Rethinking Education

The Blind Leading the Blind

The sizes of classes in schools has been a hot topic for years. Clearly 40 children in a class together don’t get enough instruction time from the teacher, giving rise to calls to invest more heavily in education so that class sizes can be reduced. I am inclined to think that this does not go far enough – it is a sticking plaster over a much deeper problem with the entire system of education which we rely on.

Class sizes even as low as 15-20 still leave the teachers in either a lecturing role, or spending less than half an hour per day with each student one to one.  This results in the vast majority of interactions at schools being between children and other children. Every year, school takes up around 20% of all the waking hours of children (1140/5840 hours), but add to this after school clubs, extra curricular activities and going out to play with friends, it is easy to see that children can spend a majority of their interaction time interacting with other children, rather than adults. This results in children developing their own independent cultures and norms, further distancing themselves from adults and adult interaction (not to mention exacerbating the perception by adults of children as silly and immature).

Such a significant proportion of children’s interactions being solely with other children, with only a small amount of adult guidance seems very much like children raising children, or the blind leading the blind.  Effectively, this introduces a huge element of randomness into children’s development. Which cultural norms will prevail in any given classroom? Parents and teachers can try to instil children with kindness, a strong work ethic and a love of knowledge, but a few children in a class that are strongly opposed to such things can change the whole dynamic of the class, making fostering such virtues an uphill struggle.  The more children in a class that learn that they won’t be bullied if they don’t try so hard, or that they will be more popular if they pick on others, the harder it becomes for educators to reverse this cultural shift. Children have an enormous capacity to be awful to each other, and these tendencies need to be removed, rather than passively allowed to continue.

Interacting with your Equals

Much is made of the need for children to interact with people their own age, but what are the reasons for this?  Once people become adults, they won’t interact with children in the same way that they did when they were children, and they won’t interact with adults in that way either.  Perhaps high-school politics gives children some experience of politics in the real world, but this feels tenuous. Being good at interacting with adults as equals is a valuable skill that many people only really get significant practice at once they leave school, and become adults themselves.  Integrating with wider society at this point can be a culture shock, and this difficulty may be a reason why there appears to be such generational stratification of culture in modern times. 

One convincing justification is that, as Laszlo Polgar wrote in his book “Raise a Genius” (archive), the issue with many adult-child interactions is the status difference – children need to interact with people that they do not feel subordinate to, in order to develop properly.  An absence of this could result in them never learning to interact as equals, being overly deferential as an adult. This does not however necessitate a majority of children’s time being spent being raised by other children.  Laszlo Polgar writes about the importance of parents and teachers of children respecting and listening to their views, and allowing them to achieve mastery in certain areas so that they feel that they have something valuable to contribute.  He then stresses the need for children to interact with people with a similar level of skill to their own, again so that they feel they have value to contribute, and can most effectively learn from their own and other’s mistakes.

This idea of interacting with people with a similar level of skill could be contorted and simplified into the idea of children interacting with people their own age.  This seems like an easy way out though, allowing for false economies to be made within the education system, and avoiding the need for parents and teachers to give respect to the views of children.  Any education system that is able to increase a child’s amount of non-subordinating adult interaction time at the expense of some unstructured child interaction time seems likely to yield great benefits.  Of course, some unstructured play is generally understood to be highly beneficial for children’s development, but this is usually raised as a criticism of ‘Tiger Parenting’, in which every second of a child’s day is scheduled and structured, allowing children no free time and no freedom of expression.  Despite the nature of school being largely children raising children, it is not free or unstructured time, therefore reducing the element of children raising children within education should not inherently raise concerns about the amount of unstructured play.

Bloom’s 2-sigma problem

It is no bad thing that modern education is universal, affording a certain base level of understanding, literacy and numeracy to everyone in society, however the sheer universality of it is problematic.  There are people with the means to give their children a much more thorough, in depth and personalised education than even a private school would provide, and yet it is so entrenched in our cultural norms, that for many it is unthinkable.

It was the Romans who invented the idea of classroom teaching, and Roman classrooms were renowned for being soul-crushingly boring, involving lots of learning by rote.  Before this became the norm, wealthy Romans would hire instructors to teach their children one to one, and it is notable that after public schooling became common in the Roman empire, many of the greatest thinkers (Marcus Aurelius for instance) were still educated outside of this system using the old method of hiring private tutors.

In modern times, this phenomenon has been quantified – Bloom’s 2-sigma problem is the seemingly inescapable reality that education does not benefit from economies of scale.  The so named “problem” states that one to one education, using a technique of ensuring thorough understanding of a topic before moving on to the next, appears to fairly reliably outperform classroom education to an almost unbelievable extent. On average, such students tend to perform in the top 2% of an equivalent classroom (2 standard deviations above the classroom mean).

This effect means that it is a great shame that people are generally not willing or able to step outside of the traditional education system.  Further to this, as Taleb, the author of Antifragile would no doubt assert – the nationally standardised curriculum also leaves the entire education system highly fragile to mistakes that can affect entire generations of people.  One example of a fairly common (though not necessarily universal) failure is that due to pressures within the system, schools often ‘‘teach to the test”, rather than ensuring good understanding of the material. Although private schools sometimes do things slightly differently, only very wealthy people can usually afford to send their children to them (the average cost of private school in the UK is £17,000 per student per year), and they still fall foul of Bloom’s 2-sigma problem.

A Rough Costing of Homeschooling

Parents with a good understanding of the material should be capable of teaching up to high school level, and whilst this may be a significant amount of work (the job of teacher is by no means easy), it is likely to be more straight-forward than trying to teach a class of 30 children.  Any gaps in the parents’ understanding could be resolved by hiring a tutor for that area, but used sparingly this solution could still be relatively inexpensive. A couple with two children could potentially both work part time, and educate their children themselves, allowing them to be on the right side of Bloom’s 2-sigma problem, whilst probably being significantly less expensive than the £34,000 per annum cost of private school.  As a more specific example, if each parent worked 3 days per week for 60% of their full-time salary, and a state school were willing to have their children in school for only 1 day per week (ensuring that both children get at least some ‘socialisation’ with their peer group), this would allow the parents to home-school their children 4 days per week, whilst still having weekends as a family.

Given Bloom’s 2-sigma problem, it is possible that this could yield better results than sending them to the average private school, and it would be cheaper as long as their post-tax full-time combined household income was less than £85,000 per annum (85k x 40% = 34k).  In fact, assuming both parents were earning a similar salary and were taxed in the UK, the tipping point where taking this strategy with 2 children became more expensive than the average private school would actually be a pre-tax full-time combined household income of £140,000.  Each parent earning £70,000 would take home about £49,000 after tax and NI, whereas at 60% pay, each would be on £42,000 which would be £32,000 after tax – an after tax difference of £17,000 each. I think it is safe to say that many very well educated parents have gross household incomes much less than £140,000 per annum, so this option would be less expensive for them than private school.

Is there a risk that this could go wrong?  Yes. Is there also a risk that a normal school could leave a child traumatised or underperforming due to bullying?  Yes. Does homeschooling have a bad reputation due to extreme religious groups and overbearing narcissists? Yes. Could your child fall in with a bad crowd at school and go off the rails?  Yes.

However, just because something is the norm, and very easy to simply go along with, doesn’t mean that any attempt to deviate from that norm is destined to end in disaster.  Learning should be fun, and teaching children that learning is fun is a good way to encourage life-long learning.