Will poorer families benefit from the return of Grammar Schools?

On Monday the outgoing chief Ofsted inspector Sir Michael Wilshaw said, “The notion that the poor stand to benefit from the return of grammar schools strikes me as quite palpable tosh and nonsense” (gov.uk, Sept 2015).  In a House of Lords debate, Lord Nash responded by saying, “there is no clear evidence to support his views” (Hansard, Sept 2015).  So who is correct, Wilshaw or Nash?

Every grammar school can provide anecdotal evidence of alumni who came from humble origins but went on to achieve great things thanks to the schooling they received but the real questions which need answering are, whether such opportunities are available equally to children from all socioeconomic backgrounds and whether selective admissions policies prejudice those on lower incomes.

Unless it’s also being argued that there is a link between wealth and intelligence then there should be no significant difference in the socioeconomic profiles of both selective and non-selective schools.  The evidence not only shows Wilshaw was completely justified in dismissing claims that grammar schools promote social mobility but also that selective admission policies, per se, appear to be incompatible with the legal requirements to ensure that admissions our publicly funded schools do not disadvantage unfairly, either directly or indirectly, a child from a particular social or racial group.


Most studies use free school meals (FSM) as an indication of income however this only provides a single measurement point.  Since 2008 the National Pupil Database has included a specific socio-economic indicator called the Income Deprivation Affecting Children Index (IDACI) listing the probability, ranging from 0 to 100%, that a child is living in poverty.

To compare the profiles of selective and non-selective admissions policies, over half a million individual pupil records were taken from the 2014/15 census and the proportion of these falling in each 5% range of IDACI scores was calculated then plotted.


Both distributions are skewed to the right because the numbers of children decreases as poverty levels reach 100% but the grammar school values, shown in pink, are far more skewed.  Selective schools admit twice as many children in the leftmost, most affluent, IDACI band when compared to non-selective schools.  Wilshaw previously said grammar schools are, “stuffed full of middle class kids” (OECD 2013) and if we define middle class as those living in areas of 10% or less poverty then the graph demonstrates that this happens at the expense of those living in areas between 20 – 70% poverty where the blue, non-selective, bars are all far higher than the grammars.

Selective schools don’t purposely set out to discriminate so we need to understand why their admission processes are so biased against children from poorer backgrounds.  There may be a number of contributory factors such as the cost of commuting long distances or schools pressuring parents to make donations but the one factor they all have in common is children are selected through an entrance test.  It’s very hard to see how a couple of hours’ of multiple choice questions can ever be capable of providing a reliable and accurate measure of an individual’s aptitude and the obvious explanation for children from wealthy families being more successful in grammar school tests is they have been tutored to pass them.

For many decades the tobacco industry argued that a correlation between smoking and lung cancer didn’t prove causality.  There is dearth of objective research into the effect of tutoring however the circumstantial evidence, in the form of a thriving cottage industry in tutoring, is overwhelming.  Parents are clearly in no doubt whatsoever that tutoring increases the chance of their child getting a place.  CEM who provide about half the 11+ tests have withdrawn promotional material suggesting their tests are resistant to tutoring and recently stated, “Without extensive and expensive research, it is not possible to quantify the impact of coaching on the results from our tests.”  The test providers are best placed to carry out any research into the effects of tutoring but like the tobacco industry before them, have no incentive to undertake research which would threaten the millions of pounds they currently earn each year from administering the tests.

‘Parental Choice’ will always be the preserve of who can afford it, whether through buying a house in the right area, paying to send a child long distances to attend a school or tutoring to pass a test but the evidence here is conclusive and quantifiable; selective schools are systematically disadvantaging children who live in areas of 20% poverty and above.


I am very grateful to National Pupil Database team for their help and guidance in applying to use and providing their data.

About the data

The data were taken from the KS4 Pupil 2015 data linked to KS2 Autumn Census 2014/15.

The Selective schools dataset consisted of 22,474 individual pupil records which matched KS4_ADMPOL=2 (selective schools).

The non-selective dataset consisted of 519,521 individual pupil records which matched either KS4_ADMPOL=1 (comprehensive) or KS4_ADMPOL=3 (Secondary Modern).


Wilshaw M, 5 Sept 2016, Speech at the London Councils education summit. Published on gov.uk website https://www.gov.uk/government/speeches/sir-michael-wilshaws-speech-at-the-london-councils-education-summit

Nash J, 8 Sept 2016, Debate in House of Lords.  Published on www.parliament.uk website http://hansard.parliament.uk/Lords/2016-09-07/debates/16090727000539/GrammarSchools

Crawford C and Greaves, E, 2013, A comparison of commonly used socio-economic indicators: their relationship to educational disadvantage and relevance to Teach First.  The Institute for Fiscal Studies, London.  http://www.ifs.org.uk/comms/r79.pdf


This entry was posted in Uncategorised. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *