This is one of several blogposts in Datalab’s ‘Who’s left’ series of posts. The full series can be found here.

Three school censuses are carried out each year – in October, January and May – recording which pupils are on-roll at which schools.

When it comes to inclusion in a mainstream secondary school’s league table results, everything hinges on the census carried out in the January of a child’s Year 11 year.

Broadly speaking, if a child is recorded as being on roll at a particular school, they count in that school’s results, and if they’re not, they don’t.

There are exceptions. If a pupil arrived in the country in the last two years and comes from a country where English is not the official language they can be omitted from results.

Likewise if a pupil has been permanently excluded from one school in the last two years, another school that accepts them does not have to count that child’s results in their overall results – intended to mean schools aren’t disincentivised from accepting pupils who have been excluded.

If a child is permanently excluded but does not join the roll of another education institution, the excluding school can also ask for that pupil’s results to be attributed to them.

Schools are also allowed to request amendments, in the small number of cases where a pupil has left England permanently, or died.

There are also something called penalty add-backs, which affects how the results of pupils reaching age 16 before Year 11 are accounted for. These pupils are always reported in a school’s results the year after which they have been in Year 10, meaning there is no possibility of them repeatedly being counted as not having reached the end of secondary school.

But, fundamentally it’s a case of: if a child was recorded as having been on-roll at a mainstream secondary school on 15 January 2015, they would have counted in that school’s results in summer 2015.

Our reweighting approach

Under our reweighting approach, following a methodology developed for a report we published at the end of 2015 [PDF], school league tables are reweighted to take account of every pupil who had spent time at a school, in proportion to the amount of time they spent there.

For most secondary schools, which admit in Year 7, a pupil can have spent a maximum of 15 terms there – three terms in each of five years – before reaching the end of secondary school.

So we allocate pupils’ results to the institutions which they have spent time on-roll at as part of their secondary education in proportion to the amount of time they spent on-roll there – 100% if they spent all fifteen terms there; 40% if they spent only six terms there, say.

Crucially, then, where a pupil is recorded as being on-roll in January of their Year 11 year does not have the great importance that it does under league tables as they stand now.

That means that the minority of schools that might be inclined to try and game the league tables by losing pupils before this date wouldn’t have such a strong incentive to do so – as the pupil’s results would still contribute to how they themselves were judged.

In our work we looked at four cohorts of pupils: those finishing secondary school in 2011/12, 2012/13, 2013/14 and 2014/15

Our population of around 2,900 schools was produced by applying the following criteria:

  • all mainstream, state-funded secondaries that admit children in Year 7 were taken as a starting point;
  • schools where a large number of pupils – two classes-worth – joined after Year 7 were excluded;
  • the school had to have been open for the entire five-year period ending with the year under consideration (taking into account any predecessor school that exists);
  • the school had at least 11 pupils recorded as completing KS4 and counting towards league table results for that school for the relevant academic year.

Where a school changed status – becoming an academy part way through the period being looked at – records were linked together and the school treated as one entity.

Under our approach, we are adding back in pupils who may legitimately have left the English education system – leaving the country permanently, or in some cases dying – and are counting these pupils as if they did not achieve five good GCSEs or equivalents. But we think these are very much a minority of cases, with the majority of children who disappear from schools rolls and don’t show as having achieved any qualifications having entered informal private or home-schooling arrangements and not taken any qualifications there. It’s also worth remembering that under our reweighting approach, these pupils only count to the extent that they spent time on a school’s roll – so if they only spent a year there they would count relatively little.

And if the Department for Education were to adopt our suggestion of reweighting league tables, it wouldn’t be too difficult to filter out these small number of cases, using established processes.

Now read the rest of Datalab’s ‘Who’s left’ series of posts.