Attainment gap between better-off and poor pupils disappears when schools select only the brightest poor

Janet Downs's picture

It’s obvious – if schools choose only the brightest poor pupils, children eligible for free schools meals (FSM) who pass the 11+, then there will be no attainment gap between FSM pupils and the rest.

The attainment gap between better-off and poor pupils would  be eliminated at 11 by selection.  There will be no gap to close.

This blindingly obvious fact is not shared by the Education Secretary Justine Greening who told the 2016 Tory party conference:

'Grammar schools have a track record of closing the attainment gap between children on free school meals and their better off classmates.'

 A Department for Education on 21 November 2016 (now removed) made the same claim and linked to the short survey asking for opinions about government proposals to increase selection.   The questionnaire allows multiple, fictitious responses - someone sitting at a computer can answer the survey as many times as they like using made-up names.

 At the weekend I submitted two empty responses in the names of Scrooge and Marley.   I thought the DfE might be alerted to the problem and start asking for email addresses.  But this isn’t the case.  It’s still possible for respondents to send as many multiple, fictitious submissions as they want to in order to manipulate the results one way or the other.  I submitted another empty survey this morning in the name of Thomas Gradgrind, parent in Manchester. 

Those ultimately responsible for this bogus survey are the Education Secretary Justine Greening and her ministers.  The DfE might be able to get away with a couple of rogue tweets but not a survey which is part of a consultation.

The survey’s a sham – it should be abandoned.  

UPDATE 12.27  It appears Nick Gibb hasn't grasped the fact that if a school eliminates the attainment gap by selection, then there will be no gap to close.  In a Commons debate yesterday he said:

'We know that grammar schools are vehicles of social mobility for the pupils who attend them, almost eliminating the attainment gap between pupils from disadvantaged backgrounds and their peers.'

Gibb preceded this remark by citing the recent ResPublica report on Knowsley:

'Reintroducing grammar schools is potentially a transformative idea for working-class areas'.

ResPublica's opinion was produced as evidence by several Tory MPs during the debate.  But it's an opinion which is wrong as I explain here.

 UPDATE 16 February 2017.   The original article contained the DfE tweet of 21 November 2016.  This has now disappeared from the DfE twitter feed.  I have replaced the tweet with a quote from Justine Greening.  I have referred to the tweet but can no longer link to it.




Share on Twitter Share on Facebook

Be notified by email of each new post.


Roger Titcombe's picture
Wed, 23/11/2016 - 12:25

It’s obvious – if schools choose only the brightest poor pupils, children eligible for free schools meals (FSM) who pass the 11+, then there will be no attainment gap between FSM pupils and the rest.

Absolutely right Janet. There is no 'gap' at other levels of cognitive ability either, as determined by CATs results, as I have been arguing for years.

Alan MIlburn, the Sutton Trust and the like are hooked on the false premise that the school attainment gap between pupils from affluent and poor homes a result of class/ethnic discrimation by schools and teachers, combined with neglectful failures in parenting on the part of those that love their children and do their best.

This results in 'closing the gap strategies' that concentrate attainment raising strategies onto pupils from poorer backgrounds. These approaches are aimed at improving C grade pass rates at English and maths. The most cost effective methods are those of the 'knowledge-based learning movement of behaviourist cramming, backed up with the abusive approaches to discipline needed top force the necessary degree of compliance. The target pupils will also get more lesson time for English maths so narrowing their curriculum and removing other pedagogic opportunities for the development of cognitive ability, which is where the educational needs truly reside. See

From the earliest nursery years teaching should be focussed on the delopment of cognitive and other abilities. The theoretical basis has long been clearly established by Piaget.

The other day I met a highly qualified and experience Nursery Teacher who was bewailing the de-skilling of nursery education. All you need to get such a job or indeed run a private nursery is a Level 2 BTEC in 'childcare', when graduate study in child development is what is needed and what happens in Scandanavia, Germany etc. Even Labour's education policy seems more concerned with 'childcare liberating women for employment' than with proper nursery education run by fully qualified early years teachers, rather than low paid childminders.

As for nursery education, so for all the Key Stages including KS4. The emphasis should be on cognitive and other development. The appropriate pedagogy is well establed with many examples given in Part 5 of 'Learning Matters'.

Dangleze's picture
Sat, 10/12/2016 - 15:00

Yes this is a shocking abuse of statistics. However it would be interesting to compare the percentage of FSM pupils gaining 5 good GCSEs by county (paticularly Lincs as it is a poorer grammar county) to see how well (or not) grammars are really doing for poorer children.

I'm sure the govt are aware of their flawed analysis but it's a case of maintaining popularity with stupid ideas rather than thought through policy.

Janet Downs's picture
Sun, 11/12/2016 - 11:20

Dangleze - it's not really possible to compare the performance of FSM pupils at GCSE in grammar schools with those in other schools because the numbers involved are too small.  For example, in Lincolnshire four grammars had fewer than 5 pupils in the 2015 cohort so their results were suppressed.  Seven had between 6 and 10 FSM pupils in the cohort and three had between 11 and 13.  These make a tiny proportion of the overall cohort for each school.   For example, at Kesteven and Sleaford High just eleven of 108 pupils in the 2015 cohort were FSM.  All eleven (100%) achieved 5 GCSEs A*-C including Maths and English.  They actually outperformed the non FSM pupils - 97% of them achieved the benchmark.  However, it would be misleading to claim Kesteven and Sleaford High were especially good with FSM pupils based on just eleven results and the fact that these eleven were particularly chosen fo their ability in the first place.

The second reason why comparison would be flawed is that the grammars educate only high ability pupils while the non-selective schools educate the full ability range.  So it's not comparing like with like.

Add new comment

Already a member? Click here to log in before you comment. Or register with us.