Article

England’s primary school accountability and assessment system is not working well for disadvantaged children.

By the end of primary school, disadvantaged children are, on average, 10 months behind their peers, and that gap is growing. In 2026, 4 per cent of disadvantaged pupils met the higher standard in reading, writing and maths (combined), while 11 per cent of other pupils met the higher standard. In other words, non-disadvantaged pupils are nearly three times more likely to achieve the highest marks in SATs than their disadvantaged peers.

Change is needed. Building on the evidence outlined in our IPPR paper “Primary is what comes first”: How end of key stage 2 exams impact disadvantaged children and a series of roundtables and interviews with experts, in this paper we outline recommendations to build a more equitable, accessible, functional and reliable system. 

We set out four recommendations for immediate implementation from the 2027/28 academic year.

Recommendation 1: The Department for Education should establish an independent ‘statutory accessibility and bias audit’ for all key stage 2 assessments. 

Recommendation 2: The Department for Education should remove the Spelling, Punctuation and Grammar (SPaG) test.

Recommendation 3: The Department for Education should change how results are published and reported by:

  • ending the publication of school-level key stage 2 results, which would also end the publication of school-level league tables
  • using rolling averages (for instance, three- or five-year averages) of attainment data from the department and from Ofsted for internal monitoring and inspection.

Recommendation 4: The Department for Education should introduce a competency framework for pupil results in place of scaled scores.
However, reform must go further to build a truly equitable system that properly supports disadvantaged pupils. Therefore, we also set out a further recommendation to act as the foundation of a longer-term, more significant reform to primary school assessment.

Recommendation 5: The Department for Education should introduce a random sampling approach to assessment across years 3–6.