What is the point of testing children at 10?
Updated: Jun 11, 2018
In a previous commentary piece I announced my separation from the world of evidence. I asked those who support selecting children by test at the age of 10 to question what values they draw on to support their stance. Well, after three headline hitting pieces of research I have been seduced by evidence again. I am compelled to reconcile my relationship with the evidence for testing children at the age of 10 all over again.
I acknowledge that educational research is a messy and complex affair but will simply report what is well known already, what has been learnt in recent weeks and finally what conclusions can be drawn. Be patient with me as I reacquaint myself with the evidence.
This will not be a pretty piece of writing.
What did we know before (the break-up.)
Family socioeconomic status (income) is linked to academic achievement and future earning potential. Fewer pupils eligible for Free School Meals (FSM) attend selective schools. Where 11+ selection exists those eligible for FSM are less likely to attend grammar schools, even if they have proven academic ability. Children born earlier in the academic year are less likely to attend grammar schools. Past performance indicates future performance. Selection tests largely self-select children that have already shown themselves to be academically able. Schools in countries and counties that have selection have a poor social mix and indicate social and ethnic segregation. Other effects of selection at 10 for non-selective schools include resource and teacher recruitment implications, different civic knowledge and expectations, increased emotional and behavioural problems, compound impact of low SES, low expectations and low achievement, and a negative influence on teacher/pupil relationships. There is little academic advantage in attending any one particular type of school. Selection at 10 does not increase the social mobility of working-class children.
What do we know now (towards reconciliation.)
The 11-plus test does not select on ability alone. The decision to take an 11-plus test is linked to prior attainment and (family) motivation. There is a strong correlation between private tuition, either in core subjects or for the 11-plus, and passing the 11-plus.
The environmental influence of type of school on academic achievement is 1%.
The difference in performance between selective and non-selective schools is mainly due to past performance and ‘individual-level’ factors. There is more variation in academic achievement within school types than between school types. Grammar schools are no more or any less effective than any other type of school. Grammar school demographic is less representative of their geographical area than other areas of England. Those attending grammar schools live in more affluent areas and will be older in the school year. Grammar schools take very few FSM pupils and even fewer pupils who have been on FSM for long periods of time. The length of time a pupil is eligible for FSM has a significant impact on academic achievement. Non-selective secondary modern schools receive less funding from government, or parental fund-raising.
Any positive impact grammar schools may have is easily countered by the negative impact on the larger number of non-selective schools and pupils. Selection at 11 increases social and economic segregation. Separating children into groups of ‘most-able and the rest’ has little impact – even from a very early age.
What has changed (hope for the future).
What has not changed is the knowledge that grammar schools get higher levels of academic achievement compared to non-selective schools. This is well known.
But what has changed and goes a long way to ameliorating my estranged relationship with evidence is the clear link with previous academic achievement and the ‘individual factors’ that each learner brings to their educational experience. The evidence has enabled me to see that pupils and teachers do matter; it is just that the TYPE of school does not matter. All children need access to good schools and as far as academic achievement goes no one type of school does any better, or worse, than any other. So let us desist from the awful spectacle of having 10 year old children sit a test whose only outcome is to threaten social cohesion. If policy makers are happy to include children with additional needs in mainstream school, then why not do so for the more able? Unless it is all down to values!
Gorard, S.A.C. & Siddiqui, N. (2018). Grammar schools in England: a new analysis of social segregation and academic outcomes. British Journal of Sociology of Education http://dro.dur.ac.uk/24067/1/24067.pdf
Jerrim, J, and Sims, S. (2018). Why so few low and middle-income children attend a grammar school? New evidence from the Millennium Cohort Study https://johnjerrim.files.wordpress.com/2018/03/working_paper_nuffield_version_clean.pdf
Smith-Woodley, E. et al. (2018) Differences in exam performance between pupils attending selective and non-selective schools mirror the genetic differences between them. Npj/Science of Learning. https://www.nature.com/articles/s41539-018-0019-8