Skip to content

Blog post

Exploring the free school meals measure in the English National Pupil Database

Tammy Campbell, Director for Early Years at Education Policy Institute Kerris Cooper, Senior Researcher at Education Policy Institute

Pupils’ recorded eligibility for free school meals (FSM) within the English National Pupil Database (NPD) has long been utilised across the country’s research and policy communities. It is employed in analyses, policymaking, resource allocation and accountability frameworks. Many uses of the FSM measure assume or require a level of stability: that the same ‘types’ of pupils are consistently picked up, over time and place.

In a project funded by the Nuffield Foundation, we begin to scrutinise this assumption: to explore FSM’s nuances, (in)accuracies, biases, heterogeneities, uses and impacts. Our first output is an evidence review and synthesis (Campbell & Cooper, 2024). We examine the history of FSM, eligibility criteria and linked entitlements – processes through which children are recorded as FSM-eligible in the NPD – and how this has changed over time.

We critically consider uses of the FSM (and derived Pupil Premium) measures in practice and policymaking, and intended and unintended consequences. We detail previous research on FSM’s reliability and validity. We discuss possibilities for good practice in utilisations of FSM, alongside alternative measures of disadvantage.


Pragmatically, there are several strengths to the FSM measure.

  • It is easily and consistently available within the NPD and requires no additional burden on schools for collection.
  • It requires no additional disclosure of personal information from families to schools.
  • It has been widely used and therefore has currency in conveying messages about how some children who are more disadvantaged are faring under different educational and wider social policy regimes.
  • The existing body of research into the measure and ways in which it can be used can be utilised in understanding, interpreting and improving research and policymaking involving FSM.


Notwithstanding these arguments about pragmatism in using FSM, our review highlights evidence showing that – though families are entitled to apply for FSM based on individual-level benefits receipt, and low-income – registration of children for FSM varies across place and depends on numerous contextual factors. They include:

  • global economic and societal conditions that impact families’ work and income (such as the 2008 financial crisis, and the Covid-19 pandemic)
  • welfare benefits regimes and policies (because receipt of benefits determines entitlement for FSM)
  • local and national incentives to sign up to be registered as FSM eligible
  • disincentives to signing up for FSM (at the social level – for example, stigma – and the practical level – if the applications process is not straightforward)
  • the methods through which schools, local authorities and governments promote and enable families to register their children as FSM eligible
  • area demographics and group cultural/social norms.

This means that the composition of the group denoted as FSM in the NPD varies and depends to some extent on time and place. The types of disadvantages experienced by children recorded as FSM eligible have differed over the years.

‘The composition of the group denoted as FSM in the National Pupil Database varies and depends to some extent on time and place.’

We also find indications that procedures for identifying children as ‘FSM eligible’ have increasingly dissociated from practices and decisions in terms of providing actual free school meals. Furthermore, children registered for FSM receive different support and interventions depending on the area in which they live. Again, this adds to variation within the FSM-denoted group.

Other limitations of the FSM measure include its blunt binary nature; that the proportion of pupils recorded FSM eligible is increasingly misaligned with poverty rates (a significant proportion of children in poverty are not recorded as FSM eligible); and an imperfect overlap with other indicators of disadvantage related to social and educational outcomes (including mothers’ education, social class, gradated income-level, welfare benefits receipt, parental employment and housing stability/mobility). This is problematic when FSM is used in research and policymaking to proxy some of these variables, because it does not pick up all intended/assumed children.

Next steps

In the remainder of this project, we will continue to explore the FSM measure and its uses, through qualitative research with local authorities and schools, quantitative analyses of linked administrative/survey datasets, and deliberative discussion and synthesis.

Our intention is to build capacity in accurate examination of the experiences of different groups of children within the English education system: particularly those from disadvantaged and/or marginalised groups. We are also motivated by a desire to make the best possible use of existing administrative information, including that on FSM, within the NPD.

We welcome conversations and debate on these issues, in the spirit of knowledge-building and collaboration within the research and policy communities: please do get in touch if this interests you: @_TammyCampbell, @CooperKerris.


Campbell, T., & Cooper, K. (2024). What’s cooking? A review of evidence and discussion on the free school meals (FSM) measure in the National Pupil Database. Education Policy Institute.