Who are free schools really serving?

Thousands of children return to school this week as school holidays draw to a close. However, for some pupils, this new academic year signals more change as they will join one of the 55 new free schools opening this week.

In the past year, we have observed a heated debate over whether free schools affect certain social groups more than others.

With seemingly contradictory claims published in today's Financial Times and Guardian, Full Fact decided to have a closer look to try and provide some much needed clarity.

For its part, the Guardian pointed to its own research which indicates "that the majority of free schools are being established in wealthier areas." Following a link to the relevant article (dated 31 August 2011), we learn that:

"Research shows that the 10-minute commuting area around the first wave of free schools is dominated by middle-class households, appearing to undermine coalition claims that they are empowering working class families. The areas have 57% of better-off, educated and professional households compared with the English average of 42.8%. There are also a higher-than-average proportion of Asian homeowners in the free school catchment areas — 5.3% — compared with 1% in England as a whole. Just 29.1% are categorised as "hard-pressed" or of "moderate means", compared with 36.9% for the country."

What today's Guardian fails to mention, however, is that within a week of the paper publishing its findings, the Department for Education released its own research on this issue.

According to the Department, of the 24 schools opened last year as part of the first wave of free schools:

• "Over a third (9 schools) are located in the 20% most deprived communities."
• "Half the schools (12 schools) are located in the 30% most deprived communities"

Clearly these figures paint a rather different picture to that of the Guardian. So what's behind the divergence?

Crucially, the DfE's methodology is rather different to the Guardian's. Rather than focussing on catchment areas by the time taken to commute as the paper did, the DfE examined the relative deprivation (measured by the Indices of Multiple Deprivation) of the immediate surroundings of each free school (known as the Lower Super Output Area - or LSOA).

Will the new free schools change this?

With 55 new free schools opening this week, the above figures are, of course somewhat out-of-date.

However, a headline in today's FT would suggest that contemporary data on this issue is indeed available, as the paper's headline suggested that the:

"Bulk of free schools open in poor areas."

Delving into the article itself, we learn that an analysis undertaken by the paper reveals that 30 of the 79 free schools now in operation "are in the poorest fifth of neighbourhoods, measured by the number of benefit claims."

The article itself gives no further methodological details, but the author was able to give us some further information.

Like the DfE, the FT used LSOAs as their geographical measure. Next, they looked up the Income Deprivation Affecting Children Index value for each LSOA, thereby enabling them to identify how many free schools were in the poorest fifth of all LSOAs.

How many deprived children attend free schools?

It should be stressed that none of these analyses tell us how many children from deprived families are attending free schools. In fact, today's Guardian claims that figures for existing free schools show that:

"only 9.4% of their pupils are on free school meals — a key indicator of poverty — compared with a national average of 16.7%."

These figures come from a November 2011 post from blogger SchoolDuggery, which pulled together data published by the first wave free schools concerned and compared it to the national averages released by the DfE.

Again, adding the new wave of free schools opening this term to might change the outlook, as could a new intake of pupils in those schools celebrating their first anniversaries.

The social composition of the 79 free schools now open is therefore an open question depending upon what you measure. It's also one where the results are rapidly changing, as new schools are opened and more pupils start to study there.