Percentage of School Voucher Funds Going to Religious Schools by State

By The School Choice Index Editorial Team Last verified: August 1, 2026 · Dataset version 1.0.0

The evidence reviewed through August 1, 2026 contains no comparable official 50-state figure for the percentage of school voucher funds going to religious schools by state. The strongest current whole-program dollar result is North Carolina: about 78% of roughly $589 million in 2025–26 Opportunity Scholarship money went to schools Carolina Journal classified as religious. New Hampshire’s 86% figure measures religious schools’ share of tuition spending; because tuition was 56.9% of reported family expenditures, the corresponding share of reported family spending was about 48.9%.

Both New Hampshire numbers are correct. They answer different questions. That 37.1-point gap inside one program and one year is the first thing to understand before quoting any percentage on this page.

Original finding from the 51-jurisdiction audit: nine jurisdictions have a publishable benchmark in the current evidence set, and Texas has participating-school context. Only North Carolina has a current full-program, full-period dollar share classified as D1. No valid national average can be calculated from the state figures because the measures are not interchangeable.

---

Percentage of school voucher funds going to religious schools by state: what the data shows

The state evidence does not form one ranking. The table contains full-program dollars, partial dollars, reported-expenditure calculations, recipient shares, a school share, and one modeled estimate. The class beside each value tells you what can be quoted without changing the meaning.

Source: The School Choice Index evidence audit, verified August 1, 2026. Dollar classes, recipient classes, school classes, and estimates are not interchangeable and must not be averaged. The Washington Post figures were verified against the original article, not downstream restatements.

The cleanest current answer is North Carolina’s approximately 78% D1 result. The largest number in the table is not automatically the largest funding share: Missouri’s approximately 98% covers one month, Indiana’s 98% counts vouchers, and Texas’s 80% counts schools.

---

Why can the same state have two different religious-share numbers?

Every percentage has a denominator. A dollar share answers where money went; a recipient share describes who participated; a school share describes the participating supply; and a tuition-only share covers one category of spending. Those measures can differ sharply inside the same program and year.

Source: The School Choice Index methodology, dataset version 1.0.0.

Two rules matter:

1. With the same religious-school numerator, a larger denominator produces a lower or equal share. If an ESA denominator expands from tuition to all reported family expenditures, non-tuition purchases enter the denominator and the percentage cannot rise unless the numerator changes too.
2. Recipient and dollar shares tend to be closer when awards are similar, but they still are not identical by definition. Wisconsin’s 2025–26 awards were $10,877 for K–8, $13,371 for grades 9–12, and $16,049 for the Special Needs Scholarship Program. A 96% student share cannot be relabeled as a 96% dollar share.

---

Why are New Hampshire’s 86% and 48.9% figures both correct?

The 86% figure uses tuition as its denominator; the 48.9% figure uses all reported family expenditures. Families spent 56.9% of reported expenditures on tuition, so the calculation is 86% × 56.9% = 48.934%, rounded to 48.9%. That figure does not include administrator expenses or money awarded but not spent.

Source: Concord Monitor analysis of expenditure data released by Children’s Scholarship Fund New Hampshire. Tuition’s 56.9% share is also reported in the administrator’s 2024–25 program materials. The final row is The School Choice Index calculation: 0.86 × 0.569 = 0.48934.

The arithmetic cross-checks against the rounded dollar figures. Applying 56.9% to $20.2 million produces about $11.49 million in tuition; 86% of that is about $9.88 million, consistent with “nearly $10 million.” The percentage should therefore be quoted as about 48.9% of reported family expenditures, not “49% of the whole program.”

The distinction is not cosmetic. “Program money” can mean an appropriation, an amount awarded to accounts, an amount available after administration, an amount spent by families, or one category such as tuition. This page names the denominator every time.

---

Why did North Carolina fall from 86% to about 78%?

The two figures cover consecutive school years and come from different analysts using the same underlying NCSEAA program records. Public Schools First NC reported 86% for 2024–25; Carolina Journal reported about 78% for 2025–26. The eight-point difference is real in the published results, but the visible source pages do not provide a complete reproducible school-by-school affiliation protocol, so the audit does not assign the change to one cause.

Sources: Public Schools First NC, Carolina Journal, and the NCSEAA Opportunity Scholarship data hub. Public Schools First NC opposes voucher expansion; Carolina Journal is published by the John Locke Foundation, which supports school choice. The figures above are the sources’ published values; no dollar totals were reconstructed from rounded percentages.

In 2024–25, religious schools were approximately 73% of participating schools but received 86% of program dollars, a 13-point gap. That does not establish why the gap exists. It establishes that a school count and a dollar count answer different questions.

Anyone using either North Carolina percentage should keep the year, analyst, and denominator attached to it. Dropping any one of those changes the claim.

---

What does the full 51-jurisdiction audit show?

The audit covers all 50 states and the District of Columbia. It finds nine benchmark jurisdictions, one Texas school-share row, 24 U1 jurisdictions, 15 N0 jurisdictions, and two scope exclusions. A blank is never a zero, and “not calculated” does not mean that no records exist.

Source: The School Choice Index evidence audit, August 1, 2026. The program universe was seeded from the EdChoice current program inventory and checked against public program sources used in this audit. EdChoice is a school-choice advocacy organization; its inventory is used to identify programs, not to establish religious-school percentages. Puerto Rico is outside the 50-state-and-District-of-Columbia count.

The categories add to 51:

Source: The School Choice Index count of Table 5.

The U1 rows are research gaps, not zeros. Some may already have school-level payments but no religious-affiliation field; others may publish only program totals, organization totals, recipient counts, or records that cannot yet be reconciled across a complete period.

---

Which states publish enough data to calculate a religious-school dollar share?

A direct calculation needs school-level payment amounts and a defensible affiliation match for the same period and program. Publishing one side without the other leaves the percentage unfinished. Pennsylvania is changing its reporting rules, while Texas is too new to have a complete operating year.

Source: The School Choice Index review of the program and reporting sources listed on this page.

Pennsylvania’s reporting has changed

The 2022–23 Pennsylvania figure remains an E1 estimate: Keystone Research Center estimated that at least 78%, approximately $240 million, of EITC and OSTC scholarship funds reached religious schools. Historical reporting did not provide a complete school-level observed calculation.

The current reporting rule is different. Pennsylvania’s Department of Community and Economic Development says that reporting for the 2025–26 school year must include each scholarship’s dollar amount, the recipient’s district of residence, and the school attended; renewal reports are due in November 2026. Once those reports are available, a direct calculation may become possible if the records are complete and schools can be matched to a verified affiliation source. See the official EITC and OSTC reporting pages.

New Hampshire’s required disclosure arrived late

New Hampshire Administrative Code Ed 803.02(f) requires the program administrator to publish each approved provider’s name, eligible services, fee amounts received, and number of EFA students served. The Concord Monitor reported that Children’s Scholarship Fund New Hampshire had failed to publish the required provider information for the prior two years before releasing and reposting data. That delayed disclosure is what made the 2024–25 tuition and expenditure analysis possible.

---

How were schools classified as religious?

The School Choice Index did not newly classify any school for this version of the dataset. Every published percentage retains the original analyst’s classification, and the dataset identifies who made it. That avoids passing a name-pattern guess off as a verified affiliation.

For future school-level joins, the evidence hierarchy is:

1. NCES Private School Universe Survey religious orientation, when the school can be matched confidently and the survey period is appropriate.
2. An explicit state or program directory field, when the administering agency actually publishes one.
3. The school’s official site or an official denominational directory.
4. Documented manual review, with the evidence URL, decision, date, and reviewer note retained.

The source methods differ:

• The Washington Post analyzed original state records and reported the state percentages directly.
• The Concord Monitor combined administrator expenditure data with New Hampshire school classification information.
• The Texas Observer used school websites, handbooks, survey responses, and calls to school leaders.
• Carolina Journal and Public Schools First NC each published a North Carolina classification result, but their visible pages do not contain a complete school-by-school decision rule.
• Keystone Research Center produced a model because the historical Pennsylvania records did not support a direct school-level observation.

That variation is why the page names the analyst instead of pretending every percentage was produced by one uniform national system.

---

What does the federal private-school baseline say?

NCES estimates that 66.7% of U.S. private schools and 75.8% of private-school students were religiously oriented in 2023–24. Those are national private-school statistics, not voucher percentages. They describe the sector available to families, not where any specific program’s money went.

Source: NCES, Characteristics of Private Schools in the United States: Results From the 2023–24 Private School Universe Survey, NCES 2026-015, Table C-2.

The same weighted totals produce three useful comparisons that are not printed as a single block in the federal report:

• Religious private schools enrolled about 190 students per school on average: 3,863,203 ÷ 20,385 = 189.5.
• Nonsectarian private schools enrolled about 121 students per school on average: 1,233,162 ÷ 10,169 = 121.3.
• The average religious private school therefore enrolled about 1.56 times as many students as the average nonsectarian private school.

That size difference explains why the religious share of students is nine points higher than the religious share of schools. Again, institutions and people are different denominators.

Roman Catholic schools enrolled 1,715,733 students, or 33.7% of all private-school students. Other religious schools enrolled the remaining 42.1% of all private-school students, calculated as 75.8% minus the Catholic share using the underlying weighted counts. These figures provide national context; they do not predict any state program’s denominational mix.

How much uncertainty is in the NCES classification?

The Private School Universe Survey is the most consistent national source for religious orientation, but its published totals are weighted estimates. NCES reports 22,510 completed interviews and 6,809 nonrespondents, an unweighted response rate of 76.8% and a base-weighted response rate of 76.0%.

NCES’s nonresponse adjustment reduced the mean estimated absolute relative bias from 13.91% before adjustment to 2.05% after adjustment. The share of tested variable categories with statistically significant bias fell from 87.50% to 28.13%. Selected pre-adjustment affiliation biases included Pentecostal −86.07%, Mennonite −33.74%, Amish −18.20%, Jewish −12.05%, and nonsectarian regular −18.49%.

Those technical notes do not make the published estimates unusable. They explain why the federal figures should be described as estimates and why a direct state payment join still needs careful matching and documented unknowns.

---

How much voucher money goes to Catholic schools specifically?

Only North Carolina and New Hampshire in the current benchmark set provide a published denominational split precise enough to print here. North Carolina’s 2025–26 analysis assigns at least 64% of all Opportunity Scholarship dollars to Protestant Christian schools and 11% to Catholic schools. New Hampshire’s 2024–25 tuition analysis assigns 43% to independent Christian schools and 43% to parochial schools.

Sources: Carolina Journal and Concord Monitor. The denominators differ and the rows should not be compared as one ranking.

Missouri’s source identifies Catholic, Christian, Jewish, and Islamic recipients but does not publish a denominational dollar split. The Washington Post state summaries used here do not provide a denomination-by-dollar table for Arizona, Florida, Indiana, Ohio, or Wisconsin.

---

Can public tuition-assistance programs include religious schools?

Yes. The U.S. Supreme Court has held that once a state creates a generally available private-education benefit, it cannot disqualify otherwise eligible private schools solely because they are religious. That rule comes from Espinoza v. Montana Department of Revenue and was applied to Maine tuition assistance in Carson v. Makin.

The Court’s formulation is direct: “A State need not subsidize private education. But once a State decides to do so, it cannot disqualify some private schools solely because they are religious.” Read the official opinions in Espinoza and Carson.

This is a general description of the federal holdings, not legal advice. Program rules, state constitutional claims, administration, and litigation can still differ by jurisdiction.

---

Why is Texas included without a spending percentage?

Texas has a useful school-share snapshot but no complete-year religious-school spending share. The Texas Observer reported that 80% of more than 1,700 enrolled private schools were affiliated with or owned by religious or faith-based groups as of February 4, 2026. Its earlier figure above 90% covered a narrower sample of 291 state-selected schools serving students beyond kindergarten.

The official roster kept growing. The Texas Comptroller reported 2,200 participating schools by March 17, 2026, so the Observer’s 80% remains a dated snapshot rather than a final roster percentage. The state then announced that nearly 73,000 accounts would receive initial funding July 1, 2026; private-school students receive 25% on July 1, 25% on October 1, and the remaining 50% on February 1, 2027, subject to continued enrollment. See the Comptroller’s March 17 enrollment update and June 30 funding announcement.

A complete religious-school dollar share requires a completed expenditure period and school/provider payments. Until then, Texas remains S1, not D1 or D2.

---

How was this dataset produced?

We audited 50 states and the District of Columbia, recorded the best publishable religious-school measure for each jurisdiction, and kept every figure attached to its denominator, period, program coverage, source, and evidence class. We performed four original calculations from published components: New Hampshire’s 48.9% reported-expenditure share and the three NCES size/composition comparisons shown above. We did not newly classify any school.

Scope

The strict school-directed scope includes:

• direct vouchers;
• education savings accounts when private-school tuition or required fees can be isolated;
• tax-credit education savings accounts;
• tax-credit scholarships;
• refundable education credits when a school-level payment can be identified.

Parent-side individual deductions or credits that cannot be traced to a recipient school are excluded from the main payment measure. Illinois is therefore X1. Alaska’s correspondence allotment is also X1 because the Alaska Department of Law’s official guidance permits discrete private-school materials and services that support a public correspondence education but says public allotment funds cannot pay tuition for full-time private-school enrollment.

Denominator rules

For direct vouchers, the preferred denominator is tuition and required-fee payments to eligible private schools for a complete program period.

For ESAs, three denominators must remain separate:

1. private-school tuition and required fees;
2. all family expenditures from accounts;
3. total appropriations or awards, including unspent balances and administration.

For tax-credit scholarships, the preferred denominator is scholarship awards used at recipient schools, not the face value of tax credits authorized and not scholarship-organization administration.

Release rules

A percentage is publishable only when:

• the numerator and denominator cover the same period and program scope;
• the source states what the figure measures;
• partial periods and vendor subsets are visible beside the value;
• suppressed values are not replaced with zero;
• a modeled estimate is labeled E1;
• a recipient or school share is not relabeled as a dollar share;
• unknown or unmatched records cannot silently change the rounded result.

Multiple programs

Programs are calculated separately first. A state rollup is valid only when periods, payment definitions, and coverage are compatible and duplicate students or payments can be ruled out. Otherwise the page prints separate program rows or keeps the state at U1.

Program-universe source

The audit used EdChoice’s current program inventory to identify possible programs, then relied on administering agencies, official reports, court opinions, and the original data producers or original analysts for the actual claims. EdChoice openly identifies itself as a school-choice advocacy organization; its policy judgments are not treated as neutral findings.

Data fields

The downloadable dataset records:

jurisdiction
program
reported_value
evidence_class
what_it_measures
period
program_coverage
publisher
publisher_type
religious_classification_by
underlying_data_source
source_url
last_verified
publish_status
notes

Download the CSV dataset.

---

What does this data show, and what does it not show?

It shows the best sourced state evidence in the current audit and, just as importantly, what each percentage does not measure. It does not establish school quality, student outcomes, the amount of religious instruction received, family motives, legal compliance, or whether a program should exist.

The figures cannot be summed or averaged. An arithmetic mean of 78%, 98%, 87%, 96%, 91%, 98%, 82%, and 78% would mix full-program dollars, one-month dollars, partial-vendor tuition, recipients, vouchers, students, and a model. The result would have no coherent numerator or denominator.

The page also does not assume that missing means secret. A U1 row can reflect unavailable records, incompatible records, a new program, an unfinished affiliation join, or a calculation that has not met the release rules. The only honest value in that situation is “not calculated.”

---

What will change next?

The dataset will change when complete source periods and new reporting files arrive, not merely because the calendar changes. Pennsylvania’s first enhanced reports are due in November 2026, and Texas’s first program-year payments continue through February 2027. Those are concrete recalculation triggers.

The visible “Last verified” date changes only after the relevant sources are actually checked. Earlier dataset versions remain available so that a previously quoted figure can still be reconstructed.

---

Limitations

The limitations are part of the data, not fine print. The biggest risk is not that a percentage has a caveat; it is that someone removes the caveat and turns one measure into another.

• Religious affiliation was classified by different analysts. We identify each analyst and do not pretend the methods were identical.
• The benchmark sources have different institutional positions. Public Schools First NC opposes voucher expansion; Carolina Journal is published by a foundation that supports school choice; Keystone Research Center argues for stronger voucher accountability. Their numbers are retained only with their methods and labels.
• The Washington Post figures are original, directly verified reporting. Arizona is a partial vendor/tuition result; Florida, Indiana, Ohio, and Wisconsin are recipient or voucher measures.
• New Hampshire’s 48.9% is a share of reported family expenditures. It is not a share of appropriations, all awards, administrator costs, or unspent balances.
• The years differ. The benchmark table spans 2022–23 through 2025–26 and is not a same-year ranking.
• NCES totals are weighted estimates. The report’s response rates and nonresponse adjustments are disclosed above.
• U1 rows may become calculable. “Not calculated” is not zero and is not a permanent judgment.
• Tax-credit scholarships are not direct appropriations. The source and accounting route remain visible in every relevant row.
• A school’s affiliation does not reveal how an individual student used or experienced the program. This is a funding and participation audit.

We would rather you knew these than found them yourself.

---

Frequently asked questions

These answers keep the denominator attached to the number. None turns a recipient, school, partial-period, or estimated figure into a full-program dollar share.

What percentage of all school voucher money nationwide goes to religious schools?

No defensible national percentage can be calculated from the available state figures because they cover different programs, years, denominators, and levels of completeness. The 75.8% national figure from NCES describes private-school students, not voucher spending.

Which state sends the highest percentage of voucher funds to religious schools?

The evidence does not support a valid ranking. Missouri’s approximately 98% covers one month of one direct-aid stream, Indiana’s 98% is a voucher-recipient share, and North Carolina’s approximately 78% is a full-program dollar share.

Is the percentage of voucher students the same as the percentage of voucher dollars?

No. Award amounts can differ by grade, income, disability status, program, and school, so a student or voucher share is not a dollar share.

Why are New Hampshire’s 86% and 48.9% figures both correct?

The 86% figure is the religious share of tuition spending. Tuition was 56.9% of reported family expenditures, so 86% multiplied by 56.9% equals about 48.9% of reported family expenditures.

Can school voucher programs include religious schools?

Yes. Under Espinoza v. Montana Department of Revenue and Carson v. Makin, a state that offers a generally available private-school tuition benefit may not disqualify otherwise eligible schools solely because they are religious. Program rules and other legal questions still vary.

Are education savings accounts included in this dataset?

Yes, when identifiable private-school tuition or required fees can be isolated. Non-school ESA purchases are not silently counted as school tuition.

Are tax-credit scholarships counted as vouchers here?

They are included in the broad audit because “voucher” is often used as an umbrella term, but each row identifies the actual program type. Tax-credit scholarships are never silently merged with direct appropriations.

Why do some states show no percentage?

The current evidence set may lack a complete numerator, denominator, religious-affiliation match, or compatible reporting period. Not calculated is not zero.

Are all religious schools in these programs Christian?

No. The published analyses include Catholic, Protestant and other Christian schools as well as Jewish and Islamic schools. The available denomination mix varies by state and source.

Why is Texas included without a dollar percentage?

Texas launched funding in July 2026 and has no complete-year spending share. Its 80% figure is a February 4 snapshot of enrolled private schools classified as religious or faith-based, so it is labeled S1 rather than a funding percentage.

---

How to cite this page

The citation below identifies the editorial team, page title, publication, update date, and permanent URL. It is neutral attribution information for readers who independently reference the dataset.

The School Choice Index Editorial Team. “Percentage of School Voucher Funds Going to Religious Schools by State.” The School Choice Index. Last verified August 1, 2026. https://theschoolchoiceindex.com/research/percentage-of-school-voucher-funds-going-to-religious-schools-by-state/

Dataset: School Voucher Funds Going to Religious Schools by State — CSV, version 1.0.0.

---

Who produced this and why?

The School Choice Index Editorial Team produced the page from public program records, original data reporting, federal statistics, and a documented source-year audit. The School Choice Index is an independent, non-partisan editorial reference for U.S. K–12 school-choice programs. Its research section publishes sourced datasets and explanatory resources about those programs.

How: We reviewed current program inventories, administering-agency publications, official reporting rules, original state-data analyses, the NCES Private School Universe Survey, and the Supreme Court opinions listed below. Each percentage retains its denominator, period, program scope, source, and evidence class. Original arithmetic is shown where used.

Why: Claims about religious-school voucher funding routinely mix dollars, students, schools, tuition categories, partial periods, and estimates. This dataset exists to keep those measures separate and make every published value reproducible from its source trail.

Related research

• School voucher payments by private school — school-level payment records and disclosure formats.

---

Primary and original source references

These are the controlling sources for the claims and calculations on the page. State agencies and federal sources establish program records, reporting rules, national statistics, and legal holdings; original analysts establish the religious-affiliation percentages they calculated from those records.

• North Carolina State Education Assistance Authority — Opportunity Scholarship summary data
• Carolina Journal — 2025–26 North Carolina disbursement analysis
• Public Schools First NC — 2024–25 North Carolina voucher analysis
• Concord Monitor — 2024–25 New Hampshire EFA expenditure analysis
• Children’s Scholarship Fund New Hampshire — program and financial reports
• Missouri Independent/KSMU — August 2025 MOScholars invoice analysis
• Washington Post — Arizona, Florida, Indiana, Ohio, and Wisconsin state-data analysis
• Wisconsin Watch — 2025–26 voucher participation analysis
• Keystone Research Center — 2022–23 Pennsylvania estimate
• Pennsylvania DCED — current EITC reporting requirements
• Pennsylvania DCED — current OSTC reporting requirements
• Texas Observer — February 2026 participating-school analysis
• Texas Comptroller — March 17, 2026 participating-school update
• Texas Comptroller — June 30, 2026 initial-funding announcement
• NCES — 2023–24 Private School Universe Survey, NCES 2026-015
• EdChoice — current program inventory
• Alaska Department of Law — correspondence allotment and private-school-use guidance
• Illinois Department of Revenue — Education Expense Credit rules
• New Hampshire Administrative Code Ed 803.02 — provider disclosure requirements
• U.S. Supreme Court — Espinoza v. Montana Department of Revenue
• U.S. Supreme Court — Carson v. Makin

---

Structured data

This JSON-LD describes only content visible on this page and the CSV distribution listed above. The FAQ entries match the visible questions, and no spreadsheet distribution is marked up because no public XLSX URL is part of version 1.0.0.

{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "Article",
      "@id": "https://theschoolchoiceindex.com/research/percentage-of-school-voucher-funds-going-to-religious-schools-by-state/#article",
      "headline": "Percentage of School Voucher Funds Going to Religious Schools by State",
      "description": "Nine state benchmarks on voucher money reaching religious schools, with dates, denominators, sources, and a 51-jurisdiction audit.",
      "url": "https://theschoolchoiceindex.com/research/percentage-of-school-voucher-funds-going-to-religious-schools-by-state/",
      "mainEntityOfPage": {
        "@type": "WebPage",
        "@id": "https://theschoolchoiceindex.com/research/percentage-of-school-voucher-funds-going-to-religious-schools-by-state/"
      },
      "datePublished": "2026-08-01",
      "dateModified": "2026-08-01",
      "inLanguage": "en-US",
      "articleSection": "Research",
      "author": {
        "@type": "Organization",
        "name": "The School Choice Index Editorial Team",
        "url": "https://theschoolchoiceindex.com/authors/editorial-team/"
      },
      "publisher": {
        "@type": "Organization",
        "name": "The School Choice Index",
        "url": "https://theschoolchoiceindex.com/"
      },
      "about": [
        {
          "@type": "Thing",
          "name": "School vouchers"
        },
        {
          "@type": "Thing",
          "name": "Religious schools"
        },
        {
          "@type": "Thing",
          "name": "Education savings accounts"
        },
        {
          "@type": "Thing",
          "name": "Tax-credit scholarships"
        }
      ],
      "isPartOf": {
        "@type": "CreativeWorkSeries",
        "name": "The School Choice Index Research",
        "url": "https://theschoolchoiceindex.com/research/"
      }
    },
    {
      "@type": "Dataset",
      "@id": "https://theschoolchoiceindex.com/research/percentage-of-school-voucher-funds-going-to-religious-schools-by-state/#dataset",
      "name": "School Voucher Funds Going to Religious Schools: State Evidence Audit",
      "description": "A 50-state and District of Columbia source-year audit of reported dollar, recipient, school, and estimated shares associated with religious private schools, with every figure classified by what it measures.",
      "url": "https://theschoolchoiceindex.com/research/percentage-of-school-voucher-funds-going-to-religious-schools-by-state/",
      "version": "1.0.0",
      "datePublished": "2026-08-01",
      "dateModified": "2026-08-01",
      "creator": {
        "@type": "Organization",
        "name": "The School Choice Index Editorial Team",
        "url": "https://theschoolchoiceindex.com/authors/editorial-team/"
      },
      "publisher": {
        "@type": "Organization",
        "name": "The School Choice Index",
        "url": "https://theschoolchoiceindex.com/"
      },
      "spatialCoverage": {
        "@type": "Place",
        "name": "United States: 50 states and District of Columbia"
      },
      "temporalCoverage": "2022-07/2026-08",
      "measurementTechnique": "Published payment, expenditure, recipient, and participating-school analyses were catalogued and classified by numerator, denominator, program coverage, period, and evidence class. The New Hampshire reported-expenditure share and three NCES comparison statistics were calculated from published components with the arithmetic shown. No school’s religious affiliation was newly classified by the publisher.",
      "variableMeasured": [
        {
          "@type": "PropertyValue",
          "name": "Reported religious-school share"
        },
        {
          "@type": "PropertyValue",
          "name": "Evidence class",
          "description": "D1, D2, D3, R1, S1, E1, U1, N0, or X1"
        },
        {
          "@type": "PropertyValue",
          "name": "Program type"
        },
        {
          "@type": "PropertyValue",
          "name": "Reporting period"
        },
        {
          "@type": "PropertyValue",
          "name": "Program coverage"
        },
        {
          "@type": "PropertyValue",
          "name": "Calculation status"
        }
      ],
      "distribution": [
        {
          "@type": "DataDownload",
          "encodingFormat": "text/csv",
          "contentUrl": "https://theschoolchoiceindex.com/research/data/school-voucher-religious-schools-by-state.csv"
        }
      ]
    },
    {
      "@type": "FAQPage",
      "@id": "https://theschoolchoiceindex.com/research/percentage-of-school-voucher-funds-going-to-religious-schools-by-state/#faq",
      "mainEntity": [
        {
          "@type": "Question",
          "name": "What percentage of all school voucher money nationwide goes to religious schools?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "No defensible national percentage can be calculated from the available state figures because they cover different programs, years, denominators, and levels of completeness. The 75.8% national figure from NCES describes private-school students, not voucher spending."
          }
        },
        {
          "@type": "Question",
          "name": "Which state sends the highest percentage of voucher funds to religious schools?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "The evidence does not support a valid ranking. Missouri’s approximately 98% covers one month of one direct-aid stream, Indiana’s 98% is a voucher-recipient share, and North Carolina’s approximately 78% is a full-program dollar share."
          }
        },
        {
          "@type": "Question",
          "name": "Is the percentage of voucher students the same as the percentage of voucher dollars?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "No. Award amounts can differ by grade, income, disability status, program, and school, so a student or voucher share is not a dollar share."
          }
        },
        {
          "@type": "Question",
          "name": "Why are New Hampshire’s 86% and 48.9% figures both correct?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "The 86% figure is the religious share of tuition spending. Tuition was 56.9% of reported family expenditures, so 86% multiplied by 56.9% equals about 48.9% of reported family expenditures."
          }
        },
        {
          "@type": "Question",
          "name": "Can school voucher programs include religious schools?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "Yes. Under Espinoza v. Montana Department of Revenue and Carson v. Makin, a state that offers a generally available private-school tuition benefit may not disqualify otherwise eligible schools solely because they are religious. Program rules and other legal questions still vary."
          }
        },
        {
          "@type": "Question",
          "name": "Are education savings accounts included in this dataset?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "Yes, when identifiable private-school tuition or required fees can be isolated. Non-school ESA purchases are not silently counted as school tuition."
          }
        },
        {
          "@type": "Question",
          "name": "Are tax-credit scholarships counted as vouchers here?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "They are included in the broad audit because voucher is often used as an umbrella term, but each row identifies the actual program type and tax-credit scholarships are never merged silently with direct appropriations."
          }
        },
        {
          "@type": "Question",
          "name": "Why do some states show no percentage?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "The current evidence set may lack a complete numerator, denominator, religious-affiliation match, or compatible reporting period. Not calculated is not zero."
          }
        },
        {
          "@type": "Question",
          "name": "Are all religious schools in these programs Christian?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "No. The published analyses include Catholic, Protestant and other Christian schools as well as Jewish and Islamic schools. The available denomination mix varies by state and source."
          }
        },
        {
          "@type": "Question",
          "name": "Why is Texas included without a dollar percentage?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "Texas launched funding in July 2026 and has no complete-year spending share. Its 80% figure is a February 4 snapshot of enrolled private schools classified as religious or faith-based, so it is labeled S1 rather than a funding percentage."
          }
        }
      ]
    }
  ]
}