Contributors: Christos Tsoukalis and Emily Jolliffe
Our Future of Britain initiative sets out a policy agenda for a new era of invention and innovation, based on radical-yet-practical ideas and genuine reforms that embrace the tech revolution. The solutions developed by our experts will transform public services and deliver a greener, healthier, more prosperous UK.
Britain’s school system has made significant progress over the past 25 years, something reflected in results, in comparative studies like the OECD’s Programme for International Student Assessment (PISA) tables and in Ofsted reports.
However, as in all nations, it should now embrace a new era of reform, driven by the necessities and opportunities of the technological revolution.
Children learn at different speeds in different subjects, with a wide variation in aptitude that can account for more than 40 per cent of the difference in GCSE results.[_] Attempts have been made to address this, but with consequences. The old division between grammar schools and secondary modern schools resulted in a highly arbitrary separation of students at age 11 that entrenched social inequity. Streaming then tried to account for individual differences within the comprehensive-school system, with poor outcomes for disadvantaged students.[_]
The reality remains that there is no guarantee of a quality education. This must – and can – change.
Today, the tools exist to provide every child with a personalised education, with learning tailored to individual needs and differentiated by subject. Bringing the best and most innovative educational materials into the classroom would free up teachers’ time so they can focus on giving each student the right mix of challenge and support to help them fulfil their potential. Education technology can be embedded in classroom and home learning, and artificial intelligence (AI) used to revolutionise the experience of pupils and teachers.
Significantly, the power to accumulate and analyse vast amounts of data can change completely the context in which the system assesses performance, promotes success and corrects failure.
None of this obviates the need for “the basics” to be in place: essential literacy and numeracy skills, supplemented by strong digital competencies, and a conducive environment for learning, including good behaviour, respect and high-quality leadership.
But all of it demands radical change.
To realise this vision of a personalised education for every child, we propose the following:
1. Introduce a digital learner ID for every pupil that would:
contain all educational information, including formal test results, attendance records, week-by-week assessments, marked homework, records of non-academic achievement and more.
become a hub of digital learning, connecting learners with apps to supplement traditional teaching – educational games for younger age groups, times-table practise for primary-school children or challenging maths problems for older students, for example.
give pupils and parents control of their data and provide them with useful insights from the information, such as suggestions for further study or employment opportunities, or assistance in the selection of schools or nurseries.
2. Establish an independent body to secure storage and appropriate use of this information.
With this in place, anonymised and aggregated data could be used by educators to analyse trends and real-time shifts in nurseries and schools. Over the long term, this would then provide the evidence and reliable data infrastructure needed to improve the assessment system.
3. Radically upgrade the Ofsted system so that it regains the confidence of schools and parents.
This system has been essential for school accountability over the past quarter of a century, but it needs modernising. Rather than occasional lightning visits, it can utilise technology and data from schools to assess performance, monitor progress and mandate intervention to improve its judgement of school effectiveness.
4. Increase parent choice and access to quality education to all by:
giving schools in England the freedom and funding to provide services, including hybrid lessons in subjects where they have particular expertise, to other schools anywhere in the country, and to boost the range of their local offerings through cooperation with other settings. This should include private providers who may have developed new ways of learning.
giving parents and pupils the right to request online classes delivered by other schools elsewhere in the country, increasing the incentives for schools to improve performance.
5. Reduce teachers’ workloads to free up their time to teach.
Technology and AI can provide new ways of organising the classroom and working days, and supporting marking, lesson planning and coordination. The introduction of team teaching, including through changes to the career-progression framework, would help teachers share workloads and improve professional-development opportunities.
Introducing a digital learner ID is at the heart of this change. This learner ID would securely collate information about a child as they move through the education system, starting with early-years education. It would provide the data needed to maximise the impact of new tools such as personalised artificial-intelligence (AI) learning and create a real-time understanding of educational progress – both in terms of the child, and more broadly in terms of their school, their area and even the country. It would give learners and their parents control over their data, while underpinning a host of necessary reforms to assessment and accountability.
Improving the education system begins with understanding how it is functioning. Less guesswork and more real-time insight are needed so that information about what is working well and what needs to change can be acted upon. Fully unlocking the power of data in education is the way to achieve this.
The Department for Education (DfE) already collects data from schools that include learner characteristics, information on safeguarding and special educational needs and disabilities (SEND), attendance records, exam results and exclusions. It uses this information to monitor student attainment, hold schools to account against narrow performance metrics and determine school-funding allocations.
However, the current data-collection process adds to schools’ administrative load and wastes resources. There is no interoperability, meaning local governments, for example, can’t use these data to provide targeted support. The information that is collected lacks the scope, depth and real-time qualities needed to support individual learning or to drive accountability in a more meaningful way.
None of it is readily available to the people it concerns the most: learners and their parents. Parents can ask to view their child’s record,[_] which takes up to 15 school days. If they want their own copy, schools can print it out for a fee of up to £50 for a record of 500 pages. Some settings, like academies and free schools, are not obliged to respond. This is a threat to parents’ and children’s digital rights.
Tech apps (like Google Classroom) also collect large amounts of data which parents and learners can rarely access or control. Moreover, the transformative potential of these data is limited by vendor lock-in; much of it remains siloed. This makes it hard for educators to experiment with different tools and use insights derived from them to support their work.
An alternative approach is possible. Rich, real-time data could be captured efficiently, shared securely and made available to parents and learners. Both the information teachers document every day, for example through marking and behavioural reports, and the data that apps record but do not currently share could be much better used. These data streams could be supplemented with new sources: wellbeing surveys for students, teachers and parents, for example. Collectively, this information could be used to generate deep insights, improve pupils’ and schools’ performance, and reduce teachers’ workloads. It would become a remarkable tool for policymakers to ensure education constantly innovates.
To deliver this, a secure, functional and interoperable digital learner-ID system is needed. This should include a unique identifier and an ecosystem of tools to link it with information about learning, wherever it is held. Both elements are needed to unlock quality learning for everyone, everywhere, putting parents and learners in control and empowering educators to do what they do best – and to keep getting better.
1. Introduce a Digital Learner ID for Every Pupil
Using a single portal, children and their parents would have access to a learner profile displaying key insights derived from data linked to the learner. This profile would contain all educational information, including formal test results, attendance records, week-by-week assessments, marked homework, records of non-academic achievement and more. It would paint a full picture of the student’s abilities, strengths and weaknesses, including insights from every stage of their educational journey, and would help educators develop a deeper understanding of each pupil and target learning accordingly.
This system would also serve as a hub of digital learning, connecting children with apps that supplement traditional teaching, for example by suggesting educational games suitable for early-years age groups, allowing primary-school children to practise times tables or challenging secondary-school students to tackle maths problems. The learner ID should work as a single sign-on for these tools and they should be able to talk to each other, allowing learners and educators to switch between platforms and choose the ones that work best for them.
Combining data from these tools with the information generated in schools within the learner ID would help build a more rounded picture of each student’s development and enable the provision of more personalised learning programmes. Students have different capabilities and learning preferences. Educators could use the most effective learning methods to provide challenging-but-rewarding experiences for every child, as well as support when needed.
In nurseries, practitioners could capture live updates about each child’s development and instantly share them with parents, giving parents more opportunities to engage with early-years learning. In schools, adaptive learning would become the norm in every subject, using data collated from different apps to create assignments that challenge, stimulate and engage the learner without leaving them behind. Each learner would have access to a personal learning map, with AI-supported tuition, and could revisit any content when needed.
Expanded use of data would also facilitate the introduction of robust, sophisticated methods of ongoing and end-point assessment. The current emphasis on examinations as the lynchpin of assessment and accountability is harming England’s education system by restricting the curriculum, limiting innovation and hampering life-skills development.[_] According to a recent poll by The Times and YouGov, 62 per cent of parents believe the emphasis on tests has gone too far.[_]
Instead of this, a multimodal approach for assessment over time that would more accurately reflect the development of students’ knowledge and abilities could be adopted. This could combine comparative assessment (especially in difficult-to-assess areas such as oracy), peer grading, self-reflection, feedback on project work, learning data from adaptive tools and, where appropriate, end-point exams. It would also enable a better understanding of the extent to which schools are helping students develop life skills such as communication, collaboration and critical thinking.[_]
The ultimate owners of this information would be the learner and their parents; they would have full visibility of data linked to the learner ID, including information on how and when it was collected, by who and for what purpose. If information is shared through the learner ID, parents and pupils should be able to see who is accessing it and why.[_]
While some information may need to be shared to meet statutory requirements (for example, for safeguarding purposes), learner-ID owners should be able to decline sharing in all other cases. They should also have access to an easy way to challenge and correct inaccurate information. Parents and learners should be supported to make best use of the information collated in the learner ID through the allocation of dedicated funding to schools and family hubs to provide free digital-skills training and skilled coaches.
Through the learner ID, students could receive personalised recommendations in areas such as work experience or career opportunities, courses to apply to, local extracurricular activities or educational support. They could apply directly to further study or jobs, choosing what information to share, or obtain guidance on the skills that are in highest demand locally and how they could acquire them.
Proposals for a new system of learning using a digital learner ID:
Issue a unique learner ID to every child and link all education data to it.
Create a national data infrastructure that includes a single sign-on, a parent-and-child-facing profile and privacy-preserving aggregate databases for accountability purposes.
Replace obsolete analogue exams with more sophisticated tech-driven assessments for the AI era, including progress checks for life skills such as collaboration, communication, critical thinking and creativity.
Provide training for educators, leaders and regulators on working with learning data, and digital-skills education for parents and learners so they can make the most of the insights available to them through the learner ID.
2. Establish an Independent Body to Manage the Digital Learner ID
A new body should be established to secure the storage and appropriate use of this information. To avoid political interference, the system should be managed by an independent designated data body rather than the DfE.
This body should work with the private sector and privacy experts to develop interoperability standards for the real-time exchange of education data and require private vendors who want to sell to the public sector to adopt them. This move, which would increase the amount of data feeding into learner IDs and allow students to move seamlessly across platforms, would mirror the success of the UK’s Open Banking standards.[_]
Anonymised and aggregated data could then be used by educators to analyse trends and real-time shifts in nurseries and schools. Over the long term this would provide the evidence to improve assessments and a robust data infrastructure that would safeguard their validity and reliability.
Fully anonymised data, shared with explicit parental permission, could be made available to innovators in the education-technology (edtech) space, making Britain the go-to country for those who want to positively transform education.[_]
Proposals for establishing an independent body to manage the digital learner ID:
Select an independent designated body to set up and operate a privacy-preserving national data infrastructure, regulating the collection of data and access to them by different parties.
Establish and enforce interoperability standards, based on the same principles as the Open Banking initiative, and require private vendors to adopt them.
Create and regulate access to unique public-good data sets based on anonymised, opt-in education data.
3. Use the Data to Radically Upgrade the Ofsted School-Inspection System
Too often, parents find themselves in education “cold spots” in which there is no good school or where their first choice is oversubscribed. One in five parents don’t get their first choice of school. TBI analysis shows that in six local authorities, fewer than half the schools are rated “good” by Ofsted. In eight local authorities, at least 20 per cent of schools are rated “inadequate”. Just three of the top 15 local authorities for primary-school performance are outside London; for secondary schools, none of the top 15 are outside the capital.
One way to break the link between geography and educational quality is to use technology to raise standards across the board. A digital learner ID could help drive continuous improvement in nurseries and schools.
Every educational leader would have a wealth of information at their fingertips, including feedback from students and parents, as well as data on academic performance, skills development, behaviour, safeguarding and governance. They could use the data to benchmark performance against similar settings, gaining a view of how they are doing and informing improvement plans.
In aggregate, these data would allow local authorities to provide joined-up support to young people in need, commission provision such as nursery places when there is a shortage, improve take-up of services and prevent people falling out of education.[_]
Most importantly, it would enable a positive overhaul of Ofsted’s defunct approach to school inspections and accountability. This must be radically reformed and upgraded to make much better use of the accurate real-time data that the digital learner ID would provide.
A new approach must rest on a much deeper analysis and more thorough investigation of a school’s performance, based on a complete picture that develops over time instead of relying on snapshot inspections and broad-brush judgements. The guesswork of a flying visit can be replaced with a live view of emerging challenges as well as lasting strengths.
This would allow robust recommendations on improvements to be made and progress to be monitored using accurate, benchmarked data. Leaders and regulators could have a shared understanding of how a school is performing compared to similar settings, jointly decide on specific steps for improvement and schools could be transparently assessed on their implementation.
This would include inviting other leaders who had successfully tackled similar challenges in similar circumstances to co-design plans for improvement, replicating the impact of shared data use and peer-to-peer support that led to the success of the London Challenge programme.[_]
The use of rich, real-time data could also help address concerns that Ofsted’s overall “good” or even “outstanding” judgements can mask specific gaps in performance. Data from digital learner IDs would provide visibility of schools slipping behind on specific measures and allow pre-emptive and prompt targeted interventions.
To reflect this new approach and remit, and reset the relationship with the sector, Ofsted should be renamed the Office for Accountability, Improvement and Development in Education (OfAIDE). It should become genuinely independent of the DfE and secure the funding, technical capability and skilled staff to accurately assess rich digital information about nurseries, schools and trusts.
Part of OfAIDE’s approach should be to engage in ongoing dialogue with nursery, school and trust leaders and facilitate targeted peer-to-peer support from experienced heads, similar to the NHS’s successful Getting It Right First Time programme.[_] It should also retain the power to intervene directly in cases of sustained failure by re-brokering academies or converting maintained schools.
Proposals to strengthen accountability:
Radically reform and upgrade Ofsted’s approach to draw on data-driven insights and ongoing dialogue with leaders to monitor the quality of educational provision in real-time, instead of through rare flying visits.
To support its broader remit and mission, rename Ofsted the Office for Accountability, Improvement and Development in Education (OfAIDE), make it independent from the DfE and hire a digitally literate workforce with the skills to make good use of anonymised data aggregated from digital learner IDs.
Empower OfAIDE to intervene pre-emptively and promptly where necessary, working constructively with schools, trusts and local authorities that are slipping behind on key indicators, including by brokering counsel from heads and experts who have successfully overcome similar challenges in comparable circumstances.
4. Use the Data to Give Parents Greater Choice and Improve Access to Education
A digital learner ID would give parents the information they need to make decisions that are in the best interests of their children.
Instead of relying on arbitrary league tables to choose a school or nursery, they could create personalised selections based on factors or values they care about (distance from home, schools’ performance in particular subjects, students’ wellbeing levels, sports facilities, SEND support and more). This would improve the current approach, with more personalised and easy-to-understand information.
Digital tools, including those built by private-sector companies, could recommend schools or nurseries based on the student’s information as well as real-time data on availability. Generative AI would make it easy to ask questions and receive clear responses. This provision of timely and relevant insights would take the guesswork out of choosing for parents.
A digital-first education system could also allow pupils to take online classes delivered by state or independent schools elsewhere in the country, further breaking the link between geography and outcomes. Good schools should be incentivised to provide live online lessons and, using the digital learner ID, pupils’ progress would be captured in one place regardless of setting so they could seamlessly combine studies.
Students could learn online full-time, with the option of joining lessons from a dedicated safe building with adult supervision, good broadband, quiet spaces, meals and sports facilities, or part-time (from their “main school”). This kind of system would allow parents to augment existing educational provision or change where learning happens without having to move house.
Proposals to use insights to increase parent choice:
Develop personalised league tables and a recommendation engine for nursery and school applications, allowing parents to rank their choices based on factors they care most about.
Allow parents and pupils to request online classes delivered by schools elsewhere in the country, breaking the link between geography and available options.
Incentivise high-quality state and independent schools to offer online lessons to other schools, for example where they have a particular strength in a subject, and collaborate with them on the development of the most efficient learning methods.
5. Use the Data to Reduce Teachers’ Workloads, Freeing Up Their Time to Teach
Collecting information more easily, improving interoperability and expanding the use of edtech enabled by the digital learner ID would help to address one of the root causes of England’s teacher-retention crisis: unsustainable workloads.
Seventy-two per cent of teachers say their workload is unacceptable and 66 per cent say they spend more than half of their time on work that takes place outside the classroom.[_] Almost half (48 per cent) say they spend too much time planning lessons, 46 per cent too much time marking and 53 per cent too much time recording and inputting data.
Introducing a digital learner ID would significantly reduce this workload. Capturing data in real time in a consistent format across different tools would greatly reduce the need for time-consuming manual “data drops”. A “once-only” principle should be put in place to ensure the same information does not need to be submitted on separate occasions. Where data collection would represent an additional burden, extra resources should be provided to support settings in meeting this.
Expanded use of data would also help automate analysis of students’ strengths and weaknesses, simplify coordination between teachers and allow them to use AI tools to plan lessons that are engaging and challenging for all pupils.
Most importantly, it would allow teachers to shift repetitive marking tasks to adaptive-learning systems. A move to a real-time data environment would mean AI could mark a class’s work in less than a second and provide personalised feedback. This wouldn’t replace teachers but instead free them up to do what they do best: teach.
Educators could use their pupils’ learner-ID profiles to tailor their teaching, giving timely feedback, encouraging independent learning when appropriate and stepping in to support as needed. When children move schools, their profiles would be used by their new teachers to get to know them quickly and ease the transition.
Using the digital learner ID to identify students’ needs on a day-to-day basis would also make it possible to remove class-size limits in favour of team teaching using mixed-learning methods. With several qualified educators in the room, lessons could include a mix of whole-class activities and small-group or one-to-one tuition.
Teachers’ administrative load would be reduced through joint lesson-planning and automated marking, freeing them up to provide individual feedback to pupils and work through problem areas in small groups. The digital learner ID would allow teachers to collaborate more effectively as it would contain a record of each educational intervention and allow colleagues to build on each other’s work. This approach would also open up new professional-development opportunities, with junior colleagues able to learn by observing experienced teachers and vice versa.
Proposals to reduce workloads and free up teachers’ time to teach:
Reduce administrative work by providing tools that allow educators to collaborate on planning, create timetables, automate marking and collect important information about children’s progress in the flow of work rather than as a separate data-collection exercise.
Transform classrooms by ditching class-size limits in favour of mixed-learning methods – large groups, one-to-one learning, remote learning – according to student needs that day. Reform the career-progression framework for teachers, making team-teaching the norm to fast-track future leaders and retain the best teachers.
UK schools currently offer an industrial-scale “one-size-fits-all” approach. But today, we have the power to personalise education, allowing students to learn in the way that’s most effective for them. By giving every student a digital learner ID, which contains all their educational information, we can teach differently, embrace innovative new tools and materials, and use AI and data to revolutionise the experience of both students and teachers. The ability to accumulate and analyse vast amounts of data can also be used to radically overhaul the Ofsted approach to school accountability so that it is based on real-time insights and geared towards the continuous improvement of standards across the education system.