Grammar schools: selection by
206. Selection by general ability is designed to
identify, for the purposes of school admissions, pupils of high
Some schools will recruit some or all of their pupils in this
way. The strategy has the effect of narrowing the range of abilities
within a school, or at least to raise the median level of ability.
It is therefore the case that in areas where selective and non-selective
schools coexist, those schools which do not select necessarily
receive either no or a reduced proportion of pupils at the top
end of the ability range. The larger the proportion of the age
group in an area that is selected, the greater this effect will
be. As a matter of arithmetic, where 25% of an age group are defined
as being of high ability and all go to one or more selective schools,
75% of the age group must necessarily go to schools which lack
any pupils of high academic ability.
207. This is a particularly significant factor in
areas where pupil numbers are declining. In these areas, if selective
schools continue to admit to their capacity, albeit from a wider
ability range, non-selective schools will carry the full burden
of the decline in pupil numbers and suffer a further reduction
in the proportion of able or even average ability pupils they
are able to recruit. Policy-makers and protagonists must therefore
be aware that the unavoidable consequence of selection in areas
of declining school population is that many non-selective schools,
already with significantly skewed intakes, will have fewer pupils
able to achieve at the highest level.
208. There are 164 grammar schools in England and
no provision for more to be created. This has not prevented the
expansion of those schools. The Secretary of State for Education
and Skills, Charles Clarke MP recently told the House of Commons
that the number of pupils in grammar schools has increased from
117,147 (3.1% of the age group) in 198384 to 150,750 (4.6%)
in 2003-04. This
means that 33,603 more pupils are in grammar schools today than
was the case in 1983. It is also the case that 22,029 more pupils
are in grammar schools than in 1997 when the proportion of the
school population in grammar schools was 4.3%.
These increases are relatively small in absolute terms. However,
as grammar schools are not distributed evenly, but clustered in
particular areas, the local effects of growth in the proportion
of able pupils selected out of mainstream secondary education
can be considerable, particularly when coupled with the impact
of falling rolls.
209. In Opposition the Labour Party's then education
spokesman promised "read my lips: no [more] selection, either
by examination or interview, under a Labour government."
 The fact remains
that the numbers and proportion of pupils selected into grammar
schools have increased in recent years. Ministers have claimed
that there has been no "acceleration"
in this rate of increase but it is undeniable that selection has
increased and, in areas where the school population is falling,
may well increase further.
210. Although we recognise that the data on the
impact of selection is open to alternative interpretations, we
received evidence from three eminent academics with different
research interests, all of whom felt that selection by academic
ability had an adverse impact on educational standards and post-16
participation rates. The written submission from the DfES
quoted the most recent research from the NFER which demonstrated,
as Professor Jesson's work had done previously, that the most
academically able 25% of the ability range performed equally well,
if not slightly better, in non-selective schools. We are aware
of the recent Ofsted/Audit Commission report on Buckinghamshire
draws attention to the large disparities in the funding of selective
and non-selective schools in the county.
211. In addition, we were concerned by the conclusions
of the recent Ofsted report on Kent which concludes that:
"When national comparisons are made, the proportion
of high achieving schools (A* against national benchmarks) is
substantially greater [in Kent] than nationally, probably reflecting
the number of grammar schools in the county.
The proportion of low achieving schools (E and
E* against national benchmarks)is substantially higher than
nationally, again probably reflecting the number of secondary
modern schools in the county."
212. The report also notes that the evidence from
inspection suggests that there are fewer schools in Kent judged
to be "very good" than nationally and that Kent schools
are substantially more likely to require special measures or to
have serious weaknesses than those nationally. The report also
observes that all of the Kent secondary schools that require special
measures are secondary modern schools.
213. We are aware of no research evidence that indicates
that schools which select wholly by academic ability help to raise
standards or post-16 participation rates or that they have a positive
effect on the coherence of the local education system or a benevolent
effect on social inclusion. We invited as witnesses two supporters
of grammar schools, neither of whom were able to furnish any statistical
information to support their case.
214. All forms of selection at one set of schools
have, as a matter of arithmetic, consequences for other schools.
A government that permits the continuing expansion of selection,
by ability or by aptitude, can only be understood to approve of
both the practice of selection and its outcomes. If that is the
position of the present Government it should be publicly stated.
215. We believe that it is time for Ministers
to engage in an informed debate about the role of selection in
secondary education and its impact across the education system
as a whole. The Government needs to explain how it reconciles
its insistence that there will be no return to selection with
its willingness to retain and increase selection where it already
exists. Without an honest and robust engagement with this issue
the Government's policy on selection will continue to appear ad
hoc and without principle.