Let’s not argue over why we can’t make a difference

November 4, 2017

Despite all the talk among progressives of needing new types of education, one of the dividing lines in contemporary debate tends to be about the potential for massive improvements in education. And I do mean “tends” here. This dividing line is a blurry one and there are many exceptions. However, traditionalists often see our education system as failing, and even if we don’t use the language of failure, we do usually argue that massive improvements can be made. We are usually confident that evidence based teaching of early reading can cause a large change in literacy. We do tend to argue that if more knowledge was taught, taught better and taught in a way that would be retained, academic standards could be transformed. We have often explained how better discipline in schools can be the difference between kids getting an education, and many kids being in school, but not actually learning. Meanwhile, progressives are more likely to argue that this line, even when it comes from teachers, is “teacher bashing” and if politicians would just leave the education system alone, provide lots of money, make society fair and equal, and let educationalists and managers get on with their job unhindered by accountability or change, everything would be just fine, or at the very least, the problems will not be those caused by schools.

This has been the background to interpretation of research that shows that, on average, schools don’t make that much difference to outcomes. There are two lines of research that show this. One type shows huge correlations between social class and educational outcomes. Another type of research shows, on average, outcomes can be predicted based on one’s genetic inheritance, either by showing that who you share genes with matters more than who raises you, or by attempting to measure “innate” rather than learned abilities. Both types of research has been used to show schools make little difference on average. Both types of research have been used to justify similar ideas. If differences between students matter more than differences between schools, people argue:

  1. there can be no type of education or knowledge suitable for all;
  2. that the levels of educational failure we see, are, from a school’s point of view inevitable and schools cannot be held responsible for them;
  3. that there is no way to improve our schools other than making them better at catering to student difference.

Roughly speaking, these positions can be described as “determinist”. In my experience, modern progressives are more likely to be social determinists. They are often on the political left and believe that as social class determines outcomes, the important thing for schools to do is to lead the oppressed to political maturity. However, historically, many progressives were genetic determinists. Before the second world war, eugenics was seen as a progressive cause,  accepted by enlightened leftists like the Fabian Society, and opposed by regressive conservatives (like the Catholic writer G.K. Chesterton). There is an excellent chapter in Left Back: A Century of Battles over School Reform about the influence ideas of measuring innate ability had on some progressive educationalists.

As you can imagine, the history here is pretty ugly and there are few advocates of progressive education now who justify their position in terms of genetics, although talk of naturally “academic” and “non-academic” kids is still part of progressive discourse. The nearest you can find now to somebody using genetics to justify progressive education would probably be behaviourial geneticist Robert Plomin, who is quoted in the Guardian as concluding the following from his work on how genes determine outcomes:

“Education is still focused on a one-size-fits-all approach and if genetics tells us anything it’s that children are different in how easily they learn and what they like to learn. Forcing them into this one academic approach is going to make some children confront failure a lot and it doesn’t seem a wise approach. It ought to be more personalised,”

The arguments over which is the right sort of determinism have become vicious. Some now consider the idea that genes influence outcomes, or that we might measure or study innate ability, to be utterly abhorrent and only to be mentioned by racists and eugenicists. Sometimes, even those of us who merely challenge social determinism, without advocating genetic determinism, are tarred with the same brush. The situation is not helped by the fact that the empirical evidence seems to favour the genetic determinists, but there are many problems with those empirical methods (not the least of which is that correlation is not causation) and even more problems with drawing out practical lessons from that evidence. This means that, a lot of the time, people are arguing over what the empirical evidence actually means, whether it is ethical to consider empirical results in the first place and, inevitably, whether the views they don’t like should ever be expressed.

For those educational traditionalists who believe that much can be done to improve schools, the issue is educationally irrelevant. It does not matter what, on average, causes educational outcomes if, on average, outcomes aren’t good enough. If we believe schools can make a huge difference to education, then arguing over what determines outcomes when schools aren’t making a huge difference, is a waste of time. Imagine if people had decided there was no point inventing a polio vaccine until we knew whether the effects of polio were best predicted by social class or genetic make up. To us, it’s like Liliputians and Belifuscudians arguing over which end of a boiled egg to crack. The important thing is not whether we are writing kids off because of their genes or because of their social class, but whether we are writing off kids at all. I’ll weigh into the debate if I see people arguing for censorship or doing their maths incorrectly, but I don’t really care about the core issue except in how it affects how we end up treating people. I’ll happily oppose both eugenics and totalitarian social engineering. I have no interest in choosing between Brave New World and 1984.


  1. This feels a little like a rather Sisyphean dirge on the human condition: we are not going to make a difference, but we just have keep rolling the stone up the hill anyway. But if teachers don’t make a difference, then why bother? Why not pack it in and become a banker?

    I am surprised, in all the recent, heated debate on this subject, that I see so few references to Dan Willingham’s interesting take on the subject. One of the likely reasons that research appears to show a large influence for genetic heritability, Willingham argues, is due to the amplifying effect of environment. Children with relatively small differences in genetic predisposition select very different environments for themselves. This amplifying effect can be massively reduced by introducing an element of compulsion in educational programmes.

    The argument also needs to be linked to one of the strong early findings of the PISA reports that the western tendency to blame poor attainment on innate incapacity was a key reason why we were under-performing Asian jurisdictions, which emphasize the potential of every individual to improve through effort.

    I recommend Brian Simon’s “The State & Educational Change” which traces the deterministic philosophy of education through eugenics, psychometric testing & Piagetian theories of developmental psychology. If determinism has been a mistake, then it has been a massively influential one.

    One final point (part of my one-man campaign against what I think is an almost universal misinterpretation on this subject). Correlation is the only evidence we ever have for causation. It just needs to be triangulated: a single correlation is rarely sufficient.

  2. Academically, some kids don’t get it. So what? Teachers still have a huge roll in developing them as good citizens and helping them achieve their personal goals. So what if those goals have nothing to do with English, Maths and Science. Damn, swore I would stay away from this crap when I walked away from teaching because of the perverse direction education has taken.

  3. The idea that schools make no difference seems odd to me.

    Separate identical twins. Give one group a series of good schools. Give the other no schooling at all. See if schooling makes a difference.

    We couldn’t even run that experiment because everyone knows what the outcome would be and it would be considered massively unethical.

    And while the highly edumacated pontificate about not having any evidence that some schools make a difference to education, almost everyone knows that they do. Which is why very expensive private schools exist and those that can’t afford them fret about which school zone to buy their next house in.

    Sometimes you have to be very severely educated to be stupid.

    • I think we run that experiment all the time, not with twins (usually!). It’s not so much schooling or innate intelligence ; it’s that intelligent people talk to their children, read to them, control what they watch on TV and so on. They help them with homework. They don’t threaten their teachers automatically if their children complain they “weren’t doing nothing wrong”.
      It’s difficult to say if ability is innate or not when the damage may already be done before they come anywhere near a school.

  4. It is strange that we go to such lengths to deny that there is a significant genetic component to intelligence while at the same time recognising that genes are the major determinant in virtually every other variation in living organisms. If nothing else, no one seriously proposes that the Flynn Effect has genetic roots.
    This said, our fixation with IQ is surely a significant factor in our belief that school should be trying to make kids ‘smarter’, and the consequent emphasis on teaching thinking skills. Ironically, nothing could be more damaging to less-able children, nearly all of whom are more than capable of learning masses of declarative and procedural knowledge.
    This is no trivial matter. Anyone who has served in the military knows that no force is better than its NCOs, most of whom are of average intelligence at best. Junior officers are little more than messenger boys between corporals and colonels. If they are wise, company commanders never make a move without consulting their company sergeant major, whose wealth of declarative and procedural knowledge of all things pertaining to the company, its men and its mission will enable him to make wiser decisions than the brightest lieutenant or captain.
    I’d go so far as to argue that the greatest weakness of contemporary society is the extent to which we’ve hollowed out our organisational structures. Nowhere is this more apparent that in teaching, where bright young NQTs waste no time in getting post-graduate degrees in education, which put them on the fast track for an assistant headship, where they can lord it over working teachers whose wealth of experience makes them much more effective in the classroom.

  5. I think the idea is that the specific school you go to doesn’t matter very much, not that you shouldn’t go to School. Andrew feel free to clarify.

  6. Reblogged this on The Echo Chamber.

  7. You cite two lines of research that find correlations between student background and characteristics and academic outcomes.
    But there is another line of research out there that asks whether there are schools that break out of that pattern and if so what are they like? Sir Michael Rutter’s Fifteen Thousand Hours back in 1979 found that similar children who attended different schools in London had very different outcomes that depended largely on the leadership of their schools. Ronald Edmonds at Harvard found the same thing.
    My observations have led me to the same conclusion they drew: what schools do matters. But we can’t simply keep doing the same things and expect different results. We have to organize somewhat differently in order to get different results.
    To hear a discussion of the research, listen to Episode 1 of ExtraOrdinary Districts podcast (and then stay for the other episodes). Get on iTunes or other places you get podcasts — or go to http://www.edtrust.org/extraordinarydistricts.
    Or, if you can throw a tiny bit of money around, buy my book, Schools that Succeed (Harvard Education Press, 2017).

  8. […] to racism, misogyny, homophobia or eugenics. Some of this has been based on the idea discussed here that if you don’t think that educational outcomes are caused by socioeconomic status , then […]

  9. […] Let’s not argue over why we can’t make a difference […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: