Since setting up one of England’s first free schools in 2011, I’ve become interested in what schools can and cannot achieve. Six years ago, I shared the optimism that characterises most graduates entering the education sector for the first time and talked passionately about the transformative impact that good schools can have. But six years later I’m a little more realistic. I now like to quote the opening verse of the Serenity Prayer when talking about this subject:
God grant me the serenity/to accept the things I cannot change;/courage to change the things I can;/and the wisdom to know the difference.
So what are the things that schools cannot change? Having immersed myself in psychology, particularly psychometrics, I’ve reluctantly come to the conclusion that it is naïve to think schools can do much to ameliorate the effects of inequality. I don’t just mean socio-economic inequality; I also mean differences in intelligence. A child’s general cognitive ability is the strongest single predictor of how well they do in their GCSEs, with differences in IQ accounting for more than half of the variance in exam results. See this 2007 study, for instance, which involved tracking 70,000 English schoolchildren over a five-year period. It’s a finding that has been replicated several times.
Can schools do anything to raise children’s general cognitive ability? The answer is maybe, but we haven’t yet discovered how to do it. Intelligence is a highly heritable characteristic, which is to say that more than half the variance in IQ at a population level is due to genetic differences, with less than half due to environmental differences. It’s true that the heritability of IQ is lower among children than it is among adults, with the environment playing a bigger role during adolescence. But the impact of the environment on children’s attainment, even during these formative years, is still fairly negligible – lower than most educationalists believe. Overall, children’s genes account for between 60 and 70 per cent of the variance in GCSE results, with IQ accounting for about half that genetic influence.
Paradoxically, schools do appear to have an effect on the [itals] mean [itals] IQ scores of large populations. As a general rule, the better a country’s public education system, the higher its average IQ. Not only that, but the political scientist James Flynn has demonstrated that the mean IQ of populations in the more affluent parts of the world has increased since 1930, an effect he partly attributes to better schooling. (For more on the Flynn Effect, see here. Interestingly, Flynn now believes IQ across the developed world has started to fall.)
But what schools cannot do, or haven’t been able to do up to now, is raise the IQs of individual students. In particular, they haven’t been able to reduce the differences in IQ among their pupils by raising the general cognitive ability of those who start out below average. A fairly common misunderstanding among educationalists is thinking that if you make schools more equal, you will equalise attainment. In fact, if every school is equally good, you may succeed in reducing some of the differences in GCSE results due to environmental differences, but by doing that you will automatically accentuate the variation due to differences in natural ability, including genetic differences when it comes to conscientiousness and other personality traits linked with attainment. Looked at this way, school improvement may actually [itals] increase [itals] inequality of school outcomes rather than reduce it.
So what can schools do? The good news is that environmental differences still account for between 30 and 40 per cent of the variance in GCSE results, and some of that is linked to the quality of the school. The bad news is that differences between schools, such as the amount of resources a school receives, the number of children in a class, the quality of the teachers, etc., account for no more than 10 per cent of the variance in exam results after you control for variables like students’ IQ and parental socio-economic status.
Now, the fact that ‘school effects’ are quite small shouldn’t be a reason to despair. Good teachers and good schools can still make a difference for key attributes like motivation, attitudes toward learning and self-confidence – see the impact of No Excuses charter schools on raising the attainment of minority students in America’s inner-cities, for instance. And I believe it’s possible – even likely – that we will eventually discover how to boost children’s IQs. By this, I don’t mean that teachers will become better at instilling a ‘Growth Mindset’ – see here for a wide-ranging discussion of the shortcomings of that approach. Rather, I mean that as our understanding of the neuro-biology of intelligence deepens, we may be able to develop pharmacological interventions that boost children’s intelligence. Smart drugs that actually make you smarter – permanently. As I say, I think that could happen, probably within the next 25-50 years. (For more on this, see The Neuroscience of Intelligence by Richard Haier.) Of course, the risk is that affluent parents will be the first to take advantage of this technology, thereby increasing inequality.
In the meantime, we should acknowledge the limitations of what schools can do. As the Serenity Prayer says, it takes courage to change the things you can – and fortitude to keep on going when you know those changes are bound to be quite modest.