to Remake American Schooling. New York: Oxford University Press, 2013.

Labaree, David F. Someone Has to Fail: The Zero-Sum Game of Public Schooling. Cambridge, MA: Harvard University Press, 2010.

Chapter 5 — The Case For and Against Public Schools

Ravitch, The Death and Life of the Great American School System.

Kozol, Jonathan. Savage Inequalities: Children in America’s Schools. New York: Crown, 1991.

Hanushek, Eric A., and Paul E. Peterson, eds. Endangering Prosperity: A Global View of the American School. Washington, DC: Brookings Institution Press, 2013.

Jackson, C. Kirabo, Rucker C. Johnson, and Claudia Persico. “The Effects of School Spending on Educational and Economic Outcomes.” Quarterly Journal of Economics 131, no. 1 (2016).

Chapter 6 — Evaluating a Public School

GreatSchools.org. School ratings and information. greatschools.org.

U.S. News & World Report. Best High Schools Rankings, annual.

National Assessment of Educational Progress (NAEP). nationsreportcard.gov.

Chapter 7 — Private Schools

National Center for Education Statistics. Private School Universe Survey, biennial.

Bryk, Anthony S., Valerie E. Lee, and Peter B. Holland. Catholic Schools and the Common Good. Cambridge, MA: Harvard University Press, 1993.

Garnett, Nicole Stelle. Lost Classroom, Lost Community: Catholic Schools’ Importance in Urban America. Chicago: University of Chicago Press, 2014.

National Association of Independent Schools. nais.org.

Chapter 8 — The Case For and Against Private Schools

Coleman, James S., Thomas Hoffer, and Sally Kilgore. High School Achievement: Public, Catholic, and Private Schools Compared. New York: Basic Books, 1982.

Wolf, Patrick, et al. Evaluations of the D.C. Opportunity Scholarship Program. Institute of Education Sciences reports, various years.

Dynarski, Mark, Ann Nichols-Barrer, Philip Gleason, and others. Evaluations of the Louisiana Scholarship Program and related voucher programs.

Lubienski, Sarah Theule, and Christopher Lubienski. The Public School Advantage: Why Public Schools Outperform Private Schools. Chicago: University of Chicago Press, 2013.

Chapter 9 — Evaluating a Private School

National Association of Independent Schools. Data and resources at nais.org.

National Catholic Educational Association. ncea.org.

Association of Christian Schools International. acsi.org.

Chapter 10 — Charter Schools: The Experiment

Whitmire, Richard. The Founders: Inside the Revolution to Invent (and Reinvent) America’s Best Charter Schools. San Francisco: Jossey-Bass, 2016.

Junge, Ember Reichgott. Zero Chance of Passage: The Pioneering Charter School Story. Edina, MN: Beaver’s Pond Press, 2012.

National Alliance for Public Charter Schools. publiccharters.org.

Chapter 11 — The Research on Charter Outcomes

Center for Research on Education Outcomes (CREDO). As a Matter of Fact: The National Charter School Study III 2023. Stanford, CA: Stanford University, 2023.

CREDO. Earlier national studies (2009, 2013) and state and city studies. credo.stanford.edu.

Angrist, Joshua D., Sarah Cohodes, Susan Dynarski, Parag Pathak, and Christopher Walters. “Stand and Deliver: Effects of Boston’s Charter High Schools on College Preparation, Entry, and Choice.” Journal of Labor Economics 34, no. 2 (2016).

Harris, Douglas N. Charter School City: What the End of Traditional Public Schools in New Orleans Means for American Education. Chicago: University of Chicago Press, 2020.

Mathematica Policy Research. KIPP Middle Schools: Impacts on Achievement and Other Outcomes. Various reports.

National Education Policy Center. “CREDO Report Makes Overstated Claims of Charter School Gains.” NEPC Review, September 2023.

Chapter 12 — Magnet Schools and Other Specialized Public Options

Finn, Chester E., Jr., and Jessica A. Hockett. Exam Schools: Inside America’s Most Selective Public High Schools. Princeton, NJ: Princeton University Press, 2012.

Magnet Schools of America. magnet.edu.

Association for Career and Technical Education. acteonline.org.

National Association for Gifted Children. nagc.org.

Chapter 13 — Homeschooling: Growth and Variety

Bauer, Susan Wise, and Jessie Wise. The Well-Trained Mind: A Guide to Classical Education at Home. 4th ed. New York: W. W. Norton, 2016.

Gaither, Milton. Homeschool: An American History. 2nd ed. New York: Palgrave Macmillan, 2017.

Home School Legal Defense Association. hslda.org.

Johns Hopkins Homeschool Research Lab. Data and publications at education.jhu.edu.

Chapter 14 — The Case For and Against Homeschooling

Kunzman, Robert, and Milton Gaither. “Homeschooling: A Comprehensive Survey of the Research.” Other Education 2, no. 1 (2013).

Kunzman, Robert. Write These Laws on Your Children: Inside the World of Conservative Christian Homeschooling. Boston: Beacon Press, 2009.

Murphy, Joseph. Homeschooling in America: Capturing and Assessing the Movement. Thousand Oaks, CA: Corwin, 2012.

Ray, Brian D. “Research Facts on Homeschooling.” National Home Education Research Institute, various years.

Coalition for Responsible Home Education. responsiblehomeschool.org.

Chapter 15 — Practical Realities

Bauer, The Well-Trained Mind.

Duffy, Cathy. 101 Top Picks for Homeschool Curriculum. Latest ed.

Bell, Debra. The Ultimate Guide to Homeschooling. Nashville: Thomas Nelson, latest ed.

Binz, Lee. The HomeScholar series on homeschool high school.

Chapter 16 — Vouchers, ESAs, and the Public Policy Debate

EdChoice. The ABCs of School Choice, annual.

Ballotpedia. “School choice in the United States.” Updated regularly.

FutureEd. “Legislative Tracker: 2025 State Private-School Choice Bills.” 2025.

Wolf, Patrick J., et al. D.C. Opportunity Scholarship Program evaluations. Institute of Education Sciences.

Dynarski, Mark, Ann Nichols-Barrer, and Philip Gleason. Louisiana Scholarship Program evaluations.

Chapter 17 — Special Needs, Gifted, and Non-Typical Learners

Rogers, Karen B. Re-Forming Gifted Education. Scottsdale, AZ: Great Potential Press, 2002.

Assouline, Susan G., Nicholas Colangelo, Joyce VanTassel-Baska, and Ann Lupkowski-Shoplik. A Nation Empowered: Evidence Trumps the Excuses Holding Back America’s Brightest Students. Iowa City, IA: Belin-Blank Center, 2015.

Wrightslaw. wrightslaw.com. Resources on special education law.

Council of Parent Attorneys and Advocates. copaa.org.

Chapter 18 — What Actually Matters for Outcomes

Coleman, Equality of Educational Opportunity.

Chetty, Raj, John N. Friedman, and Jonah E. Rockoff. “Measuring the Impacts of Teachers I: Evaluating Bias in Teacher Value-Added Estimates” and “Measuring the Impacts of Teachers II: Teacher Value-Added and Student Outcomes in Adulthood.” American Economic Review 104, no. 9 (2014).

Hanushek, Eric A. Various. hanushek.stanford.edu.

Wexler, Natalie. The Knowledge Gap: The Hidden Cause of America’s Broken Education System—and How to Fix It. New York: Avery, 2019.

Hanford, Emily. Sold a Story (podcast series). American Public Media, 2022.

Tough, Paul. How Children Succeed: Grit, Curiosity, and the Hidden Power of Character. Boston: Houghton Mifflin Harcourt, 2012.

Chapter 19 — Matching Child to School

Kurcinka, Mary Sheedy. Raising Your Spirited Child. Updated eds.

Gray, Peter. Free to Learn: Why Unleashing the Instinct to Play Will Make Our Children Happier, More Self-Reliant, and Better Students for Life. New York: Basic Books, 2013.

Chapter 20 — Making the Decision

Pondiscio, Robert. How the Other Half Learns: Equality, Excellence, and the Battle Over School Choice. New York: Avery, 2019.

Ripley, Amanda. The Smartest Kids in the World: And How They Got That Way. New York: Simon & Schuster, 2013.

Kohn, Alfie. The Schools Our Children Deserve: Moving Beyond Traditional Classrooms and “Tougher Standards.” Boston: Houghton Mifflin, 1999.

Hirsch, E. D., Jr. Why Knowledge Matters: Rescuing Our Children from Failed Educational Theories. Cambridge, MA: Harvard Education Press, 2016.

— end of guide —