School types, places, dates and access to legal routes differ. The 12 reprieves come from nine grouped decision episodes, and some local decisions are still linked. Neither the cases nor the graphic order measure effectiveness.
Read the complete method and evidence limitations
Research cutoff: 22 September 2026. Version 2.0.
This selection cannot show a national survival rate or the chance that Kew Riverside Primary School stays open.
What this catalogue contains
The catalogue covers twelve selected English state-funded primary schools. In each, a proposed closure or site merger stopped, or provision continued through an academy successor, in decisions between December 2019 and June 2026. All twelve successor/current URNs were checked directly against the Department for Education's Get Information about Schools (GIAS) register on 22 September 2026. All were listed open. Hazlewood has a renewed closure proposal reported in September 2026. An open register entry does not prove a sustainable budget, enrolment recovery or protection against a future proposal.
Four more closure comparisons are included: Cowes, Broad Oak, St Bartholomew's and De Beauvoir. All four closure dates were checked directly in GIAS. Cowes also completes the original six-school 2024 Isle of Wight programme. It is counted only once among the sixteen named schools and is not counted among the twelve reprieves.
The comparison cases were chosen after the reprieve research. Broad Oak and Cowes link to authorities/programmes already covered. St Bartholomew's gives a different adjudication outcome, and De Beauvoir an urban community-school comparison. These comparisons were chosen on purpose to give context. They are not a matched control group or a complete set of unsuccessful campaigns. Adding them does not remove selection bias.
“Did not prevent closure” describes the observed outcome in a case where an effort is documented. It does not establish that the effort was useless or that it reduced the chance of success. St Bartholomew's secured an explicitly reasoned extra term for transition. Other benefits, counterfactual outcomes, total campaign spending and effort intensity are not measured. Council assessments are attributed, not presented as independently audited findings.
Selection and searches
Targeted searches used terms including primary school closure withdrawn, rejected, not to proceed, saved from closure and scrapped. Searches were refined to named schools, councils, dioceses, official adjudicator decisions and follow-up records. London cases and cases illustrating the initial consultation stage were prioritised. Council records and OSA determinations were preferred over campaign accounts. This is a purposive search, not a registered systematic review and not a census of every eligible school.
England was selected because Kew Riverside Primary School is an English maintained community school. Independent schools, Scottish/Welsh/Northern Irish schools, mothballed schools and academic-performance turnarounds are not pooled into the analytical set. A news report describing a school as saved is not itself evidence of a lasting recovery. St George's, Langton is retained as an explicitly labelled reporting-based case because it is a recent early-stage withdrawal.
Units and denominators
- Exhibit 1: 12 schools. Initial consultation 3; local decision after statutory notice 3; external intervention 6. An academy consultation is grouped by broad stage, not claimed to be the identical statutory process.
- Exhibit 2: 9 grouped decision episodes. Fenstanton and Holy Trinity share one adjudication. Brading, Oakfield and Wroxall share one Cabinet episode. These groups reduce obvious duplication, but other Isle of Wight and Lambeth episodes remain related.
- Exhibit 3: all 6 original schools in one named local programme. One removed before statutory notice, three retained by Cabinet, one retained after adjudication and Cowes closed on 31 August 2025. This denominator is defined locally; it is not representative of national proposals.
- Exhibit 4: a single school's contemporaneous financial forecasts, in pounds. These are forecast year-end balances, not annual funding needs or actual year-end results. The 2021/22 interval is an optimistic/cautious pupil scenario, not a confidence interval. The reported £74,368 support total combines £44,368 fundraising, £20,000 trust pledge and £10,000 parish council resolution. Do not label the total cash received or attribute every budget change solely to it.
- Exhibit 5: one longitudinal caution, Hazlewood; it does not estimate a repeat-proposal rate.
- Exhibit 6: an editorial priority order, without effectiveness scores, budgets, probabilities or claimed causal effects.
- Exhibit 7: four selected closure comparisons. Efforts, assessments and outcomes are documented separately; no tactic failure rate is calculated.
- Exhibit 8: online petition signature counts for all six original Isle of Wight schools, reported in paragraph 92 of the March 2025 Cabinet report. The online counts total 3,771. Paper counts are shown separately, never added to online counts because overlap is unknown. The snapshot relates to the initial autumn 2024 consultation, not final lifetime campaign totals. The outcome horizon is 31 August 2025. No adjustment is made for school size, campaign duration or signatory residence, and no effective threshold or causal effect can be identified.
The Hackney-wide 95% opposition figure is rounded from 207 of 219 clearly classifiable proposal-level responses. The process received 175 submissions; an individual could address several proposals. It is not 95% of 175 people, a De Beauvoir-only percentage, or a representative opinion poll. S25 sections 4.24-4.28 provide the denominator; S30 is the council's narrative summary.
Reason coding
One analyst reviewed the source materials and coded six overlapping themes for each grouped episode. Each row records its sources and a coding explanation in `data/catalogue.json`.
- 2: an explicit ground in the decision or the adopted recommendation/rationale. A finding of 2 does not mean the factor caused success independently of other factors.
- 1: a matter was raised, considered, or part of the institutional route, but was not established as a ground for the favourable outcome.
- 0: not established in the reviewed record. It does not mean the factor was absent or ineffective.
No independent second coding was performed. Do not present coding as inter-rater-validated research. For the three Isle of Wight Cabinet reprieves, the minutes establish the outcomes but do not isolate a school-specific explanation. The matrix therefore records the shared discussion without inventing a decisive cause. St George's has no factor coded 2 because attributed reporting does not expose a sufficiently examined decision rationale. In St John the Divine's case, mixed-age classes and community provision were discussed but were not necessary to the adjudicator's central conclusion.
The matrix covers mechanisms and institutional circumstances, not merely campaign tactics. For example, rural status is not something Kew can acquire through campaign effort. Partner support can be a route to adjudication without constituting the merits on which a proposal was rejected.
Evidence provenance
Eleven schools have primary decision/recommendation or judgment evidence; St George's has secondary reports quoting the trust. Fletching's proposed recommendation is paired with the council's later report confirming the 23 December 2019 decision. Pooles Park's High Court judgment was read through indexed text; direct retrieval was blocked. Its November 2023 ministerial decision is the catalogue's event date. Continued provision is verified through the current academy successor URN, not by treating the predecessor's administrative closure as lost provision.
The complete appendix lists all 45 sources used, including individual GIAS entries, with full URLs, publication/record dates, access dates, locators, inspection methods and claim IDs. The 49-entry ledger records case outcomes, observed efforts, follow-up, exhibits, contextual facts and editorial interpretations. Completeness means coverage of this pack, not every available source about these schools or all closure literature. Sources and claims are also supplied as JSON.
The register describes the consulted online versions; search-engine publication metadata was not substituted for dates printed within historical documents. Downloaded public-source PDFs and browser test intermediates are held outside the distributable package. The package includes original summaries, links, coding and the chart masters. Source URLs can change; recheck them before later publication.
Limits on inference
Selection initially focused on favourable outcomes. Four purposively selected closure comparisons broaden the evidence but do not provide a national denominator, matched failed-campaign dataset, representative sampling frame or standard follow-up horizon. School type, geography, funding arrangements, date, reasons for closure and access to remedies differ. Different sources also report some issues much more fully than others. The presence of a tactic cannot establish that it increased the probability of success. The absence of a source mention cannot show that a tactic was unused.
So do not calculate campaign success odds, p-values, causal rankings or predicted returns on effort from this dataset. The valid use is to identify concrete hypotheses and evidence to test for Kew. To estimate a rate, first define a jurisdiction, period, entry stage and outcome horizon. Then gather all proposals, including closures, withdrawals, mergers, conversions, deferrals and unresolved cases. Preserve successor links and follow each case to the same horizon before calculating a denominator.
Application to Kew Riverside Primary School
The proposal page retrieved on 22 September describes an initial/pre-statutory consultation. The current DfE closing-schools guidance applies from 1 September 2026. Historic legal processes should not be copied without checking present guidance and the school's type. Kew is a community school. Do not assume that a voluntary-aided governing body's or diocese's referral route gives every parent or community school a general appeal right. Do not assume that the rural-school presumption applies either.
The suggested priorities are a costed operating alternative and an audit of demand and net closure savings. They also include evidence of effects on families, formal participation and a specific process check where justified. The historical evidence does not show that any Kew alternative is already viable. School-specific budgets, partner commitments, pupil assumptions and continuing PFI obligations still need to be checked.
