Education · England · state-funded schools · 2024/25

Strong English and maths GCSE passes: one in four disadvantaged pupils, half of other pupils.

England, state-funded schools, 2024/25: 25.8% of disadvantaged pupils achieved grade 5 or above in both English and maths, compared with 53.1% of other pupils, the same rates as in 2023/24. The 27.3-point gap is wider than the 25.2 points of 2018/19, the last pre-pandemic exam year. These thresholds are not a complete measure of learning.

Priority: Improve learning and opportunity, especially for pupils missing effective support. What would change it →

Evidence last reviewed 10 October 2026

School measures cover England and the stated pupil cohort; international comparisons and UK youth outcomes are labelled separately.

Is the disadvantage gap already there before secondary school?

Disadvantage is visible at both primary school and GCSE

Share of pupils reaching each stage’s standard, disadvantaged pupils and all other pupils, England, 2024/25. Two separate measures on two separate cohorts.

England · Academic year 2024/25
Disadvantaged pupilsAll other pupils

Age 11: expected standard in reading, writing and maths, 2024/25

Year 6 pupils, key stage 2. Published to whole percentages.

Age 16: grade 5 or above in English and maths GCSE, 2024/25

Year 11 pupils, key stage 4. Grade 5 is a “strong pass”.

By age 11, 47% of disadvantaged pupils reach the expected standard in reading, writing and maths, against 69% of their peers. At 16 the gap in strong English and maths passes is 27.3 points. Both gaps are wider than in 2018/19, the last exam year before the pandemic. This shows how far apart the groups are, not why: disadvantage overlaps with special educational needs, attendance and where pupils live, and the chart does not separate those.

England · Academic year 2024/25 · Official statistics (key stage 2: revised; key stage 4: final)

Sources, definitions and data — Disadvantage is visible at both primary school and GCSE
Geography:
England — Pupils in state-funded schools at the end of key stage 2 (age 10–11) and key stage 4 (age 15–16)
Measure:
% of pupils in each group meeting that panel’s standard; gap in percentage points
Period:
Academic year 2024/25
Status:
Official statistics (key stage 2: revised; key stage 4: final)
Calculation:
Gap = other pupils’ rate minus disadvantaged pupils’ rate, in percentage points. Rates as published by DfE.
Compare with:
2018/19, the last exam year before the pandemic: 20.0-point gap at 11, 25.2-point gap at 16
Not included / breaks:
Disadvantaged means eligible for free school meals at any point in the past six years, looked after, or adopted from care — a proxy for household income, not a measure of it. The two panels are different children and different standards, so the gaps cannot be compared as one child’s journey. DfE’s provisional 2025/26 key stage 2 figures (49% against 70%, which DfE reports as a 22-point gap) are not yet in this chart; they exclude some looked-after children.

Source: DfE, Key stage 4 performance, academic year 2024/25, National characteristics data file (disadvantage status; SEN provision) (released 23 April 2026 (updated 4 August 2026)). Chart EDU-01.

View the data as a table
MeasureYearDisadvantagedOther pupilsGap (points)
Age 11: expected standard, reading, writing and maths2024/2547%69%22.0
Age 11: expected standard, reading, writing and maths2018/1951%71%20.0
Age 16: grade 5+ in English and maths2024/2525.8%53.1%27.3
Age 16: grade 5+ in English and maths2018/1924.7%49.9%25.2

Scroll the table sideways to see every column.

Overview

The short version

Three key measures

Who feels this most?

The disadvantage gap is not the only divide. Pupils with special educational needs are far less likely to leave school with strong English and maths: 7.6% of those with an education, health and care plan did in 2024/25, against 45.4% of all pupils. Many pupils are in both groups. Whether support for special needs arrives in time is a question of its own — the SEND chapter takes it up.

The evidence

From early disadvantage to attendance and attainment

The gap starts early and survives into secondary school

These charts follow successive cohorts, not the same children. A thirty-year perspective is useful for policy history, but today’s attainment measures do not have thirty years of comparable results. Explore the reforms and measurement breaks.

Has the gap narrowed?

The primary gap persists across successive cohorts

% reaching the expected standard in reading, writing and maths

England · 2015/16–2024/25
% of pupilsAssessments disrupted0%50%100%20162020202447%69%
End-of-primary pupils in England. The current expected-standard measure starts in 2016; earlier levels are not spliced in. Writing assessment changed in 2018, so the line breaks there. Tests were cancelled in 2019/20 and 2020/21. Each year follows different children; these are not individual progress trajectories.

England · 2015/16–2024/25 · Official statistic

Sources, definitions and data — The primary gap persists across successive cohorts
Geography:
England
Measure:
% reaching the expected standard in reading, writing and maths
Period:
2015/16–2024/25
Status:
Official statistic
Calculation:
Published percentages; missing years left blank. Academic years plotted at year end.
Not included / breaks:
End-of-primary pupils in England. The current expected-standard measure starts in 2016; earlier levels are not spliced in. Writing assessment changed in 2018, so the line breaks there. Tests were cancelled in 2019/20 and 2020/21. Each year follows different children; these are not individual progress trajectories.

Source: DfE, Key stage 2 attainment, academic year 2024/25 (revised), ks2 by pupil characteristics data file. Chart EDU-04.

View the data as a table
Academic yearDisadvantaged (%)Other pupils (%)
2015/163960
2016/174767
2017/185170
2018/195171
2019/20: no comparable observationNot availableNot available
2020/21: no comparable observationNot availableNot available
2021/224366
2022/234466
2023/244667
2024/254769

Has the gap narrowed?

Strong GCSE passes remain far less common among disadvantaged pupils

% achieving grade 5+ in both English and maths

England · 2018/19–2024/25
% of pupilsAssessments disrupted0%50%100%201920202021202220232024202526%53%
End-of-KS4 pupils in state-funded schools, England. Grade 5 began in 2017; this checked dataset starts in 2019. Earlier archived group values require separate definition review and are not shown. Teacher-assessed 2019/20 and 2020/21 omitted; 2021/22 grading was transitional. Free-school-meal protections also change group composition.

England · 2018/19–2024/25 · Official statistic

Sources, definitions and data — Strong GCSE passes remain far less common among disadvantaged pupils
Geography:
England
Measure:
% achieving grade 5+ in both English and maths
Period:
2018/19–2024/25
Status:
Official statistic
Calculation:
Published percentages; missing years left blank. Academic years plotted at year end.
Not included / breaks:
End-of-KS4 pupils in state-funded schools, England. Grade 5 began in 2017; this checked dataset starts in 2019. Earlier archived group values require separate definition review and are not shown. Teacher-assessed 2019/20 and 2020/21 omitted; 2021/22 grading was transitional. Free-school-meal protections also change group composition.

Source: DfE, Key stage 4 performance, academic year 2024/25, National characteristics data file (disadvantage status; SEN provision) (released 23 April 2026 (updated 4 August 2026)). Chart EDU-05.

View the data as a table
Academic yearDisadvantaged (%)Other pupils (%)
2018/1924.749.9
2019/20: no comparable observationNot availableNot available
2020/21: no comparable observationNot availableNot available
2021/2229.657
2022/2325.252.4
2023/2425.853.1
2024/2525.853.1
Explore annual values: The primary gap persists across successive cohorts

End-of-primary pupils in England. The current expected-standard measure starts in 2016; earlier levels are not spliced in. Writing assessment changed in 2018, so the line breaks there. Tests were cancelled in 2019/20 and 2020/21. Each year follows different children; these are not individual progress trajectories. A dash means no comparable observation is supplied; it is not zero. Values describe separate cohorts.

1996–2026; % reaching the expected standard in reading, writing and maths. No change is calculated across definition or assessment breaks.

The primary gap persists across successive cohorts; % reaching the expected standard in reading, writing and maths
YearDisadvantaged pupilsOther pupils
1996——
1997——
1998——
1999——
2000——
2001——
2002——
2003——
2004——
2005——
2006——
2007——
2008——
2009——
2010——
2011——
2012——
2013——
2014——
2015——
20163960
20174767
20185170
20195171
2020——
2021——
20224366
20234466
20244667
20254769
2026——
Explore annual values: Strong GCSE passes remain far less common among disadvantaged pupils

End-of-KS4 pupils in state-funded schools, England. Grade 5 began in 2017; this checked dataset starts in 2019. Earlier archived group values require separate definition review and are not shown. Teacher-assessed 2019/20 and 2020/21 omitted; 2021/22 grading was transitional. Free-school-meal protections also change group composition. A dash means no comparable observation is supplied; it is not zero. Values describe separate cohorts.

1996–2026; % achieving grade 5+ in both English and maths. No change is calculated across definition or assessment breaks.

Strong GCSE passes remain far less common among disadvantaged pupils; % achieving grade 5+ in both English and maths
YearDisadvantaged pupilsOther pupils
1996——
1997——
1998——
1999——
2000——
2001——
2002——
2003——
2004——
2005——
2006——
2007——
2008——
2009——
2010——
2011——
2012——
2013——
2014——
2015——
2016——
2017——
2018——
201924.749.9
2020——
2021——
202229.657
202325.252.4
202425.853.1
202525.853.1
2026——
What does the “19-month disadvantage gap” mean?

The Education Policy Institute’s 2026 report estimates a 19.0-month secondary attainment gap in 2025, against 18.8 months in 2024 on its revised EverFSM definition. These are attainment units, not months of teaching missed. Its earlier 19.1-month estimate for 2024 used Ever6 FSM: the two report vintages must not be joined into a trend. Read EPI’s method and findings.

What those pressures can mean in everyday life

Three illustrative situations, not real people, testimony or predictions. Illustration generated with AI.

Editorial illustration of a pupil studying, a family discussing support, and a young adult considering study and practical training.

A pupil trying to catch up

A disadvantaged Year 10 pupil attends lessons but needs extra practice in English and maths.

Possible response: Assess specific learning gaps, then offer structured small-group or individual tuition linked to classroom teaching.

Limits and a test of success

Average trial effects cannot predict this pupil’s grades. Attendance, teaching quality and the match to their needs matter.

Track learning against a baseline and participation, without promising a pass.

Attainment and tutoring evidence

A family waiting for support

A child needs specialist support while the family waits for an assessment and, if required, an EHC plan.

Possible response: Coordinate school, health and local-authority support around the child, and track what is actually provided.

Limits and a test of success

A plan issued within 20 weeks is a timeliness measure. It does not show whether provision starts or meets the child’s needs.

Monitor time to support, provision delivered and family experience alongside statutory deadlines.

The SEND pathway and evidence

A young person choosing a route

A 17-year-old is weighing college, an apprenticeship and the cost of travelling to a placement.

Possible response: Provide clear route advice, access to placements and support to complete the chosen qualification.

Limits and a test of success

Apprenticeship starts are not completions. UK 16–24 NEET estimates include older young adults and economic inactivity.

Follow sustained participation, completion and destinations for the same cohort.

Routes from classroom to work

Parents’ reported confidence · England · May–June 2025

Would school listen to a concern about additional needs?

DfE asked parents whether they were confident their child’s school would listen to concerns about a potential special educational need. This measures confidence, not whether support was delivered or children’s needs were met.

Primary school parents

80% very or fairly confident

Published 95% interval: 77.2–83.2%. Unweighted sample: 1,138.

Secondary school parents

72% very or fairly confident

Published 95% interval: 69.0–75.3%. Unweighted sample: 1,694.

Question, sample and limitations

Exact question: “How confident are you that [YPFnam]’s school would listen to any concerns you may have about a potential special educational need affecting your child's learning or development?”

Parents of pupils in state-funded primary and secondary schools, years 1–11; fieldwork 21 May–13 June 2025. Online panel; probability recruitment from National Pupil Database; subsequent email/SMS invitations. Recruitment weights plus wave-specific response/attrition weighting; technical report pp.30–35.

Published 95% confidence intervals, calculated with SPSS Complex Samples; not a simple-random-sample approximation. Self-reported confidence, not measured school provision. Special-school parents excluded from this question. Nonresponse and panel attrition can remain after weighting. One dated wave; no trend or causal comparison.

Source table: senidentification_listen. Estimates use the published combined response and its interval; the intervals for separate answers were not added together.

DfE survey release and methods · Original tables

Absence is a separate warning sign

Pupils cannot learn in lessons they miss. Persistent absence roughly doubled after the pandemic and has come down only part of the way: in 2024/25, 18.1% of pupils missed at least one day in ten. It is shown here on its own, not laid over the attainment gap, because the two measure different things and the data do not show that one drives the other.

Is persistent absence back to where it was before the pandemic?

Persistent absence has eased from its peak but remains well above 2018/19

Pupil enrolments missing 10% or more of their possible sessions, state-funded primary, secondary and special schools, England, 2015/16 to 2024/25. Each academic year is plotted at the year it ends (2025 = 2024/25).

England · 2015/16 to 2024/25 (no data for 2019/20)

1996–2026 displayed window · observations only. Shading marks years outside the displayed record.

% of pupil enrolmentsNo observations shown0%10%20%2000201020202018/19: 10.9%18.1%2021/22 peak 22.5%
Persistent absence rose from 10.9% of pupils in 2018/19 to 22.5% in 2021/22, and has fallen in each year since, to 18.1% in 2024/25. Severe absence — missing half of sessions or more — has kept rising, to 2.4% (176 thousand enrolments), three times the 2018/19 rate. Absence and low attainment go together, but this chart does not show that one causes the other.

England · 2015/16 to 2024/25 (no data for 2019/20) · Official statistics

Sources, definitions and data — Persistent absence has eased from its peak but remains well above 2018/19
Geography:
England — Pupil enrolments in state-funded primary, secondary and special schools, full academic year
Measure:
% of pupil enrolments
Period:
2015/16 to 2024/25 (no data for 2019/20)
Status:
Official statistics
Calculation:
As published by DfE. The chart starts in 2015/16, when the current definition (10% of each pupil’s own possible sessions) began.
Compare with:
2018/19, before the pandemic: 10.9%
Not included / breaks:
No figures for 2019/20; 2020/21 was affected by school closures and is not a normal year. A termly release (autumn 2025/26, 18.10%) is not comparable with full-year figures and is not plotted.

Source: DfE, Pupil absence in schools in England, 2024/25, Absence for persistent and severe absentees (time series) (released 26 March 2026). Chart EDU-02.

View the data as a table
Academic yearPersistent absence (%)
2015/1610.5
2016/1710.8
2017/1811.2
2018/1910.9
2019/20no data
2020/2112.1
2021/2222.5
2022/2321.2
2023/2420.0
2024/2518.1

The questions the chapters follow

A sequence of questions, not a measured causal chain: each step has its own data, population and date, and the arrows do not mean one step causes the next.

Questions explored — not a quantified causal model

  1. 1

    Being in school, with the right support

    Persistent absence 18.1% (2024/25); 46.1% of new EHC plans on time (2025).

    More on this step
  2. 2

    Qualifications

    At 16, a 27.3-point gap in strong English and maths passes between disadvantaged pupils and the rest.

    More on this step
  3. 3

    Post-16 destinations

    Sixth form, college, apprenticeships, university or none — and whether pupils stay there. Measured separately, years later.

    More on this step
  4. 4

    Work and pay

    Young people not in education, employment or training are counted in a different survey, for the whole UK.

    More on this step

Who feels this most?

Which pupils are least likely to leave school with strong English and maths?

Fewer than one in ten pupils with an EHC plan gets a strong pass in English and maths

Pupils achieving grade 5 or above in English and maths GCSE, by disadvantage and by special educational needs, England, 2024/25 (filled dot) and 2018/19 (open dot).

England · 2024/25 and 2018/19
2024/252018/19All pupils 45.4%
  • Not known to be disadvantaged53.1%
  • Disadvantaged25.8%
  • No identified SEN51.8%
  • SEN support22.5%
  • EHC plan7.6%
Disadvantage and special educational needs both mark wide divides in GCSE results. 22.5% of pupils on SEN support and 7.6% of pupils with an EHC plan got a strong pass, against 51.8% of pupils with no identified need. The groups overlap — pupils with SEN are more likely to be disadvantaged — so these are not independent effects. Every group did better than in 2018/19, but disadvantaged pupils gained least — which is why the gap widened.

England · 2024/25 and 2018/19 · Official statistics

Sources, definitions and data — Fewer than one in ten pupils with an EHC plan gets a strong pass in English and maths
Geography:
England — Pupils at the end of key stage 4 in state-funded schools
Measure:
% of pupils in each group
Period:
2024/25 and 2018/19
Status:
Official statistics
Calculation:
As published by DfE.
Compare with:
All pupils: 45.4%
Not included / breaks:
Two separate breakdowns of the same cohort: a pupil can be both disadvantaged and on SEN support, so the rows overlap and must not be added. Some pupils with an EHC plan follow programmes where GCSEs are not the goal.

Source: DfE, Key stage 4 performance, academic year 2024/25, National characteristics data file (disadvantage status; SEN provision) (released 23 April 2026 (updated 4 August 2026)). Chart EDU-03.

View the data as a table
Group2024/252018/19
Not known to be disadvantaged53.1%49.9%
Disadvantaged25.8%24.7%
No identified SEN51.8%48.2%
SEN support22.5%16.8%
EHC plan7.6%5.5%

Local figures differ again: how do Hampshire’s schools and SEND services compare?

Better support needs people, time and usable funding

Are trained teachers staying?

Recruitment must replace teachers who leave

Share still teaching five years later; horizontal axis is qualification year

England · Qualifying cohorts 1997–2020

1996–2026 displayed window · observations only. Shading marks years outside the displayed record.

% retained after five years0%50%100%20002010202067%
A teacher who changes school can remain in this measure. Leaving state-funded teaching in England is not necessarily leaving teaching altogether. Recent qualifying cohorts have not yet completed five years and are not treated as zero.

England · Qualifying cohorts 1997–2020 · Official statistic

Sources, definitions and data — Recruitment must replace teachers who leave
Geography:
England
Measure:
% of qualifying cohort
Period:
Qualifying cohorts 1997–2020
Status:
Official statistic
Calculation:
As published; horizontal axis is qualification year, not observation year.

Source: Teacher retention (released 2026-06-04). Chart EDU-07.

View the data as a table
Qualification cohort and observation yearRetained (%)
Qualified 1997; observed 200271.1
Qualified 1998; observed 200371.6
Qualified 1999; observed 200471.2
Qualified 2000; observed 200571.9
Qualified 2001; observed 200670.5
Qualified 2002; observed 200771.4
Qualified 2003; observed 200871.7
Qualified 2004; observed 200971.6
Qualified 2005; observed 201073.9
Qualified 2006; observed 201173.6
Qualified 2007; observed 201273.8
Qualified 2008; observed 201373.7
Qualified 2009; observed 201472.8
Qualified 2010; observed 201570.8
Qualified 2011; observed 201670.1
Qualified 2012; observed 201768.5
Qualified 2013; observed 201867.9
Qualified 2014; observed 201967.2
Qualified 2015; observed 202068.8
Qualified 2016; observed 202169.1
Qualified 2017; observed 202269
Qualified 2018; observed 202367.9
Qualified 2019; observed 202468
Qualified 2020; observed 202567.4

How much funding is available for each pupil?

Follow real funding alongside the pressures it must cover

Revenue funding per pupil, 2025/26 prices

England · 2010/11–2026/27
£ per pupil, constant prices£0£5k£10k20102014201820222026£8,390
Funding is the amount allocated, not what schools actually spent. The GDP deflator measures economy-wide prices rather than school costs. Changes in SEND demand, pay and estate condition affect what the money can provide; this chart cannot establish the effect of spending on attainment.

England · 2010/11–2026/27 · Official statistics; latest settlement includes estimates

Sources, definitions and data — Follow real funding alongside the pressures it must cover
Geography:
England — State-funded education, ages 5–16
Measure:
£ per pupil, GDP-deflated
Period:
2010/11–2026/27
Status:
Official statistics; latest settlement includes estimates
Calculation:
DfE published real-terms series; no additional deflation.
Not included / breaks:
Revenue funding for state-funded education ages 5–16, not expenditure or a school-specific cost index. Pupil counts are January preceding the financial year. Latest settlement contains estimates.

Source: Revenue funding to state-funded schools in England for pupils aged 5-16, in cash and real terms, 2010-11 to 2026-27 (released 2026-01-29). Chart EDU-06.

View the data as a table
Financial year£ per pupil (2025/26 prices)Pupils FTEStatus
2010/1177606761000published funding
2011/1277606788000published funding
2012/1377606834000published funding
2013/1477406902000published funding
2014/1577206967000published funding
2015/1677607063000published funding
2016/1776007183000published funding
2017/1875007305000published funding
2018/1974807404000published funding
2019/2075307502000published funding
2020/2175607581000published funding
2021/2278307613000published funding
2022/2378207673000published funding
2023/2479507745000published funding
2024/2581407772000published funding
2025/2684007725000published funding
2026/2783907685000estimate

Scroll the table sideways to see every column.

Which subjects meet recruitment targets?

Teacher recruitment differs sharply by subject

Postgraduate initial teacher training entrants against targets, 2025/26

England · 2025/26
2025/26Target met
  • Chemistry120%
  • Computing81%
  • English107%
  • Mathematics114%
  • Physics78%
A rate above 100% means the target was exceeded. Targets themselves change with forecast need, and entrants are not yet qualified teachers. The November vacancy census and five-year retention answer different questions about capacity.

England · 2025/26 · Official statistic; current intake provisional

Sources, definitions and data — Teacher recruitment differs sharply by subject
Geography:
England
Measure:
% of subject recruitment target
Period:
2025/26
Status:
Official statistic; current intake provisional
Calculation:
100 × postgraduate entrants / modelled target. Five selected core/STEM subjects; not a full ranking.

Source: DfE, Initial teacher training census. Chart EDU-08.

View the data as a table
SubjectEntrantsTargetTarget met (%)
Chemistry878730120.3
Computing72389580.8
English20831950106.8
Mathematics26282300114.3
Physics1095141077.7

Scroll the table sideways to see every column.

SEND: count need, waiting and provision separately

+215.0%Jan 2010–Jan 2026England · plans
No observationsshown0400K800K19962026
228,221 → 718,838Jan 2010–Jan 2026
England, ages 0–25. Statements became EHC plans from 2014 and the age range expanded; growth is not simply a measure of prevalence.
Sources and data

Statements were replaced by EHC plans from September 2014, which can run to age 25 (statements usually ended at 16–19). Part of the rise since 2015 is that wider age range, mostly young people in further education. 2014 reform: statements → EHC plans, age range extended to 0–25.

Source dataset →
PeriodplansStatus
Jan 2010228,221Outturn
Jan 2011229,017Outturn
Jan 2012230,156Outturn
Jan 2013233,431Outturn
Jan 2014237,111Outturn
Jan 2015240,183Outturn
Jan 2016256,315Outturn
Jan 2017287,290Outturn
Jan 2018319,819Outturn
Jan 2019353,995Outturn
Jan 2020390,109Outturn
Jan 2021430,697Outturn
Jan 2022473,255Outturn
Jan 2023517,048Outturn
Jan 2024576,474Outturn
Jan 2025638,745Outturn
Jan 2026718,838Outturn

Plan numbers describe caseload. Timeliness describes the assessment process. Neither establishes how much support children receive or how well it works.

Plans, specialist capacity and support

Buildings and budgets are part of delivery

The National Audit Office’s school-buildings audit identifies condition and safety risks. Its dated estimates are not a current count of unsafe schools. Funding allocations, actual expenditure and estate condition need separate scrutiny.

Read the NAO audit

Funding and expenditure · Workforce detail

Four education systems, a shared international test

PISA 2022 assessed samples of 15-year-olds. These means and approximate 95% confidence intervals show uncertainty, not a precise league table. All four UK nations carry OECD sampling cautions. The UK aggregate is omitted here to avoid counting the same pupils twice.

maths

PISA 2022 score · axis 440–530 (not zero)

  • England *492

    95% interval 487.1–497.3

  • Northern Ireland *475

    95% interval 469.3–481.0

  • Scotland *471

    95% interval 465.5–475.8

  • Wales *466

    95% interval 459.5–471.9

Dashed reference: OECD average 472.4.

reading

PISA 2022 score · axis 440–530 (not zero)

  • England *496

    95% interval 490.9–501.8

  • Northern Ireland *485

    95% interval 478.6–491.8

  • Scotland *493

    95% interval 486.1–499.4

  • Wales *466

    95% interval 458.3–472.9

Dashed reference: OECD average 475.6.

science

PISA 2022 score · axis 440–530 (not zero)

  • England *503

    95% interval 497.5–508.2

  • Northern Ireland *488

    95% interval 482.0–494.7

  • Scotland *483

    95% interval 477.2–489.4

  • Wales *473

    95% interval 465.5–480.2

Dashed reference: OECD average 484.6.

View PISA data as a table
PISA 2022: published scores, standard errors and sampling cautions
SubjectNationMean scoreStandard errorSampling caution
mathsEngland492.222.59Yes
mathsNorthern Ireland475.142.97Yes
mathsScotland470.622.63Yes
mathsWales465.713.17Yes
readingEngland496.382.78Yes
readingNorthern Ireland485.203.38Yes
readingScotland492.783.40Yes
readingWales465.623.74Yes
scienceEngland502.842.73Yes
scienceNorthern Ireland488.373.24Yes
scienceScotland483.313.11Yes
scienceWales472.813.75Yes
Source, uncertainty and limits

Intervals calculated as mean ± 1.96 × published standard error. Overlap is not a test of whether two means differ. * One or more sampling standards were not met; intervals do not account for possible non-response bias. Scores across subjects are different constructs. These results cannot establish the effect of funding, curriculum or a particular policy. PISA 2022 Results, Volumes I and II, published 2023-12-05.

Compare with Europe and inspect the full data · University funding across countries

Shareable visual summary

Visual summary

Image held

The image is not offered for download or sharing until the reasons below are resolved. The chart here is drawn from today’s records.

Strong English and maths GCSE passes: one in four disadvantaged pupils, half of other pupils.

In 2024/25, 25.8% of disadvantaged pupils got grade 5+ in both English and maths at 16, against 53.1% of other pupils: a gap of 27.3 percentage points. At age 11 the gap was 22 points (reading, writing and maths). Different cohorts and assessments: not one widening gap.

Share reaching the threshold, England, 2024/25

GroupDisadvantagedAll other pupilsGap
Age 11: expected standard in reading, writing and maths47%69%22 points
Age 16: grade 5+ in English and maths25.8%53.1%27.3 points

Why the supplied image is held

  • The age-11 figures in the image (47% against 69%) are from 2024/25; DfE’s provisional 2025/26 figures (49% against 70%) are newer. The gap is 22 points in both.
Text and data version

Strong English and maths GCSE passes: one in four disadvantaged pupils, half of other pupils. In 2024/25, 25.8% of disadvantaged pupils got grade 5+ in both English and maths at 16, against 53.1% of other pupils: a gap of 27.3 percentage points. At age 11 the gap was 22 points (reading, writing and maths). Different cohorts and assessments: not one widening gap.

Records behind this visual summary
MeasureValuePeriodGeographyTypeSource
KS2 disadvantage threshold gapCalculation: other pupils − disadvantaged pupils (threshold gap, percentage points)22 points2024-25Englandderived calculationDepartment for Education, Key stage 2 attainment
KS4 disadvantage threshold gap (grade 5+ English and maths)Calculation: other pupils − disadvantaged pupils (threshold gap, percentage points)27.3 points2024-25Englandderived calculationDepartment for Education, Key stage 4 performance

All visual summaries, their source records and their publication state are listed on Shareable visuals.

Go deeper

  • Who gets left behind, and has the gap moved?
    -3.7%2010-11–2024-25England · index
    No observationsshown02.5519962026
    4.1 → 3.92010-11–2024-25
    Lower means a narrower attainment gap. An index, not a percentage; grading and assessment changed during the pandemic.
    Sources and data

    The disadvantage gap index is a rank-based measure on a 0 to 10 scale, where a higher number means a wider gap. It is NOT a percentage, not a percentage-point difference, and not a difference in scores. An index of 3.92 does not mean 3.92 of anything — it is a position on a constructed scale, designed to be robust to the grading changes that make raw score gaps incomparable over time. 2019/20 and 2020/21 grades were teacher-assessed (Covid); the index for those years is not comparable.

    Source dataset →
    PeriodindexStatus
    2010-114.07Outturn
    2011-123.89Outturn
    2012-133.8Outturn
    2013-143.74Outturn
    2014-153.79Outturn
    2015-163.78Outturn
    2016-173.66Outturn
    2017-183.68Outturn
    2018-193.7Outturn
    2019-203.66Outturn
    2020-213.79Outturn
    2021-223.84Outturn
    2022-233.94Outturn
    2023-243.93Outturn
    2024-253.92Outturn
    Outcomes, the disadvantage index explained once, and what it cannot show.Read the chapter → Who gets left behind, and has the gap moved?
  • How many children are missing school?
    -1.2percentage points2006-07–2024-25England · %
    No observationsshown012.52519962026
    19.3% → 18.1%2006-07–2024-25
    Missing at least 10% of sessions. No comparable full-year measure for 2019/20; school closures also affect 2020/21.
    Sources and data

    Pupils missing 10% or more of sessions (persistent) or 50% or more (severe), state-funded primary, secondary and special schools. No figures for 2019/20 (Covid). No data for 2019/20; 2020/21 affected by Covid closures.

    Source dataset →
    Period%Status
    2006-0719.3Outturn
    2007-0818.5Outturn
    2008-0918.5Outturn
    2009-1017.4Outturn
    2010-1116.3Outturn
    2011-1213.5Outturn
    2012-1313.6Outturn
    2013-1410.7Outturn
    2014-1511Outturn
    2015-1610.5Outturn
    2016-1710.8Outturn
    2017-1811.2Outturn
    2018-1910.9Outturn
    2020-2112.1Outturn
    2021-2222.5Outturn
    2022-2321.2Outturn
    2023-2420Outturn
    2024-2518.1Outturn
    Persistent and severe absence, exclusions, and children not on any roll.Read the chapter → How many children are missing school?
  • SEND: does support arrive in time?
    -14.3percentage points2019–2025England · %
    No observationsshown0408019962026
    60.4% → 46.1%2019–2025
    Plans issued within 20 weeks, excluding exceptions. This does not measure when support begins.
    Sources and data

    Calendar year. New EHC plans issued in the year, counted from the date the request was received to the date the final plan was issued. The legal limit is 20 weeks (SEND Regulations 2014, reg. 13). The headline excludes cases where a statutory exception applied. It covers plans that were issued — not requests refused, withdrawn or still open — and says nothing about whether the support in the plan was delivered.

    Source dataset →
    Period%Status
    201960.4Outturn
    202058Outturn
    202159.9Outturn
    202249.2Outturn
    202350.3Outturn
    202446.4Outturn
    202546.1Outturn
    Plans, SEN support, delays, provision, appeals and funding.Read the chapter → SEND: does support arrive in time?

All thirteen chapters

The full section, in order. Most school-system measures cover England; each chapter labels its population and geography. Compare all four UK nations with Europe: Britain vs Europe.

  1. 1Eighty years of education
  2. 2What education costs
  3. 3Are children learning more?
  4. 4Who runs our schools?
  5. 5Special educational needs and disabilities
  6. 6Attendance and missing education
  7. 7Behaviour and exclusion
  8. 8Teachers and the workforce
  9. 9Who gets left behind?
  10. 10From classroom to work
  11. 11The price of a degree
  12. 12Where education goes next
  13. 13Childcare and early years

Evidence → choices → outcomes

What could improve it?

A narrower gap between disadvantaged pupils and the rest in England, with better learning and attendance for all.

These options link to their national assessment, including funding, delivery constraints and remaining uncertainties.

  • Assessed option · provisional

    Skills in the building trades

    8/10 criteria scored

    Coverage of the appraisal, not an impact score.

    Delivery: Amber

    Whole-life costs incomplete

    Evidence and delivery constraints

    Historical apprenticeship returns support consideration, not the marginal return from this shortage-trades expansion.

    Identify trades and locations; secure paid placements, trainers, facilities and employer demand; measure completion and retention.

    Growth and prosperity listEvidence and assessment

  • Assessed option · provisional

    Make childcare places work for working parents

    Test additional capacity, affordability and parental employment.

    8/10 criteria scored

    Coverage of the appraisal, not an impact score.

    Delivery: Amber

    Whole-life costs incomplete

    Evidence and delivery constraints

    IFS finds employment effects depend on the hours offered. This is indirect evidence for filling today’s local supply gaps. A September 2026 process evaluation exists; it is not a causal national impact estimate.

    Verify unmet demand, existing awards and provider viability; cost places, workforce, premises, operating subsidy and parental fees. Track access and sustained parental work. Proposed owners: councils, DfE and providers in England; separate devolved arrangements.

    Growth and prosperity listEvidence and assessment

  • Assessed option · provisional

    Close the early-language support gap

    Reach eligible children with evidence-based language support.

    9/10 criteria scored

    Coverage of the appraisal, not an impact score.

    Delivery: Amber

    Whole-life costs incomplete

    Evidence and delivery constraints

    EEF effectiveness trial supports language outcomes; the transfer to a gap-closing delivery programme and persistence of effects need checking. Programme availability alone is not a verified access gap.

    Audit reach, eligibility and delivery fidelity; cost trained staff, release time, screening, materials and follow-up. Protect SEND provision. Measure language outcomes and access. Proposed owners: schools, education bodies and councils as appropriate.

    Equality and opportunity listEvidence and assessment

  • Assessed option · provisional

    Build family hubs around what worked

    Evaluated Sure Start features, weakest-access areas first, and a phased, evaluable roll-out of the family hubs.

    9/10 criteria scored

    Coverage of the appraisal, not an impact score.

    Delivery: Amber

    Whole-life costs incomplete

    Evidence and delivery constraints

    Sure Start evidence supports defined features; it does not validate all current hub models or immediate benefits.

    Specify service intensity and target reach; staff health/family services; measure uptake, fidelity and long-term outcomes.

    Wellbeing listEvidence and assessment

  • Assessed option · provisional

    School mental health: symptoms, functioning, attendance and equitable access

    A comparison-based evaluation and a route for greater needs.

    8/10 criteria scored

    Coverage of the appraisal, not an impact score.

    Delivery: Amber

    Whole-life costs incomplete

    Evidence and delivery constraints

    Artifact describes early implementation evidence without a comparator; universal classroom programmes and targeted treatment differ.

    Specify intervention and referral threshold; secure qualified staff and escalation to specialist care; evaluate outcomes.

    Wellbeing listEvidence and assessment

This domain in the UK SWOT →UK National Action Plan →

How we assessed these options

What would it take to improve this?

What would move the needle?

The outcome we want: A narrower gap between disadvantaged pupils and the rest in England, with better learning and attendance for all.

The main constraint: The gap opens before school, and the interventions with the best evidence depend on trained staff time that schools must take from other work — so the binding choice is how schools and councils use scarce staff, not only how much money is announced.

Constraint type: workforce, money, evidence.

The options on the table

What is funded now, and the changes assessed against it — the current baseline and the alternatives, before the position today and the actions.

Provisional single-reviewer assessments from the 6 October 2026 review; not approved business cases, and not government policy.

  • Skills in the building tradesDelivery RAG: Amber

    Funded now
    Baseline: a construction skills package of over £600m (March 2025), ten construction Technical Excellence Colleges, £287m for post-16 places including construction courses (July 2026), and free apprenticeship training for eligible under-25s from August 2026. Completion support targeted at shortage trades is a UK Facts proposal, not government policy.
    The change assessed
    Not yet defined.

    Range if completed: 15.0–21.0 out of 30 (equal weights; not yet assessed: expected incremental benefit (Bn), incremental affordability (Co))

  • Childcare places that parents can actually useDelivery RAG: Amber

    Funded now
    Funded baseline (England): 30 hours a week for 38 weeks for working parents of children from 9 months (since September 2025), the universal 15 hours for 3–4-year-olds, and 15 hours for disadvantaged 2-year-olds, with £9.5bn for early years in 2026–27; school-based nursery and capital grants add places. Commissioning extra places only where a local gap is shown is a UK Facts proposal, not government policy.
    The change assessed
    Commission specified additional places/hours in demonstrated local supply gaps beyond the funded baseline; adult-carer support remains a separate option.

    Range if completed: 16.0–22.0 out of 30 (equal weights; not yet assessed: expected incremental benefit (Bn), incremental affordability (Co))

  • Close early-language support delivery gapsDelivery RAG: Amber

    Funded now
    DfE funds NELI free for every state-funded school with a reception class, confirmed to academic year 2028–29 subject to spending rounds (reception improvement offer, updated July 2026); The proposal is protected teaching-assistant time to deliver all 20 weeks — not a replacement for statutory SEND assessment.
    The change assessed
    Deliver the evaluated Reception NELI model to eligible children currently missing suitable support, with protected trained-staff time.

    Range if completed: 20.0–23.0 out of 30 (equal weights; not yet assessed: incremental affordability (Co))

Assessed with another page as home

These options also bear on this topic. The full assessment sits on the home page named under each.

  • Family hubs built on Sure Start featuresDelivery RAG: Amber

    Home: Health.

    Funded now
    Best Start Family Hubs are being rolled out to every local authority in England from April 2026, “backed by over £500 million”, with up to 1,000 hubs by the end of 2028 (DfE, July 2025); by March 2026 the stated backing was £900m over three years, including £200m for Healthy Babies services, and on 30 September 2026 the Burnham government said the Healthy Babies programme would reach every hub. They are not automatically equivalent to the 2000s Sure Start centres that were evaluated. The proposal is to hold the hubs to the features that were tested, start where access is weakest, and phase the roll-out so it can be evaluated.
    The change assessed
    Evaluated Sure Start features, weakest-access areas first, and a phased, evaluable roll-out of the family hubs.

    Range if completed: 17.0–20.0 out of 30 (equal weights; not yet assessed: incremental affordability (Co))

  • School mental healthDelivery RAG: Amber

    Home: Health.

    Funded now
    Teams covered 60% of pupils and learners in 2025–26, and the Spending Review 2025 commits to every school by 2029–30. The proposal is to make the remaining roll-out evaluable and to fill the needs gap, not to start a new service.
    The change assessed
    A comparison-based evaluation and a route for greater needs.

    Range if completed: 15.0–21.0 out of 30 (equal weights; not yet assessed: expected incremental benefit (Bn), incremental affordability (Co))

  • Employer-linked shortage training beyond constructionDelivery RAG: Amber

    Home: Jobs & pay.

    Funded now
    Not yet stated in the record.
    The change assessed
    Appraise a defined adult or technical training cohort with employer placements in a verified non-construction shortage.

    Range if completed: 15.0–21.0 out of 30 (equal weights; not yet assessed: expected incremental benefit (Bn), incremental affordability (Co))

Alternatives already on record: 25 options listed under the points in the position today (15 without an action record), and 3 earlier options under other options compared. They have not been scored in the review.

The position today

  • What we can build on (2)
  • What is holding us back (5)School transport is an access and delivery constraint, especially for children with SEND. The NAO found rising costs and risks from reductions in discretionary services, but insufficient data to measure the effect on attendance. (NAO audit, October 2025 · England)
  • What could improve things (1)Targeting tested support at diagnosed gaps — early language in reception, small-group tuition later — and rebuilding integrated early-years help could address the gap at different ages.
  • What could make it worse (1)

Which quadrant a point sits in is our analytical judgement, not a statistical category. Each point says whether it is a measured fact, our synthesis or a risk.

What we can build on Strengths

  • Some help for children who are behind is well tested: an early-language programme added three months’ language progress in a 193-school randomised trial.

    Measured factConfidence: High· 2018–19 trial· England

    Source: Education Endowment Foundation / RAND / University of Oxford, Nuffield Early Language Intervention (re-grant): evaluation report (2020) and addendum (2023) (Report Table 1)

    Options assessed against this point: Close early-language support delivery gaps (Amber delivery RAG)

    What government has committed to

    Funding: DfE funds the Nuffield Early Language Intervention free for every state-funded school with a reception class, with funding confirmed until academic year 2028 to 2029, subject to future spending rounds.

    Who and when:
    Starmer government (to July 2026), 6 July 2026.
    Since 20 July 2026:
    Not confirmed either way since the change of Prime Minister on 20 July 2026. No Burnham-era statement specific to NELI found. The related 75% good-level-of-development ambition was re-stated in a written answer on 8 Sep 2026.
    Source:
    DfE, Reception improvement offer (Section 'Speech and language development through NELI': 'funding is confirmed until academic year 2028 to 2029, subject to future spending rounds')

    Checked 4 October 2026. A commitment is not delivery.

    Options (2)

    • ShortlistedOffer NELI to reception pupils with the weakest language, with protected staff time

      Delivering the full programme to screened children adds language progress before gaps widen.

      Who: Schools and trusts; DfE (funding) · Evidence: Experimental or quasi-experimental evaluation · Status: Established · Works through SEND: Delivering the full programme to screened children adds language progress before gaps widen.

      Source: DfE, Reception improvement offer (Section 'Speech and language development through NELI': 'funding is confirmed until academic year 2028 to 2029, subject to future spending rounds')

    • Use early identification in Best Start Family Hubs to refer children to language support

      Hub inclusion practitioners identify emerging needs before reception so support starts earlier.

      Who: Local authorities; DfE/DHSC · Evidence: Untested proposal · Status: Funded programme

      Source: UK Parliament (written questions API; DfE answer), Written question 24677: Pre-school Education ('the government's ambition for 75% of children to reach a good level of development in Reception by the end of 2028'; Best Start Inclusion Practitioner in every hub)

  • Small-group tuition adds about four months’ progress on average across 62 studies, less in secondary schools.

    Measured factConfidence: Moderate· Review updated July 2021· Studies in several countries

    Source: Education Endowment Foundation, Small group tuition (Headline impact; “Behind the average”)

    What government has committed to

    Funding: Pupil premium (2026-27: £1,550 per primary and £1,100 per secondary pupil) must be spent on evidence-informed activities in DfE's 'menu of approaches'; DfE guidance names tutoring as 'an effective and well-evidenced targeted approach'. The subsidised National Tutoring Programme has ended.

    Who and when:
    Starmer government (to July 2026), 27 March 2026.
    Since 20 July 2026:
    Kept by the Burnham government. Acted on under Burnham: 'Using pupil premium' guidance updated 24 Aug 2026 and September allocations published 23 Sep 2026. White paper plan for government-funded pupil premium strategy reviews re-stated 8 Sep 2026. Source: DfE, Using pupil premium: guidance for school leaders (Section 'Tutoring': 'an effective and well-evidenced targeted approach').

    Checked 4 October 2026. A commitment is not delivery.

    Options (2)

What is holding us back Weaknesses

What could improve things Opportunities

What could make it worse Threats

The actions worth examining

Our shortlist for this page. Research quality, policy status, cost and funding are judged separately; none of these is government policy unless its status says so.Our shortlist. None is government policy unless its status says so.

  1. Raise completion in construction apprenticeships in the shortage trades — not yet costed / funding not identified
  2. Commission childcare places where councils can show parents cannot use the funded hours — not yet costed / funding not identified
  3. Meet early language needs before children fall further behind — programme funding exists; project allocation unconfirmed

And 2 more in the full list.

  • Action 1A55Policy status: external proposal

    Raise completion in construction apprenticeships in the shortage trades

    Completion support for construction apprentices in shortage trades — mentoring, placement guarantees and employer incentives tied to completion — turns more starts into qualified workers.

    The thread: issue, evidence, action

    1. Issue Weakness: Construction apprenticeships have the lowest achievement rate of any subject area — 58.3% in 2024/25 — and about 24,500 started that year, while the industry training board estimates 41,200 extra workers a year are needed from 2026 to 2030 (a need across all routes into the trade, not an apprenticeship target).
    2. Evidence Construction apprenticeship achievement rate
    3. Action A55, this action

    Assessment (draft, 6 October review)

    Who decides:
    Department for Education and Skills England, with colleges, training providers and employers (England)
    Evidence:
    Observational analysis. DfE estimates of returns compare completers with non-completers; apprenticeship reviews find employment effects generally positive; no evaluation shows that construction training raises housing output.
    Expected effect:
    Direction plausible; size not established.
    Cost and funding

    Not yet costed

    Funding: Not yet costed / funding not identified

    Why it is shortlisted here: The national building-trades option (Growth list): construction apprenticeships have the lowest completion of any subject; completion support is the bounded change, measured by achievement rates.

    What it takes
    The problem

    Construction apprenticeships have the lowest achievement rate of any subject area, 58.3% in 2024/25, and starts have stayed around 24,500 a year while the industry estimates it needs 41,200 extra workers a year to 2030.

    Who is affected, and where

    Construction apprentices in England in shortage trades (such as bricklaying, carpentry and electrical work), starting in Technical Excellence College areas. Territory: England. Apprenticeships are devolved in Scotland, Wales and Northern Ireland; CITB forecasts cover all four nations. Not assessed outside England.

    Policy status

    Proposed by researchers or others, not government policy. Baseline: a construction skills package of over £600m (March 2025), ten construction Technical Excellence Colleges, £287m for post-16 places including construction courses (July 2026), and free apprenticeship training for eligible under-25s from August 2026. Completion support targeted at shortage trades is a UK Facts proposal, not government policy.

    The evidence
    What the research found:
    Construction apprenticeships carry a 22% wage premium at Level 2 (no employment gain) and 14% with 2 points of employment at Level 3; the report advises caution because 24% of completers become self-employed.
    Evaluated on:
    Learners completing or starting Level 2 and 3 qualifications in England; England, 2021–22
    Transfer to this case:
    Average returns, not the return on an extra completer; no link to housebuilding output is established.
    Our recommendation (separate from the research):
    Fund completion support in shortage trades and measure achievement rates and entry to construction work against similar programmes without it.
    Strongest counter-evidence:
    Completion support may raise qualifications without adding workers if completers would have worked in construction anyway; overseas recruitment and pay may matter more.
    Cost and funding

    Cost completeness: Not yet costed. Funding: Not yet costed / funding not identified.

      Who pays:
      DfE; employers through the apprenticeship levy and CITB levy.
      Financing proposed:
      From the construction skills package if redirected; not identified.
      Gap remaining:
      Unknown: the cost per extra trade completer or retained worker, and trainer and supervision costs.
      What delivery needs
      Skills:
      Trainers and assessors.
      Operational capacity:
      Employer placements.
      Infrastructure:
      College workshops.
      Permissions:
      None.
      Law:
      None.
      Public consent:
      Employer participation.
      First step and stages

      First step: DfE and Skills England publish construction achievement rates by trade and provider and fund completion support in Technical Excellence College areas, with comparison providers. (owner: DfE with Skills England)

      1. Publish achievement rates by trade — DfE (proposed stage, not an official timetable)
      2. Fund completion support — DfE and colleges (proposed stage, not an official timetable)
      3. Compare achievement and entry to construction work — DfE (evaluation) (proposed stage, not an official timetable)

      Depends on: Employer placements; Trainer supply

      Trade-offs
      Who gains and who loses:
      Young people and adults entering construction trades, often from lower-income backgrounds.
      What else the money or effort could do:
      Money for completion support competes with more starts or other skills.
      Strongest counterargument:
      Pay and conditions in construction, not training, may drive non-completion.
      Risks:
      Paying for completions that would have happened; measuring qualifications rather than work.
      How we would judge it
      Baseline:
      2024/25: 24,500 construction apprenticeship starts; 58.3% achievement rate (lowest of any subject area); June–August 2026: 29,000 construction vacancies (UK).
      Leading indicator:
      Achievement rates in supported trades.
      Outcome:
      Completers in construction work a year later against comparison providers.
      Continue, adapt or stop:
      Continue where achievement and entry to construction work rise relative to comparisons.
      Sources

      Source published 16 July 2026 (DfE apprenticeships); February 2025 (DfE returns); June 2026 (CITB); evidence reviewed 2026-10-07. Reconsider if: DfE apprenticeships full-year 2025/26 (November 2026); CITB Construction Workforce Outlook 2027.

    • Action 2A66Policy status: external proposal

      Commission childcare places where councils can show parents cannot use the funded hours

      Councils use their sufficiency duty to commission specified extra places or hours where codes are issued but not used and where providers cannot meet demand, so eligible parents can take up the hours and work the hours they want.

      The thread: issue, evidence, action

      1. Issue Weakness: The funded childcare hours reach fewest of the children with the greatest need: 71.6% of eligible disadvantaged two-year-olds were registered in January 2026 (74.8% in 2024), against an estimated 8 in 10 eligible under-threes of working parents.
      2. Evidence Eligible disadvantaged two-year-olds registered for funded early learning; Eligible children aged 9 months to 2 years registered for working-parent hours
      3. Action A66, this action

      Assessment (draft, 6 October review)

      Who decides:
      Councils (Childcare Act 2006, section 6 sufficiency duty) with the Department for Education and providers (England)
      Evidence:
      Experimental or quasi-experimental evaluation. The best English evidence on free childcare and parents’ work is a regression-discontinuity study, but it is indirect: it tests part-time hours and full-time school entry, not commissioned places in local gaps. The DfE process evaluation of the expansion is self-reported, and its impact evaluation is not due until late 2027.
      Expected effect:
      Direction plausible; size not established.
      Cost and funding

      Not yet costed

      Funding: Not yet costed / funding not identified

      Why it is shortlisted here: The national childcare option (standalone addition, 7 October 2026): the funded hours reach fewest disadvantaged two-year-olds, and councils report gaps for SEND, atypical hours and rural areas; commissioning places only where a gap is shown is the bounded test. The chapter is /education/childcare.

      What it takes
      The problem

      The funded hours are reaching most eligible families of working parents, but not evenly: take-up is lowest for disadvantaged two-year-olds and babies under one, and councils report too little childcare for children with SEND, parents working atypical hours and rural areas.

      Who is affected, and where

      Families in England with children aged 9 months to 4 years who are eligible for funded hours but cannot find a usable place, in council areas with a demonstrated local gap — starting with under-twos, children with SEND, parents working atypical hours and rural areas. Territory: England. Childcare is devolved. Scotland funds up to 1,140 hours a year for 3–5-year-olds and some 2-year-olds; Wales runs the Childcare Offer (up to 30 hours for 48 weeks for 3–4-year-olds of working parents); Northern Ireland has a 15% Childcare Subsidy Scheme for working parents. None was assessed here.

      Policy status

      Proposed by researchers or others, not government policy. Funded baseline (England): 30 hours a week for 38 weeks for working parents of children from 9 months (since September 2025), the universal 15 hours for 3–4-year-olds, and 15 hours for disadvantaged 2-year-olds, with £9.5bn for early years in 2026–27; school-based nursery and capital grants add places. Commissioning extra places only where a local gap is shown is a UK Facts proposal, not government policy.

      The evidence
      What the research found:
      Part-time free childcare had no effect on mothers’ employment (2.1 points more searching for work); full-time provision at school entry raised mothers’ employment by 3.5 points by the end of the first year. Free hours largely replaced care parents already used. In 2026 the DfE process evaluation found 83% of councils but 57% of providers saying demand is met, 63% of providers restricting funded hours and 18% of councils reporting fewer places for disadvantaged 2-year-olds.
      Evaluated on:
      Mothers in England whose youngest child became eligible for free part-time nursery or full-time school (Labour Force Survey, 2000–2013); councils, providers and parents in the 2024–26 expansion; England
      Transfer to this case:
      The employment effects come from 3–4-year-olds and school entry before 2014; effects for under-twos, and of filling gaps rather than adding hours, are not established.
      Our recommendation (separate from the research):
      Commission extra places only where councils can show eligible parents cannot use the hours (codes issued but not used, waiting lists, SEND and atypical-hours gaps), and compare take-up and parents’ hours with similar areas that do not.
      Strongest counter-evidence:
      Free hours mostly paid for care families already used, so extra places may add little work; and if staff and providers are the constraint, commissioning money may not create places at all.
      Cost and funding

      Cost completeness: Not yet costed. Funding: Not yet costed / funding not identified.

        Who pays:
        Central government through council early years funding; councils for commissioning and sufficiency work; parents for hours beyond the entitlement.
        Financing proposed:
        None identified. A capacity grant from DfE or a share of entitlement funding held back for local gaps would need a decision.
        Gap remaining:
        Unknown: there is no published cost per commissioned place or hour, no council-level count of unmet demand since October 2024, and no council-level take-up rate for working-parent hours.
        What delivery needs
        Skills:
        Qualified early years staff, including SEND and level 3 practitioners.
        Operational capacity:
        Providers willing to open or extend places; council sufficiency teams.
        Infrastructure:
        Premises, including school-based nurseries.
        Permissions:
        Ofsted registration of new provision.
        Law:
        None new: councils already hold the sufficiency duty.
        Public consent:
        Provider participation at the funding rate offered.
        First step and stages

        First step: DfE publishes codes issued and validated by council and age as a local gap measure; three to five councils with the largest gaps commission specified places, with comparison councils named in advance. (owner: DfE with councils)

        1. Publish council-level gap measures — DfE (proposed stage, not an official timetable)
        2. Commission places in named gap areas — Councils (proposed stage, not an official timetable)
        3. Compare take-up and parents’ hours with comparison areas — DfE (evaluation) (proposed stage, not an official timetable)

        Depends on: Council-level gap data (codes issued against validated, waiting lists); Early years workforce supply; The CMA market study (interim report December 2026)

        Trade-offs
        Who gains and who loses:
        Aimed at families who cannot use hours they are entitled to — disproportionately lower-income, rural, SEND and shift-working families — rather than at families already using funded places.
        What else the money or effort could do:
        Money and staff used for commissioned places are not available for higher funding rates for existing places, which providers say do not cover costs.
        Strongest counterargument:
        Raising the hourly rate for every place may do more for supply than targeted commissioning, and councils may not be able to identify gaps reliably.
        Risks:
        Displacing existing places (including disadvantaged 2-year-old places); paying for places that would have opened anyway; staff shortages.
        How we would judge it
        Baseline:
        January 2026: 80% of eligible children aged 9 months to 2 years registered for working-parent hours (46% at 9–11 months); 71.6% of eligible disadvantaged 2-year-olds registered (74.8% in 2024).
        Leading indicator:
        Codes issued but not validated, by council and age; commissioned places opened.
        Outcome:
        Take-up by eligible children in commissioned areas against comparison areas; parents’ hours of work.
        Continue, adapt or stop:
        Continue where take-up rises without disadvantaged 2-year-old take-up falling; stop where commissioned places replace existing ones.
        Sources

        Source published 2 July 2026 (DfE take-up); September 2026 (DfE process evaluation); 2022 (Brewer et al.); evidence reviewed 2026-10-07. Reconsider if: DfE take-up release, July 2027; The CMA market study interim report (December 2026); The DfE impact evaluation of the expansion (late 2027); The Budget on 28 October 2026.

      • Action 3A11Policy status: established

        Meet early language needs before children fall further behind

        Trained teaching assistants deliver a tested 20-week language programme to reception children with the weakest language, with protected time.

        The thread: issue, evidence, action

        1. Issue Weakness: 25.8% of disadvantaged pupils got grade 5+ in English and maths GCSE, against 53.1% of other pupils — a 27.3 points gap, against 25.2 points in 2018/19. Also: threat, opportunity, strength
        2. Evidence Disadvantage gap: expected standard in reading, writing and maths at 11; Disadvantage gap: grade 5+ in English and maths at 16
        3. Action A11, this action

        Assessment (draft, 6 October review)

        Who decides:
        Schools and trusts, with councils and DfE (England)
        Evidence:
        Experimental or quasi-experimental evaluation. A large randomised trial with the highest evidence security; the measured outcome is language, not SEND demand.
        Expected effect:
        +3 months (effect size 0.26): oral language of reception pupils with the weakest language (193 schools in England). A language outcome; not evidence of fewer EHC plans or lower SEND demand. The 2023 follow-up lost 55.5% of pupils.
        Cost and funding

        Not yet costed

        Funding: Programme funding exists; project allocation unconfirmed

        Why it is shortlisted here: The strongest trial evidence on this page, acting before gaps widen.

        What it takes
        The problem

        Children who start school with weak language fall behind in reading and learning, and some wait a long time for specialist help.

        Who is affected, and where

        Reception-age children with weak oral language in England. Territory: England. The trial was in England; devolved nations would need their own decisions.

        Policy status

        Already in place. DfE funds NELI free for every state-funded school with a reception class, confirmed to academic year 2028–29 subject to spending rounds (reception improvement offer, updated July 2026); The proposal is protected teaching-assistant time to deliver all 20 weeks — not a replacement for statutory SEND assessment.

        The evidence
        What the research found:
        Three months’ additional language progress (effect size 0.26, 95% CI 0.17 to 0.35). A 2023 follow-up found about two months on early reading but lost 55.5% of pupils.
        Evaluated on:
        1,156 reception pupils with the lowest language scores in 193 schools; England, 2018–19
        Our recommendation (separate from the research):
        Fund the teaching-assistant time needed to deliver it well, and keep referral routes open for children who need specialist help.
        Strongest counter-evidence:
        Staff felt it worked less well for pupils with SEN; there is no evidence it reduces EHC plans or overall SEND demand.
        Effects on record
        • Oral language of reception pupils with the weakest language: +3 months (effect size 0.26) (randomised control schools; 2018–19 school year; effect size 95% CI 0.17 to 0.35; 5 of 5 padlocks). evaluated, high confidence; 193 schools in England. A language outcome; not evidence of fewer EHC plans or lower SEND demand. The 2023 follow-up lost 55.5% of pupils.
        Cost and funding

        Cost completeness: Historical price only; current cost and staffing to establish. Funding: Programme funding exists; project allocation unconfirmed.

        • NELI materials and training (recurring spending; paid by central government): Not yet costed. Trial price: £870 per one-form-entry school (£58 per child a year over three years), covering materials and training only. Current prices must be checked with the provider.
        • Teaching-assistant delivery time, cover and supervision (recurring spending; paid by central government): Not yet costed. Existing staff time is not free: it is taken from other work.
        • DfE-funded NELI offer to every state school with a reception class: amount not established — not committed (programme envelope, not a project award). No amount is published on the page read. Teaching-assistant time (C-A11-STAFF) is not covered.
        Who pays:
        Schools, from core or inclusion budgets.
        Financing proposed:
        A named school or inclusion allocation; an announced national grant does not confirm local staff time is covered.
        Gap remaining:
        Unknown: current prices and staff time are not costed.
        What delivery needs
        Skills:
        Trained teaching assistants.
        Operational capacity:
        Protected delivery time over about 20 weeks.
        Infrastructure:
        None.
        Permissions:
        None.
        Law:
        None.
        Public consent:
        Family involvement.
        First step and stages

        First step: Assess language needs and confirm staffing before choosing the programme. (owner: Schools)

        1. Screen language in reception — Schools (proposed stage, not an official timetable)
        2. Deliver the 20-week programme — Schools (proposed stage, not an official timetable)
        3. Assess and refer children who need more — Schools and SENCOs (proposed stage, not an official timetable)

        Depends on: Teaching-assistant capacity

        Trade-offs
        Who gains and who loses:
        Gains for children with weak language, often disadvantaged; staff time taken from other work.
        What else the money or effort could do:
        Teaching-assistant hours spent here are not spent on other support.
        Strongest counterargument:
        A generic programme cannot substitute for individual specialist provision.
        Risks:
        Screening that delays statutory assessment; diluted delivery without protected time.
        How we would judge it
        Baseline:
        Reception language scores at screening.
        Leading indicator:
        Programme completion at fidelity.
        Outcome:
        Language progress and later reading, including outcomes for children with greater needs.
        Continue, adapt or stop:
        Continue where progress matches the trial; review where it does not.
        Sources

        Source published May 2020 (trial); March 2023 (follow-up); evidence reviewed 2026-10-04. Reconsider if: A new evaluation of NELI at scale; Changes to SEND legislation.

        Also on: SEND — one shared record, so its cost is counted once.

      • Action 4A10Policy status: established

        Target tutoring at the gaps pupils actually have

        Diagnose each pupil’s gaps, then teach them in groups of two to five, linked to classroom teaching, for a defined block.

        The thread: issue, evidence, action

        1. Issue Weakness: 25.8% of disadvantaged pupils got grade 5+ in English and maths GCSE, against 53.1% of other pupils — a 27.3 points gap, against 25.2 points in 2018/19. Also: opportunity, strength
        2. Evidence Disadvantage gap: grade 5+ in English and maths at 16
        3. Action A10, this action

        Assessment (draft, 6 October review)

        Not assessed in the 6 October review; this action stays in the topic plan with its own evidence.

        Who decides:
        Schools and academy trusts, within DfE pupil premium conditions (England)
        Evidence:
        Synthesis of robust evaluations. A synthesis of 62 studies with moderate security; effects vary by age, subject and delivery.
        Expected effect:
        +4 months (primary); +2 months (secondary): attainment progress of pupils receiving small-group tuition (Studies in several countries). An average across studies; not a predicted change in the national GCSE gap.
        Cost and funding

        Not yet costed

        Funding: Not yet costed / funding not identified

        Why it is shortlisted here: Schools already fund tuition from the pupil premium; targeting it at diagnosed gaps is the change with the most direct evidence for older pupils.

        What it takes
        The problem

        Disadvantaged pupils fall behind in English and maths and general catch-up time is spread too thinly to close specific gaps.

        Who is affected, and where

        Disadvantaged pupils who have fallen behind, in England. Territory: England. Education is devolved; the devolved nations fund schools separately.

        Policy status

        Already in place. Schools can already fund tutoring from the pupil premium; the subsidised National Tutoring Programme has ended. The recommendation is about targeting, not a new programme.

        The evidence
        What the research found:
        Four months’ additional progress on average (two months in secondary); effects fall in groups above six or seven.
        Evaluated on:
        Pupils in small-group tuition studies, mostly primary; Studies in several countries
        Our recommendation (separate from the research):
        Buy sessions only for a defined cohort with diagnosed gaps, and stop provision that is not moving progress.
        Strongest counter-evidence:
        Many studies were not independent evaluations, and secondary-school evidence is thinner and smaller.
        Effects on record
        • Attainment progress of pupils receiving small-group tuition: +4 months (primary); +2 months (secondary) (pupils taught as usual; over a school year; review updated July 2021; moderate evidence security (3 of 5 padlocks); 62 studies). evaluated, moderate confidence; Studies in several countries. An average across studies; not a predicted change in the national GCSE gap.
        Cost and funding

        Cost completeness: Not yet costed. Funding: Not yet costed / funding not identified.

        • Tutor time, preparation, training and materials (recurring spending; paid by central government): Not yet costed. The EEF rates the approach “low cost”; its example of £70–£100 per pupil for a 15-hour block came from the subsidised first year of the National Tutoring Programme and is not a full 2026 cost.
        Who pays:
        Schools, from the pupil premium or core budgets.
        Financing proposed:
        Reallocation within school budgets; a recorded school allocation would show what it displaces.
        Gap remaining:
        A national expansion is not costed.
        What delivery needs
        Skills:
        Trained tutors or teaching assistants.
        Operational capacity:
        Timetable space.
        Infrastructure:
        None.
        Permissions:
        None.
        Law:
        None.
        Public consent:
        Within schools’ discretion.
        First step and stages

        First step: Identify a defined cohort and the specific gaps before buying sessions. (owner: Schools and trusts)

        1. Diagnose gaps — Schools (proposed stage, not an official timetable)
        2. Deliver a defined block — Schools (proposed stage, not an official timetable)
        3. Assess progress and adapt — Schools and trusts (proposed stage, not an official timetable)

        Depends on: Attendance: pupils must be in school to benefit

        Trade-offs
        Who gains and who loses:
        Gains for pupils who are behind; time out of other lessons.
        What else the money or effort could do:
        Pupil premium spent on tutoring is not spent on other support.
        Strongest counterargument:
        Average research effects do not guarantee better GCSE results in a given school.
        Risks:
        Tutor shortages, large groups and indefinite low-impact provision.
        How we would judge it
        Baseline:
        Disadvantage gap at grade 5+ in English and maths (DfE KS4).
        Leading indicator:
        Matched progress of tutored pupils over the block.
        Outcome:
        Attainment gaps at 11 and 16 over several cohorts.
        Continue, adapt or stop:
        Continue blocks that show matched progress; stop those that do not.
        Earlier option on this page

        This card replaces the option “Targeted small-group tuition and evidence-based teaching” (operating: Pupil premium in 2026–27: £1,550 per eligible primary pupil and £1,100 per secondary pupil.). Its evidence: The EEF Toolkit estimates about four months’ additional progress from high-quality small-group tuition, rated moderate impact for low cost on moderate evidence; groups above six or seven pupils lose effect. These are averages across studies, not a guaranteed result in every school. Education Endowment Foundation, Teaching and Learning Toolkit: small group tuition · DfE, Pupil premium 2026 to 2027: technical note

        Sources

        Source published July 2021 (review update); evidence reviewed 2026-10-04. Reconsider if: The EEF updates the toolkit entry; A new national tutoring subsidy.

      • Action 5A13Policy status: funded programme

        Rebuild integrated early-years help where access is weakest

        Provide accessible, integrated health, parenting and early-education support for young children, with outreach, in areas where access is weakest.

        The thread: issue, evidence, action

        1. Issue Weakness: 25.8% of disadvantaged pupils got grade 5+ in English and maths GCSE, against 53.1% of other pupils — a 27.3 points gap, against 25.2 points in 2018/19. Also: opportunity
        2. Evidence Disadvantage gap: grade 5+ in English and maths at 16
        3. Action A13, this action

        Assessment (draft, 6 October review)

        Who decides:
        DfE, DHSC and councils (England)
        Evidence:
        Experimental or quasi-experimental evaluation. A strong quasi-experimental evaluation of the 2000s programme; transfer to today’s services is uncertain.
        Expected effect:
        +0.9 percentage points (about 2%): five GCSEs at A*–C (or equivalent) at 16 (England). 2000s Sure Start, not today’s family hubs. 1 more result in “What it takes”.
        Cost and funding

        Not yet costed

        Funding: Programme funding exists; project allocation unconfirmed

        Why it is shortlisted here: The gap opens before school; the best UK evaluation of integrated early-years services shows lasting gains, with honest limits on transfer.

        What it takes
        The problem

        The attainment gap opens before school, and integrated early-years services have thinned out in many areas.

        Who is affected, and where

        Children under five and their families in areas with weak access, in England. Territory: England. Sure Start was an England programme; devolved nations run separate early-years services.

        Policy status

        A funded programme already exists. Best Start Family Hubs are being rolled out to every local authority in England from April 2026, “backed by over £500 million”, with up to 1,000 hubs by the end of 2028 (DfE, July 2025); by March 2026 the stated backing was £900m over three years, including £200m for Healthy Babies services, and on 30 September 2026 the Burnham government said the Healthy Babies programme would reach every hub. They are not automatically equivalent to the 2000s Sure Start centres that were evaluated. The proposal is to hold the hubs to the features that were tested, start where access is weakest, and phase the roll-out so it can be evaluated.

        The evidence
        What the research found:
        Five A*–C GCSEs up 0.9 percentage points; hospital admissions down 8.5% by age 15 (up 10% at age 1); no significant effect on EHC plans; more cautions by 16. Central appraisal: £2.7bn cost per cohort against £5.5bn discounted benefits (£2.4bn to government), 2023–24 prices.
        Evaluated on:
        Children in England with access to a Sure Start centre, cohorts born around 1990–2007; England, 1999–2010 roll-out
        Transfer to this case:
        Today’s family hubs differ in services, funding and staffing from 2000s Sure Start.
        Our recommendation (separate from the research):
        Specify the service mix and outreach first, then fund it where access is weakest, and evaluate cohorts rather than claiming the old results.
        Strongest counter-evidence:
        Updated methods removed an earlier fall in EHC plans; crime results are mixed (more cautions); benefits are largely modelled lifetime earnings. Wales’s Flying Start raised service contact but found no measurable difference in child development by about age three.
        Effects on record
        • Five GCSEs at A*–C (or equivalent) at 16: +0.9 percentage points (about 2%) (children without access, before and after centres opened; cohorts born around 1990–2003; quasi-experimental; method update changed some earlier findings). evaluated, moderate confidence; England. 2000s Sure Start, not today’s family hubs.
        • Hospital admissions by age 15: −8.5% by age 15 (areas with fewer centres per child; cohorts born 1990s–2000s; admissions rose 10% at age 1 before falling). evaluated, moderate confidence; England. Not a cashable NHS saving.
        • Criminal record by age 16: A mixed effect; size not established. evaluated, moderate confidence; England. More young people had a criminal record by 16, driven by an 11% rise in cautions; a fall in convictions was not significant. A trade-off to watch, not a benefit.
        Cost and funding

        Cost completeness: Published programme envelope; project costs still needed. Funding: Programme funding exists; project allocation unconfirmed.

        • Integrated early-years service package and workforce (recurring spending; paid by central government, councils, NHS): Not yet costed. The IFS appraisal of 2000s Sure Start is a benchmark, not today’s rollout price; see the separate appraisal chart.
        • Best Start Family Hubs roll-out (programme envelope) (recurring spending; paid by central government): at least £500m in total, Announced 6 July 2025 (“over £500 million”), hubs in every local authority from April 2026 to the end of 2028; fiscal years and price basis not stated in the announcement., price basis not stated by the source. A programme envelope for the hubs, not a costed package of the Sure Start features the evaluation tested; an individual area’s offer is funded only once its allocation and specification are confirmed.
        • Best Start Family Hubs: “over £500 million” to the end of 2028: amount not established — not committed (programme envelope, not a project award). Recorded as “at least £500m” in the cost record; the amount is not repeated here to avoid implying a precise figure.
        Who pays:
        Central government and councils; the NHS for health visiting.
        Financing proposed:
        Best Start Family Hubs allocations (a programme envelope of over £500m) and public-health and health-visiting budgets; extra coverage or staffing beyond the hubs needs a new allocation or explicit displacement.
        Gap remaining:
        Unknown until the service package is specified and costed. Lifetime benefits cannot fund upfront delivery.
        What delivery needs
        Skills:
        Early-years practitioners, health visitors, outreach workers.
        Operational capacity:
        Local commissioning.
        Infrastructure:
        Accessible buildings.
        Permissions:
        None.
        Law:
        None.
        Public consent:
        Families’ trust and use of services.
        First step and stages

        First step: Map service access and specify the offer before choosing buildings or branding. (owner: Councils with DfE)

        1. Map access and specify services — Councils (proposed stage, not an official timetable)
        2. Build provision and track reach — Councils and NHS (proposed stage, not an official timetable)
        3. Evaluate cohorts over years — DfE with researchers (proposed stage, not an official timetable)

        Depends on: Early-years workforce

        Trade-offs
        Who gains and who loses:
        Largest gains for disadvantaged children; costs fall now, benefits mostly later and partly to other budgets.
        What else the money or effort could do:
        Funds and staff are not available for childcare places or other services.
        Strongest counterargument:
        Effects are modest per child and arrive slowly; diluted provision may not reproduce them.
        Risks:
        Workforce shortages and attributing every childhood outcome to one programme.
        How we would judge it
        Baseline:
        Good level of development at age five (DfE EYFS profile).
        Leading indicator:
        Reach among families in target areas.
        Outcome:
        Child development, later attainment and admissions, kept separate from projections.
        Continue, adapt or stop:
        Expand where reach and early outcomes improve; review where reach is low.
        Sources

        Source published 22 May 2025 (IFS); December 2020 (Marmot); evidence reviewed 2026-10-04. Reconsider if: An evaluation of family hubs or Best Start services; A change in early-years funding.

        Also on: Inequality, Health — one shared record, so its cost is counted once.

      How the options are assessed

      One shared record per option — the same one the national plan reads. Delivery RAG, gate and list membership are separate judgements.

      • Skills in the building trades — Amber delivery RAG; gate pending
      • Childcare places that parents can actually use — Amber delivery RAG; gate pending
      • Close early-language support delivery gaps — Amber delivery RAG; gate pending

      Provisional single-reviewer assessments from the 6 October 2026 review; not approved business cases, and not government policy.

      • A55Delivery RAG: AmberGate: pending

        Skills in the building trades

        Existing action A55, assessed at the same scope.

        Provisional single-reviewer assessment; not an approved business case.

        Who and where
        Construction apprentices in England in shortage trades (such as bricklaying, carpentry and electrical work), starting in Technical Excellence College areas. (from the action record) Jurisdiction: England.
        Result so far
        Range if completed: 15.0–21.0 out of 30 (equal weights; not yet assessed: expected incremental benefit (Bn), incremental affordability (Co))
        Funded now (baseline)
        Baseline: a construction skills package of over £600m (March 2025), ten construction Technical Excellence Colleges, £287m for post-16 places including construction courses (July 2026), and free apprenticeship training for eligible under-25s from August 2026. Completion support targeted at shortage trades is a UK Facts proposal, not government policy. (from the action record)
        The change assessed (marginal)
        Marginal change not yet defined
        What the evidence supports—and the next decision

        A training place is not a qualified worker retained on a building site. Announced construction funding already belongs in the baseline.

        Codex — AI-assisted analytical follow-up · 8 October 2026. Proposed appraisal design; the attributed scores below are unchanged. No funding or delivery agreement is implied.

        Define the change
        Specify the occupation, college catchment, cohort and employer placements missing from existing construction provision.
        Test whether it works
        Completion, occupational competence and employment in the trade after 6 and 12 months; compare with similar eligible applicants and track displacement of existing apprentices.
        Establish the cost
        Trainer recruitment, tuition, equipment, trainee support and employer supervision, net of existing awards. Cost each additional sustained placement, not each start.
        Protect against harm
        Require safe placements and access for applicants who cannot afford unpaid training.

        Read with the outcome evidence and sources in the full assessment. Monitoring proposals are not existing baseline measurements. A numerical benefit or cost score needs evidence for the defined increment, not just the scale of the problem.

        Full assessment of Skills in the building trades: ten criteria, scenarios, delivery, costs and monitoring
        The ten criteria (0–3)

        Each criterion is scored 0 to 3 by the reviewer. “Not yet assessed” means unknown: it is never counted as zero.

        1. Scale and severity (Sc)2 of 3 — Serious harm to a defined group or widespread moderate harm

          Reviewer’s reason: Provisional problem-importance judgement from linked site pages; not the scale of achievable benefit.

        2. Increment beyond baseline (Ad)2 of 3 — Specific material change to existing delivery

          Reviewer’s reason: Identify trades and locations; secure paid placements, trainers, facilities and employer demand; measure completion and retention. A missing rating requires a defined marginal change.

        3. Confidence in the proposed action (Ev)1 of 3 — Indirect, mixed, modelled or weak causal support

          Reviewer’s reason: Historical apprenticeship returns support consideration, not the marginal return from this shortage-trades expansion.

        4. Expected incremental benefit (Bn)Not yet assessed

          Reviewer’s reason: Unknown unless a reasonably defined target-population benefit can be judged; no bonus merely for spanning domains.

        5. Residual harms (Hm)2 of 3 — Manageable harm with specified safeguards

          Reviewer’s reason: Displacement of existing trainees; low completion; insufficient trainers.

        6. Incremental affordability (Co)Not yet assessed

          Reviewer’s reason: Additional training, trainee support, employer supervision and trainer recruitment; levy eligibility is not an allocation.

        7. Delivery capacity (Ca)1 of 3 — Binding scarce resource

          Reviewer’s reason: Identify trades and locations; secure paid placements, trainers, facilities and employer demand; measure completion and retention.

        8. Distribution (Fa)3 of 3 — Benefits concentrated on disadvantaged people with assessed burdens

          Reviewer’s reason: Provisional distribution judgement for the stated target; verify who pays, who gains and excluded groups.

        9. Meaningful first outcome (Ti)2 of 3 — More than one and up to three years

          Reviewer’s reason: First meaningful outcome, not announcement: Completions, sustained trade employment and employer vacancies.

        10. Readiness (Rd)2 of 3 — Named commissioning/allocation/coordination step

          Reviewer’s reason: DfE, skills bodies, colleges and employers; nation-specific systems; Appraise employer-linked expansion

        Four weightings

        Normalised as 10 × weighted sum ÷ sum of weights, on a 0–30 scale. Where a criterion is not yet assessed, the range is the lowest and highest result if it were later scored 0 or 3 — not a score, a confidence interval or a rank.

        • Equal weights: Range if completed: 15.0–21.0 out of 30. Not yet assessed: expected incremental benefit (Bn), incremental affordability (Co). Funding gate: not passed.
        • Evidence doubled: Range if completed: 14.5–20.0 out of 30. Not yet assessed: expected incremental benefit (Bn), incremental affordability (Co). Funding gate: not passed.
        • Scale doubled: Range if completed: 15.5–20.9 out of 30. Not yet assessed: expected incremental benefit (Bn), incremental affordability (Co). Funding gate: not passed.
        • Cost ignored (the funding gate still applies): Range if completed: 16.7–20.0 out of 30. Not yet assessed: expected incremental benefit (Bn), incremental affordability (Co). Funding gate: not passed.

        Funding gate: Not passed — the cost is not yet assessed or no allocation is committed for this scope. Ignoring cost in a weighting never removes the gate.

        Delivery RAG and gate

        Delivery RAG shown: AmberReview’s claim: Amber. Best the recorded evidence allows: Amber.

        Review’s reason: Additional training, trainee support, employer supervision and trainer recruitment; levy eligibility is not an allocation.

        Why it is not Green yet

        • Scope (population, place and increment) not verified against the delivery baseline.
        • No verified owner with the powers in place.
        • Delivery capacity not verified.
        • Lifetime incremental costs not complete.
        • No credible funding (a committed allocation for the scope).
        • Safeguards not specified.
        • Monitoring lacks a metric, a data source or a baseline.

        Green needs verified scope, a verified owner with the powers, verified capacity, complete lifetime costs with credible funding, specified safeguards and monitoring with a metric, source and baseline. Amber means appraisal gaps remain; Red means a material delivery obstacle. Red does not mean unimportant.

        Gate: Pending — appraisal incomplete (review label: “Pending: incomplete appraisal”).

        Why: Scores incomplete; Funding gate not passed (cost unknown or no credible funding).

        The gate is separate from national list membership: being listed does not make an option eligible, and an eligible option need not be listed.

        Cost and funding

        Lifetime costing: Not established. Review: “Full incremental costing not established in reviewed material”.

        Still to cost: Additional training, trainee support, employer supervision and trainer recruitment; levy eligibility is not an allocation.

        Funding status: not identified.

        Owner, powers and capacity

        Proposed owner: DfE, skills bodies, colleges and employers; nation-specific systems (proposed, unverified).

        Powers: needs a decision or allocation. Capacity: binding constraint.

        Powers and capacity read from the review’s Rd and Ca anchors; no proposed owner has agreed to act (the review: “Proposed owners have not agreed to act”).

        England:
        Department for Education and Skills England, with colleges, training providers and employers (action record)
        Scotland:
        not assessed
        Wales:
        not assessed
        Northern Ireland:
        not assessed

        Apprenticeships are devolved in Scotland, Wales and Northern Ireland; CITB forecasts cover all four nations. Not assessed outside England.

        Dependencies, safeguards and risks

        Depends on: None recorded.

        Safeguards

        None specified yet.

        Risks and residual harms: Displacement of existing trainees; low completion; insufficient trainers.

        Monitoring and measured results

        Metric: Completions, sustained trade employment and employer vacancies.

        Data source: Not yet named. Baseline: Not yet established.

        Measures already on the domain page

        A page measure is not necessarily the review’s metric; linking the two is an editorial decision still to be made.

        Evidence for the outcome

        Outcome: Completions, sustained trade employment and employer vacancies.

        Historical apprenticeship returns support consideration, not the marginal return from this shortage-trades expansion.

        Review record

        Assessment version: working-review-2026-10-06 (A55@working-review-2026-10-06).

        Reviewer: Author of the review blueprint (single reviewer, not named in uk-facts-complete-plan.json). Reviewed: 6 October 2026.

        Scored in the working review of 6 October 2026 on the shared ten-criterion scale (one reviewer); not independently calibrated.

        First 90 days (review): Produce a signed scope and current-delivery baseline; Identify trades and locations; secure paid placements, trainers, facilities and employer demand; measure completion and retention. Obtain incremental costs and an accountable decision before expansion.

        Decision route: Appraise employer-linked expansion

        An editorial screening framework by one reviewer, not a formal appraisal or business case. Proposed owners have not agreed to act.

      • A66Delivery RAG: AmberGate: pending

        Childcare places that parents can actually use

        Existing action A66, assessed at the same scope.

        Not stated in the review record; treated as a provisional single-reviewer assessment, not an approved business case.

        Who and where
        Families in England with children aged 9 months to 4 years who are eligible for funded hours but cannot find a usable place, in council areas with a demonstrated local gap — starting with under-twos, children with SEND, parents working atypical hours and rural areas. (from the action record) Jurisdiction: England.
        Result so far
        Range if completed: 16.0–22.0 out of 30 (equal weights; not yet assessed: expected incremental benefit (Bn), incremental affordability (Co))
        Funded now (baseline)
        Funded baseline (England): 30 hours a week for 38 weeks for working parents of children from 9 months (since September 2025), the universal 15 hours for 3–4-year-olds, and 15 hours for disadvantaged 2-year-olds, with £9.5bn for early years in 2026–27; school-based nursery and capital grants add places. Commissioning extra places only where a local gap is shown is a UK Facts proposal, not government policy. (from the action record)
        The change assessed (marginal)
        Commission specified additional places/hours in demonstrated local supply gaps beyond the funded baseline; adult-carer support remains a separate option. (from the 6 October review)
        What the evidence supports—and the next decision

        The effect of older free-childcare policies cannot simply be applied to the under-three expansion or to places parents cannot use.

        Codex — AI-assisted analytical follow-up · 8 October 2026. Proposed appraisal design; the attributed scores below are unchanged. No funding or delivery agreement is implied.

        Define the change
        Commission additional places in verified local gaps, specifying hours, age, SEND access and fees beyond current entitlements.
        Test whether it works
        Usable places taken up, parental employment and hours, child development and quality; compare similar eligible families and separate substitution for care already purchased.
        Establish the cost
        Premises, qualified staff, ratios, operating subsidy, fees and evaluation; test provider viability and continuing demand.
        Protect against harm
        Protect quality and access for disadvantaged families, children with SEND and parents working atypical hours.

        Read with the outcome evidence and sources in the full assessment. Monitoring proposals are not existing baseline measurements. A numerical benefit or cost score needs evidence for the defined increment, not just the scale of the problem.

        Full assessment of Childcare places that parents can actually use: ten criteria, scenarios, delivery, costs and monitoring
        The ten criteria (0–3)

        Each criterion is scored 0 to 3 by the reviewer. “Not yet assessed” means unknown: it is never counted as zero.

        1. Scale and severity (Sc)3 of 3 — Severe harm across a broad population

        2. Increment beyond baseline (Ad)2 of 3 — Specific material change to existing delivery

        3. Confidence in the proposed action (Ev)1 of 3 — Indirect, mixed, modelled or weak causal support

        4. Expected incremental benefit (Bn)Not yet assessed

        5. Residual harms (Hm)2 of 3 — Manageable harm with specified safeguards

        6. Incremental affordability (Co)Not yet assessed

        7. Delivery capacity (Ca)1 of 3 — Binding scarce resource

        8. Distribution (Fa)3 of 3 — Benefits concentrated on disadvantaged people with assessed burdens

        9. Meaningful first outcome (Ti)2 of 3 — More than one and up to three years

        10. Readiness (Rd)2 of 3 — Named commissioning/allocation/coordination step

        Four weightings

        Normalised as 10 × weighted sum ÷ sum of weights, on a 0–30 scale. Where a criterion is not yet assessed, the range is the lowest and highest result if it were later scored 0 or 3 — not a score, a confidence interval or a rank.

        • Equal weights: Range if completed: 16.0–22.0 out of 30. Not yet assessed: expected incremental benefit (Bn), incremental affordability (Co). Funding gate: not passed.
        • Evidence doubled: Range if completed: 15.5–20.9 out of 30. Not yet assessed: expected incremental benefit (Bn), incremental affordability (Co). Funding gate: not passed.
        • Scale doubled: Range if completed: 17.3–22.7 out of 30. Not yet assessed: expected incremental benefit (Bn), incremental affordability (Co). Funding gate: not passed.
        • Cost ignored (the funding gate still applies): Range if completed: 17.8–21.1 out of 30. Not yet assessed: expected incremental benefit (Bn), incremental affordability (Co). Funding gate: not passed.

        Funding gate: Not passed — the cost is not yet assessed or no allocation is committed for this scope. Ignoring cost in a weighting never removes the gate.

        Delivery RAG and gate

        Delivery RAG shown: AmberReview’s claim: Amber. Best the recorded evidence allows: Amber.

        Review’s reason: Verify unmet demand, existing awards and provider viability; cost places, workforce, premises, operating subsidy and parental fees. Track access and sustained parental work. Proposed owners: councils, DfE and providers in England; separate devolved arrangements.

        Why it is not Green yet

        • Scope (population, place and increment) not verified against the delivery baseline.
        • No verified owner with the powers in place.
        • Delivery capacity not verified.
        • Lifetime incremental costs not complete.
        • No credible funding (a committed allocation for the scope).
        • Safeguards not specified.
        • Monitoring lacks a metric, a data source or a baseline.

        Green needs verified scope, a verified owner with the powers, verified capacity, complete lifetime costs with credible funding, specified safeguards and monitoring with a metric, source and baseline. Amber means appraisal gaps remain; Red means a material delivery obstacle. Red does not mean unimportant.

        Gate: Pending — appraisal incomplete (review label: “Pending baseline verification, incremental costing and funding; not a rollout approval”).

        Why: Scores incomplete; Funding gate not passed (cost unknown or no credible funding).

        The gate is separate from national list membership: being listed does not make an option eligible, and an eligible option need not be listed.

        Cost and funding

        Lifetime costing: Not established. Review: “Not fully costed in this review”.

        Still to cost: Verify unmet demand, existing awards and provider viability; cost places, workforce, premises, operating subsidy and parental fees. Track access and sustained parental work. Proposed owners: councils, DfE and providers in England; separate devolved arrangements.

        Funding status: not identified.

        Owner, powers and capacity

        Proposed owner: councils, DfE and providers in England; separate devolved arrangements (proposed, unverified).

        Powers: needs a decision or allocation. Capacity: binding constraint.

        Powers and capacity read from the review’s Rd and Ca anchors; no proposed owner has agreed to act (the review: “Proposed owners have not agreed to act”).

        England:
        Councils (Childcare Act 2006, section 6 sufficiency duty) with the Department for Education and providers (action record)
        Scotland:
        not assessed
        Wales:
        not assessed
        Northern Ireland:
        not assessed

        Childcare is devolved. Scotland funds up to 1,140 hours a year for 3–5-year-olds and some 2-year-olds; Wales runs the Childcare Offer (up to 30 hours for 48 weeks for 3–4-year-olds of working parents); Northern Ireland has a 15% Childcare Subsidy Scheme for working parents. None was assessed here.

        Dependencies, safeguards and risks

        Depends on: None recorded.

        Safeguards

        None specified yet.

        Monitoring and measured results

        Metric: Define and baseline the outcome described in delivery requirements before approval.

        Data source: Not yet named. Baseline: Not yet established.

        Measures already on the domain page

        A page measure is not necessarily the review’s metric; linking the two is an editorial decision still to be made.

        Evidence for the outcome

        Outcome: Define and baseline the outcome described in delivery requirements before approval.

        IFS finds employment effects depend on the hours offered. This is indirect evidence for filling today’s local supply gaps. A September 2026 process evaluation exists; it is not a causal national impact estimate.

        Sources: ifs.org.uk (not yet in the source register)

        Review record

        Assessment version: working-review-2026-10-06 (A66@working-review-2026-10-06).

        Reviewer: Author of the review blueprint (single reviewer, not named in uk-facts-complete-plan.json). Reviewed: 6 October 2026.

        Scored in the working review of 6 October 2026 on the shared ten-criterion scale (one reviewer); not independently calibrated.

        An editorial screening framework by one reviewer, not a formal appraisal or business case. Proposed owners have not agreed to act.

      • A11Delivery RAG: AmberGate: pending

        Close early-language support delivery gaps

        Existing action A11, assessed at the same scope.

        Not stated in the review record; treated as a provisional single-reviewer assessment, not an approved business case.

        Who and where
        Reception-age children with weak oral language in England. (from the action record) Jurisdiction: England.
        Result so far
        Range if completed: 20.0–23.0 out of 30 (equal weights; not yet assessed: incremental affordability (Co))
        Funded now (baseline)
        DfE funds NELI free for every state-funded school with a reception class, confirmed to academic year 2028–29 subject to spending rounds (reception improvement offer, updated July 2026); The proposal is protected teaching-assistant time to deliver all 20 weeks — not a replacement for statutory SEND assessment. (from the action record)
        The change assessed (marginal)
        Deliver the evaluated Reception NELI model to eligible children currently missing suitable support, with protected trained-staff time. (from the 6 October review)
        What the evidence supports—and the next decision

        The evaluated NELI model is specific. Its results do not justify an unspecified early-language programme or prove future earnings gains.

        Codex — AI-assisted analytical follow-up · 8 October 2026. Proposed appraisal design; the attributed scores below are unchanged. No funding or delivery agreement is implied.

        Define the change
        Find eligible Reception children missing suitable support and deliver the evaluated model with trained staff and protected time.
        Test whether it works
        Language outcomes, reach and fidelity with appropriate comparison and follow-up; report SEND and English-as-an-additional-language access.
        Establish the cost
        Screening, training, materials, staff delivery and release time. A published resource price is not the full cost of school delivery.
        Protect against harm
        Protect existing specialist provision and avoid extending trial results to untested ages or models.

        Read with the outcome evidence and sources in the full assessment. Monitoring proposals are not existing baseline measurements. A numerical benefit or cost score needs evidence for the defined increment, not just the scale of the problem.

        Full assessment of Close early-language support delivery gaps: ten criteria, scenarios, delivery, costs and monitoring
        The ten criteria (0–3)

        Each criterion is scored 0 to 3 by the reviewer. “Not yet assessed” means unknown: it is never counted as zero.

        1. Scale and severity (Sc)2 of 3 — Serious harm to a defined group or widespread moderate harm

        2. Increment beyond baseline (Ad)2 of 3 — Specific material change to existing delivery

        3. Confidence in the proposed action (Ev)2 of 3 — Relevant evaluated evidence with limitations

        4. Expected incremental benefit (Bn)2 of 3 — Material for the target population

        5. Residual harms (Hm)2 of 3 — Manageable harm with specified safeguards

        6. Incremental affordability (Co)Not yet assessed

        7. Delivery capacity (Ca)2 of 3 — Stretched but plausibly obtainable

        8. Distribution (Fa)3 of 3 — Benefits concentrated on disadvantaged people with assessed burdens

        9. Meaningful first outcome (Ti)3 of 3 — Within one year

        10. Readiness (Rd)2 of 3 — Named commissioning/allocation/coordination step

        Four weightings

        Normalised as 10 × weighted sum ÷ sum of weights, on a 0–30 scale. Where a criterion is not yet assessed, the range is the lowest and highest result if it were later scored 0 or 3 — not a score, a confidence interval or a rank.

        • Equal weights: Range if completed: 20.0–23.0 out of 30. Not yet assessed: incremental affordability (Co). Funding gate: not passed.
        • Evidence doubled: Range if completed: 20.0–22.7 out of 30. Not yet assessed: incremental affordability (Co). Funding gate: not passed.
        • Scale doubled: Range if completed: 20.0–22.7 out of 30. Not yet assessed: incremental affordability (Co). Funding gate: not passed.
        • Cost ignored (the funding gate still applies): Range if completed: 22.2–22.2 out of 30. Not yet assessed: incremental affordability (Co). The missing criteria carry no weight here, so the ends meet; the record is still incomplete. Funding gate: not passed.

        Funding gate: Not passed — the cost is not yet assessed or no allocation is committed for this scope. Ignoring cost in a weighting never removes the gate.

        Delivery RAG and gate

        Delivery RAG shown: AmberReview’s claim: Amber. Best the recorded evidence allows: Amber.

        Review’s reason: Audit reach, eligibility and delivery fidelity; cost trained staff, release time, screening, materials and follow-up. Protect SEND provision. Measure language outcomes and access. Proposed owners: schools, education bodies and councils as appropriate.

        Why it is not Green yet

        • Scope (population, place and increment) not verified against the delivery baseline.
        • No verified owner with the powers in place.
        • Delivery capacity not verified.
        • Lifetime incremental costs not complete.
        • No credible funding (a committed allocation for the scope).
        • Safeguards not specified.
        • Monitoring lacks a metric, a data source or a baseline.

        Green needs verified scope, a verified owner with the powers, verified capacity, complete lifetime costs with credible funding, specified safeguards and monitoring with a metric, source and baseline. Amber means appraisal gaps remain; Red means a material delivery obstacle. Red does not mean unimportant.

        Gate: Pending — appraisal incomplete (review label: “Pending baseline verification, incremental costing and funding; not a rollout approval”).

        Why: Scores incomplete; Funding gate not passed (cost unknown or no credible funding).

        The gate is separate from national list membership: being listed does not make an option eligible, and an eligible option need not be listed.

        Cost and funding

        Lifetime costing: Not established. Review: “Not fully costed in this review”.

        • NELI materials and training (historical price only; C-A11-NELI)
        • Teaching-assistant delivery time, cover and supervision (not costed; C-A11-STAFF)

        Still to cost: Audit reach, eligibility and delivery fidelity; cost trained staff, release time, screening, materials and follow-up. Protect SEND provision. Measure language outcomes and access. Proposed owners: schools, education bodies and councils as appropriate.

        Funding status: programme route, not allocated.

        Owner, powers and capacity

        Proposed owner: schools, education bodies and councils as appropriate (proposed, unverified).

        Powers: needs a decision or allocation. Capacity: unverified.

        Powers and capacity read from the review’s Rd and Ca anchors; no proposed owner has agreed to act (the review: “Proposed owners have not agreed to act”).

        England:
        Schools and trusts, with councils and DfE (action record)
        Scotland:
        not assessed
        Wales:
        not assessed
        Northern Ireland:
        not assessed

        The trial was in England; devolved nations would need their own decisions.

        Dependencies, safeguards and risks

        Depends on: None recorded.

        Safeguards

        • Protect SEND provision (proposed, not yet specified)
        • Deliver the evaluated Reception NELI model to eligible children currently missing suitable support, with protected trained-staff time (proposed, not yet specified)
        Monitoring and measured results

        Metric: Define and baseline the outcome described in delivery requirements before approval.

        Data source: Not yet named. Baseline: Not yet established.

        Measures already on the domain page

        A page measure is not necessarily the review’s metric; linking the two is an editorial decision still to be made.

        Evidence for the outcome

        Outcome: Define and baseline the outcome described in delivery requirements before approval.

        EEF effectiveness trial supports language outcomes; the transfer to a gap-closing delivery programme and persistence of effects need checking. Programme availability alone is not a verified access gap.

        Sources: educationendowmentfoundation.org.uk (S07)

        Effect records for this action

        • Oral language of reception pupils with the weakest language: +3 months (effect size 0.26) (evaluated, high confidence; 193 schools in England)
        Review record

        Assessment version: working-review-2026-10-06 (A11@working-review-2026-10-06).

        Reviewer: Author of the review blueprint (single reviewer, not named in uk-facts-complete-plan.json). Reviewed: 6 October 2026.

        Scored in the working review of 6 October 2026 on the shared ten-criterion scale (one reviewer); not independently calibrated.

        An earlier assessment exists on the 4 October 2026 selection table (nine H/M/L bands; superseded 7 October 2026). Its bands are kept, not converted into these scores.

        An editorial screening framework by one reviewer, not a formal appraisal or business case. Proposed owners have not agreed to act.

      Decision briefs

      • Build the new family hubs like the Sure Start centres that worked, starting where access is weakest
      • Spend tutoring money on the gaps pupils actually have, in short fixed blocks — and stop blocks that do not move progress

      The leading option set out as a decision someone could take: who decides, who benefits, the best evidence, cost and funding, and the main trade-off. The full research record is one click away.

      Decision briefExisting programmeReviewed 2026-10-04

      Build the new family hubs like the Sure Start centres that worked, starting where access is weakest

      The thread: issue, evidence, action

      1. Issue Weakness: 25.8% of disadvantaged pupils got grade 5+ in English and maths GCSE, against 53.1% of other pupils — a 27.3 points gap, against 25.2 points in 2018/19. Also: opportunity
      2. Evidence Disadvantage gap: grade 5+ in English and maths at 16
      3. Action Rebuild integrated early-years help where access is weakest
      Decision required
      Whether DfE specifies the hub offer from the evaluated Sure Start features, puts the weakest-access areas first and phases the remaining roll-out so it can be evaluated.
      Who and where
      Children under five and their families, especially in disadvantaged areas with weak access to services. Applies in: England.
      Accountable body
      Department for Education
      Intervention
      Offer integrated health, parenting and early-education support close to families, with outreach, where access is weakest.
      Best evidence, and where it applies
      Living near a 2000s Sure Start centre raised the chance of five good GCSEs by 0.9 points and cut hospital admissions by 8.5% by age 15 (IFS); Wales’s Flying Start raised contact but showed no child-development difference by age three.Studied in: Children in England born around 1990–2007 (Sure Start roll-out 1999–2010; outcomes to age 16).How closely the comparison transfers to this decision: Wales — partial.
      Extra cost of this change
      Capital: none identified in the brief.Recurring: Best Start Family Hubs roll-out — at least £500m in total, Announced 6 July 2025 (“over £500 million”), hubs in every local authority from April 2026 to the end of 2028; fiscal years and price basis not stated in the announcement., price basis not stated by the source; Evaluated Sure Start service mix and workforce — Not yet costed.One-off or mixed: Phased-roll-out evaluation — not yet established.
      Funding
      Not funded yet: a possible funding route exists, but nothing is allocated to this change.Committed to this change: none.Possible source, not an award: Best Start Family Hubs: “over £500 million” to the end of 2028.Unfunded: The cost of meeting the specification and of the evaluation is not established; the over-£500m is a programme envelope.
      Savings and wider benefits
      Not yet specified: no cashable saving or avoided cost has been costed for this change. Expected outcomes are listed one by one in the research appendix and are never added together or counted as money available.
      First 90-day step
      Hub specification, access map and roll-out order published.The proposed first milestone; a 90-day date is not yet specified. Owner: DfE with councils. Proposed by UK Facts, not an official deadline.
      Twelve-month milestone
      Not yet specified.Time to a first useful result: Reach within a year; child development at two to three in about two years; attainment only as cohorts age.
      Outcome, and when to scale or stop
      Measure: Reach among disadvantaged families in target areas.Scale if: Reach rises and early development improves against later-starting areas.Stop if: No difference in development against later areas once the specification is met.
      Main trade-off
      Effects per child are modest and slow, and a phased roll-out means some areas wait.
      Source and review date
      The short- and medium-term effects of Sure Start on children’s outcomes (May 2025); National evaluation of Flying Start: impact report (December 2013). Reviewed 2026-10-04.

      An editorial proposal by UK Facts for discussion, not government policy. Records: A13 · effects E-A13-GCSE, E-A13-HOSPITAL, E-A13-CRIME · costs C-BSFH-2025, C-A13-SERVICE

      Research appendix: study design, effect sizes and limits, alternatives, full costing, international comparisons and how we would know

      1.The change, in plain words

      Build the new family hubs like the Sure Start centres that worked, starting where access is weakest

      2.The decision requested

      Decision
      Publish a hub specification drawn from the evaluated Sure Start features (local access within reach of families, health visiting joined with early education, outreach), prioritise the areas with the weakest access, and fix the order of the remaining roll-out so that comparable areas start at different times.
      Accountable institution
      Department for Education
      Geography
      England
      Powers
      Existing powers and the Best Start programme’s grant conditions; no new law.
      Delivery partners
      Councils; Department of Health and Social Care and NHS commissioners (health visiting); Researchers (evaluation)

      3.Who it is for, and what changes beyond today

      Target population
      Children under five and their families, especially in disadvantaged areas with weak access to services.
      Eligibility
      Universal hub access, with outreach to families least likely to use services.
      Scale
      Up to 1,000 hubs by the end of 2028 are already planned; the decision is their specification and order.
      Baseline
      Best Start Family Hubs reach every local authority from April 2026, backed by over £500m; the evaluated programme was 2000s Sure Start.

      Source: Department for Education, Government revives family services supporting 500,000 more kids (press release) (Press release text); Institute for Fiscal Studies, The short- and medium-term effects of Sure Start on children’s outcomes (Executive summary pp.6–10)

      Change beyond existing law or funding
      The hubs are funded; the change is the specification, weakest-access areas first, and a phased roll-out that allows a comparison.

      4.How it would work

      Do: Offer integrated health, parenting and early-education support close to families, with outreach, where access is weakest. Changing: More disadvantaged families use early help before gaps widen.

      • GCSE attainment at 16

        Measured for this interventionEffect record E-A13-GCSE

        Measured for 2000s Sure Start; transfer to today’s hubs is uncertain.

      • Hospital admissions by age 15

        Measured for this interventionEffect record E-A13-HOSPITAL

        Measured for 2000s Sure Start (−8.5% by 15; +10% at age 1).

      • Criminal record by age 16

        Measured for this interventionEffect record E-A13-CRIME

        Mixed: more cautions by 16.

      Outcomes are kept separate and never added into one impact figure.

      5.Best evidence

      • The short- and medium-term effects of Sure Start on children’s outcomes

        Author or institution:
        Carneiro, Cattan, Conti, Crawford, Farquharson and Ridpath, Institute for Fiscal Studies
        Published:
        May 2025
        Evaluation period:
        Sure Start roll-out 1999–2010; outcomes to age 16
        Design:
        Quasi-experimental (variation in distance to and timing of centres)
        Sample or population:
        Children in England born around 1990–2007
        Effect and uncertainty:
        Five A*–C GCSEs +0.9 points; hospital admissions −8.5% by 15; no effect on EHC plans; more cautions
        Limits:
        Today’s hubs differ in services, funding and staffing

        Where: Executive summary pp.6–10; Table ES.1. Affiliation is provenance, not proof of quality.

      • National evaluation of Flying Start: impact report

        Author or institution:
        Welsh Government (Ipsos MORI and SQW)
        Published:
        December 2013
        Evaluation period:
        Survey 2012–13, children aged two to four
        Design:
        Matched comparison survey
        Sample or population:
        2,116 parents in Flying Start and comparison areas
        Effect and uncertainty:
        More health-visitor contact; no significant difference in child development
        Limits:
        Comparison areas less disadvantaged; early outcomes only

        Where: Summary pp.5–8. Affiliation is provenance, not proof of quality.

      6.Alternatives, counter-evidence and who loses

      Closest feasible alternative
      Expanding funded childcare hours instead: reaches more families but is a different intervention whose child-development effects were not reviewed here.
      Strongest counter-evidence
      Flying Start in Wales raised service use without a measurable change in child development by about age three.

      Source: Welsh Government (Ipsos MORI and SQW), National evaluation of Flying Start: impact report (Summary pp.5–8); Institute for Fiscal Studies, The short- and medium-term effects of Sure Start on children’s outcomes (Section 5.2)

      Distributional effects
      Largest gains expected for disadvantaged children; costs now, benefits later and partly to other budgets.
      Main downside
      Slow, modest benefits, and some areas wait longer under a phased roll-out.

      The options compared, including current policy

      • The proposal

        Specify hubs to the evaluated features, weakest access first, phased

        Who is reached
        Disadvantaged under-fives in weak-access areas first
        Evidence
        IFS quasi-experiment; Flying Start caution
        Likely outcome and timing
        Reach in a year; development in two; attainment in 10–15
        Cost and funding
        Within the hub envelope; specification not costed
        Main trade-off
        Some areas wait

      • Current policy

        Current policy: hubs in every authority from April 2026, up to 1,000 by the end of 2028

        Who is reached
        Families in every authority
        Evidence
        No evaluation design announced
        Likely outcome and timing
        Unknown
        Cost and funding
        Over £500m envelope
        Main trade-off
        May not reproduce the evaluated features

      • Alternative

        More funded childcare hours

        Who is reached
        Working parents of young children
        Evidence
        Not reviewed here
        Likely outcome and timing
        Unknown for child development
        Cost and funding
        Separate budget
        Main trade-off
        A different intervention

      7.What other countries or the devolved nations show

      Country lesson · Wales, Programme from 2006–07; impact survey 2012–13

      Flying Start in Wales: more contact, no measured change in child development by age three

      Evidence: Observational analysisPartial match: some conditions differ

      What was changed
      An area-based offer for families with children under four in the most disadvantaged areas: enhanced health visiting, parenting support, early language support and free part-time childcare for two-to-three-year-olds.
      Study design and population
      Matched comparison survey (propensity score matching), intention to treat. 2,116 parents of children aged two to four in Flying Start and comparison areas.
      Measured outcome there
      Much higher service contact (5.7 more health-visitor contacts; take-up of parenting programmes +12.5%), but no statistically significant difference in children’s cognitive and language skills, social and emotional development or self-regulation, or in parents’ confidence or mental health.
      Cost status
      Not assessed in the impact report.
      What failed or remains uncertain
      The weak child results may reflect that Flying Start areas started more disadvantaged than comparison areas, which matching cannot fully correct; outcomes were measured early (about age three), before the school-age effects found for Sure Start in England.
      Why the UK differs
      Flying Start targeted small areas and included childcare; England’s Sure Start evaluation found school-age effects years later. The lesson for England is that reach alone does not show child benefit: the hubs need an evaluation design from the start.
      First change required here
      Phase the Best Start Family Hubs roll-out so that comparable areas start at different times, and measure child development (not just contacts) from the outset.

      Not a UK forecast: the effect above was measured in that place and population. Source: Welsh Government (Ipsos MORI and SQW), National evaluation of Flying Start: impact report (Summary pp.5–8). Reviewed 2026-10-04.

      8.What delivery needs first

      • staff — Early-years practitioners, health visitors and outreach workers. (Scarce resource shared with other priorities; owner: DfE, DHSC and councils)
      • data — An access map and an evaluation design before the remaining roll-out is scheduled. (Requirement; owner: DfE)

      9.What it would cost

      How the cost is built up: Existing hub envelope + the cost of meeting the evaluated specification in weakest-access areas + evaluation.

      • Best Start Family Hubs roll-out

        England

        at least £500m in total, Announced 6 July 2025 (“over £500 million”), hubs in every local authority from April 2026 to the end of 2028; fiscal years and price basis not stated in the announcement., price basis not stated by the source (ledger C-BSFH-2025)

        recurring; price year not stated; April 2026 to the end of 2028; Current quoted cost

      • Evaluated Sure Start service mix and workforce

        Areas of weakest access

        Not yet costed (ledger C-A13-SERVICE)

        recurring; price year not stated; Ongoing; Not yet established

      • Phased-roll-out evaluation

        National

        Not yet established per evaluation

        one-off revenue; price year not stated; Two to three years; Not yet established

      10.Who pays, and what is missing

      Programme envelope (a possible route, not an award)
      Best Start Family Hubs: “over £500 million” to the end of 2028
      Allocation committed to this change
      None
      Remaining gap
      The cost of meeting the specification and of the evaluation is not established; the over-£500m is a programme envelope.
      Proposed financing
      Best Start Family Hubs allocations; health visiting from public-health budgets; evaluation from DfE research budgets.
      Responsible payer
      central government, councils, NHS
      Budget displaced
      Childcare places or other services the same money and staff could fund.

      11.First milestone and time to a useful result

      Hub specification, access map and roll-out order published. (Owner: DfE with councils. Proposed by UK Facts, not an official deadline.)

      Time to first useful result: Reach within a year; child development at two to three in about two years; attainment only as cohorts age.

      What could delay it: Health visitors and practitioners can be recruited. Councils accept a phased order.

      12.How we would know

      Baseline
      Service use and child development at two to three in hub areas before opening.
      Comparison
      Comparable areas whose hubs open later in the phased roll-out.
      Leading indicator
      Reach among disadvantaged families in target areas.
      Follow-up
      Good level of development at five; attainment and admissions as cohorts age.
      Continue if
      Reach rises and early development improves against later-starting areas.
      Adapt if
      Reach among disadvantaged families stays low; change outreach and location.
      Stop if
      No difference in development against later areas once the specification is met.
      Thresholds
      Not yet set: From the evaluation design and power calculation for the phased roll-out.

      What is not yet known, and how to find out

      • What does the evaluated service mix cost per hub? (cost)Next step: Cost the specification in a first set of weakest-access areas.Scope: Staffing and services per hub to the specification · Data or quotes needed: Council staffing plans; health-visiting costs · Who can estimate it: DfE with councils
      • Do today’s hubs reproduce Sure Start’s effects? (effect)Next step: The phased-roll-out comparison.Defensible route: evaluation of existing delivery

      Sources last checked 4 October 2026 · Owner: Evidence editor (James Harris) · Next check: DfE early years foundation stage profile results: 2025 to 2026 (22 October 2026) or the first trigger: Best Start Family Hubs allocations, years or price basis confirmed.

      Decision briefExisting programmeReviewed 2026-10-04

      Spend tutoring money on the gaps pupils actually have, in short fixed blocks — and stop blocks that do not move progress

      The thread: issue, evidence, action

      1. Issue Weakness: 25.8% of disadvantaged pupils got grade 5+ in English and maths GCSE, against 53.1% of other pupils — a 27.3 points gap, against 25.2 points in 2018/19. Also: opportunity, strength
      2. Evidence Disadvantage gap: grade 5+ in English and maths at 16
      3. Action Target tutoring at the gaps pupils actually have
      Decision required
      An academy trust or a council’s school-improvement service agrees one rule for pupil-premium tutoring across its schools: diagnose each pupil’s specific gap, teach a fixed block in groups of two to five linked to class teaching, record attendance and progress, and change or stop blocks that do not beat similar untutored pupils.
      Who and where
      Pupils eligible for the pupil premium — recorded for free school meals in the last six years, or looked-after or previously looked-after — who are behind in English or maths, in the trust’s primary and secondary schools. Applies in: England.
      Accountable body
      Academy trust boards and headteachers (maintained schools with their local authority); the Department for Education sets the pupil premium conditions of grant
      Intervention
      Diagnose each pupil’s specific gap and teach a fixed block in groups of two to five, linked to what the class is learning.
      Best evidence, and where it applies
      Across 62 studies, small-group tuition added about four months’ progress on average and about two months in secondary schools, with moderate evidence security (Education Endowment Foundation toolkit, review updated July 2021).Studied in: 62 studies meeting the Toolkit’s inclusion criteria, with fewer in secondary schools (Studies of varying dates synthesised up to 2021).How closely the comparison transfers to this decision: Madrid and Catalonia, Spain — partial.
      Extra cost of this change
      Capital: none identified in the brief.Recurring: Tutor time, preparation, training and materials — Not yet costed; Tutor or teaching-assistant hours per block (sessions × length ÷ group size) — not yet established; Diagnostic assessment and shared progress check — not yet established.
      Funding
      Not funded yet: a possible funding route exists, but nothing is allocated to this change.Committed to this change: none.Possible source, not an award: none identified.Unfunded: No new money is proposed: the change reallocates pupil premium within schools. Tutoring beyond what a school can reallocate is unfunded, and a national expansion is not costed.
      Savings and wider benefits
      Not yet specified: no cashable saving or avoided cost has been costed for this change. Expected outcomes are listed one by one in the research appendix and are never added together or counted as money available.
      First 90-day step
      Agree the trust-wide targeting and dosage rule and the shared diagnostic and progress check, and name the first cohort in each school, before the start of a term.The proposed first milestone; a 90-day date is not yet specified. Owner: Trust or council school-improvement lead with headteachers. Proposed by UK Facts, not an official deadline.
      Twelve-month milestone
      Not yet specified.Time to a first useful result: Progress on the diagnosed gap at the end of the first block (about a term); key stage results only over several cohorts.
      Outcome, and when to scale or stop
      Measure: Sessions attended against sessions planned, and group sizes actually used.Scale if: Tutored pupils make more progress on the diagnosed gap than similar untutored pupils, with sessions delivered as planned.Stop if: Blocks delivered as planned show no more progress than similar untutored pupils across two successive blocks.
      Main trade-off
      Tutoring uses teaching-assistant or tutor time, often during other lessons, and pupil premium spent on it is not spent on anything else; an average research effect does not guarantee results in a given school.
      Source and review date
      Small group tuition (Live toolkit strand; underlying review last updated July 2021); Tutoring: guidance for education settings (16 May 2024). Reviewed 2026-10-04.

      An editorial proposal by UK Facts for discussion, not government policy. Records: A10 · effects E-A10-ATTAINMENT · costs C-A10-TUTORING

      Research appendix: study design, effect sizes and limits, alternatives, full costing, international comparisons and how we would know

      1.The change, in plain words

      Spend tutoring money on the gaps pupils actually have, in short fixed blocks — and stop blocks that do not move progress

      2.The decision requested

      Decision
      Adopt, for the schools in one trust or local-authority area, a written targeting and dosage rule for pupil-premium tutoring and a shared progress check; and report in each school’s pupil premium strategy statement what was bought, for which pupils, how much was delivered and what changed.
      Accountable institution
      Academy trust boards and headteachers (maintained schools with their local authority); the Department for Education sets the pupil premium conditions of grant
      Geography
      England
      Powers
      Existing: the pupil premium conditions of grant for 2026–27, which require spending on DfE’s menu of approaches (including one-to-one and small group tuition) and an annual strategy statement. No new law or grant.
      Delivery partners
      Teaching assistants and tutors; Subject and SEND leads in each school; Trust or council school-improvement team; Education Endowment Foundation research schools; Virtual school heads (for looked-after children)

      3.Who it is for, and what changes beyond today

      Target population
      Pupils eligible for the pupil premium — recorded for free school meals in the last six years, or looked-after or previously looked-after — who are behind in English or maths, in the trust’s primary and secondary schools.
      Eligibility
      A specific gap diagnosed from the school’s own assessment (for example reading fluency or a number topic), not general low attainment. Pupils with SEND can take part in groups of one or two, aligned with any EHC plan.
      Scale
      Set by each school from its assessment data and the staff time it can protect — for example one block a term for the pupils with the clearest diagnosed gaps — not a national number. Starting across one trust or area lets progress be compared between schools and with pupils not yet tutored.
      Baseline
      25.8% of disadvantaged pupils in England got grade 5 or above in English and maths GCSE in 2024/25, against 53.1% of other pupils — a 27.3-point gap (25.2 points in 2018/19). At 11 the gap in reaching the expected standard in reading, writing and maths was 22.0 points in 2024/25. Pupil premium is £1,550 for each eligible primary pupil and £1,100 for each secondary pupil in 2026–27; UK Facts found no national series showing how much of it buys tutoring or what dose pupils receive.

      Source: Department for Education, Key stage 4 performance, academic year 2024/25 (National characteristics data file (disadvantage status)); Department for Education, Key stage 2 attainment, academic year 2024/25 (revised) (KS2 by pupil characteristics data file); Department for Education, Pupil premium 2026 to 2027: technical note (Per-pupil rates table)

      Change beyond existing law or funding
      Schools can already fund tutoring from the pupil premium, and the subsidised National Tutoring Programme ran only from 2020/21 to 2023/24. The added change is discipline in targeting and dosage: a diagnosed gap before sessions are bought, a fixed block (sessions, weeks, group size) linked to class teaching, attendance and progress recorded, and blocks stopped when they do not move progress.

      4.How it would work

      Do: Diagnose each pupil’s specific gap and teach a fixed block in groups of two to five, linked to what the class is learning Changing: pupils close the particular gap they were behind on, instead of receiving general catch-up time spread thinly

      • Attainment progress of tutored pupils over the block

        Measured for this interventionEffect record E-A10-ATTAINMENT

        An average across 62 studies (about four months; about two in secondary schools). Not a forecast for a particular school or for the national gap.

      • Disadvantage gaps at 11 and 16 over several cohorts

        Hypothesis to test

        No study links school tutoring blocks to the national GCSE gap; the gap can only be tracked over several cohorts, alongside everything else that changes.

      • No loss of progress in the lessons that tutoring replaces

        Hypothesis to test

        A risk to monitor, not a benefit: DfE’s guidance, citing the EEF, says tutoring taken from class time should at least compensate for the teaching missed.

      Outcomes are kept separate and never added into one impact figure.

      5.Best evidence

      • Small group tuition

        Author or institution:
        Education Endowment Foundation (Teaching and Learning Toolkit)
        Published:
        Live toolkit strand; underlying review last updated July 2021
        Evaluation period:
        Studies of varying dates synthesised up to 2021
        Design:
        Synthesis (meta-analysis) of studies of small-group tuition
        Sample or population:
        62 studies meeting the Toolkit’s inclusion criteria, with fewer in secondary schools
        Effect and uncertainty:
        About four months’ additional progress on average over a year; about two months in secondary schools; moderate security (three of five padlocks); effects fall in groups above six or seven.
        Limits:
        Many studies were not independent evaluations and the secondary evidence is smaller. The toolkit’s example price of £70–£100 per pupil for a 15-hour block came from the subsidised first year of the National Tutoring Programme and is not a full 2026 cost.

        Where: Headline impact; “Behind the average”; “How secure is the evidence?”; cost section. Affiliation is provenance, not proof of quality.

      • Tutoring: guidance for education settings

        Author or institution:
        Department for Education
        Published:
        16 May 2024
        Evaluation period:
        Draws on the National Tutoring Programme years 2020/21 to 2023/24, Ofsted’s 2022/23 review and NFER’s 2022/23 evaluation
        Design:
        Non-statutory guidance summarising EEF, Ofsted and NFER findings
        Sample or population:
        Schools and colleges in England
        Effect and uncertainty:
        Repeats the EEF averages (up to five months for one-to-one, up to four months for groups of two to five) and names group size, frequency, course length and curriculum alignment as the drivers of impact; suggests groups of one or two for pupils with SEND.
        Limits:
        Guidance, not an evaluation; sets no dose or minimum standard.

        Where: Guidance PDF pp.3–6. Affiliation is provenance, not proof of quality.

      6.Alternatives, counter-evidence and who loses

      Closest feasible alternative
      Spend the same pupil premium on improving whole-class teaching (for example training in explicit instruction or reading), which reaches every disadvantaged pupil in a class but does not target a specific gap.
      Strongest counter-evidence
      The EEF rates the tuition evidence only moderate: many studies were not independent evaluations, and secondary-school evidence is thinner and smaller. In the Spanish online tutoring trial the test-score gain was imprecise (significant only at the 10% level) and larger for pupils who were higher-achieving at the start.

      Source: Education Endowment Foundation, Small group tuition (“How secure is the evidence?”); Journal of Public Economics, Online tutoring works: experimental evidence from a program with vulnerable children (Introduction (results and mechanisms))

      Distributional effects
      Gains go to disadvantaged pupils with a diagnosed gap. Pupils just above the free-school-meals line — including those eligible only for the expanded free school meals from 2026/27, who do not attract pupil premium — are not targeted unless schools choose to include them; pupils taken out of other lessons lose that teaching.
      Main downside
      Staff time is the constraint: each block uses teaching-assistant or tutor hours that schools take from other work, and blocks that drift on without diagnosis or review spend money without moving progress.

      The options compared, including current policy

      • The proposal

        Targeted tutoring blocks with a diagnosed gap and a progress check (this brief)

        Who is reached
        Current pupils who are behind, primary and secondary — the only option here that reaches pupils already near GCSE. Dosage: a fixed block in groups of two to five. Staff: tutor or teaching-assistant hours. Costing route: pupil premium.
        Evidence
        Synthesis of 62 studies: about four months’ progress (two in secondary), moderate security. UK Facts’ Europe page compares outcomes, not interventions, so it neither supports nor contradicts this choice.
        Likely outcome and timing
        Progress within a term; any effect on the GCSE gap only over several cohorts
        Cost and funding
        Not costed; paid by reallocating pupil premium, nothing new allocated
        Main trade-off
        Staff time and missed lessons; average effects may not appear in every school

      • Current policy

        Current policy: schools choose whether and how to tutor from the pupil premium

        Who is reached
        Varies by school; no national record of which pupils receive what dose
        Evidence
        Same toolkit evidence; the National Tutoring Programme subsidy ended after 2023/24
        Likely outcome and timing
        Unknown
        Cost and funding
        Pupil premium and core budgets
        Main trade-off
        Without a diagnosed gap and a stop rule, low-impact provision can continue indefinitely

      • Alternative

        Nuffield Early Language Intervention (NELI) in reception — an existing funded programme

        Who is reached
        Reception children with the weakest oral language. Dosage: about 20 weeks of small-group and one-to-one sessions. Staff: trained teaching assistants. Costing route: DfE offer free to state schools with a reception class, confirmed to 2028–29 subject to spending rounds; schools supply teaching-assistant time.
        Evidence
        Randomised trial in 193 schools: three months’ extra language progress (effect size 0.26). A language result: it is not evidence that NELI closes the national GCSE gap, which these children reach about 11 years later.
        Likely outcome and timing
        Language gains within the school year
        Cost and funding
        Programme funded by DfE; teaching-assistant time not covered
        Main trade-off
        The marginal change is take-up, fidelity and targeting, not new money; staff time comes from other work

      • Alternative

        Integrated early-years help where access is weakest — see the national decision brief

        Who is reached
        Children under five and their families in areas with weak access. Dosage: a local service package (health visiting, family support, early learning), not a fixed number of sessions. Staff: health visitors, family workers, early-years practitioners. Costing route: Best Start Family Hubs (over £500 million) and public-health budgets.
        Evidence
        IFS evaluation of 2000s Sure Start: five good GCSEs up 0.9 percentage points at 16 for children near a centre; today’s family hubs are a different service
        Likely outcome and timing
        Effects on school results take many years; the GCSE result was measured at age 16
        Cost and funding
        Programme envelope announced; area allocations and the package not costed here
        Main trade-off
        Longest wait for results; does nothing for pupils already in school

      7.What other countries or the devolved nations show

      Country lesson · Madrid and Catalonia, Spain, March to June 2021 (published 2024)

      Two-to-one online maths tutoring in Spain

      Evidence: Experimental or quasi-experimental evaluationPartial match: some conditions differ

      What was changed
      Menπores: eight weeks of after-school online maths tutoring, three 50-minute sessions a week, in fixed groups of two pupils with the same tutor, mostly paid qualified maths teachers; sessions also worked on motivation and a growth mindset.
      Study design and population
      Randomised controlled trial within blocks at 18 secondary schools in highly disadvantaged neighbourhoods; intention-to-treat analysis; an externally graded standardised maths test plus school grades and surveys. 375 pupils in the first two years of secondary school (aged 12 to 15) whom schools identified as most in need (205 assigned to tutoring, 170 to control).
      Measured outcome there
      Standardised maths test +0.26 standard deviations (significant only at the 10% level, p = 0.077); end-of-year maths grades +0.49 standard deviations; lower probability of repeating the year; higher aspirations and self-reported effort.
      Cost status
      Historic: the authors estimate about €300 per pupil for the eight-week programme (2021); not a UK price.
      What failed or remains uncertain
      The test-score effect is imprecise (a small trial; p = 0.077). There was no effect on overall well-being, grit or other subjects, and pupils became more likely to report an external locus of control. The programme worked better for pupils who were higher-achieving at baseline. Volunteer tutors could not be recruited at scale, so paid teachers were used.
      Why the UK differs
      Different curriculum, school day and grade-repetition system; online delivery needed devices and connectivity supplied to pupils. England’s tutoring is funded by schools from the pupil premium and is often delivered in school time by teaching assistants. UK Facts’ /education/europe page compares outcomes (PISA reading and maths, absence) across the four UK systems and Europe; it does not show which intervention produced a country’s result, so it cannot rank tutoring designs.
      First change required here
      Specify tutoring as a fixed dose (sessions, weeks, group size, tutor qualification) aligned with classroom teaching, and record attendance at each session, so a school can tell whether a block was delivered as designed before judging its effect.

      Not a UK forecast: the effect above was measured in that place and population. Source: Journal of Public Economics, Online tutoring works: experimental evidence from a program with vulnerable children (Abstract; section 1 (results, cost); section 3.1 (programme); randomisation and sample). Reviewed 2026-10-04.

      8.What delivery needs first

      • staff — Trained tutors or teaching assistants with protected time for sessions, preparation and liaison with class teachers. (Scarce resource shared with other priorities; owner: Headteachers and trust workforce leads)
      • data — A diagnostic assessment to identify each pupil’s gap and a common progress check, with a record of sessions planned and attended. (Requirement; owner: Trust or council school-improvement team)
      • dependency — Pupils attending school: absent pupils miss sessions as well as lessons. (Requirement; owner: Schools and local-authority attendance services)
      • infrastructure — Timetable slots and space for small groups without displacing core lessons. (Requirement; owner: Headteachers)

      9.What it would cost

      How the cost is built up: Per school: pupils tutored ÷ group size × sessions per block × session length × hourly cost of tutor time (with preparation and training) + diagnostic assessment + materials; across the trust, + the shared progress check. Set against the pupil premium allocation to show what else it displaces.

      • Tutor time, preparation, training and materials

        A defined cohort of disadvantaged pupils in a school or trust

        Not yet costed (ledger C-A10-TUTORING)

        recurring; price year not stated; Each school year; Not yet established

        Read from the ledger record, which has no amount; the old subsidised National Tutoring Programme price is not a 2026 cost.

        Source: Education Endowment Foundation, Small group tuition (Cost widget)

      • Tutor or teaching-assistant hours per block (sessions × length ÷ group size)

        Each pupil receiving a block

        Not yet established per pupil per block

        recurring; price year not stated; Per block (about a term); Not yet established

        To be priced from the trust’s own pay rates, including preparation and liaison time.

      • Diagnostic assessment and shared progress check

        All schools in the trust or area

        Not yet established per trust per year

        recurring; price year not stated; Each school year; Not yet established

      10.Who pays, and what is missing

      Programme envelope (a possible route, not an award)
      None identified
      Allocation committed to this change
      None
      Remaining gap
      No new money is proposed: the change reallocates pupil premium within schools. Tutoring beyond what a school can reallocate is unfunded, and a national expansion is not costed.
      Proposed financing
      The pupil premium (2026–27: £1,550 per eligible primary pupil, £1,100 per secondary pupil, £2,690 for looked-after and previously looked-after children) can already pay for tutoring under DfE’s menu of approaches; otherwise schools’ core budgets. There has been no separate tutoring subsidy since the National Tutoring Programme ended after 2023/24, and nothing new is allocated.
      Responsible payer
      central government
      Budget displaced
      Other pupil-premium uses (attendance and pastoral support, teaching-quality training, enrichment) and teaching-assistant time in classrooms.

      11.First milestone and time to a useful result

      Agree the trust-wide targeting and dosage rule and the shared diagnostic and progress check, and name the first cohort in each school, before the start of a term. (Owner: Trust or council school-improvement lead with headteachers. Proposed by UK Facts, not an official deadline.)

      Time to first useful result: Progress on the diagnosed gap at the end of the first block (about a term); key stage results only over several cohorts.

      What could delay it: Tutor or teaching-assistant time can be protected Pupils attend the sessions A diagnostic assessment is in place before the block Similar untutored pupils are available for comparison

      12.How we would know

      Baseline
      For each tutored pupil: the diagnosed gap and an assessment score before the block. For each school: its key stage 2 and key stage 4 disadvantage gaps and its current tutoring spend from the strategy statement.
      Comparison
      Similar pupils not yet tutored (for example a group starting the next term) and schools in the trust that start later.
      Leading indicator
      Sessions attended against sessions planned, and group sizes actually used.
      Follow-up
      End of each block; the following term, to see whether gains last; key stage results over several cohorts.
      Continue if
      Tutored pupils make more progress on the diagnosed gap than similar untutored pupils, with sessions delivered as planned.
      Adapt if
      Sessions are missed or groups grow beyond five: fix scheduling, attendance or staffing before buying more.
      Stop if
      Blocks delivered as planned show no more progress than similar untutored pupils across two successive blocks.
      Thresholds
      Not yet set: What counts as enough extra progress depends on the assessment used, the cohort size and the cost per block, which each trust must set before starting.

      What is not yet known, and how to find out

      • What does a tutoring block cost per pupil at current pay rates? (cost)Next step: Price a standard block from the trust’s pay rates before the first term and record actual hours delivered.Scope: One block per pupil (sessions, weeks, group size) in the trust’s schools · Data or quotes needed: Teaching-assistant and tutor hourly costs, preparation and training time, assessment time and materials · Who can estimate it: Trust finance and school business managers, using EEF cost guidance
      • Does a trust-wide targeting and stop rule raise progress compared with schools’ current tutoring? (effect)Next step: Run it across one trust with a staggered start and compare tutored pupils with similar pupils not yet tutored.Defensible route: bounded pilot
      • Does secondary tutoring narrow the GCSE gap for the pupils who receive it? (effect)Next step: Link block records to key stage 4 results for tutored and comparison pupils over several cohorts.Defensible route: evaluation of existing delivery
      • How much pupil premium currently buys tutoring, and at what dose? (data)Next step: Compile spend, pupils, sessions and group sizes from the trust’s published strategy statements.

      Sources last checked 4 October 2026 · Owner: Evidence editor (James Harris) · Next check: DfE key stage 4 performance, 2026 (15 October 2026) or the first trigger: A change to the pupil premium or a new tutoring grant.

      What it would take

      The order things have to happen in, who pays, and the effects and appraisals on record.

      The order things have to happen in

      1. First, the prerequisite

        Diagnose gaps: language screening in reception; attainment gaps later

        Owner: Schools

        No date: the timing depends on the steps before it.

      2. Then, the decision

        Allocate protected staff time from pupil premium or school budgets

        Owner: Headteachers and trusts

        No date: the timing depends on the steps before it.

      3. Then, delivery

        Deliver defined blocks; map and specify early-years services where access is weakest

        Owner: Schools; councils

        No date: the timing depends on the steps before it.

      4. Finally, the measured outcome

        Gaps at 11 and 16 narrow over several cohorts

        Owner: DfE statistics

        No date: the timing depends on the steps before it.

      5. First, the prerequisite

        Publish childcare gaps by council: eligibility codes issued against validated, by age

        Owner: DfE

        No date: the timing depends on the steps before it.

      6. Then, delivery

        Commission places in named gap areas, with comparison councils

        Owner: Councils

        No date: the timing depends on the steps before it.

      7. Then, delivery

        Fund completion support for construction apprentices in shortage trades

        Owner: DfE and Skills England

        No date: the timing depends on the steps before it.

      Who pays, and what is still missing

      Cost, committed funding and the remaining gap for the same scope and years. A programme envelope is not an award to a project; a proposal is not committed money. None of these costs has been established yet, so no total is shown.

      Cost and funding ledger

      • Tutor time, preparation, training and materials

        A10 · recurring spending · not costed

        Amount
        Not yet costed
        Who pays
        central government
        Committed
        None identified
        Proposed, not committed
        —
        Gap
        Not established

      • NELI materials and training

        A11 · recurring spending · historical price only

        Amount
        Not yet costed
        Who pays
        central government
        Committed
        None identified
        Proposed, not committed
        DfE-funded NELI offer to every state school with a reception class
        Gap
        Not established

      • Teaching-assistant delivery time, cover and supervision

        A11 · recurring spending · not costed

        Amount
        Not yet costed
        Who pays
        central government
        Committed
        None identified
        Proposed, not committed
        —
        Gap
        Not established

      • Best Start Family Hubs roll-out (programme envelope)

        A13 · recurring spending · published programme envelope

        Amount
        at least £500m in total, Announced 6 July 2025 (“over £500 million”), hubs in every local authority from April 2026 to the end of 2028; fiscal years and price basis not stated in the announcement., price basis not stated by the source
        Who pays
        central government
        Committed
        None identified
        Proposed, not committed
        Best Start Family Hubs: “over £500 million” to the end of 2028
        Gap
        Not established

      • Integrated early-years service package and workforce

        A13 · recurring spending · not costed

        Amount
        Not yet costed
        Who pays
        central government, councils, NHS
        Committed
        None identified
        Proposed, not committed
        —
        Gap
        Not established

      “Not yet costed” means no defensible cost has been found — not that the action is free. Lifetime benefits, benefit-cost ratios and forecast savings are never shown here as money available to spend.

      Effects on record, with evidence ratings (5 records)

      What the evidence says each action changes

      Effects on record — these measure different outcomes, so they are not drawn on one scale
      ActionOutcome and areaEffectCompared withEvidence
      Target tutoring at the gaps pupils actually haveAttainment progress of pupils receiving small-group tuition Education and children+4 months (primary); +2 months (secondary)moderate evidence security (3 of 5 padlocks); 62 studiespupils taught as usual; over a school year; review updated July 2021evaluated; moderate confidence; Studies in several countriesAn average across studies; not a predicted change in the national GCSE gap.
      Meet early language needs before children fall further behindOral language of reception pupils with the weakest language Education and children+3 months (effect size 0.26)effect size 95% CI 0.17 to 0.35; 5 of 5 padlocksrandomised control schools; 2018–19 school yearevaluated; high confidence; 193 schools in EnglandA language outcome; not evidence of fewer EHC plans or lower SEND demand. The 2023 follow-up lost 55.5% of pupils.
      Rebuild integrated early-years help where access is weakestFive GCSEs at A*–C (or equivalent) at 16 Education and children+0.9 percentage points (about 2%)quasi-experimental; method update changed some earlier findingschildren without access, before and after centres opened; cohorts born around 1990–2003evaluated; moderate confidence; England2000s Sure Start, not today’s family hubs.
      Rebuild integrated early-years help where access is weakestHospital admissions by age 15 Health and care−8.5% by age 15admissions rose 10% at age 1 before fallingareas with fewer centres per child; cohorts born 1990s–2000sevaluated; moderate confidence; EnglandNot a cashable NHS saving.
      Rebuild integrated early-years help where access is weakestCriminal record by age 16 Crime and justice · mixedSize not established—evaluated; moderate confidence; EnglandMore young people had a criminal record by 16, driven by an 11% rise in cautions; a fall in convictions was not significant. A trade-off to watch, not a benefit.
      Past appraisals, kept apart from funding (1)
      What 2000s Sure Start cost and what the IFS estimates it returned, per cohort

      Children with access to a Sure Start centre in England, cohorts born around 1990–2007. Horizon: lifetime of each cohort (benefits discounted at 3.5%). 2023-24 prices, discounted. A historical appraisal — not a budget for a new programme and not cash savings.

      • Upfront cost: £2.7bn

      • Estimated benefits: £5.5bn in total — to government (lower spending and extra tax) £2.4bn; to individuals (higher earnings) £3.1bn

      The cost is peak 2010 annual spending used as a stand-in for one cohort. Benefits rest on modelled lifetime earnings. None of this is money available to fund a new service today, and today’s family hubs are not the same programme.

      Source: Institute for Fiscal Studies, The short- and medium-term effects of Sure Start on children’s outcomes, Table ES.1, printed p.11

      How we would know

      Disadvantage gap: grade 5+ in English and maths at 16; Disadvantage gap: expected standard in reading, writing and maths at 11; Pupils persistently absent; and 3 more.

      The outcome measures to watch. A baseline is where we start, not a target; where there is no official target we say so.

      • Disadvantage gap: grade 5+ in English and maths at 16

        percentage points · England

        Baseline
        25.2 points
        2018/19
        Latest
        27.3 points
        2024/25

        Target: none — there is no official target for the gap.

        Source: Department for Education, Key stage 4 performance, academic year 2024/25 (National characteristics data file). Next release: annual (October provisional, spring revised) (check the release calendar).

      • Disadvantage gap: expected standard in reading, writing and maths at 11

        percentage points · England

        Baseline
        20.0 points
        2018/19
        Latest
        22.0 points
        2024/25

        Target: none — there is no official target for the gap.

        Source: Department for Education, Key stage 2 attainment, academic year 2024/25 (revised) (KS2 by pupil characteristics data file). Next release: annual (check the release calendar).

      • Pupils persistently absent

        % of pupils · England

        Baseline
        10.9%
        2018/19
        Latest
        18.1%
        2024/25

        Target: none — no numerical national target is set.

        Source: Department for Education, Pupil absence in schools in England, 2024/25 (Persistent absentees time series). Next release: annual (March) (check the release calendar).

      • Eligible disadvantaged two-year-olds registered for funded early learning

        % of eligible children · England

        Baseline
        74.8%
        January 2024
        Latest
        71.6%
        January 2026

        Target: none — no official take-up target.

        Source: Department for Education, Funded early education and childcare, reporting year 2026 (Data file “Headline figures” (early learning for 2-year-olds)). Next release: annual (July 2027) (check the release calendar).

      • Eligible children aged 9 months to 2 years registered for working-parent hours

        % of eligible children (estimate) · England

        Baseline
        72%
        January 2025
        Latest
        80%
        January 2026

        Target: none — no official take-up target.

        Source: Department for Education, Funded early education and childcare, reporting year 2026 (Data file “Headline figures” (working parents, 9 months to 2 years)). Next release: annual (July 2027; DfE plans to revise 2026) (check the release calendar).

      • Construction apprenticeship achievement rate

        % of apprentices leaving who achieved · England

        Baseline
        Not published
        Latest
        58.3%
        2024/25

        Target: none — no official target for this subject area.

        Source: Department for Education, Apprenticeships, academic year 2025/26 (National achievement rates tables – impact of sector subject area). Next release: annual (national achievement rates tables) (check the release calendar).

      Measured results

      What each option would be judged on, and the latest value of the measures already on this page. A measure moving is not proof that an option caused it.

      Where the ideas come from

      Research, counter-evidence and methods (18 sources; 4 claims held back)
      • Nuffield Early Language Intervention (re-grant): evaluation report (2020) and addendum (2023) — Education Endowment Foundation / RAND / University of Oxford (Sashka Dimova, Sonia Ilie, Elena Rosa Brown, Miriam Broeks, Andreas Culora, Alex Sutherland (RAND Europe); addendum: Merrilyn Groom, Elena Rosa Brown, Lydia Lymperis), 2020-05 (report); 2023-03 (addendum).Two-arm cluster randomised trial (97 treatment and 96 control schools), June 2018 to November 2019; 193 schools, 1,156 reception pupils with the lowest language scores in each class (1,071 in the primary analysis); England. Full text checked (read from an Internet Archive copy), 2026-10-04. Limits: No quantitative SEND or EHC-plan finding; staff felt it worked less well for pupils with SEN. Cost of £870 per one-form-entry school (£58 per child a year over three years) covers materials and training only, not teaching-assistant time.
      • Small group tuition — Education Endowment Foundation, Live toolkit; review last updated July 2021.Evidence synthesis (Teaching and Learning Toolkit); 62 studies. Full text checked (read from an Internet Archive copy), 2026-10-04. Limits: Lost a padlock because many studies were not independently evaluated. An average across studies, not a promise for any school.
      • Home to school transport — National Audit Office, date not stated.Public audit; England. Relevant passages checked, 2026-10-07.
      • Key stage 4 performance, academic year 2024/25 — Department for Education, 23 April 2026 (updated 4 August 2026).Official statistics; England. Relevant passages checked, 2026-10-04.
      • Pupil absence in schools in England, 2024/25 — Department for Education, 26 March 2026.Official statistics; England. Relevant passages checked, 2026-10-04.
      • Funded early education and childcare, reporting year 2026 — Department for Education, 2026-07-02.Accredited official statistics; England. Full text checked, 2026-10-07. Limits: January census. Eligible-children estimates for the working-parent entitlement are survey-based, rounded and to be revised (DfE: treat with caution); the universal 3–4 percentage uses ONS population estimates and projections. Stored as early-years-* series (scripts/state-of-nation/ingest-childcare.py). Updated 29 July 2026; next release July 2027.
      • Apprenticeships, academic year 2025/26 — Department for Education, 2026-07-16.Official statistics; England. Relevant passages checked, 2026-10-07. Limits: The 2025/26 in-year figure covers August to April only and must not be read as a full year; the full year is due November 2026.
      • Construction Workforce Outlook 2026–2030 — CITB, 2026-06-17.Industry training board forecast; United Kingdom. Relevant passages checked, 2026-10-04.
      • The short- and medium-term effects of Sure Start on children’s outcomes — Institute for Fiscal Studies (Pedro Carneiro, Sarah Cattan, Gabriella Conti, Claire Crawford, Christine Farquharson, Nick Ridpath), 2025-05-22.Difference-in-differences using the staggered roll-out of Sure Start centres (access = living within 2.5km of a centre), with a cost-benefit analysis; England cohorts born around 1990–2007 (varies by outcome); National Pupil Database, Hospital Episode Statistics, police and court records; England. Full text checked (read from an Internet Archive copy), 2026-10-04. Limits: Updated methods changed earlier findings (the 2024 paper’s fall in EHC plans disappears). Today’s family hubs are not automatically equivalent to 2000s Sure Start.
      • Build Back Fairer: The COVID-19 Marmot Review — UCL Institute of Health Equity / Health Foundation (Michael Marmot, Jessica Allen, Peter Goldblatt, Eleanor Herd, Joana Morrison), 2020-12.Review of evidence and data on COVID-19 and inequalities; England. Relevant passages checked (read from an Internet Archive copy), 2026-10-04. Limits: A framework of recommendations; no cashable return and no specific fuel-poverty recommendation in the executive summary.
      • National evaluation of Flying Start: impact report — Welsh Government (Ipsos MORI and SQW), 2013-12.Intention-to-treat survey of families in Flying Start and comparison areas, matched by propensity score on family characteristics.; 2,116 parents: 1,033 in Flying Start areas and 1,083 in comparison areas; Wales. Relevant passages checked (Read from the copy laid before the Senedd Children, Young People and Education Committee.), 2026-10-04. Limits: Flying Start areas were more disadvantaged than comparison areas, which matching cannot fully correct; the evaluators suggest parity with less disadvantaged areas may itself be a gain. Outcomes measured by about age three.
      • Does more free childcare help parents work more? — Labour Economics 74 (2022); IFS working paper W16/22 (Brewer M, Cattan S, Crawford C, Rabe B), 2022.Regression discontinuity on date-of-birth eligibility cut-offs for free part-time nursery places and full-time school entry.; England. Relevant passages checked (The IFS page returned HTTP 403 on 7 October 2026; read through RePEc and the ISER summary (https://www.iser.essex.ac.uk/?p=22032).), 2026-10-07. Limits: The free hours largely replaced care parents already used (about 54 extra hours a year of formal care against 570 free hours); effects come from the school-entry setting, not under-3s.
      • Expanded childcare entitlement: process evaluation report — Department for Education (IFF Research), 2026-09.Three waves of surveys and interviews with councils (88 in wave 3), providers and parents.; England. Relevant passages checked, 2026-10-07. Limits: Self-reported throughout; the separate impact evaluation is due late 2027 (section 1).
      • Childcare Survey 2026 — Coram Family and Childcare, 2026-03-18.Independent survey of local authorities; England (councils). Relevant passages checked, 2026-10-07. Limits: Self-reported by councils (137 responses for most rows; 72 for rural areas); a council judging that it has enough childcare for at least 75% of children is a local assessment, not a measured count. Secondary-source analysis.
      • Preparations to extend early years entitlements for working parents in England (HC 701) — National Audit Office, 2024-04-24.Audit report; England. Relevant passages checked, 2026-10-07.
      • The net present value of further education in England, 2021 to 2022 — Department for Education, 2025-02.Compares completers with people who start the same qualification but do not finish.; England. Relevant passages checked, 2026-10-07. Limits: The report says construction premia need cautious consideration because 24% of completers become self-employed, which the data exclude; averages, not the return on an extra place.
      • Government revives family services supporting 500,000 more kids (press release) — Department for Education, 2025-07-06.Government announcement; England. Full text checked, 2026-10-04. Limits: A programme announcement: fiscal years and price basis are not stated, and no evaluation design is described.
      • Key stage 2 attainment, academic year 2024/25 (revised) — Department for Education, date not stated.Official statistics; England. Relevant passages checked, 2026-10-04.

      Counter-evidence and criticism

      • Target tutoring at the gaps pupils actually have: Many studies were not independent evaluations, and secondary-school evidence is thinner and smaller. (Education Endowment Foundation)
      • Meet early language needs before children fall further behind: Staff felt it worked less well for pupils with SEN; there is no evidence it reduces EHC plans or overall SEND demand. (Education Endowment Foundation / RAND / University of Oxford)
      • Rebuild integrated early-years help where access is weakest: Updated methods removed an earlier fall in EHC plans; crime results are mixed (more cautions); benefits are largely modelled lifetime earnings. Wales’s Flying Start raised service contact but found no measurable difference in child development by about age three. (Institute for Fiscal Studies; Welsh Government (Ipsos MORI and SQW))
      • Commission childcare places where councils can show parents cannot use the funded hours: Free hours mostly paid for care families already used, so extra places may add little work; and if staff and providers are the constraint, commissioning money may not create places at all. (Labour Economics 74 (2022); IFS working paper W16/22; National Audit Office)
      • Raise completion in construction apprenticeships in the shortage trades: Completion support may raise qualifications without adding workers if completers would have worked in construction anyway; overseas recruitment and pay may matter more. (Department for Education)

      Claims held back

      • These programmes would close a stated share of the national GCSE gap — Study effects are measured on different outcomes and populations; no model links them to the national gap. Needed: A modelled scenario with explicit assumptions, or cohort evaluation.
      • Current coverage and funding of family hubs and Best Start services — Not verified in this review. Needed: DfE programme documents and allocations.
      • The household income threshold for the disadvantaged two-year-old offer — The GOV.UK page could not be retrieved on 7 October 2026. Needed: The GOV.UK eligibility page for free education for 2-year-olds.
      • How many phase 2 school-based nursery places opened in September 2026 — No confirmation found after the September 2025 announcement. Needed: A DfE delivery update.

      “Full text checked” means we read the report itself; “relevant passages” means the sections behind each figure; “summary or abstract only” means we could not read the full study, so no number from it is published here without another source.

      Other options compared — not on the shortlist (3)

      Other options compared

      Options from earlier work on this page that are not on the shortlist. They remain valid choices; their evidence is shown as before.

      • preventionStatus: operating

        Targeted attendance support with families

        Aims to:
        Fewer pupils persistently or severely absent, especially those already behind.
        How:
        Schools and councils identify pupils at risk early, find the specific barrier (health, anxiety, transport, caring) and work with the family, rather than relying on fines alone.
        Who decides:
        Schools, councils’ attendance support teams; DfE (statutory guidance).
        Main constraint:
        The causes sit partly outside school — children’s mental health services, SEND support, poverty — so school action alone has limits.
        Evidence:
        Limited evidence — plausible mechanism, little direct evaluation
        See the trade-offs and evidence
        What the evidence shows
        The EEF’s 2022 review of 72 studies found the overall evidence on attendance interventions weak. Targeted parental engagement and responsive approaches matched to pupils’ needs showed some promise; parental communication had a positive but very small average effect.
        Policy status
        Statutory attendance guidance in force since August 2024.
        Resources
        Attendance staff in schools and councils; access to health and mental-health services.
        Consent and decisions needed
        Penalty notices for parents are contested; support-first approaches need resources.
        Time
        Termly attendance data show change within a year; attainment effects take longer.
        Cost and yield
        No official national costing of a support-first model.
        Trade-offs and who pays
        Enforcement is cheap but can damage relationships with families; intensive support is costly per pupil.
        How we would know it worked
        Leading indicator: Persistent and severe absence rates each term. Outcome: Attainment of previously absent pupils. It would not be working if: Absence falls only where enforcement rises, with no change in severe absence.
      • capacityStatus: operating

        Pay incentives to keep shortage-subject teachers in disadvantaged schools

        Aims to:
        Fewer vacancies and less turnover in maths and science in schools with many disadvantaged pupils.
        How:
        Annual payments to early-career teachers of shortage subjects who stay in eligible schools reduce the number who leave.
        Who decides:
        DfE; schools and trusts (recruitment).
        Main constraint:
        Retention is one part of supply; recruitment into training and workload also matter.
        Evidence:
        Moderate evidence — evaluated, with limits on scope or transfer
        See the trade-offs and evidence
        What the evidence shows
        An evaluation of England’s earlier retention payments for early-career maths and physics teachers found eligible teachers were 23% less likely to leave state schools while eligible, at lower cost per teacher retained than training a replacement. Effects on pupil attainment were not measured.
        Policy status
        Targeted retention incentive of £3,000–£6,000 for eligible chemistry, computing, maths and physics teachers in 2025–26.
        Resources
        Annual payments; good data on which schools and subjects are short.
        Consent and decisions needed
        Targeted pay is contested by some unions as unequal; Treasury approval for continued funding.
        Time
        Retention effects show within a year; any attainment effect would take several years.
        Cost and yield
        Set by eligibility and payment rate; no published total costing cited here.
        Trade-offs and who pays
        Money for targeted payments is not available for general pay or other support.
        How we would know it worked
        Leading indicator: Leaving rates of eligible teachers. Outcome: Vacancy and turnover rates in disadvantaged schools, and attainment in those subjects. It would not be working if: Leaving rates fall but vacancies in the targeted schools do not.
      • capacityStatus: announced

        Timely specialist support for pupils with SEND

        Aims to:
        Pupils with special educational needs get assessed and supported before they fall further behind.
        How:
        More educational psychologists and therapists, faster assessments, and support in mainstream schools without waiting for a plan.
        Who decides:
        DfE, councils, NHS commissioners and schools.
        Main constraint:
        Specialist staff take years to train; council high-needs budgets are in deficit.
        Evidence:
        Limited evidence — plausible mechanism, little direct evaluation
        See the trade-offs and evidence
        What the evidence shows
        No national data show whether the support in plans is delivered or what difference it makes. The SEND chapter compares the options in detail.
        Policy status
        February 2026 schools white paper proposals; legislation not yet introduced.
        Resources
        Specialist staff, specialist places and high-needs funding — see the SEND chapter.
        Consent and decisions needed
        Changes to legal entitlements need legislation and are contested by families.
        Time
        Years for workforce; timeliness data each June.
        Cost and yield
        See the SEND options for announced sums.
        Trade-offs and who pays
        See the SEND chapter.
        How we would know it worked
        Leading indicator: Share of new plans issued within 20 weeks. Outcome: Attainment and destinations of pupils with SEN. It would not be working if: Timeliness improves while outcomes for pupils with SEN do not.
      The options side by side
      OptionResourcesTimeEvidenceOutcome to watch
      Targeted attendance support with familiesAttendance staff in schools and councils; access to health and mental-health services.Termly attendance data show change within a year; attainment effects take longer.limitedAttainment of previously absent pupils.
      Pay incentives to keep shortage-subject teachers in disadvantaged schoolsAnnual payments; good data on which schools and subjects are short.Retention effects show within a year; any attainment effect would take several years.moderateVacancy and turnover rates in disadvantaged schools, and attainment in those subjects.
      Timely specialist support for pupils with SENDSpecialist staff, specialist places and high-needs funding — see the SEND chapter.Years for workforce; timeliness data each June.limitedAttainment and destinations of pupils with SEN.

      Plan reviewed 2026-10-07. How we judge evidence.

      Which responses belong in the national plan?

      The shared assessments below hold the costs, funding gaps, risks and ten-criterion profiles. Their current national status comes from the approved plan. Education evidence alone does not establish a national priority ranking.

      Further candidates for appraisal

      SEND provision tracking; evaluated attendance support; targeted teacher retention; estate renewal; and adult basic-skills access merit further appraisal. They are research questions, not additions to the approved lists. Each needs an incremental cost, responsible owner, delivery capacity, distributional assessment and credible outcome evaluation before scoring or selection.

      How this connects

      Sources and methods (5)

      Method notes

      • England only: education is devolved, and Scotland, Wales and Northern Ireland use different qualifications, measures and legal frameworks.
      • Disadvantaged pupils are those eligible for free school meals at any point in the past six years, looked after, or adopted from care. Free school meal eligibility rules changed from April 2018, which affects the group over time.
      • Attainment comparisons use exam years only. 2019/20 and 2020/21 GCSE grades were teacher-assessed; 2021/22 grading sat between pandemic and pre-pandemic standards; there were no key stage 2 tests in 2019/20 or 2020/21.
      • Gaps are differences in percentage points between two published rates, not percentage changes and not grade points.