Thanks to Claude AI for writing the following from my notes.

LABOUR PARTY EDUCATION POLICY

A POLICY PROPOSAL

Raising Standards in Primary Education Through Direct Instruction

Status: Draft Policy Proposal | Audience: Shadow Education Team | February 2026

Executive Summary

Around one in four children in England leave primary school unable to read at the expected level, with comparable deficits in writing and arithmetic. This failure has profound lifelong consequences: poor literacy is associated with limited employment prospects, lower earnings, poorer health outcomes, and reduced civic participation. The economic cost to the UK has been estimated at £81 billion per year.

A substantial body of research, including the largest educational experiment ever conducted in the United States, consistently identifies Direct Instruction (DI) — the systematic, scripted teaching method developed by Siegfried Engelmann — as the most effective pedagogical approach for teaching foundational literacy and numeracy in primary schools. Yet DI is used in fewer than 1% of UK primary schools.

This paper argues that a Labour government should introduce a phased, supported programme of Direct Instruction adoption in primary schools, beginning in areas with the worst educational outcomes. This is not a proposal to eliminate teacher autonomy across the board; it is a targeted, evidence-based intervention for foundational skills where the evidence for DI’s superiority is especially strong.

1. The Scale of the Problem

The failure of England’s primary education system to reliably teach foundational skills is well-documented and persistent:

• In 2024, 26% of 11-year-olds left primary school in England not meeting the expected reading standard (Department for Education, 2024).
• In the same year, 29.3% of pupils did not reach the expected standard in the combined reading, writing and maths measure at Key Stage 2 (DfE, 2025).
• Approximately 5 million working-age adults in England have below-functional literacy skills, equivalent to the expected level for an 11-year-old. Some 8 million have below-functional numeracy skills (Full Fact, citing DfE Skills for Life Survey).
• The OECD’s 2023 Survey of Adult Skills found that 18% of adults in England scored at Level 1 or below in literacy — a significant cohort unable to handle anything beyond the most basic written material.
• Poor literacy is associated with increased risk of unemployment, poor health, low self-esteem, and involvement in the criminal justice system (National Literacy Trust, 2024).

These are not marginal failures. They represent a systematic inability of the current approach to ensure that all children acquire foundational skills — skills which no child should leave primary school without.

2. What is Direct Instruction?

It is important to be precise about terminology. In educational discourse, ‘direct instruction’ (lower case) often refers loosely to teacher-led, explicit teaching of any kind. This paper uses ‘Direct Instruction’ (DI, capitalised) to refer specifically to the structured, scripted pedagogical model developed by Siegfried Engelmann and colleagues, originally at the University of Illinois in the 1960s.

The three main approaches to primary teaching can be distinguished as follows:

Approach	Core Characteristics	Degree of Structure
Ordinary Teaching	Teacher-designed, flexible, improvised. Professional judgement governs pacing, examples, and content.	Low
Explicit Instruction	Systematic, intentional teaching of specific skills. Structured sequences with clear modelling and guided practice, but teacher-designed.	Moderate
Direct Instruction (DI)	Highly scripted commercial programme (e.g. Engelmann’s DISTAR/Reading Mastery). Word-for-word scripts, fixed sequencing, controlled examples, pre-tested lesson structures, high rates of student response and immediate feedback.	Very High

The defining characteristics of Engelmann-style DI include:

• Small-group ability-based groupings, with children seated close to the teacher without desks
• Scripted lesson delivery specifying exactly what the teacher says and in what order
• Carefully sequenced instructional examples, designed to prevent ambiguity in concept formation
• High rates of group and individual student responding, cued by the teacher
• Immediate corrective feedback, preventing errors from becoming embedded
• Mastery-based progression: students do not advance until prerequisite skills are secure

Engelmann originally developed the principles of DI while working in advertising — a context which demands that messages communicate precisely and unambiguously with the target audience. He applied the same discipline to instructional design, treating teaching as a communication problem requiring rigorous engineering.

3. The Evidence Base

3.1 Project Follow Through (1968–1977)

The most comprehensive evaluation of educational interventions ever undertaken remains Project Follow Through, a US federal programme that compared 20 different educational models across approximately 700,000 disadvantaged children (ages 5–9) in 170 communities over nine years. Data were collected by Stanford Research Institute and independently analysed by Abt Associates.

Project Follow Through tested three broad categories of intervention:

• Affective models — focused on raising self-esteem as a precondition for learning
• Cognitive models — emphasising higher-order thinking and problem-solving, on the assumption that basic skills would follow
• Basic skills models — including Engelmann’s Direct Instruction, which taught foundational skills systematically from the ground up

The results were unambiguous: Direct Instruction outperformed every other model on all three outcome measures — academic achievement, cognitive development, and — strikingly — affective outcomes including self-esteem and self-confidence. This last finding confounded critics who argued that DI’s structured approach would harm children’s motivation or wellbeing. In fact, DI students scored higher on self-concept measures than students in programmes specifically designed to raise self-esteem (Watkins, 1996).

Longitudinal follow-up studies found that children who received DI through to Grade 3 had significantly higher high school graduation rates (60% vs 40%), lower dropout rates, more grade promotions, and higher college acceptance rates than matched comparison groups (Darch, Gersten & Taylor, 1987; Meyer, Gersten & Gutkin, 1983).

Despite these results, the findings were largely suppressed by the US Office of Education, which declined to disseminate them. The educational establishment — including schools of education, teacher unions, and commercial publishers — argued against the data. As Engelmann noted, the approaches that performed worst in Project Follow Through were those most strongly promoted by the major teaching associations (Engelmann, 1992).

3.2 Meta-Analyses and Systematic Reviews

Project Follow Through was not an isolated finding. Subsequent decades of research have consistently replicated and extended its conclusions:

• Adams & Engelmann (1996) meta-analysed 25 years of DI research and found an overall effect size of d = 0.97 — roughly equivalent to a full year of additional schooling. Effect sizes for reading outcomes were d = 0.69 and for spelling d = 1.33.
• Stockard et al. (2018) conducted a meta-regression of DI studies and found an average effect on test scores of approximately d = 0.60 standard deviations — consistently positive across diverse student populations.
• John Hattie’s Visible Learning (2009), a synthesis of over 800 meta-analyses covering 42,000+ students, found DI to have an average effect size of 0.59 — significantly larger than any other curriculum model in the study.
• Kirschner, Sweller & Clark (2006) argued on theoretical grounds that minimally guided instruction is broadly inferior to explicit, structured teaching, particularly for novice learners whose working memory is easily overloaded without expert scaffolding.

As the authors of a 2021 review in PMC concluded: ‘Over the past half-century, no other curriculum model has been more heavily researched than DI. Syntheses of this research tell us that by equipping teachers with DI curricular materials, and emphasising high fidelity of implementation, the effects of schools can be substantial’ (Stockard et al., cited in PMC, 2021).

3.3 The Self-Esteem Finding

One of the most counter-intuitive and important findings from the DI evidence base concerns pupil wellbeing. Critics often argue that scripted, structured teaching is joyless and harmful to children’s self-concept. The data consistently show the opposite.

A plausible explanation is that a large proportion of disruptive classroom behaviour is ‘escape behaviour’ — children acting out to avoid tasks they find confusing, humiliating, or impossible. DI removes this trigger by teaching in small, unambiguous steps, ensuring high rates of correct responses, preventing pupils from getting stuck, and providing immediate corrective feedback. Children who succeed at tasks feel competent; children who feel competent develop positive self-concept.

4. Why Hasn’t Direct Instruction Been Adopted?

Given the strength of the evidence, the near-total absence of DI from UK primary schools demands explanation. Several factors are relevant:

4.1 Teacher Resistance and Professional Identity

Teacher resistance is the most widely documented barrier. DI’s scripted nature is perceived by many teachers as a challenge to their professional identity, autonomy, and creativity. Teachers trained within constructivist or inquiry-based frameworks often experience DI as philosophically misaligned with their beliefs about how children learn.

Teaching unions in the UK strongly defend teachers’ right to exercise professional judgement in the classroom. Scripted programmes are sometimes characterised as ‘deskilling’, treating teachers as technicians rather than professionals. This framing, while understandable, conflates the freedom to improvise with professional effectiveness.

There is also a stigma problem: DI is often (mis)associated with ‘drill and kill’ methods or with provision for low-attaining and SEND pupils only. In fact, Project Follow Through demonstrated DI’s superiority for all demographic groups — poor students, non-poor students, urban and rural students, and across all ethnicities studied.

4.2 The Educational Establishment’s Interests

It is worth acknowledging a structural tension that the evidence base itself cannot fully resolve. Educational researchers, teacher trainers, curriculum designers, and pedagogy academics all have professional and institutional interests in teaching being understood as complex, creative, and requiring sophisticated expert judgement. A finding that scripted, systematic instruction outperforms artisanal teaching is professionally inconvenient for a large portion of the education sector.

This is not to suggest bad faith — but it is to note that the suppression of Project Follow Through’s findings, and the continuing resistance to DI in schools of education, cannot be explained by evidence alone. As Engelmann himself observed, the methods that performed worst in Follow Through have continued to be most strongly promoted by professional teaching organisations.

Where possible, this policy proposal relies on outcome data and research from outside the core educational establishment — including economics of education research, independent meta-analyses, and government attainment statistics — to reduce the influence of this potential bias.

4.3 Implementation Demands

DI requires genuine fidelity to be effective. Teachers must be trained properly, implementation must be monitored, and schools must commit to the programme with adequate time and support. In a policy environment already crowded with initiatives, DI can feel like ‘one more thing’. A serious adoption programme must address these implementation barriers directly through funded training, dedicated support structures, and realistic timelines.

5. Policy Proposal

5.1 Principles

This proposal is guided by three principles:

• Evidence should drive pedagogy. Where the evidence overwhelmingly supports a particular method for teaching foundational skills, that evidence should be acted upon, not deferred to professional preference.
• No child should leave primary school without functional literacy and numeracy. These are not optional outcomes; they are a prerequisite for everything else education offers.
• Investment in effective early teaching reduces the long-term cost of remediation. The per-pupil cost of specialist intervention for pupils who fall behind significantly exceeds the cost of effective whole-class instruction.

5.2 Proposed Measures

Phase 1: Targeted Introduction in High-Need Areas

In the first parliamentary term, the government will introduce a funded DI adoption programme in primary schools in the 25% of local authority areas with the worst Key Stage 2 reading and numeracy outcomes. Schools in these areas will be offered:

• Fully funded training in Engelmann-style DI for literacy and numeracy, delivered by accredited trainers
• Access to approved DI curriculum materials at subsidised cost
• Dedicated DI implementation support staff, embedded in schools for the first year of adoption
• Ofsted inspection frameworks that treat evidence-based fidelity to DI as a positive indicator, not a mark of reduced professional autonomy

Phase 2: National Rollout with Local Flexibility

Following evaluation of Phase 1 outcomes (using value-added measures, not raw attainment scores), a broader national programme will be developed. This will not mandate DI for all subjects or all ages — the evidence base is strongest for foundational literacy and numeracy in Key Stages 1 and 2, and that is where the policy focus will remain.

Schools that can demonstrate equivalent or superior outcomes through alternative explicit instruction approaches will not be required to adopt Engelmann-style scripted programmes. The goal is outcomes, not uniformity.

Phase 3: Initial Teacher Training Reform

The longer-term success of this programme depends on changing what teachers are taught in initial training. The government will work with universities and training providers to ensure that:

• All primary teacher trainees receive substantive instruction in explicit and Direct Instruction methods
• The evidence base for different pedagogical approaches — including Project Follow Through and subsequent meta-analyses — is a required component of teacher training
• Constructivist approaches are taught as one option among several, with their evidence base presented honestly alongside alternatives

5.3 Addressing Teacher Concerns

This policy does not propose the elimination of teacher professional judgement across the curriculum. Teachers will retain full autonomy in subjects and age groups where the evidence does not clearly favour scripted instruction. The proposal is narrowly targeted at foundational literacy and numeracy in Key Stages 1 and 2, where the stakes of failure are highest and the evidence is clearest.

We recognise that asking teachers to adopt scripted programmes requires trust, support, and genuine engagement with their concerns. The implementation programme will include:

• Consultation with teachers and union representatives from the outset
• Clear communication of the evidence base, including the surprising self-esteem findings
• Recognition that skilled delivery of DI programmes is itself a professional competence requiring genuine expertise
• Pathways for experienced DI teachers to become trainers and mentors, creating career progression within the model

6. Addressing Likely Objections

Objection: DI is only for disadvantaged or low-attaining pupils.

Response: Project Follow Through demonstrated DI’s superiority across all demographic groups — the evidence does not support this characterisation. DI produces the largest gains for the most disadvantaged pupils, but it produces gains across the ability range.

Objection: Scripted teaching stifles creativity and harms children’s love of learning.

Response: The data show the opposite. DI students in Project Follow Through scored higher on self-concept and self-esteem measures than students in programmes designed specifically to enhance wellbeing. Children who learn to read and calculate competently are more motivated, not less.

Objection: The Project Follow Through findings are old and have been superseded.

Response: They have been extensively replicated by subsequent meta-analyses spanning five decades of research. No competing study of comparable scale or rigour has produced contrary findings. The evidence base for DI is among the strongest in educational research.

Objection: This amounts to de-professionalising teaching.

Response: Medicine is not ‘de-professionalised’ by requiring doctors to follow evidence-based treatment protocols. Professional judgement is exercised in how a protocol is implemented with individual patients — or pupils. Requiring literacy teaching to follow methods shown to work is consistent with professionalism properly understood.

Objection: The synthetic phonics policy shows that scripted reading instruction has already failed.

Response: England’s synthetic phonics policy is not the same as Engelmann-style DI. UCL researchers have argued that the narrow focus on synthetic phonics alone, without integration of whole-text reading, may not represent optimal practice (Wyse & Bradbury, 2022). This proposal advocates for the full DI model, which integrates phonics within a comprehensive literacy programme — not for a phonics-only approach.

7. Conclusion

The failure of a quarter of English children to achieve functional literacy by age 11 is not inevitable. It is the product of a system that has, for decades, preferred pedagogical orthodoxies over evidence. Direct Instruction is not a new idea, nor an untested one. It is the most thoroughly researched teaching method in the history of educational science, with a consistent and substantial evidence base that spans more than half a century.

A Labour government committed to opportunity and social justice should be willing to act on this evidence. The children who most need effective teaching — those from disadvantaged backgrounds, those at risk of falling behind — are precisely those who benefit most from DI. Failing them in the name of teacher autonomy or philosophical preference is not progressive. It is a choice to prioritise the comfort of the system over the futures of its pupils.

This proposal asks for targeted, supported, evidence-based action on foundational skills. It is realistic, costed, and reversible. The question is not whether the evidence justifies it. It does. The question is whether there is the political will to act on it.

References and Further Reading

Adams, G. L. & Engelmann, S. (1996). Research on Direct Instruction: 25 Years Beyond DISTAR. Seattle: Educational Achievement Systems.

Darch, C., Gersten, R. & Taylor, R. (1987). Evaluation of Williamsburg County Direct Instruction Program: Factors Leading to Student Success in an Urban Setting. School Psychology Review, 16(4).

Department for Education (2024). Key Stage 2 Attainment Statistics. London: DfE. Available at: explore-education-statistics.service.gov.uk

Engelmann, S. (1992). War Against the Schools’ Academic Child Abuse. Portland: Halcyon House.

Engelmann, S. (2007). Teaching Needy Kids in Our Backward System. Eugene: ADI Press.

Full Fact (2015). Counting the Cost of Poor Literacy and Numeracy Skills. Available at: fullfact.org

Hattie, J. (2009). Visible Learning: A Synthesis of Over 800 Meta-Analyses Relating to Achievement. London: Routledge.

Kirschner, P. A., Sweller, J. & Clark, R. E. (2006). Why Minimal Guidance During Instruction Does Not Work: An Analysis of the Failure of Constructivist, Discovery, Problem-Based, Experiential, and Inquiry-Based Teaching. Educational Psychologist, 41(2), 75–86.

Meyer, L., Gersten, R. & Gutkin, J. (1983). Direct Instruction: A Project Follow Through Success Story in an Inner-City School. Elementary School Journal, 84, 241–252.

National Literacy Trust (2024). Children and Young People’s Reading and Writing Survey. London: NLT.

OECD (2023). Survey of Adult Skills (PIAAC). Paris: OECD Publishing. Available at: gpseducation.oecd.org

Stockard, J., Wood, T. W., Coughlin, C. & Rasplica Khoury, C. (2018). The Effectiveness of Direct Instruction Curricula: A Meta-Analysis of a Half Century of Research. Review of Educational Research, 88(4), 479–507.

Stockard, J. et al. (2021). Just How Effective is Direct Instruction? PMC (PubMed Central). DOI: 10.1007/s40614-021-00291-1

Watkins, C. L. (1996). Follow Through: Why Didn’t We? Effective School Practices, 15(1), 57–66.

Wyse, D. & Bradbury, A. (2022). Reading wars or reading reconciliation? A critical examination of robust research evidence, curriculum policy and teachers’ practices for teaching phonics and reading. Review of Education, 10(1). UCL IOE.