But what I have realised is that selection does not by itself improve a school. Rather, grammar schools help bright children precisely because if you put bright children together, they individually do better. And quite obviously, that comes at the cost of a reduction in the number of the brightest, most motivated children in the ordinary comprehensives. As I said earlier, I believe I did well at least partly because most of my peers were equally bright and motivated, and from the same sort of middle-class-background, and so I had to work to compete. It is that competition that I would have missed most had I failed my 11+.
This highlights an uncomfortable truth: affluence, motivation and intelligence (and the three usually come together) are not just individual strengths; their mere presence drag up results for others. I may have done less well at a comprehensive, but someone else – another bright child who failed the test – might have done a little better. Statistically, this shows. In Kent, which has one of the most extensive grammar systems left, 55 per cent of the poorest pupils get GCSE results in the bottom 20 per cent nationally. On that measure, it is close to being the worst performing areas in the country.
We have to admit it: there is a trade-off here. The extreme benefits of grammar schools for those few who attend come partly at the expense of those who don't. Selection at 11 leaves too many bright children behind, while adding to the segregation of our schools by class. Our education system needs reform and Michael Gove is doing an excellent job of trying to change attitudes in much of the state system. But he should be very wary of embracing selection. Tests don't improve schools by themselves; they just redistribute the children.
