Using Free School Meals data in education research: potential and pitfalls

Free School Meals (FSM) data is commonly used in education research here in the UK, and is collected annually at the time of the school census for both primary and secondary schools. There are a number of relevant FSM data columns in the publically available census data – both numbers and percentages of those who take up the offer of free school meals, as well as the numbers and percentages of those learners who are eligible. In addition there is a third metric – those who are eligible (for performance tables) which is the most robust as this is an average of three years’ worth of data.

To be eligible for FSM the family must be in receipt of one of a number of state benefits, including jobseekers, disability, housing etc. As such, the FSM eligibility data points have been independently verified and as such are a very robust and useful to use as a proxy to determine the numbers of learners in low income households at a particular school.

Interestingly there is a differential between those learners who are eligible for FSMs and those who take advantage of the offer. This discrepancy varies considerably between Local Authority (LA) areas, and many LAs run programmes and drives to ensure those who are eligible take up the offer, as this also has an impact on the pupil premium funding that the school receives from central government.

Using London as an example, based on the latest school census (collected January 2017) we can see that there are differences in the numbers of those who are eligible for FSM, those who are actually taking up free school meals, and the metric of FSM used for the performance tables.

Graph 1: Comparison of the three FSM metrics by Local Authority (London)

London FSM data all cols percent 2017


When the latest school census was released, it was noted by various commentators that the numbers of those who were in receipt of school meals were fewer than in the previous years. In fact, across all school types there were 14% of learners eligible for FSM provision, the lowest proportion since 2001 when records began.

While this is true – it does not reflect fewer learners from low income households – rather this reflects the continuation of austerity economics whereby many benefits have been cut, and where the earning threshold for state assistance has also been lowered. As FSM eligibility depends on the receipt of income related benefits – when there are fewer benefits claimants the proportion of learners who are eligible for FSMs also decreases. The following graph shows the average FSM data for those learners who are eligible for and who claimed FSMs at maintained nursery and state-funded primary schools, state funded secondary schools, special schools and pupil referral units.

Graph 2: Percentage of learners’ eligible for, and claiming FSM

Percent learners eligible for and taking FSM


Graph 3: Percentage of pupils and claiming FSM, secondary and primary level

FSM primary secondary since 2001

So while FSM data is still a robust proxy for low income households, it is increasingly clear that there are many households who continue to be economically impoverished but no longer are counted by this data point. To understand the recent history of FSM figures, we need to examine the previous coalition government’s policy which began to cap benefits that a household could claim in a year. The first benefits cap was introduced in 2013 and was set at £26,000 (the average income for a family in the UK at the time). In November 2016 there was a further benefits cap introduced by the conservative government which further reduced the income rate at which a household could claim state benefits. This was lowered to £23,000 in London, and £20,000 for the rest of the country. These caps have affected thousands of households with children, the majority of which are in London.

It is clear then that FSM data should no longer be used alone as a proxy for low income households, but instead should be used as one of a bundle of measures to establish the numbers of learners from low income households.

This does present a number of problems for the schools’ data analyst. The other data sets available to establish household income are not based on school level data and these data are not collected annually. The Indices of Deprivation – in particular the Index of Multiple Deprivation (IMD) and the income deprivation affecting children index (IDACI) are geographically based series of data which although robust, are difficult to analyse alongside school based data as the areas of geography do not necessarily correspond to where learners live in relation to the school postcode. Alongside this the latest IMD data is from 2015, so there will be a mismatch of years. While there will be more technical skill involved, and time taken to marry these data sets, it is clear that relying solely on the FSM data as an indicator of poverty at school level is problematic. Certainly any FSM data since 2013, and particularly after 2016 needs to be understood within the wider context of the change in state benefits and therefore no longer adequately reflects the numbers of young people from low income households.


Graph 3 taken from Department for Education, Statistics: school and characteristics of schools and pupils. Main Text: SFR28/2017, page 1, available from:



This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s